Fabric OS Administrator’s Guide V7.2.0 1507967232powerconnect B Dcx4s Administrator Guide6 En Us

User Manual: DELL POWERCONNECT B-DCX-4S BACKBONE pdf | FreeUserManuals.com

Open the PDF directly: View PDF PDF.
Page Count: 694

DownloadFabric OS Administrator’s Guide V7.2.0  1507967232powerconnect-b-dcx4s Administrator Guide6 En-us
Open PDF In BrowserView PDF
53-1002920-02
9 September 2013

Fabric OS
Administrator’s Guide
Supporting Fabric OS 7.2.0

®

Copyright © 2013 Brocade Communications Systems, Inc. All Rights Reserved.
ADX, AnyIO, Brocade, Brocade Assurance, the B-wing symbol, DCX, Fabric OS, ICX, MLX, MyBrocade, OpenScript, VCS, VDX, and
Vyatta are registered trademarks, and HyperEdge, The Effortless Network, and The On-Demand Data Center are trademarks of
Brocade Communications Systems, Inc., in the United States and/or in other countries. Other brands, products, or service names
mentioned may be trademarks of their respective owners.
Notice: This document is for informational purposes only and does not set forth any warranty, expressed or implied, concerning
any equipment, equipment feature, or service offered or to be offered by Brocade. Brocade reserves the right to make changes to
this document at any time, without notice, and assumes no responsibility for its use. This informational document describes
features that may not be currently available. Contact a Brocade sales office for information on feature and product availability.
Export of technical data contained in this document may require an export license from the United States government.
The authors and Brocade Communications Systems, Inc. shall have no liability or responsibility to any person or entity with
respect to any loss, cost, liability, or damages arising from the information contained in this book or the computer programs that
accompany it.
The product described by this document may contain “open source” software covered by the GNU General Public License or other
open source license agreements. To find out which open source software is included in Brocade products, view the licensing
terms applicable to the open source software, and obtain a copy of the programming source code, please visit
http://www.brocade.com/support/oscd.

Brocade Communications Systems, Incorporated
Corporate and Latin American Headquarters
Brocade Communications Systems, Inc.
130 Holger Way
San Jose, CA 95134
Tel: 1-408-333-8000
Fax: 1-408-333-8101
E-mail: info@brocade.com

Asia-Pacific Headquarters
Brocade Communications Systems China HK, Ltd.
No. 1 Guanghua Road
Chao Yang District
Units 2718 and 2818
Beijing 100020, China
Tel: +8610 6588 8888
Fax: +8610 6588 9999
E-mail: china-info@brocade.com

European Headquarters
Brocade Communications Switzerland Sàrl
Centre Swissair
Tour B - 4ème étage
29, Route de l'Aéroport
Case Postale 105
CH-1215 Genève 15
Switzerland
Tel: +41 22 799 5640
Fax: +41 22 799 5641
E-mail: emea-info@brocade.com

Asia-Pacific Headquarters
Brocade Communications Systems Co., Ltd. (Shenzhen WFOE)
Citic Plaza
No. 233 Tian He Road North
Unit 1308 – 13th Floor
Guangzhou, China
Tel: +8620 3891 2000
Fax: +8620 3891 2111
E-mail: china-info@brocade.com

Document History
Title

Publication number Summary of changes

Date

Fabric OS Administrator’s Guide 53-1002920-01

Added Fabric OS v7.2.0 software features
and support for embedded switches:
Brocade 5431, M6505, and 6547.

July 2013

Fabric OS Administrator’s Guide 53-1002920-02

Corrections and additions for the Fabric OS
7.2.0a release.

September 2013

Contents (High Level)

Section I

Standard Features

Chapter 1

Understanding Fibre Channel Services . . . . . . . . . . . . . . . . . . . . . . . . . 45

Chapter 2

Performing Basic Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Chapter 3

Performing Advanced Configuration Tasks . . . . . . . . . . . . . . . . . . . . . . 83

Chapter 4

Routing Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115

Chapter 5

Buffer-to-Buffer Credits and Credit Recovery . . . . . . . . . . . . . . . . . . . .135

Chapter 6

Managing User Accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151

Chapter 7

Configuring Protocols. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .195

Chapter 8

Configuring Security Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .231

Chapter 9

Maintaining the Switch Configuration File . . . . . . . . . . . . . . . . . . . . . .277

Chapter 10

Installing and Maintaining Firmware . . . . . . . . . . . . . . . . . . . . . . . . . .289

Chapter 11

Managing Virtual Fabrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .309

Chapter 12

Administering Advanced Zoning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .337

Chapter 13

Traffic Isolation Zoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .379

Chapter 14

Optimizing Fabric Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .413

Chapter 15

Bottleneck Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .427

Chapter 16

In-flight Encryption and Compression . . . . . . . . . . . . . . . . . . . . . . . . .445

Chapter 17

Diagnostic Port. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .459

Chapter 18

NPIV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .473

Chapter 19

Fabric-Assigned PWWN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .479

Chapter 20

Managing Administrative Domains . . . . . . . . . . . . . . . . . . . . . . . . . . .485

Section II

Licensed Features

Chapter 21

Administering Licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .515

Chapter 22

Inter-chassis Links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .543

Chapter 23

Monitoring Fabric Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .551

Chapter 24

Managing Trunking Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . .569

Fabric OS Administrator’s Guide
53-1002920-02

3

4

Chapter 25

Managing Long-Distance Fabrics . . . . . . . . . . . . . . . . . . . . . . . . . . . .587

Chapter 26

Using FC-FC Routing to Connect Fabrics . . . . . . . . . . . . . . . . . . . . . . .593

Appendix A

Port Indexing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .641

Appendix B

FIPS Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .645

Appendix C

Hexadecimal Conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .657

Fabric OS Administrator’s Guide
53-1002920-02

Contents

About This Document
Supported hardware and software . . . . . . . . . . . . . . . . . . . . . . . . . . 35
What’s new in this document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Document conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Notice to the reader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Additional information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Getting technical help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Document feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Section I
Chapter 1

Standard Features
Understanding Fibre Channel Services
Fibre Channel services overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Management server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Platform services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Platform services and Virtual Fabrics. . . . . . . . . . . . . . . . . . . . . 47
Enabling platform services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Disabling platform services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Management server database. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Displaying the management server ACL. . . . . . . . . . . . . . . . . . . 48
Adding a member to the ACL. . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Deleting a member from the ACL . . . . . . . . . . . . . . . . . . . . . . . . 49
Viewing the contents of the management server database . . . 50
Clearing the management server database . . . . . . . . . . . . . . . 51
Topology discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Displaying topology discovery status . . . . . . . . . . . . . . . . . . . . . 51
Enabling topology discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Disabling topology discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Device login . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Principal switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
E_Port login process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Fabric login process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Port login process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
RSCNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Duplicate Port World Wide Name . . . . . . . . . . . . . . . . . . . . . . . . 55
High availability of daemon processes . . . . . . . . . . . . . . . . . . . . . . . 55

Fabric OS Administrator’s Guide
53-1002920-02

5

Chapter 2

Performing Basic Configuration Tasks
Fabric OS overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Fabric OS command line interface. . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Console sessions using the serial port. . . . . . . . . . . . . . . . . . . . 58
Telnet or SSH sessions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Getting help on a command . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Viewing a history of command line entries . . . . . . . . . . . . . . . . 61
Password modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Default account passwords. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
The switch Ethernet interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Virtual Fabrics and the Ethernet interface . . . . . . . . . . . . . . . . . 65
Management Ethernet port bonding . . . . . . . . . . . . . . . . . . . . . 65
Displaying the network interface settings . . . . . . . . . . . . . . . . . 66
Static Ethernet addresses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
DHCP activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
IPv6 autoconfiguration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Date and time settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Setting the date and time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Time zone settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Network time protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Domain IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Displaying the domain IDs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Setting the domain ID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Switch names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Customizing the switch name . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Chassis names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Customizing chassis names . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Fabric name. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Configuring the fabric name . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
High availability considerations for fabric names . . . . . . . . . . . 78
Upgrade and downgrade considerations for fabric names. . . . 78
Switch activation and deactivation . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Disabling a switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Enabling a switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Disabling a chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Enabling a chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Switch and Backbone shutdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Powering off a Brocade switch . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Powering off a Brocade Backbone . . . . . . . . . . . . . . . . . . . . . . . 81
Basic connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Device connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Switch connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

6

Fabric OS Administrator’s Guide
53-1002920-02

Chapter 3

Performing Advanced Configuration Tasks
Port identifiers (PIDs) and PID binding overview . . . . . . . . . . . . . . . 83
Core PID addressing mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Fixed addressing mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
10-bit addressing mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
256-area addressing mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
WWN-based PID assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Ports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Port Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Backbone port blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Setting port names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Port identification by slot and port number . . . . . . . . . . . . . . . . 89
Port identification by port area ID. . . . . . . . . . . . . . . . . . . . . . . . 90
Port identification by index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Configuring a device-switch connection . . . . . . . . . . . . . . . . . . . 90
Swapping port area IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Port activation and deactivation . . . . . . . . . . . . . . . . . . . . . . . . . 92
Port decommissioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Setting port modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Setting port speeds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Setting all ports on a switch to the same speed . . . . . . . . . . . . 94
Setting port speed for a port octet . . . . . . . . . . . . . . . . . . . . . . . 95
Blade terminology and compatibility . . . . . . . . . . . . . . . . . . . . . . . . . 95
CP blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Core blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Port and application blade compatibility . . . . . . . . . . . . . . . . . . 98
FX8-24 compatibility notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Enabling and disabling blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Enabling blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Disabling blades. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Blade swapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
How blades are swapped . . . . . . . . . . . . . . . . . . . . . . . . . . . . .100
Swapping blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102
Disabling switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .102
Power management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103
Powering off a port blade . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103
Powering on a port blade . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103
Equipment status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .104
Checking switch operation . . . . . . . . . . . . . . . . . . . . . . . . . . . .104
Verifying High Availability features (Backbones only) . . . . . . .104
Verifying fabric connectivity. . . . . . . . . . . . . . . . . . . . . . . . . . . .105
Verifying device connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . .105
Viewing the switch status policy threshold values. . . . . . . . . .105
Setting the switch status policy threshold values . . . . . . . . . .106
Audit log configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Verifying host syslog prior to configuring the audit log . . . . . .109
Configuring an audit log for specific event classes . . . . . . . . .109

Fabric OS Administrator’s Guide
53-1002920-02

7

Duplicate PWWN handling during device login . . . . . . . . . . . . . . . .110
Setting 0, First login precedence . . . . . . . . . . . . . . . . . . . . . . .110
Setting 1, Second login precedence. . . . . . . . . . . . . . . . . . . . .110
Setting 2, Mixed precedence . . . . . . . . . . . . . . . . . . . . . . . . . .110
Setting the behavior for handling duplicate PWWNs. . . . . . . .111
Enabling forward error correction . . . . . . . . . . . . . . . . . . . . . . . . . .111
FEC Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112
Using the portCfgFec command . . . . . . . . . . . . . . . . . . . . . . . .112

Chapter 4

Routing Traffic
Routing overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115
Paths and route selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . .116
FSPF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .116
Fibre Channel NAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Inter-switch links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .118
Buffer credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .119
Congestion versus over-subscription . . . . . . . . . . . . . . . . . . . .119
Virtual channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .119
Gateway links. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .120
Configuring a link through a gateway . . . . . . . . . . . . . . . . . . . .121
Routing policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .122
Displaying the current routing policy . . . . . . . . . . . . . . . . . . . .122
Port-based routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123
Exchange-based routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123
Device-based routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123
Dynamic Path Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .124
AP route policies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .124
Route selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125
Dynamic Load Sharing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .125
Frame order delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .126
Forcing in-order frame delivery across topology changes . . . .127
Restoring out-of-order frame delivery across topology
changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .127
Using Frame Viewer to understand why frames are dropped .127
Lossless Dynamic Load Sharing on ports . . . . . . . . . . . . . . . . . . . .129
Lossless core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .130
Configuring Lossless Dynamic Load Sharing . . . . . . . . . . . . . .131
Lossless Dynamic Load Sharing in Virtual Fabrics . . . . . . . . .131
Frame Redirection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .132
Creating a frame redirect zone . . . . . . . . . . . . . . . . . . . . . . . . .132
Deleting a frame redirect zone . . . . . . . . . . . . . . . . . . . . . . . . .133
Viewing frame redirect zones . . . . . . . . . . . . . . . . . . . . . . . . . .133

8

Fabric OS Administrator’s Guide
53-1002920-02

Chapter 5

Buffer-to-Buffer Credits and Credit Recovery
Buffer credit management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .135
Buffer-to-buffer flow control . . . . . . . . . . . . . . . . . . . . . . . . . . .135
Optimal buffer credit allocation . . . . . . . . . . . . . . . . . . . . . . . .136
Fibre Channel gigabit values reference definition. . . . . . . . . .137
Buffer credit allocation based on full-size frames. . . . . . . . . .137
Allocating buffer credits based on average-size frames . . . . .140
Configuring buffers for a single port directly . . . . . . . . . . . . . . 141
Configuring buffers using frame size . . . . . . . . . . . . . . . . . . . . 141
Calculating the number of buffers required given the
distance, speed, and frame size. . . . . . . . . . . . . . . . . . . . . . . .142
Allocating buffer credits for F_Ports . . . . . . . . . . . . . . . . . . . . .142
Monitoring buffers in a port group . . . . . . . . . . . . . . . . . . . . . .142
Buffer credits switch or blade model . . . . . . . . . . . . . . . . . . . .143
Maximum configurable distances for Extended Fabrics . . . . .144
Downgrade considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . .145
Configuring credits for a single VC . . . . . . . . . . . . . . . . . . . . . .146
Buffer credit recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .146
Buffer credit recovery over an E_Port. . . . . . . . . . . . . . . . . . . . 147
Buffer credit recovery over an F_Port. . . . . . . . . . . . . . . . . . . . 147
Buffer credit recovery over an EX_Port. . . . . . . . . . . . . . . . . . .148
Enabling and disabling buffer credit recovery . . . . . . . . . . . . .148
Credit loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .149
Back-end credit loss detection and recovery support on
Brocade 5300 switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .149
Back-end credit loss detection and recovery support on
Brocade 6520 switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .149
Enabling back-end credit loss detection and recovery . . . . . .150

Chapter 6

Managing User Accounts
User accounts overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151
Role-Based Access Control . . . . . . . . . . . . . . . . . . . . . . . . . . . .152
Management channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .154
Managing user-defined roles . . . . . . . . . . . . . . . . . . . . . . . . . .154
Local database user accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155
Default accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .156
Local account passwords . . . . . . . . . . . . . . . . . . . . . . . . . . . . .157
Local user account database distribution. . . . . . . . . . . . . . . . . . . .158
Distributing the local user database . . . . . . . . . . . . . . . . . . . .158
Accepting distributed user databases on the local switch . . .158
Rejecting distributed user databases on the local switch . . .159
Password policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .159
Password strength policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .159
Password history policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .160
Password expiration policy . . . . . . . . . . . . . . . . . . . . . . . . . . . .161
Account lockout policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .161

Fabric OS Administrator’s Guide
53-1002920-02

9

The boot PROM password . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .163
Setting the boot PROM password for a switch with a
recovery string . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .163
Setting the boot PROM password for a Backbone with a
recovery string . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .164
Setting the boot PROM password for a switch without a
recovery string . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .165
Setting the boot PROM password for a Backbone without a
recovery string . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .166
Remote authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .167
Remote authentication configuration. . . . . . . . . . . . . . . . . . . .167
Setting the switch authentication mode . . . . . . . . . . . . . . . . . 171
Fabric OS user accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
Fabric OS users on the RADIUS server . . . . . . . . . . . . . . . . . . .172
Setting up a RADIUS server. . . . . . . . . . . . . . . . . . . . . . . . . . . .175
LDAP configuration and Microsoft Active Directory . . . . . . . . .181
LDAP configuration and OpenLDAP . . . . . . . . . . . . . . . . . . . . .184
TACACS+ service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .189
Remote authentication configuration on the switch . . . . . . . .192
Configuring local authentication as backup. . . . . . . . . . . . . . .194

Chapter 7

Configuring Protocols
Security protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .195
Secure Copy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .196
Setting up SCP for configuration uploads and downloads . . .197
Secure Shell protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .197
SSH public key authentication . . . . . . . . . . . . . . . . . . . . . . . . .198
Secure Sockets Layer protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . .200
Browser and Java support . . . . . . . . . . . . . . . . . . . . . . . . . . . . .200
SSL configuration overview . . . . . . . . . . . . . . . . . . . . . . . . . . . .201
The browser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .204
Root certificates for the Java plugin . . . . . . . . . . . . . . . . . . . . .205
Simple Network Management Protocol . . . . . . . . . . . . . . . . . . . . . .206
SNMP Manager . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .206
SNMP Agent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .206
Management Information Base (MIB) . . . . . . . . . . . . . . . . . . .207
Basic SNMP operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .207
Understanding MIBs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .208
Access to MIB variables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .208
SNMP support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .209
Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .209
Loading Brocade MIBs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .212
Access Gateway and Brocade MIBs . . . . . . . . . . . . . . . . . . . . .216
Firmware upgrades and enabled traps . . . . . . . . . . . . . . . . . .216
Support for Administrative Domains . . . . . . . . . . . . . . . . . . . .216
Support for Role-Based Access Control . . . . . . . . . . . . . . . . . .216
Support for IPv6 addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Support for Virtual Fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Configuring SNMP using CLI . . . . . . . . . . . . . . . . . . . . . . . . . . .218

10

Fabric OS Administrator’s Guide
53-1002920-02

Telnet protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .226
Blocking Telnet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .227
Unblocking Telnet. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .228
Listener applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .228
Ports and applications used by switches . . . . . . . . . . . . . . . . . . . .229
Port configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .229

Chapter 8

Configuring Security Policies
ACL policies overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .231
How the ACL policies are stored . . . . . . . . . . . . . . . . . . . . . . . .231
Policy members . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .232
ACL policy management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .232
Displaying ACL policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .233
Saving changes without activating the policies . . . . . . . . . . . .233
Activating ACL policy changes . . . . . . . . . . . . . . . . . . . . . . . . . .233
Deleting an ACL policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .233
Adding a member to an existing ACL policy . . . . . . . . . . . . . . .234
Removing a member from an ACL policy . . . . . . . . . . . . . . . . .234
Abandoning unsaved ACL policy changes . . . . . . . . . . . . . . . .234
FCS policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .235
FCS policy restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .235
Ensuring fabric domains share policies . . . . . . . . . . . . . . . . . .236
Creating an FCS policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .236
Modifying the order of FCS switches . . . . . . . . . . . . . . . . . . . .237
FCS policy distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .238
Device Connection Control policies . . . . . . . . . . . . . . . . . . . . . . . . .238
DCC policy restrictions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .239
Creating a DCC policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .239
Deleting a DCC policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .240
DCC policy behavior with Fabric-Assigned PWWNs . . . . . . . . . 241
SCC Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .242
Creating an SCC policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .243
Authentication policy for fabric elements . . . . . . . . . . . . . . . . . . . .243
E_Port authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .244
Device authentication policy . . . . . . . . . . . . . . . . . . . . . . . . . . .246
AUTH policy restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Authentication protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .248
Secret key pairs for DH-CHAP . . . . . . . . . . . . . . . . . . . . . . . . . .249
FCAP configuration overview. . . . . . . . . . . . . . . . . . . . . . . . . . .251
Fabric-wide distribution of the authorization policy. . . . . . . . .253

Fabric OS Administrator’s Guide
53-1002920-02

11

IP Filter policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .253
Creating an IP Filter policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . .254
Cloning an IP Filter policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . .254
Displaying an IP Filter policy . . . . . . . . . . . . . . . . . . . . . . . . . . .254
Saving an IP Filter policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .255
Activating an IP Filter policy. . . . . . . . . . . . . . . . . . . . . . . . . . . .255
Deleting an IP Filter policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . .255
IP Filter policy rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .255
IP Filter policy enforcement. . . . . . . . . . . . . . . . . . . . . . . . . . . .258
Adding a rule to an IP Filter policy. . . . . . . . . . . . . . . . . . . . . . .259
Deleting a rule from an IP Filter policy . . . . . . . . . . . . . . . . . . .259
Aborting an IP Filter transaction . . . . . . . . . . . . . . . . . . . . . . . .259
IP Filter policy distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . .260
Policy database distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .260
Database distribution settings . . . . . . . . . . . . . . . . . . . . . . . . .261
ACL policy distribution to other switches . . . . . . . . . . . . . . . . .262
Fabric-wide enforcement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .263
Notes on joining a switch to the fabric . . . . . . . . . . . . . . . . . . .264
Management interface security . . . . . . . . . . . . . . . . . . . . . . . . . . . .266
Configuration examples. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .267
IPsec protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .269
Security associations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .269
Authentication and encryption algorithms . . . . . . . . . . . . . . . .269
IPsec policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .270
IKE policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
Creating the tunnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .272
Example of an end-to-end transport tunnel mode. . . . . . . . . . 274

Chapter 9

Maintaining the Switch Configuration File
Configuration settings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .277
Configuration file format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .278
Configuration file backup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .279
Uploading a configuration file in interactive mode . . . . . . . . .279
Configuration file restoration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .280
Restrictions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .281
Configuration download without disabling a switch . . . . . . . .282
Configurations across a fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . .284
Downloading a configuration file from one switch to
another switch of the same model . . . . . . . . . . . . . . . . . . . . . .284
Security considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .284
Configuration management for Virtual Fabrics . . . . . . . . . . . . . . . .285
Uploading a configuration file from a switch with
Virtual Fabrics enabled . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .285
Restoring a logical switch configuration using
configDownload . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .285
Restrictions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .286
Brocade configuration form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .287

12

Fabric OS Administrator’s Guide
53-1002920-02

Chapter 10

Installing and Maintaining Firmware
Firmware download process overview . . . . . . . . . . . . . . . . . . . . . . .289
Upgrading and downgrading firmware . . . . . . . . . . . . . . . . . . .291
Considerations for FICON CUP environments . . . . . . . . . . . . .291
HA sync state . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .291
Preparing for a firmware download . . . . . . . . . . . . . . . . . . . . . . . . .292
Obtaining and decompressing firmware . . . . . . . . . . . . . . . . .293
Connected switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .293
Firmware download on switches . . . . . . . . . . . . . . . . . . . . . . . . . . .294
Switch firmware download process overview. . . . . . . . . . . . . .294
Firmware download on a Backbone. . . . . . . . . . . . . . . . . . . . . . . . .296
Backbone firmware download process overview. . . . . . . . . . .296
Firmware download from a USB device . . . . . . . . . . . . . . . . . . . . . .299
Enabling the USB device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .299
Viewing the USB file system . . . . . . . . . . . . . . . . . . . . . . . . . . .299
Downloading from the USB device using the relative path. . .300
Downloading from the USB device using the absolute path. .300
FIPS support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .300
Public and private key management . . . . . . . . . . . . . . . . . . . .300
The firmwareDownload command . . . . . . . . . . . . . . . . . . . . . .301
Power-on firmware checksum test . . . . . . . . . . . . . . . . . . . . . .302
Testing and restoring firmware on switches . . . . . . . . . . . . . . . . . .302
Testing a different firmware version on a switch . . . . . . . . . . .302
Testing and restoring firmware on Backbones . . . . . . . . . . . . . . . .304
Testing different firmware versions on Backbones . . . . . . . . .304
Validating a firmware download . . . . . . . . . . . . . . . . . . . . . . . . . . . .306

Chapter 11

Managing Virtual Fabrics
Virtual Fabrics overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .309
Logical switch overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
Default logical switch. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
Logical switches and fabric IDs. . . . . . . . . . . . . . . . . . . . . . . . .311
Port assignment in logical switches . . . . . . . . . . . . . . . . . . . . .312
Logical switches and connected devices . . . . . . . . . . . . . . . . .313
Management model for logical switches. . . . . . . . . . . . . . . . . . . . . 314
Logical fabric overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .315
Logical fabric and ISLs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .315
Base switch and extended ISLs . . . . . . . . . . . . . . . . . . . . . . . . 316
Account management and Virtual Fabrics . . . . . . . . . . . . . . . . . . .319
Supported platforms for Virtual Fabrics . . . . . . . . . . . . . . . . . . . . .320
Supported port configurations in the fixed-port switches. . . .320
Supported port configurations in Brocade Backbones . . . . . .321
Virtual Fabrics interaction with other Fabric OS features . . . .322

Fabric OS Administrator’s Guide
53-1002920-02

13

Limitations and restrictions of Virtual Fabrics . . . . . . . . . . . . . . . .322
Restrictions on XISLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .323
Restrictions on moving ports . . . . . . . . . . . . . . . . . . . . . . . . . .324
Enabling Virtual Fabrics mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . .324
Disabling Virtual Fabrics mode . . . . . . . . . . . . . . . . . . . . . . . . . . . .325
Configuring logical switches to use basic configuration values. . .326
Creating a logical switch or base switch . . . . . . . . . . . . . . . . . . . . .326
Executing a command in a different logical switch context . . . . . .328
Deleting a logical switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .329
Adding and moving ports on a logical switch . . . . . . . . . . . . . . . . .329
Displaying logical switch configuration . . . . . . . . . . . . . . . . . . . . . .330
Changing the fabric ID of a logical switch . . . . . . . . . . . . . . . . . . . .331
Changing a logical switch to a base switch . . . . . . . . . . . . . . . . . . .331
Setting up IP addresses for a logical switch . . . . . . . . . . . . . . . . . .333
Removing an IP address for a logical switch . . . . . . . . . . . . . . . . . .333
Configuring a logical switch to use XISLs . . . . . . . . . . . . . . . . . . . .333
Changing the context to a different logical fabric . . . . . . . . . . . . . .334
Creating a logical fabric using XISLs . . . . . . . . . . . . . . . . . . . . . . . .334

Chapter 12

Administering Advanced Zoning
Zone types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .337
Zoning overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .338
Approaches to zoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .339
Zone objects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .340
Zone configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .341
Zoning enforcement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .342
Considerations for zoning architecture . . . . . . . . . . . . . . . . . .342
Best practices for zoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .343
Broadcast zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .343
Broadcast zones and Admin Domains . . . . . . . . . . . . . . . . . . .344
Broadcast zones and FC-FC routing . . . . . . . . . . . . . . . . . . . . .345
High availability considerations with broadcast zones . . . . . .346
Loop devices and broadcast zones . . . . . . . . . . . . . . . . . . . . .346
Broadcast zones and default zoning mode . . . . . . . . . . . . . . .346
Zone aliases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .346
Creating an alias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .347
Adding members to an alias . . . . . . . . . . . . . . . . . . . . . . . . . . .347
Removing members from an alias . . . . . . . . . . . . . . . . . . . . . .348
Deleting an alias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .349
Viewing an alias in the defined configuration . . . . . . . . . . . . .349

14

Fabric OS Administrator’s Guide
53-1002920-02

Zone creation and maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . .350
Displaying existing zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . .350
Creating a zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .350
Adding devices (members) to a zone . . . . . . . . . . . . . . . . . . . .351
Removing devices (members) from a zone . . . . . . . . . . . . . . .352
Replacing zone members . . . . . . . . . . . . . . . . . . . . . . . . . . . . .353
Deleting a zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .355
Viewing a zone in the defined configuration . . . . . . . . . . . . . .356
Viewing zone configuration names without case distinction .356
Validating a zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .358
Default zoning mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .360
Setting the default zoning mode. . . . . . . . . . . . . . . . . . . . . . . .361
Viewing the current default zone access mode . . . . . . . . . . . .361
Zone database size. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .362
Zone configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .362
Creating a zone configuration . . . . . . . . . . . . . . . . . . . . . . . . . .363
Adding zones to a zone configuration . . . . . . . . . . . . . . . . . . .363
Removing members from a zone configuration. . . . . . . . . . . .364
Enabling a zone configuration . . . . . . . . . . . . . . . . . . . . . . . . .364
Disabling a zone configuration . . . . . . . . . . . . . . . . . . . . . . . . .365
Deleting a zone configuration . . . . . . . . . . . . . . . . . . . . . . . . . .365
Abandoning zone configuration changes . . . . . . . . . . . . . . . . .366
Viewing all zone configuration information . . . . . . . . . . . . . . .366
Viewing selected zone configuration information . . . . . . . . . .367
Viewing the configuration in the effective zone database . . .367
Clearing all zone configurations . . . . . . . . . . . . . . . . . . . . . . . .367
Zone object maintenance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .368
Copying a zone object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .368
Deleting a zone object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .369
Renaming a zone object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .370
Zone configuration management. . . . . . . . . . . . . . . . . . . . . . . . . . .370
Security and zoning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
Zone merging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
Fabric segmentation and zoning. . . . . . . . . . . . . . . . . . . . . . . .373
Zone merging scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .373
Concurrent zone transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
Viewing zone database transactions . . . . . . . . . . . . . . . . . . . .377

Chapter 13

Traffic Isolation Zoning
Traffic Isolation Zoning overview . . . . . . . . . . . . . . . . . . . . . . . . . . .379
TI zone failover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .380
Additional considerations when disabling failover . . . . . . . . .381
FSPF routing rules and traffic isolation . . . . . . . . . . . . . . . . . .383
Enhanced TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .384
Illegal configurations with enhanced TI zones. . . . . . . . . . . . .385

Fabric OS Administrator’s Guide
53-1002920-02

15

Traffic Isolation Zoning over FC routers . . . . . . . . . . . . . . . . . . . . . .386
TI zones within an edge fabric . . . . . . . . . . . . . . . . . . . . . . . . .388
TI zones within a backbone fabric . . . . . . . . . . . . . . . . . . . . . .389
Limitations of TI zones over FC routers . . . . . . . . . . . . . . . . . .390
Fabric-Level Traffic Isolation in a backbone fabric . . . . . . . . . . . . .390
Fabric-Level TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .391
Failover behavior for Fabric-Level TI zones . . . . . . . . . . . . . . .392
Creating a separate TI zone for each path . . . . . . . . . . . . . . . .392
Creating a single TI zone for all paths . . . . . . . . . . . . . . . . . . .393
General rules for TI zones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .394
Traffic Isolation Zone violation handling for trunk ports . . . . .395
Supported configurations for Traffic Isolation Zoning . . . . . . . . . .396
Additional configuration rules for enhanced TI zones . . . . . . .396
Trunking with TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .397
Limitations and restrictions of Traffic Isolation Zoning . . . . . . . . .398
Admin Domain considerations for Traffic Isolation Zoning . . . . . .398
Virtual Fabrics considerations for Traffic Isolation Zoning . . . . . . .399
Traffic Isolation Zoning over FC routers with Virtual Fabrics . . . . .401
Creating a TI zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .402
Creating a TI zone in a base fabric . . . . . . . . . . . . . . . . . . . . . .404
Modifying TI zones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .405
Changing the state of a TI zone . . . . . . . . . . . . . . . . . . . . . . . . . . . .406
Deleting a TI zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .407
Displaying TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .407
Troubleshooting TI zone routing problems . . . . . . . . . . . . . . . . . . .408
Setting up TI zones over FCR (sample procedure) . . . . . . . . . . . . .409

Chapter 14

Optimizing Fabric Behavior
Adaptive Networking overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . .413
Ingress Rate Limiting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414
Virtual Fabrics considerations. . . . . . . . . . . . . . . . . . . . . . . . . . 414
Limiting traffic from a particular device . . . . . . . . . . . . . . . . . . 415
Disabling Ingress Rate Limiting . . . . . . . . . . . . . . . . . . . . . . . . 415
QoS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .415
License requirements for QoS. . . . . . . . . . . . . . . . . . . . . . . . . . 416
CS_CTL-based frame prioritization. . . . . . . . . . . . . . . . . . . . . . . . . . 416
Supported configurations for CS_CTL-based frame
prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
High availability considerations for CS_CTL-based frame
prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
Enabling CS_CTL-based frame prioritization on ports . . . . . . . 417
Disabling CS_CTL-based frame prioritization on ports . . . . . .418
Using CS_CTL auto mode at the chassis level . . . . . . . . . . . . .418
Considerations for using CS_CTL-based frame prioritization .418

16

Fabric OS Administrator’s Guide
53-1002920-02

QoS zone-based traffic prioritization . . . . . . . . . . . . . . . . . . . . . . . . 419
QoS zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
QoS on E_Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .421
QoS over FC routers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .421
Virtual Fabrics considerations for QoS zone-based traffic
prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .422
High-availability considerations for QoS zone-based traffic
prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .422
Supported configurations for QoS zone-based traffic
prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .423
Limitations and restrictions for QoS zone-based traffic
prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .424
Setting QoS zone-based traffic prioritization. . . . . . . . . . . . . . . . . .424
Setting QoS zone-based traffic prioritization over FC routers . . . .426
Disabling QoS zone-based traffic prioritization. . . . . . . . . . . . . . . .426

Chapter 15

Bottleneck Detection
Bottleneck detection overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . .427
Types of bottlenecks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .428
How bottlenecks are reported. . . . . . . . . . . . . . . . . . . . . . . . . .428
Supported configurations for bottleneck detection . . . . . . . . . . . .429
Limitations of bottleneck detection . . . . . . . . . . . . . . . . . . . . .429
High availability considerations for bottleneck detection . . . .430
Upgrade and downgrade considerations for bottleneck
detection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .430
Trunking considerations for bottleneck detection . . . . . . . . . .430
Virtual Fabrics considerations for bottleneck detection . . . . .430
Access Gateway considerations for bottleneck detection. . . .430
Enabling bottleneck detection on a switch . . . . . . . . . . . . . . . . . . .431
Displaying bottleneck detection configuration details . . . . . . . . . .431
Setting bottleneck detection alerts . . . . . . . . . . . . . . . . . . . . . . . . .433
Setting both a congestion alert and a latency alert . . . . . . . .434
Setting a congestion alert only . . . . . . . . . . . . . . . . . . . . . . . . .434
Setting a latency alert only . . . . . . . . . . . . . . . . . . . . . . . . . . . .435
Changing bottleneck detection parameters . . . . . . . . . . . . . . . . . .435
Examples of applying and changing bottleneck detection
parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .436
Advanced bottleneck detection settings . . . . . . . . . . . . . . . . . . . . .439
Excluding a port from bottleneck detection . . . . . . . . . . . . . . . . . .440
Displaying bottleneck statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . .442
Disabling bottleneck detection on a switch . . . . . . . . . . . . . . . . . .442

Fabric OS Administrator’s Guide
53-1002920-02

17

Chapter 16

In-flight Encryption and Compression
In-flight encryption and compression overview. . . . . . . . . . . . . . . .445
Supported ports for in-flight encryption and compression . . .446
In-flight encryption and compression restrictions . . . . . . . . . .446
How in-flight encryption and compression are enabled . . . . .448
Authentication and key generation for encryption and
compression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .448
Availability considerations for encryption and compression. .449
Virtual Fabrics considerations for encryption and
compression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .449
In-flight compression on long-distance ports. . . . . . . . . . . . . .450
Compression ratios for compression-enabled ports . . . . . . . .450
Configuring in-flight encryption and compression on an EX_Port .450
Configuring in-flight encryption and compression on an E_Port . .451
Viewing the encryption and compression configuration . . . . . . . .452
Configuring and enabling authentication for in-flight encryption .453
Enabling in-flight encryption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .455
Enabling in-flight compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . .456
Disabling in-flight encryption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .456
Disabling in-flight compression . . . . . . . . . . . . . . . . . . . . . . . . . . . .457

Chapter 17

Diagnostic Port
Diagnostic Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .459
Supported platforms for D_Port . . . . . . . . . . . . . . . . . . . . . . . . . . . .459
Licensing requirements for D_Port . . . . . . . . . . . . . . . . . . . . . . . . .460
Understanding D_Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .460
Advantages of D_Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .461
D_Port configuration mode and nature of test . . . . . . . . . . . .461
General limitations and considerations for D_Port . . . . . . . . .462
Supported topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .463
Topology 1: ISLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .463
Topology 2: ICLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .463
Topology 3: Access Gateways . . . . . . . . . . . . . . . . . . . . . . . . . .464
Topology 4: HBA to switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . .465
Using D_Port without HBAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .465
Enabling D_Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .465
Disabling D_Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .466
Using D_Port with HBAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .467
Automatic mode configuration . . . . . . . . . . . . . . . . . . . . . . . . .467
Dynamic mode configuration . . . . . . . . . . . . . . . . . . . . . . . . . .468
BCU D_Port commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .468
Limitations and considerations for D_Port with HBAs. . . . . . .468
Controlling testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .469

18

Fabric OS Administrator’s Guide
53-1002920-02

Example test scenarios and output . . . . . . . . . . . . . . . . . . . . . . . . .469
Confirming SFP and link status with an HBA . . . . . . . . . . . . . .470
Starting and stopping D_Port testing . . . . . . . . . . . . . . . . . . . .470

Chapter 18

NPIV
NPIV overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .473
Upgrade considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
Fixed addressing mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
10-bit addressing mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
Configuring NPIV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .475
Enabling and disabling NPIV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 476
Viewing NPIV port configuration information . . . . . . . . . . . . . . . . . 476
Viewing virtual PID login information . . . . . . . . . . . . . . . . . . . .478

Chapter 19

Fabric-Assigned PWWN
Fabric-Assigned PWWN overview . . . . . . . . . . . . . . . . . . . . . . . . . . .479
User- and auto-assigned FA-PWWN behavior . . . . . . . . . . . . . . . . .480
Configuring an FA-PWWN for an HBA connected to an
Access Gateway. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .481
Configuring an FA-PWWN for an HBA connected to an edge
switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .482
Supported switches and configurations for FA-PWWN. . . . . . . . . .483
Configuration upload and download considerations for
FA-PWWN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .483
Security considerations for FA-PWWN . . . . . . . . . . . . . . . . . . . . . . .483
Restrictions of FA-PWWN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .484
Access Gateway N_Port failover with FA-PWWN . . . . . . . . . . . . . . .484

Chapter 20

Managing Administrative Domains
Administrative Domains overview . . . . . . . . . . . . . . . . . . . . . . . . . .485
Admin Domain features. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .487
Requirements for Admin Domains . . . . . . . . . . . . . . . . . . . . . .487
Admin Domain access levels. . . . . . . . . . . . . . . . . . . . . . . . . . .487
User-defined Admin Domains . . . . . . . . . . . . . . . . . . . . . . . . . .488
System-defined Admin Domains. . . . . . . . . . . . . . . . . . . . . . . .488
Home Admin Domains and login . . . . . . . . . . . . . . . . . . . . . . .490
Admin Domain member types. . . . . . . . . . . . . . . . . . . . . . . . . .491
Admin Domains and switch WWNs. . . . . . . . . . . . . . . . . . . . . .492
Admin Domain compatibility, availability, and merging . . . . . .494

Fabric OS Administrator’s Guide
53-1002920-02

19

Admin Domain management for physical fabric administrators . .494
Setting the default zoning mode for Admin Domains . . . . . . .495
Creating an Admin Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . .495
User assignments to Admin Domains . . . . . . . . . . . . . . . . . . .496
Removing an Admin Domain from a user account . . . . . . . . .498
Activating an Admin Domain . . . . . . . . . . . . . . . . . . . . . . . . . . .498
Deactivating an Admin Domain . . . . . . . . . . . . . . . . . . . . . . . .499
Adding members to an existing Admin Domain . . . . . . . . . . . .499
Removing members from an Admin Domain . . . . . . . . . . . . . .500
Renaming an Admin Domain . . . . . . . . . . . . . . . . . . . . . . . . . .500
Deleting an Admin Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . .501
Deleting all user-defined Admin Domains . . . . . . . . . . . . . . . .502
Deleting all user-defined Admin Domains non-disruptively . .502
Validating an Admin Domain member list . . . . . . . . . . . . . . . .506
SAN management with Admin Domains . . . . . . . . . . . . . . . . . . . . .506
CLI commands in an AD context . . . . . . . . . . . . . . . . . . . . . . . .507
Executing a command in a different AD context . . . . . . . . . . .507
Displaying an Admin Domain configuration . . . . . . . . . . . . . . .508
Switching to a different Admin Domain context. . . . . . . . . . . .508
Admin Domain interactions with other Fabric OS features . . .509
Admin Domains, zones, and zone databases . . . . . . . . . . . . . 510
Admin Domains and LSAN zones . . . . . . . . . . . . . . . . . . . . . . .511
Configuration upload and download in an AD context . . . . . .512

Section II
Chapter 21

Licensed Features
Administering Licensing
Licensing overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .515
Brocade 7800 Upgrade license . . . . . . . . . . . . . . . . . . . . . . . . . . . .523
ICL licensing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .523
ICL 1st POD license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .523
ICL 2nd POD license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .524
ICL 8-link license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .524
ICL 16-link license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .524
Enterprise ICL license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .524
8G licensing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .525
Slot-based licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .526
Upgrade and downgrade considerations . . . . . . . . . . . . . . . . .526
Assigning a license to a slot . . . . . . . . . . . . . . . . . . . . . . . . . . .526
Removing a license from a slot . . . . . . . . . . . . . . . . . . . . . . . . .527
10G licensing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .527
Enabling 10 Gbps operation on an FC port . . . . . . . . . . . . . . .528
Enabling the 10-GbE ports on an FX8-24 blade . . . . . . . . . . .529

20

Fabric OS Administrator’s Guide
53-1002920-02

Temporary licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .530
Restrictions on upgrading temporary slot-based licenses . . .531
Date change restriction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .531
Configupload and download considerations . . . . . . . . . . . . . .531
Expired licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .531
Universal temporary licenses . . . . . . . . . . . . . . . . . . . . . . . . . .532
Extending a universal temporary license . . . . . . . . . . . . . . . . .532
Universal temporary license shelf life. . . . . . . . . . . . . . . . . . . .532
Viewing installed licenses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .532
Activating a license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .533
Adding a licensed feature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .533
Removing a licensed feature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .534
Ports on Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .535
Displaying installed licenses . . . . . . . . . . . . . . . . . . . . . . . . . . .536
Activating Ports on Demand . . . . . . . . . . . . . . . . . . . . . . . . . . .537
Dynamic Ports on Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . .537
Displaying the port license assignments . . . . . . . . . . . . . . . . .538
Enabling Dynamic Ports on Demand . . . . . . . . . . . . . . . . . . . .538
Disabling Dynamic Ports on Demand. . . . . . . . . . . . . . . . . . . .539
Reserving a port license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .540
Releasing a port from a POD set. . . . . . . . . . . . . . . . . . . . . . . .540

Chapter 22

Inter-chassis Links
Inter-chassis links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .543
License requirements for ICLs . . . . . . . . . . . . . . . . . . . . . . . . .544
ICLs for the Brocade DCX 8510 Backbone family. . . . . . . . . . . . . .544
ICL trunking on the Brocade DCX 8510-8 and DCX 8510-4 . .545
ICLs for the Brocade DCX Backbone family. . . . . . . . . . . . . . . . . . .546
ICL trunking on the Brocade DCX and DCX-4S. . . . . . . . . . . . .547
Virtual Fabrics considerations for ICLs . . . . . . . . . . . . . . . . . . . . . .547
Supported topologies for ICL connections . . . . . . . . . . . . . . . . . . .547
Mesh topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .547
Core-edge topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .549

Chapter 23

Monitoring Fabric Performance
Advanced Performance Monitoring overview . . . . . . . . . . . . . . . . .551
Types of monitors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .551
Restrictions for installing monitors . . . . . . . . . . . . . . . . . . . . . .552
Virtual Fabrics considerations for Advanced Performance
Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .552
Access Gateway considerations for Advanced Performance
Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .553

Fabric OS Administrator’s Guide
53-1002920-02

21

End-to-end performance monitoring . . . . . . . . . . . . . . . . . . . . . . . .553
Maximum number of EE monitors . . . . . . . . . . . . . . . . . . . . . .553
Supported port configurations for EE monitors . . . . . . . . . . . .554
Adding EE monitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .554
Setting a mask for an EE monitor . . . . . . . . . . . . . . . . . . . . . . .555
Deleting EE monitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .556
Displaying EE monitor counters . . . . . . . . . . . . . . . . . . . . . . . .557
Clearing EE monitor counters . . . . . . . . . . . . . . . . . . . . . . . . . .557
Frame monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .558
License requirements for frame monitoring . . . . . . . . . . . . . .558
Creating frame types to be monitored . . . . . . . . . . . . . . . . . . .559
Creating a frame monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . .559
Deleting frame types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .560
Adding frame monitors to a port. . . . . . . . . . . . . . . . . . . . . . . .560
Removing frame monitors from a port . . . . . . . . . . . . . . . . . . .560
Saving a frame monitor configuration . . . . . . . . . . . . . . . . . . .560
Displaying frame monitors . . . . . . . . . . . . . . . . . . . . . . . . . . . .561
Clearing frame monitor counters . . . . . . . . . . . . . . . . . . . . . . .562
Top Talker monitors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .562
Top Talker monitors and FC-FC routing. . . . . . . . . . . . . . . . . . .563
Limitations of Top Talker monitors . . . . . . . . . . . . . . . . . . . . . .565
Adding a Top Talker monitor to a port (port mode) . . . . . . . . .565
Adding Top Talker monitors on all switches in the fabric
(fabric mode) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .565
Displaying the top n bandwidth-using flows on a port
(port mode). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .566
Displaying top talking flows for a given domain ID
(fabric mode) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .566
Deleting a Top Talker monitor on a port (port mode) . . . . . . .567
Deleting all fabric mode Top Talker monitors. . . . . . . . . . . . . .567
Trunk monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .567
Trunk monitoring considerations . . . . . . . . . . . . . . . . . . . . . . .567
Saving and restoring monitor configurations . . . . . . . . . . . . . . . . .567
Performance data collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .568

Chapter 24

Managing Trunking Connections
Trunking overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .569
Types of trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .570
Masterless trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .570
License requirements for trunking . . . . . . . . . . . . . . . . . . . . . . 571
Port groups for trunking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 571
Supported platforms for trunking. . . . . . . . . . . . . . . . . . . . . . . . . . . 571
Supported configurations for trunking . . . . . . . . . . . . . . . . . . . . . . 571
High Availability support for trunking . . . . . . . . . . . . . . . . . . . .572
Requirements for trunk groups . . . . . . . . . . . . . . . . . . . . . . . . . . . .572
Recommendations for trunk groups . . . . . . . . . . . . . . . . . . . . . . . .572
Configuring trunk groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .573

22

Fabric OS Administrator’s Guide
53-1002920-02

Enabling trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574
Disabling trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574
Displaying trunking information . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574
Trunk Area and Admin Domains. . . . . . . . . . . . . . . . . . . . . . . . . . . . 576
Example of Trunk Area assignment on port domain,index . . . 576
ISL trunking over long-distance fabrics . . . . . . . . . . . . . . . . . . . . . . 576
EX_Port trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .577
Masterless EX_Port trunking. . . . . . . . . . . . . . . . . . . . . . . . . . .577
Supported configurations and platforms for EX_Port
trunking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .578
Configuring EX_Port trunking . . . . . . . . . . . . . . . . . . . . . . . . . .578
Displaying EX_Port trunking information . . . . . . . . . . . . . . . . .578
F_Port trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .579
F_Port trunking for Access Gateway . . . . . . . . . . . . . . . . . . . . .579
F_Port trunking for Brocade adapters . . . . . . . . . . . . . . . . . . .581
F_Port trunking considerations. . . . . . . . . . . . . . . . . . . . . . . . .582
F_Port trunking in Virtual Fabrics . . . . . . . . . . . . . . . . . . . . . . .584
Displaying F_Port trunking information . . . . . . . . . . . . . . . . . . . . . .585
Disabling F_Port trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .585
Enabling the DCC policy on a trunk area. . . . . . . . . . . . . . . . . . . . .586

Chapter 25

Managing Long-Distance Fabrics
Long-distance fabrics overview . . . . . . . . . . . . . . . . . . . . . . . . . . . .587
Extended Fabrics device limitations . . . . . . . . . . . . . . . . . . . . . . . .588
Long-distance link modes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .588
Configuring an extended ISL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .589
Enabling long distance when connecting to TDM devices . . .590
Forward error correction on long-distance links . . . . . . . . . . . . . . .591
Enabling FEC on a long-distance link . . . . . . . . . . . . . . . . . . . .591
Disabling FEC on a long-distance link . . . . . . . . . . . . . . . . . . .591

Chapter 26

Using FC-FC Routing to Connect Fabrics
FC-FC routing overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .593
License requirements for FC-FC routing . . . . . . . . . . . . . . . . . .594
Supported platforms for FC-FC routing. . . . . . . . . . . . . . . . . . .594
Supported configurations for FC-FC routing. . . . . . . . . . . . . . .595
Network OS connectivity limitations . . . . . . . . . . . . . . . . . . . . .595
Fibre Channel routing concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . .596
Proxy devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .599
FC-FC routing topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .600
Phantom domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .601
FC router authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .603
Setting up FC-FC routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .603
Verifying the setup for FC-FC routing . . . . . . . . . . . . . . . . . . . .604

Fabric OS Administrator’s Guide
53-1002920-02

23

Backbone fabric IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .605
Assigning backbone fabric IDs . . . . . . . . . . . . . . . . . . . . . . . . .606
FCIP tunnel configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .606
Inter-fabric link configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .607
Configuring an IFL for both edge and backbone connections 607
Configuring EX_Ports on an ICL . . . . . . . . . . . . . . . . . . . . . . . .611
FC router port cost configuration . . . . . . . . . . . . . . . . . . . . . . . . . . .613
Port cost considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .614
Setting router port cost for an EX_Port. . . . . . . . . . . . . . . . . . .614
Shortest IFL cost configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . .615
Configuring shortest IFL cost . . . . . . . . . . . . . . . . . . . . . . . . . . 617
EX_Port frame trunking configuration . . . . . . . . . . . . . . . . . . . . . . .619
LSAN zone configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .620
Use of Admin Domains with LSAN zones and FC-FC routing .620
Zone definition and naming . . . . . . . . . . . . . . . . . . . . . . . . . . .620
LSAN zones and fabric-to-fabric communications. . . . . . . . . .621
Controlling device communication with the LSAN . . . . . . . . . .621
Configuring backbone fabrics for interconnectivity . . . . . . . . .623
Setting the maximum LSAN count . . . . . . . . . . . . . . . . . . . . . .624
HA and downgrade considerations for LSAN zones . . . . . . . .624
LSAN zone policies using LSAN tagging . . . . . . . . . . . . . . . . . .624
LSAN zone binding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .628
Proxy PID configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .633
Fabric parameter considerations . . . . . . . . . . . . . . . . . . . . . . . . . . .633
Inter-fabric broadcast frames. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .634
Displaying the current broadcast configuration. . . . . . . . . . . .634
Enabling broadcast frame forwarding . . . . . . . . . . . . . . . . . . .634
Disabling broadcast frame forwarding . . . . . . . . . . . . . . . . . . .634
Resource monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .634
FC-FC routing and Virtual Fabrics. . . . . . . . . . . . . . . . . . . . . . . . . . .636
Logical switch configuration for FC routing . . . . . . . . . . . . . . .637
Backbone-to-edge routing with Virtual Fabrics . . . . . . . . . . . .638
Upgrade and downgrade considerations for FC-FC routing . . . . . .639
How replacing port blades affects EX_Port configuration. . . .639
Displaying the range of output ports connected to xlate domains 639

Appendix A

Port Indexing

Appendix B

FIPS Support
FIPS overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .645
Zeroization functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .645
Power-on self-tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .647
Conditional tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .647

24

Fabric OS Administrator’s Guide
53-1002920-02

FIPS mode configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .647
LDAP in FIPS mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .648
LDAP certificates for FIPS mode . . . . . . . . . . . . . . . . . . . . . . . .650
Preparing a switch for FIPS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .651
Overview of steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .652
Enabling FIPS mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .652
Zeroizing for FIPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .655
Displaying FIPS configuration . . . . . . . . . . . . . . . . . . . . . . . . . .655

Appendix C

Hexadecimal Conversion
Example conversion of the hexadecimal triplet Ox616000 . .657
Decimal-to-hexadecimal conversion table . . . . . . . . . . . . . . . .658

Index

Fabric OS Administrator’s Guide
53-1002920-02

25

26

Fabric OS Administrator’s Guide
53-1002920-02

Figures

Figure 1

Well-known addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Figure 2

Identifying the blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

Figure 3

Blade swap with Virtual Fabrics during the swap. . . . . . . . . . . . . . . . . . . . . . . . 101

Figure 4

Blade swap with Virtual Fabrics after the swap . . . . . . . . . . . . . . . . . . . . . . . . . 102

Figure 5

Principal ISLs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

Figure 6

New switch added to existing fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Figure 7

Virtual channels on a QoS-enabled ISL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120

Figure 8

Gateway link merging SANs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Figure 9

Single host and target . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

Figure 10

Windows 2000 VSA configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Figure 11

Example of a brocade.dct file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180

Figure 12

Example of the dictiona.dcm file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180

Figure 13

SNMP structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

Figure 14

SNMP query. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

Figure 15

SNMP trap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

Figure 16

Brocade MIB tree location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208

Figure 17

DH-CHAP authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244

Figure 18

Protected endpoints configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Figure 19

Gateway tunnel configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268

Figure 20

Endpoint-to-gateway tunnel configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268

Figure 21

Switch before and after enabling Virtual Fabrics . . . . . . . . . . . . . . . . . . . . . . . . 310

Figure 22

Switch before and after creating logical switches . . . . . . . . . . . . . . . . . . . . . . . 311

Figure 23

Fabric IDs assigned to logical switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312

Figure 24

Assigning ports to logical switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312

Figure 25

Logical switches connected to devices and non-Virtual Fabrics switch . . . . . . 314

Figure 26

Logical switches in a single chassis belong to separate fabrics . . . . . . . . . . . . 314

Figure 27

Logical switches connected to other logical switches through physical ISLs. . 316

Figure 28

Logical switches connected to form logical fabrics . . . . . . . . . . . . . . . . . . . . . . 316

Figure 29

Base switches connected by an XISL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317

Figure 30

Logical ISLs connecting logical switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318

Figure 31

Logical fabric using ISLs and XISLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318

Figure 32

Example of logical fabrics in multiple chassis and XISLs . . . . . . . . . . . . . . . . . 335

Figure 33

Zoning example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339

Figure 34

Broadcast zones and Admin Domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

Figure 35

Traffic Isolation zone creating a dedicated path through the fabric . . . . . . . . . 380

Figure 36

Fabric incorrectly configured for TI zone with failover disabled . . . . . . . . . . . . 382

Fabric OS Administrator’s Guide
53-1002920-02

27

28

Figure 37

Dedicated path is the only shortest path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383

Figure 38

Dedicated path is not the shortest path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384

Figure 39

Enhanced TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384

Figure 40

Illegal ETIZ configuration: two paths from one port to two devices on the
same remote domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385

Figure 41

Illegal ETIZ configuration: two paths from one port . . . . . . . . . . . . . . . . . . . . . . 386

Figure 42

Traffic Isolation Zoning over FCR. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387

Figure 43

TI zone in an edge fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388

Figure 44

TI zone in a backbone fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389

Figure 45

Fabric-level traffic isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391

Figure 46

TI zone misconfiguration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395

Figure 47

Dedicated path with Virtual Fabrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399

Figure 48

Creating a TI zone in a logical fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400

Figure 49

Creating a TI zone in a base fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400

Figure 50

Example configuration for TI zones over FC routers in logical fabrics . . . . . . . 401

Figure 51

Logical representation of TI zones over FC routers in logical fabrics . . . . . . . . 401

Figure 52

TI over FCR example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409

Figure 53

QoS traffic prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420

Figure 54

QoS with E_Ports enabled . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421

Figure 55

Traffic prioritization in a logical fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423

Figure 56

Affected seconds for bottleneck detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433

Figure 57

Encryption and compression on 16 Gbps ISLs. . . . . . . . . . . . . . . . . . . . . . . . . . 446

Figure 58

Example of a basic D_Port connection between switches . . . . . . . . . . . . . . . . 460

Figure 59

ISLs connecting multiple switches and chassis . . . . . . . . . . . . . . . . . . . . . . . . . 463

Figure 60

ICLs connecting chassis blades. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463

Figure 61

Single Access Gateway to switch. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464

Figure 62

Multiple Access Gateways cascaded to switch . . . . . . . . . . . . . . . . . . . . . . . . . 464

Figure 63

Access Gateway to HBA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464

Figure 64

HBA to switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465

Figure 65

Fabric-assigned port World Wide Name provisioning scenarios . . . . . . . . . . . . 480

Figure 66

Fabric with two Admin Domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486

Figure 67

Filtered fabric views when using Admin Domains . . . . . . . . . . . . . . . . . . . . . . . 486

Figure 68

Fabric with AD0 and AD255. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490

Figure 69

Fabric showing switch and device WWNs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493

Figure 70

Filtered fabric views showing converted switch WWNs . . . . . . . . . . . . . . . . . . . 493

Figure 71

AD0 and two user-defined Admin Domains, AD1 and AD2 . . . . . . . . . . . . . . . . 504

Figure 72

AD0 with three zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 504

Figure 73

Minimum configuration for 64 Gbps ICLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545

Figure 74

DCX-4S allowed ICL connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546

Figure 75

ICL triangular topology with Brocade DCX 8510-8 chassis . . . . . . . . . . . . . . . . 548

Figure 76

Full nine-mesh topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549

Figure 77

64 Gbps ICL core-edge topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 550

Fabric OS Administrator’s Guide
53-1002920-02

Figure 78

Setting end-to-end monitors on a port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554

Figure 79

Mask positions for end-to-end monitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 556

Figure 80

Fabric mode Top Talker monitors on FC router do not monitor any flows . . . . 564

Figure 81

Fabric mode Top Talker monitors on FC router monitor flows over the E_Port 564

Figure 82

Port group configuration for the Brocade 5100 . . . . . . . . . . . . . . . . . . . . . . . . . 571

Figure 83

Switch in Access Gateway mode without F_Port masterless trunking . . . . . . . 580

Figure 84

Switch in Access Gateway mode with F_Port masterless trunking . . . . . . . . . . 580

Figure 85

A metaSAN with inter-fabric links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 596

Figure 86

A metaSAN with edge-to-edge and backbone fabrics and LSAN zones . . . . . . 597

Figure 87

Edge SANs connected through a backbone fabric. . . . . . . . . . . . . . . . . . . . . . . 599

Figure 88

MetaSAN with imported devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600

Figure 89

Sample topology (physical topology) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 601

Figure 90

EX_Port phantom switch topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602

Figure 91

Shortest IFL solution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 617

Figure 92

Example of setting up Speed LSAN tag. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 626

Figure 93

LSAN zone binding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 629

Figure 94

EX_Ports in a base switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 637

Figure 95

Logical representation of EX_Ports in a base switch . . . . . . . . . . . . . . . . . . . . . 638

Figure 96

Backbone-to-edge routing across base switch using FC router in legacy mode 639

Fabric OS Administrator’s Guide
53-1002920-02

29

30

Fabric OS Administrator’s Guide
53-1002920-02

Tables

Table 1

Daemons that are automatically restarted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Table 2

Terminal port parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

Table 3

Help topic contents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Table 4

fabricShow fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Table 5

Ports affected when you enable or disable a switch in VF or non-VF mode. . . . 79

Table 6

Core and CP blade terminology and platform support. . . . . . . . . . . . . . . . . . . . . 95

Table 7

Port blade terminology, numbering, and platform support . . . . . . . . . . . . . . . . . 96

Table 8

Blade compatibility within Brocade Backbone families . . . . . . . . . . . . . . . . . . . . 98

Table 9

Duplicate PWWN behavior: First login takes precedence over second login . . 110

Table 10

Duplicate PWWN behavior: Second login overrides first login . . . . . . . . . . . . . 110

Table 11

Duplicate PWWN behavior: Port type determines which login takes
precedence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Table 12

Combinations of routing policy and IOD with Lossless DLS enabled . . . . . . . . 130

Table 13

Fibre Channel gigabit values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

Table 14

Fibre Channel data frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

Table 15

Total FC ports, ports per port group, and unreserved buffer credits per
port group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

Table 16

Configurable distances for Extended Fabrics . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

Table 17

Default Fabric OS roles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

Table 18

Permission types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Table 19

Maximum number of simultaneous sessions. . . . . . . . . . . . . . . . . . . . . . . . . . . 154

Table 20

Default local user accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

Table 21

LDAP options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Table 22

Authentication configuration options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Table 23

Syntax for VSA-based account roles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

Table 24

Entries in dictionary.brocade file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Table 25

Brocade custom TACACS+ attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

Table 26

Secure protocol support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

Table 27

Items needed to deploy secure protocols. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196

Table 28

Main security scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196

Table 29

SSL certificate files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203

Table 30

Brocade SNMP MIB dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214

Table 31

Access Gateway MIB support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216

Table 32

Security level options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218

Table 33

Blocked listener applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228

Table 34

Access defaults. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

Fabric OS Administrator’s Guide
53-1002920-02

31

32

Table 35

Port information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

Table 36

Valid methods for specifying policy members . . . . . . . . . . . . . . . . . . . . . . . . . . 232

Table 37

FCS policy states. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

Table 38

FCS switch operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236

Table 39

Distribution policy states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238

Table 40

DCC policy states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239

Table 41

DCC policy behavior with FA-PWWN when created using lockdown support . . 241

Table 42

DCC policy behavior when created manually with PWWN . . . . . . . . . . . . . . . . . 242

Table 43

SCC policy states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242

Table 44

FCAP certificate files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

Table 45

Supported services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256

Table 46

Implicit IP Filter rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258

Table 47

Default IP policy rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258

Table 48

Interaction between fabric-wide consistency policy and distribution settings . 261

Table 49

Supported policy databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261

Table 50

Fabric-wide consistency policy settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263

Table 51

Merging fabrics with matching fabric-wide consistency policies. . . . . . . . . . . . 265

Table 52

Examples of strict fabric merges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266

Table 53

Fabric merges with tolerant and absent combinations . . . . . . . . . . . . . . . . . . . 266

Table 54

Algorithms and associated authentication policies . . . . . . . . . . . . . . . . . . . . . . 270

Table 55

CLI commands to display or modify switch configuration information . . . . . . . 281

Table 56

Brocade configuration and connection form . . . . . . . . . . . . . . . . . . . . . . . . . . . 287

Table 57

Backbone HA sync states. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292

Table 58

Commands used for validating a firmware download . . . . . . . . . . . . . . . . . . . . 307

Table 59

Blade and port types supported on logical switches . . . . . . . . . . . . . . . . . . . . . 321

Table 60

Virtual Fabrics interaction with Fabric OS features . . . . . . . . . . . . . . . . . . . . . . 322

Table 61

Maximum number of logical switches per chassis. . . . . . . . . . . . . . . . . . . . . . . 323

Table 62

Approaches to fabric-based zoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339

Table 63

Considerations for zoning architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342

Table 64

Zone merging scenarios: Defined and effective configurations . . . . . . . . . . . . 373

Table 65

Zone merging scenarios: Different content . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374

Table 66

Zone merging scenarios: Different names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375

Table 67

Zone merging scenarios: TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375

Table 68

Zone merging scenarios: Default access mode . . . . . . . . . . . . . . . . . . . . . . . . . 376

Table 69

Zone merging scenarios: Mixed Fabric OS versions. . . . . . . . . . . . . . . . . . . . . . 376

Table 70

Traffic behavior when failover is enabled or disabled in TI zones . . . . . . . . . . 381

Table 71

Comparison between CS_CTL-based and QoS zone-based prioritization. . . . . 416

Table 72

Mapping of CS_CTL values to QoS priority for frame prioritization in
CS_CTL default mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417

Table 73

Mapping of CS_CTL values to QoS priority for frame prioritization in
CS_CTL auto mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417

Fabric OS Administrator’s Guide
53-1002920-02

Table 74

Number of ports supported for in-flight encryption and compression
at various port speeds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447

Table 75

Supported platforms for D_Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460

Table 76

D_Port configuration mode and nature of test. . . . . . . . . . . . . . . . . . . . . . . . . . 462

Table 77

Limitation on number of D_Ports for simultaneous tests . . . . . . . . . . . . . . . . . 469

Table 78

Number of supported NPIV devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 474

Table 79

AD user types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488

Table 80

Ports and devices in CLI output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507

Table 81

Admin Domain interaction with Fabric OS features . . . . . . . . . . . . . . . . . . . . . . 509

Table 82

Configuration upload and download scenarios in an AD context . . . . . . . . . . . 512

Table 83

Available Brocade licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 516

Table 84

License requirements and location name by feature . . . . . . . . . . . . . . . . . . . . 519

Table 85

Base to Upgrade license comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523

Table 86

List of available user ports when implementing PODs . . . . . . . . . . . . . . . . . . 535

Table 87

Number of logical switches that support performance monitors . . . . . . . . . . . 552

Table 88

Maximum number of frame monitors and offsets per port . . . . . . . . . . . . . . . . 558

Table 89

Predefined values at offset 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559

Table 90

Trunking over long distance for the Brocade Backbones and blades . . . . . . . 576

Table 91

F_Port masterless trunking considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 582

Table 92

PWWN format for F_Port and N_Port trunk ports. . . . . . . . . . . . . . . . . . . . . . . . 584

Table 93

Fabric-wide settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 589

Table 94

LSAN information stored in FC routers, with and without LSAN zone binding . 630

Table 95

Zeroization behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645

Table 96

FIPS mode restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 647

Table 97

FIPS and non-FIPS modes of operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 648

Table 98

Active Directory keys to modify . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 650

Table 99

Decimal-to-hexadecimal conversion table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 658

Fabric OS Administrator’s Guide
53-1002920-02

33

34

Fabric OS Administrator’s Guide
53-1002920-02

About This Document

In this chapter
• Supported hardware and software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• What’s new in this document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Document conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Notice to the reader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Additional information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Getting technical help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Document feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

35
36
36
38
38
40
41

Supported hardware and software
In those instances in which procedures or parts of procedures documented here apply to some
switches but not to others, this guide identifies exactly which switches are supported and which are
not.
Although many different software and hardware configurations are tested and supported by
Brocade Communications Systems, Inc. for Fabric OS v7.2.0, documenting all possible
configurations and scenarios is beyond the scope of this document.
The following hardware platforms are supported by this release of Fabric OS:

• Fixed-port switches:
- Brocade 300 switch
- Brocade 5100 switch
- Brocade 5300 switch
- Brocade 5410 embedded switch
- Brocade 5424 embedded switch
- Brocade 5430 embedded switch
- Brocade 5431 embedded switch
- Brocade 5450 embedded switch
- Brocade 5460 embedded switch
- Brocade 5470 embedded switch
- Brocade 5480 embedded switch
- Brocade M6505 embedded switch
- Brocade 6505 switch
- Brocade 6510 switch
Fabric OS Administrator’s Guide
53-1002920-02

35

-

Brocade 6520 switch
Brocade 6547 embedded switch
Brocade 7800 extension switch
Brocade VA-40FC
Brocade Encryption Switch

• Brocade DCX Backbone family:
- Brocade DCX
- Brocade DCX-4S
• Brocade DCX 8510 Backbone family:
- Brocade DCX 8510-4
- Brocade DCX 8510-8

What’s new in this document
Information that was modified:

• Renamed and moved the section about the two Ethernet ports on the CP blade to
“Management Ethernet port bonding” on page 65.

• Moved the section “Enabling forward error correction” from the Routing chapter to Chapter 3,
“Performing Advanced Configuration Tasks”.

• In Chapter 17, “Diagnostic Port,” updated Table 76 and added some HA considerations and
some considerations for D_Port with HBAs.

• Updated the “Setting up FC-FC routing” section and the “Configuring EX_Ports on an ICL”
section in Chapter 26, “Using FC-FC Routing to Connect Fabrics”.

• In Appendix B, “FIPS Support,” updated Table 96 on page 647 with entries for Authentication
and FC-FC routing.

Document conventions
This section describes text formatting conventions and important notice formats used in this
document.

Text formatting
The narrative-text formatting conventions that are used are as follows:
bold text

36

Identifies command names
Identifies the names of user-manipulated GUI elements
Identifies keywords and operands
Identifies text to enter at the GUI or CLI

Fabric OS Administrator’s Guide
53-1002920-02

italic text

Provides emphasis
Identifies variables
Identifies paths and Internet addresses
Identifies document titles

code text

Identifies CLI output
Identifies command syntax examples

For readability, command names in the narrative portions of this guide are presented in mixed
lettercase: for example, switchShow. In actual examples, command lettercase is often all
lowercase. Otherwise, this manual specifically notes those cases in which a command is case
sensitive.

Command syntax conventions
Command syntax in this manual follows these conventions:
command

Commands are printed in bold.

--option, option

Command options are printed in bold.

-argument, arg

Arguments.

[]

Optional element.

variable

Variables are printed in italics. In the help pages, values are underlined or
enclosed in angled brackets < >.

...

Repeat the previous element, for example “member[;member...]”

value

Fixed values following arguments are printed in plain font. For example,
--show WWN

|

Boolean. Elements are exclusive. Example: --show -mode egress | ingress

Notes, cautions, and warnings
The following notices and statements are used in this manual. They are listed below in order of
increasing severity of potential hazards.

NOTE

A note provides a tip, guidance or advice, emphasizes important information, or provides a reference
to related information.

ATTENTION
An Attention statement indicates potential damage to hardware or data.

Fabric OS Administrator’s Guide
53-1002920-02

37

CAUTION
A Caution statement alerts you to situations that can be potentially hazardous to you or cause
damage to hardware, firmware, software, or data.

DANGER
A Danger statement indicates conditions or situations that can be potentially lethal or extremely
hazardous to you. Safety labels are also attached directly to products to warn of these conditions
or situations.

Key terms
For definitions specific to Brocade and Fibre Channel, see the Brocade Glossary.
For definitions of SAN-specific terms, visit the Storage Networking Industry Association online
dictionary at:
http://www.snia.org/education/dictionary

Notice to the reader
This document may contain references to the trademarks of the following corporations. These
trademarks are the properties of their respective companies and corporations.
These references are made for informational purposes only.
Corporation

Referenced Trademarks and Products

Microsoft Corporation

Windows, Windows NT, Internet Explorer

Mozilla Corporation

Mozilla, Firefox

Netscape Communications Corporation

Netscape

Red Hat, Inc.

Red Hat, Red Hat Network, Maximum RPM, Linux Undercover

Sun Microsystems, Inc.

Sun, Solaris

Additional information
This section lists additional Brocade and industry-specific documentation that you might find
helpful.

Brocade resources
To get up-to-the-minute information, go to http://my.brocade.com and register at no cost for a user
ID and password.

38

Fabric OS Administrator’s Guide
53-1002920-02

For practical discussions about SAN design, implementation, and maintenance, you can obtain
Building SANs with Brocade Fabric Switches through:
http://www.amazon.com
For additional Brocade documentation, visit the Brocade SAN Info Center and click the Resource
Library location:
http://www.brocade.com
Release notes are available on the My Brocade website and are also bundled with the Fabric OS
firmware.

Other industry resources
For additional resource information, visit the Technical Committee T11 website. This website
provides interface standards for high-performance and mass storage applications for Fibre
Channel, storage management, and other applications:
http://www.t11.org
For information about the Fibre Channel industry, visit the Fibre Channel Industry Association
website:
http://www.fibrechannel.org

Fabric OS Administrator’s Guide
53-1002920-02

39

Getting technical help
Contact your switch support supplier for hardware, firmware, and software support, including
product repairs and part ordering. To expedite your call, have the following information available:
1. General Information

•
•
•
•
•

Switch model
Switch operating system version
Error numbers and messages received
supportSave command output
Detailed description of the problem, including the switch or fabric behavior immediately
following the problem, and specific questions

• Description of any troubleshooting steps already performed and the results
• Serial console and Telnet session logs
• syslog message logs
2. switch serial number
The switch serial number and corresponding bar code are provided on the serial number label,
as illustrated below.:

' "!&'
FT00X0054E9
The serial number label is located as follows:

• Brocade 300, 5100, 5300, 6505, M6505, 6510, 6520, 6547, 7800, VA-40FC, and Brocade
Encryption Switch—On the switch ID pull-out tab located inside the chassis on the port side on
the left

• Brocade 5410, 5424, 5430, 5431, 5450, 5460, 5470, 5480—Serial number label attached to
the module

• Brocade 6510—On the pull-out tab on the front of the switch
• Brocade DCX and DCX 8510-8—On the bottom right on the port side of the chassis
• Brocade DCX-4S and DCX 8510-4—On the bottom right on the port side of the chassis, directly
above the cable management comb
3. World Wide Name (WWN)
Use the wwn command to display the switch WWN.
If you cannot use the wwn command because the switch is inoperable, you can get the WWN
from the same place as the serial number, except for the Brocade DCX enterprise class
platform. For the Brocade DCX enterprise class platform, access the numbers on the WWN
cards by removing the Brocade logo plate at the top of the nonport side of the chassis.
For the Brocade 5424 embedded switch: Provide the license ID. Use the licenseIdShow
command to display the WWN.

40

Fabric OS Administrator’s Guide
53-1002920-02

Document feedback
Quality is our first concern at Brocade and we have made every effort to ensure the accuracy and
completeness of this document. However, if you find an error or an omission, or you think that a
topic needs further development, we want to hear from you. Forward your feedback to:
documentation@brocade.com
Provide the title and version number of the document and as much detail as possible about your
comment, including the topic heading and page number and your suggestions for improvement.

Fabric OS Administrator’s Guide
53-1002920-02

41

42

Fabric OS Administrator’s Guide
53-1002920-02

Section

Standard Features

I

This section describes standard Fabric OS features, and includes the following chapters:

• Chapter 1, “Understanding Fibre Channel Services”
• Chapter 2, “Performing Basic Configuration Tasks”
• Chapter 3, “Performing Advanced Configuration Tasks”
• Chapter 4, “Routing Traffic”
• Chapter 5, “Buffer-to-Buffer Credits and Credit Recovery”
• Chapter 6, “Managing User Accounts”
• Chapter 7, “Configuring Protocols”
• Chapter 8, “Configuring Security Policies”
• Chapter 9, “Maintaining the Switch Configuration File”
• Chapter 10, “Installing and Maintaining Firmware”
• Chapter 11, “Managing Virtual Fabrics”
• Chapter 12, “Administering Advanced Zoning”
• Chapter 13, “Traffic Isolation Zoning”
• Chapter 14, “Optimizing Fabric Behavior”
• Chapter 15, “Bottleneck Detection”
• Chapter 16, “In-flight Encryption and Compression”
• Chapter 17, “Diagnostic Port”
• Chapter 18, “NPIV”
• Chapter 19, “Fabric-Assigned PWWN”
• Chapter 20, “Managing Administrative Domains”

Fabric OS Administrator’s Guide
53-1002920-02

43

44

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

Understanding Fibre Channel Services

1

In this chapter
• Fibre Channel services overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
• Management server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
• Platform services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
• Management server database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
• Topology discovery. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
• Device login . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
• High availability of daemon processes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Fibre Channel services overview
Fibre Channel services define service functions that reside at well-known addresses. A well-known
address is a reserved three-byte address for each service. Services are provided to either nodes or
management applications in the fabric.

Figure 1

Well-known addresses

Fabric Login — The Fabric Login server assigns a fabric address to a fabric node, which allows it to
communicate with services on the switch or other nodes in the fabric. The fabric address is a 24-bit
address (0x000000) containing three 3-byte nodes. Reading from left to right, the first node
(0x000000) represents the domain ID, the second node (0x000000) the port area number of the
port where the node is attached, and the third node (0x000000) the arbitrated loop physical
address (AL_PA), if applicable.
Directory server — The directory server or name server registers fabric and public nodes and
conducts queries to discover other devices in the fabric.
Fabric controller — The fabric controller provides State Change Notifications (SCNs) to registered
nodes when a change in the fabric topology occurs.
Time server — The time server sends the time to the member switches in the fabric from either the
principal switch or, if configured, the primary fabric configuration server (FCS) switch.
Refer to Chapter 8, “Configuring Security Policies,” for additional information on FCS policies.

Fabric OS Administrator’s Guide
53-1002920-02

45

1

Management server

Management server — The management server provides a single point for managing the fabric.
This is the only service that users can configure. See “Management server” below for more details
Alias server — The alias server keeps a group of nodes registered as one name to handle multicast
groups.
Broadcast server — The broadcast server is optional. When frames are transmitted to this address,
they are broadcast to all operational N_ and NL_Ports.
When registration and query frames are sent to a well-known address, a different protocol service,
Fibre Channel Common Transport (FC-CT), is used. This protocol provides a simple, consistent
format and behavior when a service provider is accessed for registration and query purposes.

Management server
The Brocade Fabric OS management server (MS) allows a SAN management application to retrieve
information and administer interconnected switches, servers, and storage devices. The
management server assists in the autodiscovery of switch-based fabrics and their associated
topologies.
A client of the management server can find basic information about the switches in the fabric and
use this information to construct topology relationships. The management server also allows you to
obtain certain switch attributes and, in some cases, modify them. For example, logical names
identifying switches can be registered with the management server.
The management server provides several advantages for managing a Fibre Channel fabric:

• It is accessed by an external Fibre Channel node at the well-known address FFFFFAh, so an
application can access information about the entire fabric management with minimal
knowledge of the existing configuration.

• It is replicated on every Brocade switch within a fabric.
• It provides an unzoned view of the overall fabric configuration. This fabric topology view
exposes the internal configuration of a fabric for management purposes; it contains
interconnect information about switches and devices connected to the fabric. Under normal
circumstances, a device (typically an FCP initiator) queries the name server for storage devices
within its member zones. Because this limited view is not always sufficient, the management
server provides the application with a list of the entire name server database.

Platform services
By default, all management services except platform services are enabled; the MS platform service
and topology discovery are disabled.
You can activate and deactivate the platform services throughout the fabric. Activating the platform
services attempts to activate the MS platform service for each switch in the fabric. The change
takes effect immediately and is committed to the configuration database of each affected switch.
MS activation is persistent across power cycles and reboots.

NOTE
The commands msplMgmtActivate and msplMgmtDeactivate are allowed only in AD0 and AD255.

46

Fabric OS Administrator’s Guide
53-1002920-02

Management server database

1

Platform services and Virtual Fabrics
Each logical switch has a separate platform database. All platform registrations done to a logical
switch are valid only in that particular logical switch’s Virtual Fabric.
Activating the platform services on a switch activates the platform services on all logical switches in
a Virtual Fabric. Similarly, deactivating the platform services deactivates the platform service on all
logical switches in a Virtual Fabric. The msPlatShow command displays all platforms registered in a
Virtual Fabric.

Enabling platform services
When FCS policy is enabled, the msplMgmtActivate command can be issued only from the primary
FCS switch.
The execution of the msplMgmtActivate command is subject to Admin Domain restrictions that may
be in place.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the msCapabilityShow command to verify that all switches in the fabric support the MS
platform service; otherwise, the next step fails.
3. Enter the msplMgmtActivate command, as in the following example.
switch:admin> msplmgmtactivate
Request to activate MS Platform Service in progress......
*Completed activating MS Platform Service in the fabric!

Disabling platform services
Use the following procedure to disable platform services:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the msplMgmtDeactivate command.
3. Enter y to confirm the deactivation, as in the following example.
switch:admin> msplmgmtdeactivate
MS Platform Service is currently enabled.
This will erase MS Platform Service configuration
information as well as database in the entire fabric.
Would you like to continue this operation? (yes, y, no, n): [no] y
Request to deactivate MS Platform Service in progress......
*Completed deactivating MS Platform Service in the fabric!

Management server database
You can control access to the management server database.
An access control list (ACL) of WWN addresses determines which systems have access to the
management server database. The ACL typically contains those WWNs of host systems that are
running management applications.

Fabric OS Administrator’s Guide
53-1002920-02

47

1

Management server database

If the list is empty (the default), the management server is accessible to all systems connected
in-band to the fabric. For more access security, you can specify WWNs in the ACL so that access to
the management server is restricted to only those WWNs listed.

NOTE

The management server is logical switch-capable. All management server features are supported
within a logical switch.

Displaying the management server ACL
Use the following procedure to display the management server ACL:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the msConfigure command.
The command becomes interactive.
3. At the “select” prompt, enter 1 to display the access list.
A list of WWNs that have access to the management server is displayed.
Example of an empty access list
switch:admin> msconfigure
0
Done
1
Display the access list
2
Add member based on its Port/Node WWN
3
Delete member based on its Port/Node WWN
select : (0..3) [1] 1
MS Access list is empty.
0
Done
1
Display the access list
2
Add member based on its Port/Node WWN
3
Delete member based on its Port/Node WWN
select : (0..3) [1] 0
done ...

Adding a member to the ACL
Use the following procedure to add a member to the ACL:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the msConfigure command.
The command becomes interactive.
3. At the “select” prompt, enter 2 to add a member based on its port/node WWN.
4. At the “Port/Node WWN” prompt, enter the WWN of the host to be added to the ACL.
5. At the “select” prompt, enter 1 to display the access list so you can verify that the WWN you
entered was added to the ACL.
6. After verifying that the WWN was added correctly, enter 0 at the prompt to end the session.
7.

At the “Update the FLASH?” prompt, enter y.

8. Press Enter to update the nonvolatile memory and end the session.

48

Fabric OS Administrator’s Guide
53-1002920-02

Management server database

1

Example of adding a member to the management server ACL
switch:admin> msconfigure
0
Done
1
Display the access list
2
Add member based on its Port/Node WWN
3
Delete member based on its Port/Node WWN
select : (0..3) [1] 2
Port/Node WWN (in hex): [00:00:00:00:00:00:00:00] 20:00:00:20:37:65:ce:aa
*WWN is successfully added to the MS ACL.
0
Done
1
Display the access list
2
Add member based on its Port/Node WWN
3
Delete member based on its Port/Node WWN
select : (0..3) [2] 1
MS Access List consists of (14): {
20:00:00:20:37:65:ce:aa
20:00:00:20:37:65:ce:bb
20:00:00:20:37:65:ce:ff
20:00:00:20:37:65:ce:11
20:00:00:20:37:65:ce:22
20:00:00:20:37:65:ce:33
20:00:00:20:37:65:ce:44
10:00:00:60:69:04:11:24
10:00:00:60:69:04:11:23
21:00:00:e0:8b:04:70:3b
10:00:00:60:69:04:11:33
20:00:00:20:37:65:ce:55
20:00:00:20:37:65:ce:66
00:00:00:00:00:00:00:00
}
0
Done
1
Display the access list
2
Add member based on its Port/Node WWN
3
Delete member based on its Port/Node WWN
select : (0..3) [1] 0
done ...
Update the FLASH? (yes, y, no, n): [yes] y
*Successfully saved the MS ACL to the flash.

Deleting a member from the ACL
When you delete a member from the ACL, that member no longer has access to the management
server.

NOTE
If you delete the last member of the ACL, leaving the ACL list is empty, then the management server
will be accessible to all systems connected in-band to the fabric.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the msConfigure command.
The command becomes interactive.
3. At the “select” prompt, enter 3 to delete a member based on its port/node WWN.
4. At the “Port/Node WWN” prompt, enter the WWN of the member to be deleted from the ACL.

Fabric OS Administrator’s Guide
53-1002920-02

49

1

Management server database

5. At the “select” prompt, enter 1 to display the access list so you can verify that the WWN you
entered was deleted from the ACL.
6. After verifying that the WWN was deleted correctly, enter 0 at the “select” prompt to end the
session.
7.

At the “Update the FLASH?” prompt, enter y.

8. Press Enter to update the nonvolatile memory and end the session.
Example of deleting a member from the management server ACL
switch:admin> msconfigure
0
Done
1
Display the access list
2
Add member based on its Port/Node WWN
3
Delete member based on its Port/Node WWN
select : (0..3) [1] 3
Port/Node WWN (in hex): [00:00:00:00:00:00:00:00] 10:00:00:00:c9:29:b3:84
*WWN is successfully deleted from the MS ACL.
0
Done
1
Display the access list
2
Add member based on its Port/Node WWN
3
Delete member based on its Port/Node WWN
select : (0..3) [3] 1
MS Access list is empty
0
Done
1
Display the access list
2
Add member based on its Port/Node WWN
3
Delete member based on its Port/Node WWN
select : (0..3) [1] 0

Viewing the contents of the management server database
Use the following procedure to view the contents of the management server database:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the msPlatShow command.
Example of viewing the contents of the management server platform database
switch:admin> msplatshow
----------------------------------------------------------Platform Name: [9] "first obj"
Platform Type: 5 : GATEWAY
Number of Associated M.A.: 1
[35] "http://java.sun.com/products/plugin"
Number of Associated Node Names: 1
Associated Node Names:
10:00:00:60:69:20:15:71
----------------------------------------------------------Platform Name: [10] "second obj"
Platform Type: 7 : HOST_BUS_ADAPTER
Number of Associated M.A.: 1
Associated Management Addresses:
[30] "http://java.sun.com/products/1"

50

Fabric OS Administrator’s Guide
53-1002920-02

Topology discovery

1

Number of Associated Node Names: 1
Associated Node Names:
10:00:00:60:69:20:15:75

Clearing the management server database
Use the following procedure to clear the management server database:

NOTE
The command msPlClearDB is allowed only in AD0 and AD255.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the msplClearDb command.
3. Enter y to confirm the deletion.
The management server platform database is cleared.

Topology discovery
The topology discovery feature can be displayed, enabled, and disabled; it is disabled by default.
The commands mstdEnable and mstdDisable are allowed only in AD0 and AD255.

Displaying topology discovery status
Use the following procedure to display the status of the topology discovery:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the mstdReadConfig command.
switch:admin> mstdreadconfig
*MS Topology Discovery is Enabled.

Enabling topology discovery
Use the following procedure to enable topology discovery:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the appropriate following command based on how you want to enable discovery:

• For the local switch, enter the mstdEnable command.
• For the entire fabric, enter the mstdEnable all command.
Example of enabling discovery
switch:admin> mstdenable
Request to enable MS Topology Discovery Service in progress....
*MS Topology Discovery enabled locally.
switch:admin> mstdenable ALL
Request to enable MS Topology Discovery Service in progress....

Fabric OS Administrator’s Guide
53-1002920-02

51

1

Topology discovery

*MS Topology Discovery enabled locally.
*MS Topology Discovery Enable Operation Complete!!

Disabling topology discovery
Use the following procedure to disable topology discovery:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the appropriate following command based on how you want to disable discovery:

• For the local switch, enter the mstdDisable command.
• For the entire fabric, enter the mstdDisable all command.
A warning displays stating that all NID entries might be cleared.
3. Enter y to disable the Topology Discovery feature.

NOTE
Topology discovery is disabled by default.

ATTENTION
Disabling discovery of management server topology might erase all node ID entries.
If Admin Domains are enabled, you must be in the AD0 or AD255 context. Refer to Chapter 20,
“Managing Administrative Domains,” for additional information.
Example of disabling discovery
switch:admin> mstddisable
This may erase all NID entries. Are you sure?

(yes, y, no, n): [no] y

Request to disable MS Topology Discovery Service in progress....
*MS Topology Discovery disabled locally.
switch:admin> mstddisable all
This may erase all NID entries. Are you sure?

(yes, y, no, n): [no] y

Request to disable MS Topology Discovery Service in progress....
*MS Topology Discovery disabled locally.
*MS Topology Discovery Disable Operation Complete!!

52

Fabric OS Administrator’s Guide
53-1002920-02

Device login

1

Device login
A device can be storage, a host, or a switch. When new devices are introduced into the fabric, they
must be powered on and, if a host or storage device, connected to a switch. Switch-to-switch logins
(using the E_Port) are handled differently than storage and host logins. E_Ports exchange different
frames than the ones listed below with the Fabric Controller to access the fabric. Once storage and
host devices are powered on and connected, the following logins occur:
1. FLOGI—Fabric Login command establishes a 24-bit address for the device logging in, and
establishes buffer-to-buffer credits and the class of service supported.
2. PLOGI—Port Login command logs the device into the name server to register its information
and query for devices that share its zone. During the PLOGI process, information is exchanged
between the new device and the fabric. Some of the following types of information exchanges
occur:

• SCR—State Change Registration registers the device for State Change Notifications. If a
change in the fabric occurs, such as a zoning change or a change in the state of a device
to which this device has access, the device receives a Registered State Change
Notification (RSCN).

• Registration—A device exchanges registration information with the name server.
• Query—Devices query the name server for information about the device it can access.

Principal switch
In a fabric with multiple switches, and one inter-switch link (ISL) exists between any two switches, a
principal switch is automatically elected. The principal switch provides the following capabilities:

• Maintains time for the entire fabric. Subordinate switches synchronize their time with the
principal switch. Changes to the clock server value on the principal switch are propagated to all
switches in the fabric.

• Manages domain ID assignment within the fabric. If a switch requests a domain ID that has
been used before, the principal switch grants the same domain ID unless it is in use by another
switch.

E_Port login process
An E_Port does not use a FLOGI to log in to another switch. Instead, the new switch exchanges
frames with the neighboring switch to establish that the new switch is an E_Port and that it has
information to exchange. If everything is acceptable to the neighboring switch, it replies to the new
switch with an SW_ACC (accept) frame. The initializing frame is an Exchange Link Parameters (ELP)
frame that allows an exchange of parameters between two ports, such as flow control,
buffer-to-buffer credits, RA_TOV, and ED_TOV. This is not a negotiation. If one or the other port’s link
parameters do not match, a link does not occur. Once an SW_ACC frame is received from the
neighboring switch, the new switch sends an Exchange Switch Capabilities (ESC) frame. The two
switches exchange routing protocols and agree on a common routing protocol. An SW_ACC frame is
received from the neighboring switch and the new switch sends an Exchange Fabric Parameters
(EFP) frame to the neighboring switch, requesting principal switch priority and the domain ID list.
Buffer-to-buffer credits for the device and switch ports are exchanged in the SW_ACC command
sent to the device in response to the FLOGI.

Fabric OS Administrator’s Guide
53-1002920-02

53

1

Device login

Fabric login process
A device performs a fabric login (FLOGI) to determine if a fabric is present. If a fabric is detected
then it exchanges service parameters with the fabric controller. A successful FLOGI sends back the
24-bit address for the device in the fabric. The device must issue and successfully complete a
FLOGI command before communicating with other devices in the fabric.
Because the device does not know its 24-bit address until after the FLOGI, the source ID (SID) in
the frame header of the FLOGI request are zeros (0x000000).

Port login process
The steps in the port initialization process occur as the result of a protocol that functions to
discover the type of device connected and establish the port type and negotiate port speed. See
“Port Types” on page 88 for a discussion of available port types.
The Fibre Channel protocol (FCP) auto discovery process enables private storage devices that
accept the process login (PRLI) to communicate in a fabric.
If device probing is enabled, the embedded port performs a PLOGI and attempts a PRLI into the
device to retrieve information to enter into the name server. This enables private devices that do
not explicitly register with the Name Server (NS) to be entered in the NS and receive full fabric
access.
A fabric-capable device registers its information with the name server during a FLOGI. These
devices typically register information with the name server before querying for a device list. The
embedded port still performs a PLOGI and attempts a PRLI with these devices.
If a port decides to end the current session, it initiates a logout. A logout concludes the session and
terminates any work in progress associated with that session.
To display the contents of a switch’s name server, use the nsShow or nsAllShow command.
For more information about these commands, refer to the Fabric OS Command Reference.

RSCNs
A Registered State Change Notification (RSCN) is a notification frame that is sent to devices that
are zoned together and are registered to receive a State Change Notification (SCN). The RSCN is
responsible for notifying all devices of fabric changes. The following general list of actions can
cause an RSCN to be sent through your fabric:

•
•
•
•
•

A new device has been added to the fabric.
An existing device has been removed from the fabric.
A zone has changed.
A switch name has changed or an IP address has changed.
Nodes leaving or joining the fabric, such as zoning, powering on or shutting down a device, or
zoning changes.

NOTE

Fabric reconfigurations with no domain change do not cause an RSCN.

54

Fabric OS Administrator’s Guide
53-1002920-02

High availability of daemon processes

1

Duplicate Port World Wide Name
According to Fibre Channel standards, the Port World Wide Name (PWWN) of a device cannot
overlap with that of another device, thus having duplicate PWWNs within the same fabric is an
illegal configuration.
If a PWWN conflict occurs with two devices attached to the same domain, Fabric OS handles device
login in such a way that only one device may be logged in to the fabric at a time. For more
information, refer to “Duplicate PWWN handling during device login” on page 110.
If a PWWN conflict occurs and two duplicate devices are attached to the fabric through different
domains, the devices are removed from the Name Server database and a RASlog is generated.

Device recovery
To recover devices that have been removed from the Name Server database due to duplicate
PWWNs, the devices must re-login to the fabric. This is true for any device—for example, a device on
an F_Port, NPIV devices, or devices attached to a switch in Access Gateway mode.

High availability of daemon processes
Starting non-critical daemons is automatic; you cannot configure the startup process. The following
sequence of events occurs when a non-critical daemon fails:
1. A RASlog and AUDIT event message are logged.
2. The daemon is automatically started again.
3. If the restart is successful, then another message is sent to RASlog and AUDIT reporting the
successful restart status.
4. If the restart fails, another message is sent to RASlog and no further attempts are made to
restart the daemon.
Schedule downtime and reboot the switch at your convenience.
The following table lists the daemons that are considered non-critical and are automatically
restarted on failure.
Table 1

Daemons that are automatically restarted

Daemon

Description

arrd

Asynchronous Response Router, which is used to send management data to hosts when the switch is
accessed through the APIs (FA API or SMI-S).

cald

Common Access Layer daemon, which is used by manageability applications.

raslogd

Reliability, Availability, and Supportability daemon logs error detection, reporting, handling, and
presentation of data into a format readable by you and management tools.

rpcd

Remote Procedure Call daemon, which is used by the API (Fabric Access API and SMI-S).

snmpd

Simple Network Management Protocol daemon.

npd

Flow Vision daemon.

Fabric OS Administrator’s Guide
53-1002920-02

55

1

High availability of daemon processes

Table 1

56

Daemons that are automatically restarted (Continued)

Daemon

Description

traced

Trace daemon provides trace entry date and time translation to Trace Device at startup and when
date/time changed by command. Maintains the trace dump trigger parameters in a Trace Device.
Performs the trace Background Dump, trace automatic FTP, and FTP “aliveness check” if auto-FTP is
enabled.

trafd

Traffic daemon implements Bottleneck detection.

webd

Webserver daemon used for Web Tools (includes httpd as well).

weblinkerd

Weblinker daemon provides an HTTP interface to manageability applications for switch management
and fabric discovery.

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

2

Performing Basic Configuration Tasks

In this chapter
• Fabric OS overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Fabric OS command line interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Password modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• The switch Ethernet interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Date and time settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Domain IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Switch names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Chassis names. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Fabric name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Switch activation and deactivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Switch and Backbone shutdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Basic connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

57
58
63
64
72
75
76
77
77
78
80
81

Fabric OS overview
This chapter describes how to configure your Brocade SAN using the Fabric OS command line
interface (CLI). Before you can configure a storage area network (SAN), you must power up the
Backbone platform or switch and blades, and then set the IP addresses of those devices. Although
this chapter focuses on configuring a SAN using the CLI, you can also use the following methods to
configure a SAN:

• Web Tools
For Web Tools procedures, refer to Web Tools Administrator’s Guide.

• Brocade Network Advisor
For additional information, refer to the Brocade Network Advisor User Manual for the version
you have.

• A third-party application using the API
For third-party application procedures, refer to the third-party API documentation.
Because of the differences between fixed-port and variable-port devices, procedures sometimes
differ among Brocade models. As new Brocade models are introduced, new features sometimes
apply only to those models.
When procedures or parts of procedures apply to some models but not others, this guide identifies
the specifics for each model. For example, a number of procedures that apply only to variable-port
devices are found in Chapter 3, “Performing Advanced Configuration Tasks”.

Fabric OS Administrator’s Guide
53-1002920-02

57

2

Fabric OS command line interface

Although many different software and hardware configurations are tested and supported by
Brocade Communications Systems, Inc., documenting all possible configurations and scenarios is
beyond the scope of this document. In some cases, earlier releases are highlighted to present
considerations for interoperating with them.
The hardware reference manuals for Brocade products describe how to power up devices and set
their IP addresses. After the IP address is set, you can use the CLI procedures contained in this
guide. For additional information about the commands used in the procedures, refer to the Fabric
OS Command Reference.

Fabric OS command line interface
Fabric OS uses Role-Based Access Control (RBAC) to control access to all Fabric OS operations.
Each feature is associated with an RBAC role and you need to know which role is allowed to run a
command, make modifications to the switch, or view the output of the command. To determine
which RBAC role you need to run a command, review the section “Role-Based Access Control” on
page 152.
Note the following about the command display in this guide:

• Commands are shown and can be entered either in all lower case or using Java-style
capitalization. This means that while bannershow and bannerShow will both work,
BANNERSHOW and BannerShow will not.

• When command examples in this guide show user input enclosed in quotation marks, the
quotation marks are required. Example: zonecreate "zonename" requires that the value for
zonename be in quotation marks.

Console sessions using the serial port
Be aware of the following behaviors for serial connections:

• Some procedures require that you connect through the serial port; for example, setting the IP
address or setting the boot PROM password.

• Brocade DCX and DCX 8510 Backbone families: You can connect to CP0 or CP1 using either of
the two serial ports.

Connecting to Fabric OS through the serial port
Use the following procedure to connect to the Fabric OS using the serial port:
1. Connect the serial cable to the serial port on the switch and to an RS-232 serial port on
the workstation.
If the serial port on the workstation is an RJ-45 port, instead of RS-232, remove the adapter on
the end of the serial cable and insert the exposed RJ-45 connector into the RJ-45 serial port on
the workstation.
2. Open a terminal emulator application (such as HyperTerminal on a PC, TERM, TIP, or Kermit in
a UNIX environment), and configure the application as follows:

58

Fabric OS Administrator’s Guide
53-1002920-02

Fabric OS command line interface

2

• In a Windows environment enter the following parameters:
TABLE 2

Terminal port parameters

Parameter

Value

Bits per second

9600

Databits

8

Parity

None

Stop bits

1

Flow control

None

• In a UNIX environment, enter the following string at the prompt:
tip /dev/ttyb -9600

If ttyb is already in use, use ttya instead and enter the following string at the prompt:
tip /dev/ttya -9600

Telnet or SSH sessions
You can connect to the Fabric OS through a Telnet or SSH connection or by using a console session
on the serial port. The switch must also be physically connected to the network. If the switch
network interface is not configured or the switch has been disconnected from the network, use a
console session on the serial port as described in “Console sessions using the serial port” on
page 58.

NOTE

To automatically configure the network interface on a DHCP-enabled switch, plug the switch into the
network and power it on. The DHCP client automatically gets the IP and gateway addresses from the
DHCP server. The DHCP server must be on the same subnet as the switch. Refer to “DHCP
activation” on page 69.

Rules for Telnet connections
The following rules must be observed when making Telnet connections to your switch:

• Never change the IP address of the switch while two Telnet sessions are active; if you do, your
next attempt to log in fails. To recover, gain access to the switch by one of these methods:

-

You can use Web Tools to perform a fast boot. When the switch comes up, the Telnet quota
is cleared. (For instructions on performing a fast boot with Web Tools, see the Web Tools
Administrator’s Guide.)

-

If you have the required privileges, you can connect through the serial port, log in as
admin, and use the killTelnet command to identify and kill the Telnet processes without
disrupting the fabric.

• For accounts with an admin role, Fabric OS limits the number of simultaneous Telnet sessions
per switch to two. For more details on session limits, refer to Chapter 6, “Managing User
Accounts”.

Fabric OS Administrator’s Guide
53-1002920-02

59

2

Fabric OS command line interface

Connecting to Fabric OS using Telnet
Use the following procedure to connect to the Fabric OS using Telnet:
1. Connect through a serial port to the switch that is appropriate for your fabric:

• If Virtual Fabrics is enabled, log in using an admin account assigned the chassis-role
permission.

• If Virtual Fabrics is not enabled, log in using an account assigned to the admin role.
2. Verify the switch’s network interface is configured and that it is connected to the IP network
through the RJ-45 Ethernet port.
Switches in the fabric that are not connected through the Ethernet port can be managed
through switches that are using IP over Fibre Channel. The embedded port must have an
assigned IP address.
3. Log off the switch’s serial port.
4. From a management station, open a Telnet connection using the IP address of the switch to
which you want to connect.
The login prompt is displayed when the Telnet connection finds the switch in the network.
5. Enter the account ID at the login prompt.
6. Enter the password.
If you have not changed the system passwords from the default, you are prompted to change
them. Enter the new system passwords, or press Ctrl+C to skip the password prompts. For
more information on system passwords, refer to “Default account passwords” on page 63.
7.

Verify the login was successful.
The prompt displays the switch name and user ID to which you are connected.
login: admin
password: xxxxxxx

Getting help on a command
You can display a list of all command help topics for a given login level. For example, if you log in as
user and enter the help command, a list of all user-level commands that can be executed is
displayed. The same rule applies to the admin, securityAdmin, and the switchAdmin roles.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the help [|more] command with no specific command and all commands are displayed.
The optional |more argument displays the commands one page at a time.
For command-specific information, you can enter help command |more, where command is
the name of the command for which you need specific information.

60

Fabric OS Administrator’s Guide
53-1002920-02

Fabric OS command line interface

2

The commands in the following table provide help files for the indicated specific topics.

TABLE 3

Help topic contents

Topic name

Help contents description

diagHelp

Diagnostic help information

ficonHelp

FICON help information

fwHelp

Fabric Watch help information

iscsiHelp

iSCSI help information

licenseHelp

License help information

perfHelp

Performance Monitoring help information

routeHelp

Routing help information

trackChangesHelp

Track Changes help information

zoneHelp

Zoning help information

Viewing a history of command line entries
The CLI command history log file saves the last 512 commands from all users on a FIFO basis, and
this log is persistent across reboots and firmware downloads. This command is also supported for
standby CPs.
The log records the following information whenever a command ins entered in the switch CLI:

•
•
•
•
•

Timestamp
Username
IP address of the telnet session
Options
Arguments

Use the following procedure to view the CLI command log:
1. Connect to the switch and log in.
2. Enter the cliHistory command with the desired argument (see below for arguments).
Entering no specific argument displays only the command line history of the currently logged-in
user.
cliHistory
Entering the cliHistory command with no arguments displays the command line history for the
currently logged-in user only (even for the root user).
Example cliHistory command output from root login
switch:root> clihistory
CLI history
Date & Time
Thu Sep 27 04:58:00 2012
Thu Sep 27 04:58:19 2012
Thu Sep 27 05:25:45 2012

Fabric OS Administrator’s Guide
53-1002920-02

Message
root, 10.70.12.101, firmwareshow -v
root, 10.70.12.101, telnet 127.1.10.1
root, 10.70.12.101, ipaddrshow]

61

2

Fabric OS command line interface

Example cliHistory command output from admin login
switch:admin> clihistory
CLI history
Date & Time
Thu Sep 27 10:14:41 2012
Thu Sep 27 10:14:48 2012

Message
admin, 10.70.12.101, clihistory
admin, 10.70.12.101, clihistory --show

cliHistory --show
Using the “--show” argument displays the same results as entering “cliHistory” without any
arguments.
cliHistory --showuser 
Using the “--showuser ” argument displays the command line history of the named
user. This argument is available only to Root, Admin, Factory and Securityadmin RBAC roles.
Example cliHistory command output showing username
switch:root> clihistory --showuser admin
CLI history
Date & Time
Message
Thu Sep 27 10:14:41 2012
admin, 10.70.12.101, clihistory
Thu Sep 27 10:14:48 2012
admin, 10.70.12.101, clihistory --show
Thu Sep 27 10:15:00 2012
admin, 10.70.12.101, clihistory

cliHistory --showall
Using the “--showall” argument displays the command line history for all users. With this option,
admin/factory/securityadmin users can see the root user command history.
This argument is available only to Root, Admin, Factory and Securityadmin RBAC roles.
Example cliHistory showing history of all users
switch:admin> clihistory --showall
CLI history
Date & Time
Message
Thu Sep 27 04:58:00 2012
root, 10.70.12.101,
Thu Sep 27 04:58:19 2012
root, 10.70.12.101,
Thu Sep 27 05:25:45 2012
root, 10.70.12.101,
Thu Sep 27 05:25:48 2012
root, 10.70.12.101,

firmwareshow -v
telnet 127.1.10.1
ipaddrshow]
ipaddrshow

cliHistory - -help
Using the “-- help” argument displays a list of the available command arguments.
swd77:admin> clihistory --help
clihistory usage:
clihistory:
Displays the CLI History of the
clihistory --show:
Displays the CLI History of the
clihistory --showuser :
Displays the CLI History of the
clihistory --showall:
Displays the CLI History of all
clihistory --help:
Displays the command usage

62

current user
current user
given user
users

Fabric OS Administrator’s Guide
53-1002920-02

Password modification

2

Notes:
• SSH login CLI logs are not recorded in the command line history.

• The CLI command log will be collected as part of any “supportsave” operation.
The command long record of such an operation will be the equivalent of running
“cliHistory --showall”.

• For CLI commands that require a password (Examples: firmwaredownload,
configupload/download, supportsave, and so on), only the command (no arguments) is stored
(see below for an illustration).
sw0:FID128:root> firmwaredownload -s -p scp 10.70.4.109,fvt,/dist,pray4green
Server IP: 10.70.4.109, Protocol IPv4
Checking system settings for firmwaredownload...
Failed to access scp://fvt:**********@10.70.4.109//dist/release.plist
sw0:FID128:root> clihistory
Date & Time
Wed May 23 03:39:37 2012

Message
root, console, firmwaredownload

Password modification
The switch automatically prompts you to change the default account passwords after logging in for
the first time. If you do not change the passwords, the switch prompts you after each subsequent
login until all the default passwords have been changed.

NOTE
The default account passwords can be changed from their original values only when prompted
immediately following the login; the passwords cannot be changed using the passwd command later
in the session. If you skip the prompt, and then later decide to change the passwords, log out and
then back in.
The default accounts on the switch are admin, user, root, and factory. Use the “admin” account to
log in to the switch for the first time and to perform the basic configuration tasks. The password for
all of these accounts is “password”.
There is only one set of default accounts for the entire chassis. The root and factory default
accounts are reserved for development and manufacturing. The user account is primarily used for
system monitoring. For more information on default accounts, refer to “Default accounts” on
page 156.

Default account passwords
The change default account passwords prompt is a string that begins with the message “Please
change your passwords now”. User-defined passwords can have from 8 through 40 characters.
They must begin with an alphabetic character and can include numeric characters, the period (.),
and the underscore ( _ ). They are case-sensitive, and they are not displayed when you enter them
on the command line.
Record the passwords exactly as entered and store them in a secure place because recovering
passwords requires significant effort and fabric downtime. Although the root and factory accounts
are not meant for general use, change their passwords if prompted to do so and save the
passwords in case they are needed for recovery purposes.

Fabric OS Administrator’s Guide
53-1002920-02

63

2

The switch Ethernet interface

Changing the default account passwords at login
Use the following procedure to change the default account passwords:
1. Connect to the switch and log in using the default administrative account.
2. At each of the “Enter new password” prompts, either enter a new password or skip the prompt.
To skip a single prompt, press Enter. To skip all of the remaining prompts, press Ctrl-C.
Example output of changing passwords
login: admin
Password:
Please change your passwords now.
Use Control-C to exit or press 'Enter' key to proceed.
for user - root
Changing password for root
Enter new password: 
Password changed.
Saving password to stable storage.
Password saved to stable storage successfully.
(output truncated)

The switch Ethernet interface
The Ethernet (network) interface provides management access, including direct access to the
Fabric OS CLI, and allows other tools, such as Web Tools, to interact with the switch. You can use
either Dynamic Host Configuration Protocol (DHCP) or static IP addresses for the Ethernet network
interface configuration.

Brocade Backbones
On Brocade Backbones, you must set IP addresses for the following components:

• Both Control Processors (CP0 and CP1)
• Chassis management IP

Brocade switches
On Brocade switches, you must set the Ethernet and chassis management IP interfaces.
Setting the chassis management IP address eliminates the need to know which CP is active and
automatically connects the requestor to the currently active CP.
You can continue to use a static Ethernet addressing system or allow the DHCP client to
automatically acquire Ethernet addresses. Configure the Ethernet interface IP address, subnet
mask, and gateway addresses in one of the following manners:

• Using static Ethernet addresses (refer to “Static Ethernet addresses” on page 67)
• Activating DHCP (refer to “DHCP activation” on page 69)

64

Fabric OS Administrator’s Guide
53-1002920-02

The switch Ethernet interface

2

NOTE

When you change the Ethernet interface settings, open connections such as SSH or Telnet may be
dropped. Reconnect using the new Ethernet IP address information or change the Ethernet settings
using a console session through the serial port to maintain your session during the change. You
must connect through the serial port to set the Ethernet IP address if the Ethernet network interface
is not configured already. For details, refer to “Connecting to Fabric OS through the serial port” on
page 58.

Virtual Fabrics and the Ethernet interface
On the Brocade DCX and DCX-4S, the single-chassis IP address and subnet mask are assigned to
the management Ethernet ports on the front panels of the CPs. These addresses allow access to
the chassis—more specifically, the active CP of the chassis—and not individual logical switches. The
IP addresses can also be assigned to each CP individually. This allows for direct communication
with a CP, including the standby CP. On the Brocade DCX and DCX-4S Backbones, each CP has two
management Ethernet ports on its front panel. These two physical ports are bonded together to
create a single, logical Ethernet port, and it is the logical Ethernet port to which IP addresses are
assigned.
IPv4 addresses assigned to individual Virtual Fabrics are assigned to IP over Fibre Channel (IPFC)
network interfaces. In Virtual Fabrics environments, a single chassis can be assigned to multiple
fabrics, each of which is logically distinct and separate from one another. Each IPFC point of
connection to a given chassis needs a separate IPv4 address and prefix to be accessible to a
management host. For more information on how to set up these IPFC interfaces to your Virtual
Fabric, refer to Chapter 11, “Managing Virtual Fabrics”.

Management Ethernet port bonding
The two external Ethernet ports of a CP8 blade can be bound together as a single logical network
interface. This configuration uses an active-standby failover model to provide automatic failover
support for the primary Ethernet port on the blade. If the primary Ethernet port fails (due to
something other than power loss), the second Ethernet port immediately takes over to ensure link
layer communication is retained.
One of the physical Ethernet ports is selected as the active interface. The second interface is set as
the standby interface. All traffic is transmitted over the active interface. No traffic is transmitted
over the standby interface, unless the active interface is determined to be no longer connected; at
which point, the second interface is made active.
When active, all the Fabric OS kernel modules and applications on the CP8 blade will use the
logical network interface named “bond0” instead of “eth0”.

NOTE

On bootup, physical port eth0 is always made active if it is connected.
The CP8 blade contains multiple Ethernet devices (including eth0 and eth3), which map to the two
Ethernet ports on the front of the CP8 blade. Other Ethernet devices on the blade are reserved for
use by the operating system.

Fabric OS Administrator’s Guide
53-1002920-02

65

2

The switch Ethernet interface

The CP8 blade enables eth0 by default. If an error is encountered on eth0, it is treated the same as
for any other port, unless the error causes the eth0 port to go down. If eth0 goes down, the eth3
interface becomes active and will remain active even if eth0 comes back up. Use one of the
following actions to restore eth0 as the active interface.

• Unplug the network cable, wait 5 seconds, and then plug it back in.
• Perform an HA failover routine.
• Power down the switch and then power it back up again.
ATTENTION
Performing and HA failover and powering down the switch will cause a disruptive delay in content
delivery.

Supported devices
Management Ethernet port bonding is available on a CP8 blade when it is installed on a
Brocade DCX, Brocade DCX-4S, Brocade DCX 8510-8, or Brocade DCX 8510-4.

Setting up the second Ethernet port on a CP8 blade
The port speed and duplex mode between the Ethernet ports should always match. Both ports
should be set at a fixed speed or set to autonegotiate.
1. Make sure that the speed and link operating mode settings are the same for both eth3 and
eth0. Refer to “Setting port modes” on page 93 for instructions on setting port modes, and
“Setting port speeds” on page 94 for instructions on setting port speeds.
2. Physically connect the second Ethernet port to the same network as the primary Ethernet port.

Displaying the network interface settings
If an IP address has not been assigned to the network interface (Ethernet), you must connect to the
Fabric OS CLI using a console session on the serial port. For more information, see “Console
sessions using the serial port” on page 58. Otherwise, connect using SSH.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the ipAddrShow command.
ipAddrShow

Example output for a Brocade Backbone
ecp:admin> ipaddrshow
SWITCH
Ethernet IP Address: 10.1.2.3
Ethernet Subnetmask: 255.255.240.0
CP0
Ethernet IP Address: 10.1.2.3
Ethernet Subnetmask: 255.255.240.0
Host Name: ecp0
Gateway IP Address: 10.1.2.1
CP1
Ethernet IP Address: 10.1.2.4
Ethernet Subnetmask: 255.255.240.0

66

Fabric OS Administrator’s Guide
53-1002920-02

The switch Ethernet interface

2

Host Name: ecp1
Gateway IP Address: 10.1.2.3
IPFC address for virtual fabric ID 123: 11.1.2.3/24
IPFC address for virtual fabric ID 45: 13.1.2.4/20
Slot 7
eth0: 11.1.2.4/24
Gateway: 11.1.2.1
Backplane IP address of CP0 : 10.0.0.5
Backplane IP address of CP1 : 10.0.0.6
IPv6 Autoconfiguration Enabled: Yes
Local IPv6 Addresses:
sw 0 stateless fd00:60:69bc:70:260:69ff:fe00:2/64 preferred
sw 0 stateless fec0:60:69bc:70:260:69ff:fe00:2/64 preferred
cp 0 stateless fd00:60:69bc:70:260:69ff:fe00:197/64 preferred
cp 0 stateless fec0:60:69bc:70:260:69ff:fe00:197/64 preferred
cp 1 stateless fd00:60:69bc:70:260:69ff:fe00:196/64 preferred
cp 1 stateless fec0:60:69bc:70:260:69ff:fe00:196/64 preferred
IPv6 Gateways:
cp 0 fe80:60:69bc:70::3
cp 0 fe80:60:69bc:70::2
cp 0 fe80:60:69bc:70::1
cp 1 fe80:60:69bc:70::3

If the Ethernet IP address, subnet mask, and gateway address are displayed, then the network
interface is configured. Verify the information on your switch is correct. If DHCP is enabled,
the network interface information was acquired from the DHCP server.

NOTE

You can use either IPv4 or IPv6 with a classless inter-domain routing (CIDR) block notation (also
known as a network prefix length) to set up your IP addresses.

Static Ethernet addresses
Use static Ethernet network interface addresses on Brocade DCX and DCX-4S Backbones, and in
environments where DHCP service is not available. To use static addresses for the Ethernet
interface, you must first disable DHCP. You can enter static Ethernet information and disable DHCP
at the same time. For more information, refer to “DHCP activation” on page 69.
If you choose not to use DHCP or to specify an IP address for your switch Ethernet interface, you
can do so by entering “none” or “0.0.0.0” in the Ethernet IP address field.
On an application blade, configure the two external Ethernet interfaces to two different subnets.
If two subnets are not present, configure one of the interfaces and leave the other unconfigured.
Otherwise, the following message displays and blade status may go into a faulty state after a
reboot.
Neighbor table overflow.
print: 54 messages suppressed

Fabric OS Administrator’s Guide
53-1002920-02

67

2

The switch Ethernet interface

Setting the static addresses for the Ethernet network interface
Use the following procedure to set the Ethernet network interface static addresses:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Perform the appropriate action based on whether you have a switch or Backbone:

• If you are setting the IP address for a switch, enter the ipAddrSet command.
• If you are setting the IP address for a Backbone, enter the ipAddrSet command specifying
either CP0 or CP1. You must set the IP address for both CP0 and CP1.
Example of setting an IPv4 address
switch:admin> ipaddrset
Ethernet IP Address [10.1.2.3]:
Ethernet Subnetmask [255.255.255.0]:
Fibre Channel IP Address [220.220.220.2]:
Fibre Channel Subnetmask [255.255.0.0]:
Gateway IP Address [10.1.2.1]:
DHCP [OFF]: off

Example of setting an IPv6 address on a switch
switch:admin> ipaddrset -ipv6 --add 1080::8:800:200C:417A/64
IP address is being changed...Done.

For more information on setting up an IP address for a Virtual Fabric, refer to Chapter 11,
“Managing Virtual Fabrics”.
3. Enter the network information in dotted-decimal notation for the Ethernet IPv4 address or in
semicolon-separated notation for IPv6.
4. Enter the Ethernet Subnetmask at the prompt.
5. The Fibre Channel prompts are not relevant; you can skip them by pressing Enter.
The Fibre Channel IP address is used for management.
6. Enter the Gateway Address at the prompt.
7.

Disable DHCP by entering off.

Setting the static addresses for the chassis management IP interface
Use the following procedure to set the chassis management IP interface static addresses:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the ipAddrSet -chassis command.
switch:admin> ipaddrset -chassis
Ethernet IP Address [192.168.166.148]:
Ethernet Subnetmask [255.255.255.0]:
Committing configuration...Done.

3. Enter the network information in dotted-decimal notation for the Ethernet IPv4 address or in
semicolon-separated notation for IPv6.
4. Enter the Ethernet Subnet mask at the prompt.

68

Fabric OS Administrator’s Guide
53-1002920-02

The switch Ethernet interface

2

DHCP activation
Some Brocade switches have DHCP enabled by default. Fabric OS support for DHCP functionality is
only provided for Brocade fixed-port switches. These are listed in the Preface.

NOTE
The Brocade DCX and Brocade DCX-4S Backbones do not support DHCP.
The Fabric OS DHCP client supports the following parameters:

• External Ethernet port IP addresses and subnet masks
• Default gateway IP address
The DHCP client uses a DHCP vendor-class identifier that allows DHCP servers to determine that
the discover/request packet are coming from a Brocade switch. The vendor-class identifier is the
string “BROCADE” followed by the SWBD model number of the platform. For example, the
vendor-class identifier for a request from a Brocade 5300 is “BROCADESWBD64.”

NOTE
The client conforms to the latest IETF Draft Standard RFCs for IPv4, IPv6, and DHCP. DHCP can
obtain stateful IPv6 addresses.

Enabling DHCP for IPv4
When you connect a DHCP-enabled switch to the network and power on the switch, the switch
automatically obtains the Ethernet IP address, Ethernet subnet mask, and default gateway address
from the DHCP server.

NOTE
The DHCP client can only connect to a DHCP server on the same subnet as the switch. Do not enable
DHCP if the DHCP server is not on the same subnet as the switch.
Enabling DHCP after the Ethernet information has been configured releases the current Ethernet
network interface settings. These include the Ethernet IP address, Ethernet subnet mask, and
gateway IP address. The Fibre Channel IP address and subnet mask are static and are not affected
by DHCP; for instructions on setting the FC IP address, see “Static Ethernet addresses” on page 67.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the ipAddrSet command.
ipaddrset

NOTE

Alternatively, you can enable DHCP for IPv4 by entering “ipaddrset –ipv4 -add -dhcp ON“as
a single command. If you do so, you do not need to complete the following steps.
3. If already set up, you can skip the Ethernet IP address, Ethernet subnet mask, Fibre Channel IP
address, and Fibre Channel subnet mask prompts by pressing Enter.
Otherwise, enter the network information in dotted-decimal notation for the IPv4 address.
4. Enable DHCP by entering on.
5. You can confirm that the change has been made using the ipAddrShow command.

Fabric OS Administrator’s Guide
53-1002920-02

69

2

The switch Ethernet interface

Example of enabling DHCP for IPv4 interactively:
switch:admin> ipaddrset
Ethernet IP Address [10.1.2.3]:
Ethernet Subnetmask [255.255.255.0]:
Fibre Channel IP Address [220.220.220.2]:
Fibre Channel Subnetmask [255.255.0.0]:
Gateway IP Address [10.1.2.1]:
DHCP [Off]:on

Example of enabling DHCP for IPv4 using a single command:
switch:admin> ipaddrset –ipv4 -add -dhcp ON
switch:admin> ipaddrshow
SWITCH
Ethernet IP Address: 10.20.134.219
Ethernet Subnetmask: 255.255.240.0
Gateway IP Address: 10.20.128.1
DHCP: On

Disabling DHCP for IPv4
When you disable DHCP, enter the static Ethernet IP address and subnet mask of the switch and
default gateway address. Otherwise, the Ethernet settings may conflict with other addresses
assigned by the DHCP server on the network.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the ipAddrSet command.
ipaddrset

NOTE

Alternatively, you can disable DHCP for IPv4 by entering “ipaddrset –
ipv4 -add -dhcp OFF“as a single command. If you do so, you do not need to complete the
following steps.
3. Enter the network information using IPv4 dotted-decimal notation.

NOTE

If a static Ethernet address is not available when you disable DHCP, enter 0.0.0.0 at the
Ethernet IP address prompt.
4. You can skip the Fibre Channel prompts by pressing Enter.
5. When you are prompted for DHCP[On], disable it by entering off.
6. You can confirm that the change has been made using the ipAddrShow command.
Example of disabling DHCP for IPv4 interactively:
switch:admin> ipaddrset
Ethernet IP Address [10.1.2.3]:
Ethernet Subnetmask [255.255.255.0]:
Gateway IP Address [10.1.2.1]:
DHCP [On]:off

Example of disabling DHCP for IPv4 using a single command:
switch:admin> ipaddrset –ipv4 -add -dhcp OFF
switch:admin> ipaddrshow

70

Fabric OS Administrator’s Guide
53-1002920-02

The switch Ethernet interface

2

SWITCH
Ethernet IP Address: 10.20.134.219
Ethernet Subnetmask: 255.255.240.0
Gateway IP Address: 10.20.128.1
DHCP: Off

IPv6 autoconfiguration
IPv6 can assign multiple IP addresses to each network interface. Each interface is configured with
a link local address in almost all cases, but this address is only accessible from other hosts on the
same network. To provide for wider accessibility, interfaces are typically configured with at least
one additional global scope IPv6 address. IPv6 autoconfiguration allows more IPv6 addresses, the
number of which is dependent on the number of routers serving the local network and the number
of prefixes they advertise.
There are two methods of autoconfiguration for IPv6 addresses: stateless autoconfiguration and
stateful autoconfiguration. Stateless allows an IPv6 host to obtain a unique address using the
IEEE 802 MAC address; stateful uses a DHCPv6 server, which keeps a record of the IP address and
other configuration information for the host. Whether a host engages in autoconfiguration and
which method it uses is dictated by the routers serving the local network, not by a configuration of
the host. There can be multiple routers serving the network, each potentially advertising multiple
network prefixes. Thus, the host is not in full control of the number of IPv6 addresses that it
configures, much less the values of those addresses, and the number and values of addresses can
change as routers are added to or removed from the network.
When IPv6 autoconfiguration is enabled, the platform engages in stateless IPv6 autoconfiguration.
When IPv6 autoconfiguration is disabled, the platform relinquishes usage of any autoconfigured
IPv6 addresses that it may have acquired while it was enabled. This same enable or disable state
also enables or disables the usage of a link local address for each managed entity, though a link
local address continues to be generated for each nonchassis-based platform and for each CP of a
chassis-based platform because those link local addresses are required for router discovery. The
enabled or disabled state of autoconfiguration is independent of whether any static IPv6 addresses
have been configured.

Setting IPv6 autoconfiguration
Use the following procedure to enable IPv6 autoconfiguration:
1. Connect to the switch and log in using an account with admin permissions.
2. Take the appropriate following action based on whether you want to enable or disable IPv6
autoconfiguration:

• Enter the ipAddrSet -ipv6 -auto command to enable IPv6 autoconfiguration for all
managed entities on the target platform.

• Enter the ipAddrSet -ipv6 -noauto command to disable IPv6 autoconfiguration for all
managed entities on the target platform.

Fabric OS Administrator’s Guide
53-1002920-02

71

2

Date and time settings

Date and time settings
Switches maintain the current date and time inside a battery-backed real-time clock (RTC) circuit
that receives the date and time from the fabric’s principal switch. Date and time are used for
logging events. Switch operation does not depend on the date and time; a switch with an incorrect
date and time value functions properly. However, because the date and time are used for logging,
error detection, and troubleshooting, you must set them correctly.
In a Virtual Fabric, there can be a maximum of eight logical switches per Backbone. Only the
default switch in the chassis can update the hardware clock. When the date command is issued
from a non-principal pre-Fabric OS v6.2.0 or earlier switch, the date command request is dropped
by a Fabric OS v6.2.0 and later switch and the pre-Fabric OS v6.2.0 switch or earlier does not
receive an error.
Authorization access to set or change the date and time for a switch is role-based. For an
understanding of role-based access, refer to “Role-Based Access Control” on page 152.

Setting the date and time
Use the following procedure to set the device date and time:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the date command, using the following syntax:
date "mmddHHMMyy"

The values represent the following:

•
•
•
•
•

mm is the month; valid values are 01 through 12.
dd is the date; valid values are 01 through 31.
HH is the hour; valid values are 00 through 23.
MM is minutes; valid values are 00 through 59.
yy is the year, valid values are 00 through 37 and 70 through 99 (year values from 70
through 99 are interpreted as 1970 through 1999, year values from 00 through 37 are
interpreted as 2000 through 2037).

Example of showing and setting the date
switch:admin> date
Fri Sep 29 17:01:48 UTC 2007
Stealth200E:admin> date "0204101008"
Mon Feb 4 10:10:00 UTC 2008

Time zone settings
You can set the time zone for a switch by name. You can specify the setting using country and city
or time zone parameters. Switch operation does not depend on a date and time setting. However,
having an accurate time setting is needed for accurate logging and audit tracking.
If the time zone is not set with new options, the switch retains the offset time zone settings. The
tsTimeZone command includes an option to revert to the prior time zone format. For more
information about the tsTimeZone command, refer to the Fabric OS Command Reference.

72

Fabric OS Administrator’s Guide
53-1002920-02

Date and time settings

2

When you set the time zone for a switch, you can perform the following tasks:

• Display all of the time zones supported in the firmware.
• Set the time zone based on a country and city combination or based on a time zone ID,
such as PST.
The time zone setting has the following characteristics:

• Users can view the time zone settings. However, only those with administrative
permissions can set the time zones.

• The setting automatically adjusts for Daylight Savings Time.
• Changing the time zone on a switch updates the local time zone setup and is reflected in
local time calculations.

• By default, all switches are set to Greenwich Mean Time (0,0). If all switches in a fabric are
in one time zone, it is possible for you to keep the time zone setup at the default setting.

• System services that have already started reflect the time zone changes after the next
reboot.

• Time zone settings persist across failover for high availability.
• Setting the time zone on any dual domain Backbone has the following characteristics:
• Updating the time zone on any switch updates the entire Backbone.
• The time zone of the entire Backbone is the time zone of switch 0.

Setting the time zone
The following procedure describes how to set the time zone for a switch. You must perform the
procedure on all switches for which the time zone must be set. However, you only need to set the
time zone once on each switch because the value is written to nonvolatile memory.
1. Connect to the switch and log in using an account assigned to the admin role and with the
chassis-role permission.
2. Enter the tsTimeZone command.

• Use tsTimeZone with no parameters to display the current time zone setting.
• Use --interactive to list all of the time zones supported by the firmware.
• Use timeZone_fmt to set the time zone by Country/City or by time zone ID, such as Pacific
Standard Time (PST).
Example of displaying and changing the time zone to US/Central
switch:admin> tstimezone
Time Zone : US/Pacific
switch:admin> tstimezone US/Central
switch:admin> tstimezone
Time Zone : US/Central

Setting the time zone interactively
Use the following procedure to set the current time zone to PST using interactive mode:
1. Connect to the switch and log in using an account assigned to the admin role and with the
chassis-role permission.
2. Enter the tsTimeZone --interactive command.

Fabric OS Administrator’s Guide
53-1002920-02

73

2

Date and time settings

You are prompted to select a general location.
Please identify a location so that time zone rules can be set correctly.

3. Enter the appropriate number or press Ctrl-D to quit.
4. Select a country location at the prompt.
5. Enter the appropriate number at the prompt to specify the time zone region of Ctrl-D to quit.

Network time protocol
You can synchronize the local time of the principal and primary FCS switch to a maximum of eight
external Network Time Protocol (NTP) servers. To keep the time in your SAN current, it is
recommended that the principal or primary FCS switch has its time synchronized with at least one
external NTP server. The other switches in the fabric automatically take their time from the principal
or primary FCS switch, as described in “Synchronizing the local time with an external source.”
All switches in the fabric maintain the current clock server value in nonvolatile memory. By default,
this value is the local clock server (LOCL) of the principal or primary FCS switch. Changes to the
clock server value on the principal or primary FCS switch are propagated to all switches in the
fabric.
If Virtual Fabrics is enabled, all the switches in the fabric must have the same NTP clock server
configured. This includes any Fabric OS v6.2.0 or earlier switches in the fabric. This ensures that
time does not go out of sync in the logical fabric. It is not recommended to have LOCL in the server
list.
When a new switch enters the fabric, the time server daemon of the principal or primary FCS switch
sends out the addresses of all existing clock servers and the time to the new switch. When a switch
enters the fabric, it stores the list and the active servers.

NOTE
If Virtual Fabrics is enabled, multiple logical switches can share a single chassis. Therefore, the NTP
server list must be the same across all fabrics.

Synchronizing the local time with an external source
The tsClockServer command accepts multiple server addresses in IPv4, IPv6, or Domain Name
System (DNS) name formats. When multiple NTP server addresses are passed, tsClockServer sets
the first obtainable address as the active NTP server. The rest are stored as backup servers that
can take over if the active NTP server fails. The principal or primary FCS switch synchronizes its
time with the NTP server every 64 seconds.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the tsClockServer command.
switch:admin> tsclockserver "ntp1;ntp2"

In this syntax, ntp1 is the IP address or DNS name of the first NTP server, which the switch
must be able to access. The second variable, ntp2, is the second NTP server and is optional.
The operand “ntp1;ntp2” is optional; by default, this value is LOCL, which uses the local clock
of the principal or primary FCS switch as the clock server.

74

Fabric OS Administrator’s Guide
53-1002920-02

Domain IDs

2

Example of setting the NTP server
switch:admin> tsclockserver
LOCL
switch:admin> tsclockserver "10.1.2.3"

Example of displaying the NTP server
switch:admin> tsclockserver
10.1.2.3

Example of setting up more than one NTP server using a DNS name
switch:admin> tsclockserver "10.1.2.4;10.1.2.5;ntp.localdomain.net"
Updating Clock Server configuration...done.
Updated with the NTP servers

Changes to the clock server value on the principal or primary FCS switch are propagated to all
switches in the fabric.

Domain IDs
Although domain IDs are assigned dynamically when a switch is enabled, you can change them
manually so that you can control the ID number or resolve a domain ID conflict when you merge
fabrics.
If a switch has a domain ID when it is enabled, and that domain ID conflicts with another switch in
the fabric, the conflict is automatically resolved if the other switch’s domain ID is not persistently
set. The process can take several seconds, during which time traffic is delayed. If both switches
have their domain IDs persistently set, one of them needs to have its domain ID changed to a
domain ID not used within the fabric.
The default domain ID for Brocade switches is 1.

Domain ID issues
Keep the following restrictions in mind when working with domain IDs.

• Do not use domain ID 0. Using this domain ID can cause the switch to reboot continuously.
• Avoid changing the domain ID on the FCS switch in secure mode.
• To minimize downtime, change the domain IDs on the other switches in the fabric.

Displaying the domain IDs
Use the following procedure to display device domain IDs:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the fabricShow command.
Example output of fabric information, including the domain ID (D_ID)

The principal switch is determined by the arrow ( > ) next to the name of the switch. In this
output, the principal switch appears in blue boldface.
switch:admin> fabricshow
Switch ID
Worldwide Name

Fabric OS Administrator’s Guide
53-1002920-02

Enet IP Addr

FC IP Addr

Name

75

2

Switch names

------------------------------------------------------------------------2: fffc02
10:00:00:60:69:e0:01:46
10.3.220.1
0.0.0.0
"ras001"
3: fffc03
10:00:00:60:69:e0:01:47
10.3.220.2
0.0.0.0
"ras002"
5: fffc05
10:00:00:05:1e:34:01:bd
10.3.220.5
0.0.0.0
"ras005"
fec0:60:69bc:63:205:1eff:fe34:1bd
6: fffc06
10:00:00:05:1e:34:02:3e
10.3.220.6
0.0.0.0
>"ras006"
7: fffc07
10:00:00:05:1e:34:02:0c
10.3.220.7
0.0.0.0
"ras007"
(output truncated)
The Fabric has 26 switches

Table 4 displays the fabricShow fields.

TABLE 4

fabricShow fields

Field

Description

Switch ID

The switch domain_ID and embedded port D_ID. The numbers are broken down as follows:
Example 64: fffc40
64 is the switch domain_ID
fffc40 is the hexadecimal format of the embedded port D_ID.

World Wide Name The switch WWN.
Enet IP Addr

The switch Ethernet IP address for IPv4- and IPv6-configured switches. For IPv6 switches, only
the static IP address displays.

FC IP Addr

The switch Fibre Channel IP address.

Name

The switch symbolic or user-created name in quotes.

Setting the domain ID
Use the following procedure to set the domain ID:
1. Connect to the switch and log in on an account assigned to the admin role.
2. Enter the switchDisable command to disable the switch.
3. Enter the configure command.
4. Enter y after the Fabric Parameters prompt.
Fabric parameters (yes, y, no, n): [no] y

5. Enter a unique domain ID at the Domain prompt. Use a domain ID value from 1 through 239
for normal operating mode (FCSW-compatible).
Domain: (1..239) [1] 3

6. Respond to the remaining prompts, or press Ctrl-D to accept the other settings and exit.
7.

Enter the switchEnable command to re-enable the switch.

Switch names
Switches can be identified by IP address, domain ID, World Wide Name (WWN), or by customized
switch names that are unique and meaningful.

76

Fabric OS Administrator’s Guide
53-1002920-02

Chassis names

2

The following considerations apply to switch naming:

• Switch names can be from 1 through 30 characters long.
• All switch names must begin with a letter, and can contain letters, numbers, or the underscore
character.

• Switch names must be unique across logical switches.
• Changing the switch name causes a domain address format RSCN to be issued and may be
disruptive to the fabric.

Customizing the switch name
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the switchName command and enter a new name for the switch.
switch:admin> switchname newname

The prompt does not change to the new switch name until AFTER you re-login.
3. Record the new switch name for future reference.
switch:FID128:# admin> switchname myswitch
Committing configuration...
Done.
Switch name has been changed.Please re-login into the switch for the change to be
applied.
switch:FID128:# admin>

Chassis names
Brocade recommends that you customize the chassis name for each platform. Some system logs
identify devices by platform names; if you assign meaningful platform names, logs are more useful.
All chassis names supported by Fabric OS v7.0.0 and later allow 31 characters. Chassis names
must begin with an alphabetic character and can include alphabetic and numeric characters, and
the underscore ( _ ).

Customizing chassis names
Use the following procedure to customize the chassis name:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the chassisName command.
ecp:admin> chassisname newname

3. Record the new chassis name for future reference.

Fabric name
You can assign a alphanumeric name to identify and manage a logical fabric that formerly could
only be identified by a fabric ID. The fabric name does not replace the fabric ID or its usage.
The fabric continues to have a fabric ID, in addition to the assigned alphanumeric fabric name.

Fabric OS Administrator’s Guide
53-1002920-02

77

2

Switch activation and deactivation

The following considerations apply to fabric naming:

• Each name must be unique for each logical switch within a chassis; duplicate fabric names are
not allowed.

• A fabric name can be from 1 through 128 alphanumeric characters.
• All switches in a logical fabric must be running Fabric OS v7.2.0. Switches running earlier
versions of the firmware can co-exist in the fabric, but do not show the fabric name details.

• You must have admin permissions to configure the fabric name.

Configuring the fabric name
To set and display the fabric name, use the fabricName command as shown here:
switch:user> fabricname --set myfabric@1

Using the fabricName --set command without a fabric name takes the existing fabric name and
synchronizes it across the entire fabric. An error message displays if no name is configured.
To set a fabric name that includes spaces, enclose the fabric name in quotes, as shown here:
switch:user> fabricname --set "my new fabric"

To set a fabric name that includes bash special meta-characters or spaces, use the command
fabricName as shown in the following example:
switch:user> fabricname --set 'red fabric $$'

To clear the fabric name, use the fabricName --clear command.

High availability considerations for fabric names
Fabric names locally configured or obtained from a remote switch are saved in the configuration
database, and then synchronized to the standby CP on dual-CP-based systems.

Upgrade and downgrade considerations for fabric names
Fabric names are lost during a firmware downgrade. No default fabric name is provided. If a fabric
name is needed, it must be configured after the upgrade.

Switch activation and deactivation
By default, the switch is enabled after power is applied and diagnostics and switch initialization
routines have finished. You can disable and re-enable the switch as necessary.
When you enable or disable a switch, the affected ports depend on whether Virtual Fabrics is
enabled. Table 5 describes which ports are affected for each type of enable or disable operation.

78

Fabric OS Administrator’s Guide
53-1002920-02

Switch activation and deactivation

TABLE 5

2

Ports affected when you enable or disable a switch in VF or non-VF mode

Operation

Virtual Fabrics enabled

Virtual Fabrics not enabled

Enable switch

Enables all ports on logical switch

Enables all ports on physical chassis

Enable chassis

Enables all ports on physical chassis

Not allowed

Disable switch

Disables all ports on logical switch

Disables all ports on physical chassis

Disable chassis

Disables all ports on physical chassis

Not allowed

Disabling a switch
You must disable a switch before making configuration changes or before running offline
diagnostic tests.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the switchDisable command.
switch:admin> switchdisable

All Fibre Channel ports on the switch are taken offline. If the switch is part of a fabric, the fabric
is reconfigured.
If Virtual Fabrics is enabled, only the ports allocated to the logical switch are disabled. To
disable all of the ports, you must disable the entire chassis. See “Disabling a chassis” on
page 79.

Enabling a switch
The switch is enabled by default after it is powered on and switch initialization routines have
finished. You must re-enable the switch after making configuration changes or running offline
diagnostics.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the switchEnable command.
switch:admin> switchenable

All Fibre Channel ports that passed Power On Self Test (POST) are enabled. If the switch has
inter-switch links (ISLs) to a fabric, it joins the fabric.
If Virtual Fabrics is enabled, only the ports allocated to the logical switch are enabled. To
enable all of the ports, you must enable the entire chassis. See “Enabling a chassis”.

Disabling a chassis
Disabling a chassis disables all Fibre Channel ports on all logical switches in the chassis. You must
disable a chassis before making chassis-wide configuration changes or before running offline
diagnostic tests.
1. Connect to any logical switch in the chassis and log in using an account assigned to the admin
role.
2. Enter the chassisDisable command.

Fabric OS Administrator’s Guide
53-1002920-02

79

2

Switch and Backbone shutdown

switch:FID128:admin> chassisdisable
This command can cause disruption to multiple logical switches.
Are you sure you want to disable all chassis ports now? (yes, y, no, n): [no]y
switch:FID128:admin>

All Fibre Channel ports on all logical switches are taken offline. If the logical switches are in
fabrics, the fabrics are reconfigured.

NOTE
After a chassisDisable, if you want to do an haFailover, you should wait at least 30 seconds.

Enabling a chassis
Enabling a chassis enables all Fibre Channel ports on all logical switches in the chassis. The
chassis is enabled by default after it is powered on and switch initialization routines have finished.
You must re-enable the chassis after making fabric-wide configuration changes or running offline
diagnostics.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the chassisEnable command.
switch:FID128:admin> chassisenable

For all logical switches in the chassis, all Fibre Channel ports that passed Power On Self Test
(POST) are enabled. If any of the logical switches have inter-switch links (ISLs) to a fabric, it joins
the fabric.

Switch and Backbone shutdown
To avoid corrupting your file system, you must perform graceful shutdowns of Brocade switches and
Backbones.
Warm reboot (also known as graceful shutdown) refers to shutting down the switch or platform by
way of the following instructions. Cold boot (also known as a hard boot) refers to shutting down the
switch or platform by suddenly shutting down power and powering on again.

Powering off a Brocade switch
Use the following procedure to gracefully shut down a Brocade switch.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the sysShutdown command.
3. Enter y at the prompt.
switch:admin> sysshutdown
This command will shutdown the operating systems on your switch.
You are required to power-cycle the switch in order to restore operation.
Are you sure you want to shutdown the switch [y/n]?y

4. Wait until the following message displays:

80

Fabric OS Administrator’s Guide
53-1002920-02

Basic connections

2

Broadcast message from root (ttyS0) Wed Jan 25 16:12:09 2006...
The system is going down for system halt NOW !!
INIT: Switching to runlevel: 0
INIT: Sending processes the TERM signal
Unmounting all filesystems.
The system is halted
flushing ide devices: hda
Power down.

5. Power off the switch.

Powering off a Brocade Backbone
Use the following procedure to power off a Brocade Backbone device:
1. From the active CP in a dual-CP platform, enter the sysShutdown command.

NOTE

When the sysShutdown command is issued on the active CP, the active CP, the standby CP, and
any application blades are all shut down.
2. Enter y at the prompt.
3. Wait until the following message displays:
DCX:FID128:admin> sysshutdown
This command will shutdown the operating systems on your switch.
You are required to power-cycle the switch in order to restore operation.
Are you sure you want to shutdown the switch [y/n]?y
HA is disabled
Stopping blade 10
Shutting down the blade....
Stopping blade 12
Shutting down the blade....
Broadcast message from root (pts/0) Fri Oct 10 08:36:48 2008...
The system is going down for system halt NOW !!

4. Power off the switch.

Basic connections
Before connecting a switch to a fabric that contains switches running different firmware versions,
you must first set the same port identification (PID) format on all switches. The presence of
different PID formats in a fabric causes fabric segmentation.

• For information on PID formats and related procedures, refer to Chapter 3, “Performing
Advanced Configuration Tasks”.

• For information on configuring the routing of connections, refer to Chapter 4, “Routing Traffic”.
• For information on configuring extended inter-switch connections, refer to Chapter 25,
“Managing Long-Distance Fabrics”.

Fabric OS Administrator’s Guide
53-1002920-02

81

2

Basic connections

Device connection
To minimize port logins, power off all devices before connecting them to the switch. When powering
the devices back on, wait for each device to complete the fabric login before powering on the next
one.
For devices that cannot be powered off, first use the portDisable command to disable the port on
the switch, connect the device, and then use the portEnable command to enable the port.

Switch connection
See the hardware reference manual of your specific switch for ISL connection and cable
management information. The standard or default ISL mode is L0. ISL mode L0 is a static mode,
with the following maximum ISL distances:

•
•
•
•
•
•

10 km at 1 Gbps
5 km at 2 Gbps
2.5 km at 4 Gbps
1 km at 8 Gbps
1 km at 10 Gbps
1 km at 16 Gbps

For more information on extended ISL modes, which enable long distance inter-switch links, refer to
Chapter 25, “Managing Long-Distance Fabrics”.

82

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

Performing Advanced Configuration Tasks

3

In this chapter
• Port identifiers (PIDs) and PID binding overview. . . . . . . . . . . . . . . . . . . . . . 83
• Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
• Blade terminology and compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
• Enabling and disabling blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
• Blade swapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
• Disabling switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
• Power management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
• Equipment status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
• Audit log configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
• Duplicate PWWN handling during device login . . . . . . . . . . . . . . . . . . . . . . 110
• Enabling forward error correction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Port identifiers (PIDs) and PID binding overview
Port identifiers (PIDs, also called Fabric Addresses) are used by the routing and zoning services in
Fibre Channel fabrics to identify ports in the network. All devices in a fabric must use the same PID
format. When you add new equipment to the SAN, you may need to change the PID format on
legacy equipment.
Many scenarios cause a device to receive a new PID; for example, unplugging the device from one
port and plugging it into a different port as part of fabric maintenance, or changing the domain ID
of a switch, which might be necessary when merging fabrics, or changing compatibility mode
settings.
Some device drivers use the PID to map logical disk drives to physical Fibre Channel counterparts.
Most drivers can either change PID mappings dynamically, also called dynamic PID binding, or use
the WWN of the Fibre Channel disk for mapping, also called WWN binding.
Some older device drivers behave as if a PID uniquely identifies a device; they use static PID
binding. These device drivers should be updated, if possible, to use WWN binding or dynamic PID
binding instead, because static PID binding creates problems in many routine maintenance
scenarios. Fortunately, very few device drivers still behave this way. Many current device drivers
enable you to select static PID binding as well as WWN binding. You should only select static PID
binding if there is a compelling reason, and only after you have evaluated the effect of doing so.

Fabric OS Administrator’s Guide
53-1002920-02

83

3

Port identifiers (PIDs) and PID binding overview

Core PID addressing mode
Core PID is the default PID format for Brocade platforms. It uses the entire 24-bit address space of
the domain, area ID, and AL_PA to determine an object’s address within the fabric.
The Core PID is a 24-bit address built from the following three 8-bit fields:

• Domain ID, written in hex and the numeric range is from 01 through ee (1 through 239)
• Area ID, written in hex and the numeric range is from 01 through ff (1 through 255)
• AL_PA
For example, if a device is assigned an address of 0f1e00, the following would apply:

• 0f is the domain ID.
• 1e is the area ID.
• 00 is the assigned AL_PA.
From this information, you can determine which switch the device resides on from the domain ID,
which port the device is attached to from the area ID, and if this device is part of a loop from the
AL_PA number.
For more information on reading and converting hexadecimal, refer to Appendix C, “Hexadecimal
Conversion”.

Fixed addressing mode
Fixed addressing mode is the default addressing mode used in all platforms that do not have
Virtual Fabrics enabled. When Virtual Fabrics is enabled on the Brocade Backbone, fixed
addressing mode is used only on the default logical switch. With fixed addressing mode enabled,
each port has a fixed address assigned by the system based on the port number. This address
does not change unless you choose to swap the address using the portSwap command.

10-bit addressing mode
The 10-bit addressing mode is the default mode for all the logical switches created in the Brocade
Backbones. This addressing scheme is flexible to support a large number of F_Ports. In the regular
10-bit addressing mode, the portAddress --auto command supports addresses from 0x00 to 0x8F.

NOTE

The default switch in the Brocade Backbones uses the fixed addressing mode.
The 10-bit addressing mode utilizes the 8-bit area ID and the borrowed upper two bits from the
AL_PA portion of the PID. Areas 0x00 through 0x8F use only 8 bits for the port address and support
up to 256 NPIV devices. A logical switch can support up to 144 ports that can each support 256
devices. Areas 0x90 through 0xFF use an additional two bits from the AL_PA for the port address.
Therefore, these ports support only 64 NPIV devices per port.
10-bit addressing mode provides the following features:

• A PID is dynamically allocated only when the port is first moved to a logical switch and
thereafter it is persistently maintained.

• PIDs are assigned in each logical switch starting with 0xFFC0, and can go to 0x8000 in the
case of 64-port blades.

84

Fabric OS Administrator’s Guide
53-1002920-02

Port identifiers (PIDs) and PID binding overview

3

• Shared area limitations are removed on 48-port and 64-port blades.
• Any port on a 48-port or 64-port blade can support up to 256 NPIV devices (in fixed addressing
mode, only 128 NPIV devices are supported in non-VF mode and 64 NPIV devices in VF mode
on a 48-port blade).

• Any port on a 48-port blade can support loop devices.
• Any port on a 48-port or 64-port blade can support hard port zoning.
• Port index is not guaranteed to be equal to the port area ID.

256-area addressing mode
The 256-area addressing mode is available only in a logical switch on the Brocade Backbone.
In this mode, only 256 ports are supported and each port receives a unique 8-bit area address.
This mode can be used in FICON environments, which have strict requirements for 8-bit area
FC addresses.
There are two types of area assignment modes in the 256-area addressing mode: zero-based and
port-based.

• Zero-based mode assigns areas when the ports are added to the logical switch, beginning at
area 0x00. When a port is assigned to a logical switch, the next free PID starting from 0x00 is
assigned. This mode allows FICON customers to make use of the upper ports of a 48-port or
64-port blade.

• Zero-based mode is supported on the default switch.
Port-based mode is a bit more complex:

• Port-based mode is not supported on the default switch.
• 48-port cards are supported in port-based addressing mode (mode 2) on both Brocade
DCX-4S and 8510-4 devices. However, the upper 16 ports of a 64-port card are not
supported.The Brocade DCX does not support port-based addressing mode (mode 2) on the
FC8-48 blade, but does support zero-based addressing (mode 1).

• The Brocade DCX-4S supports port-based addressing (mode 2) on the FC8-48 blade.
• The Brocade 8510-4 supports port-based addressing (mode 2) on the FC16-48 blade.
• The Brocade 8510-8 does not support port-based addressing (mode 2) on the FC16-48 blade,
but does support zero-based addressing (mode 1).

ATTENTION
The Brocade DCX and 8510-8 Backbones have safeguards that disable all 49 port cards if FICON
Management Server (FMS) is enabled.
Refer to the FICON Administrator’s Guide for more details if needed.

Fabric OS Administrator’s Guide
53-1002920-02

85

3

Port identifiers (PIDs) and PID binding overview

WWN-based PID assignment
WWN-based PID assignment is disabled by default. When the feature is enabled, bindings are
created dynamically; as new devices log in, they automatically enter the WWN-based PID database.
The bindings exist until you explicitly unbind the mappings through the CLI or change to a different
addressing mode. If there are any existing devices when you enable the feature, you must manually
enter the WWN-based PID assignments through the CLI.
This feature also allows you to configure a PID persistently using a device WWN. When the device
logs in to the switch, the PID is bound to the device WWN. If the device is moved to another port in
the same switch, or a new blade is hot-plugged, the device receives the same PID (area) at its next
login.
Once WWN-based PID assignment is enabled, you must manually enter the WWN-based PID
assignments through the CLI for any existing devices.

ATTENTION
When WWN-based PID assignment is enabled, the area assignment is dynamic and does not
guarantee any order in the presence of static WWN-area binding or when the devices are moved
around.
PID assignments are supported for a maximum of 4096 devices; this includes both point-to-point
and NPIV devices. The number of point-to-point devices supported depends on the areas available.
For example, 448 areas are available on Backbones and 256 areas are available on switches.
When the number of entries in the WWN-based PID database reaches 4096 areas used up, the
oldest unused entry is purged from the database to free up the reserved area for the new FLOGI.

Virtual Fabrics considerations for WWN-based PID assignment
WWN-based PID assignment is disabled by default and is supported in the default switch on the
Brocade DCX and DCX 8510 Backbone families. This feature is not supported on application blades
such as the FS8-18, FX8-24, and the FCOE10-24. The total number of ports in the default switch
must be 256 or less.
When the WWN-based PID assignment feature is enabled and a new blade is plugged into the
chassis, the ports for which the area is not available are disabled.

NPIV
If any N_Port ID Virtualization (NPIV) devices have static PIDs configured and the acquired area is
not the same as the one being requested, the FDISC coming from that device is rejected and the
error is noted in the RASlog.
If the NPIV device has Dynamic Persistent PID set, the same AL_PA value in the PID is used. This
guarantees NPIV devices get the same PID across reboots and AL_PAs assigned for the device do
not depend on the order in which the devices come up. For more information on NPIV, refer to
Chapter 18, “NPIV”.

Enabling automatic PID assignment
NOTE

To activate the WWN-based PID assignment, you do not need to disable the switch.

86

Fabric OS Administrator’s Guide
53-1002920-02

Port identifiers (PIDs) and PID binding overview

3

Use the following procedure to enable automatic PID assignment.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the configure command.
3. At the Fabric parameters prompt, type y.
4. At the WWN Based persistent PID prompt, type y.
5. Press Enter to bypass the remaining prompts without changing them.
Example of activating PID assignments
switch: admin> configure
Configure...
Fabric parameters (yes, y, no, n): [no] y
WWN Based persistent PID (yes, y, no, n): [no] y
System services (yes, y, no, n): [no]
ssl attributes (yes, y, no, n): [no]
rpcd attributes (yes, y, no, n): [no]
cfgload attributes (yes, y, no, n): [no]
webtools attributes (yes, y, no, n): [no]
Custom attributes (yes, y, no, n): [no]
system attributes (yes, y, no, n): [no]

Assigning a static PID
Use the following procedure to assign a static PID.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the wwnAddress -bind command to assign a 16-bit PID to a given WWN.

Clearing PID binding
Use the following procedure to clear a PID binding.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the wwnAddress -unbind command to clear the PID binding for the specified WWN.

Showing PID assignments
Use the following procedure to display PID assignments.
1. Connect to the switch and log in using an account with admin permissions.
2. Based on what you want to display, enter the appropriate command:

• wwnAddress –show displays the assigned WWN-PID bindings.
• wwnAddress –findPID wwn displays the PID assigned to the specified device WWN.

Fabric OS Administrator’s Guide
53-1002920-02

87

3

Ports

Ports
Ports provide either a physical or virtual network connection point for a device. Brocade devices
support a wide variety of ports.

Port Types
The following is a list of port types that may be part of a Brocade device:

• D_Port — A diagnostic port lets an administrator isolate the inter-switch link (ISL) to diagnose
link level faults. This port runs only specific diagnostics tests and does not carry any fabric
traffic. Refer to Chapter 17, “Diagnostic Port,” for more information on this port type.

• E_Port — An expansion port that is assigned to ISL links to expand a fabric by connecting it to
other switches. Two connected E_Ports form an inter-switch link (ISL). When E_Ports are used
to connect switches, those switches merge into a single fabric without an isolation
demarcation point. ISLs are non-routed links.

• EX_Port — A type of E_Port that connects a Fibre Channel router to an edge fabric.
From the point of view of a switch in an edge fabric, an EX_Port appears as a normal E_Port.
It follows applicable Fibre Channel standards as other E_Ports. However, the router terminates
EX_Ports rather than allowing different fabrics to merge as would happen on a switch with
regular E_Ports. An EX_Port cannot be connected to another EX_Port.

• F_Port — A fabric port that is assigned to fabric-capable devices, such as SAN storage devices.
• G_Port — A generic port that acts as a transition port for non-loop fabric-capable devices.
• L_Port or FL_Port — A loop or fabric loop port that connects loop devices. L_Ports are
associated with private loop devices and FL_Ports are associated with public loop devices.

• M_Port — A mirror port that is configured to duplicate (mirror) the traffic passing between a
specified source port and destination port. This is only supported for pairs of F_Ports.
Refer to the Fabric OS Troubleshooting and Diagnostics Guide for more information on
port mirroring.

• U_Port — A universal Fibre Channel port. This is the base Fibre Channel port type, and all
unidentified or uninitiated ports are listed as U_Ports.

• VE_Port — A virtual E_Port that is a gigabit Ethernet switch port configured for an FCIP tunnel.
• VEX_Port — A virtual EX_Port that connects a Fibre Channel router to an edge fabric. From the
point of view of a switch in an edge fabric, a VEX_Port appears as a normal VE_Port. It follows
the same Fibre Channel protocol as other VE_Ports. However, the router terminates VEX_Ports
rather than allowing different fabrics to merge as would happen on a switch with regular
VE_Ports.

Backbone port blades
Because Backbones contain interchangeable port blades, their procedures differ from those for
fixed-port switches. For example, fixed-port models identify ports only by the port number, while
Backbones identify ports by slot/port notation.

NOTE
For detailed information about the Brocade DCX and DCX 8510 Backbone families, refer to the
respective hardware reference manuals.

88

Fabric OS Administrator’s Guide
53-1002920-02

Ports

3

The different blades that can be inserted into a chassis are described as follows:

• Control processor (CP) blades contain communication ports for system management, and are
used for low-level, platform-wide tasks.

• Core blades are used for intra-chassis switching as well as interconnecting two Backbones.
• Port blades are used for host, storage, and interswitch connections.
• Application (AP) blades are used for Fibre Channel Application Services and Routing Services,
FCIP, Converged Enhanced Ethernet, and encryption support.

NOTE

On each port blade, a particular port must be represented by both slot number and port number.
The Brocade DCX and DCX 8510-8 each have 12 slots that contain control processor, core, port,
and AP blades:

• Slot numbers 6 and 7 contain CPs.
• Slot numbers 5 and 8 contain core blades.
• Slot numbers 1 through 4 and 9 through 12 contain port and AP blades.
The Brocade DCX-4S and DCX 8510-4 each have 8 slots that contain control processor, core, port,
and AP blades:

• Slot numbers 4 and 5 contain CPs.
• Slot numbers 3 and 6 contain core blades.
• Slot numbers 1 and 2, and 7 and 8 contain port and AP blades.
When you have port blades with different port counts in the same Backbone (for example, 16-port
blades and 32-port blades, or 16-port blades and 18-port blades with 16 FC ports and 2 GbE ports,
or 16-port and 48-port blades), the area IDs no longer match the port numbers.
Table 7 on page 96 lists the port numbering schemes for the blades.

Setting port names
Perform the following steps to specify a port name. For Backbones, specify the slot number where
the blade is installed.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portName command.
ecp:admin> portname 1/0 trunk1

Port identification by slot and port number
The port number is a number assigned to an external port to give it a unique identifier in a switch.
To select a specific port in the Backbones, you must identify both the slot number and the port
number using the format slot number/port number. No spaces are allowed between the slot
number, the slash (/), and the port number.
Example of enabling port 4 on a blade in slot 2
ecp:admin> portenable 2/4

Fabric OS Administrator’s Guide
53-1002920-02

89

3

Ports

Port identification by port area ID
The relationship between the port number and area ID depends upon the PID format used in the
fabric. When Core PID format is in effect, the area ID for port 0 is 0, for port 1 is 1, and so forth.
For 32-port blades (FC8-32, FC8-32E, FC16-32), the numbering is contiguous up to port 15; from
port 16, the numbering is still contiguous, but starts with 128. For example, port 15 in slot 1 has a
port number and area ID of 15; port 16 has a port number and area ID of 128; port 17 has a port
number and area ID of 129.
For 48-port blades (FC8-48, FC8-48E, FC16-48), the numbering is the same as for 32-port blades
for the first 32 ports on the blade. For ports 32 through 47, area IDs are not unique and port index
should be used instead of area ID.
For the 64-port blade (FC8-64), the numbering is the same as for 32-port blades for the first 32
ports on the blade. For ports 32 through 63, area IDs are not unique and port index should be used
instead of area ID.
If you perform a port swap operation, the port number and area ID no longer match. On 48-port
blades, port swapping is supported only on ports 0 through 15.
To determine the area ID of a particular port, enter the switchShow command. This command
displays all ports on the current (logical) switch and their corresponding area IDs.

Port identification by index
With the introduction of 48-port blades, indexing was introduced. Unique area IDs are possible for
up to 255 areas, but beyond that there needed to be some way to ensure uniqueness.
A number of fabric-wide databases supported by Fabric OS (including ZoneDB, the ACL DDC, and
Admin Domain) allow a port to be designated by the use of a “D,P” (domain,port) notation. While
the “P” component appears to be the port number, for up to 255 ports it is actually the area
assigned to that port.

NOTE
The port area schema does not apply to the Brocade DCX-4S and DCX 8510-4 Backbones.

Configuring a device-switch connection
To configure an 8 Gbps (and 8 Gbps only) connection between a device and a switch, use the
portCfgFillWord command. This command provides the following configuration options:

•
•
•
•
•

Mode Link Init/Fill Word
Mode 0 IDLE/IDLE
Mode 1 ARBF/ARBF
Mode 2 IDLE/ARBF
Mode 3 If ARBF/ARBF fails, use IDLE/ARBF

ATTENTION
Although this setting only affects devices logged in at 8 Gbps, changing the mode is disruptive
regardless of the speed at which the port is operating.

90

Fabric OS Administrator’s Guide
53-1002920-02

Ports

3

The setting is retained and applied any time an 8 Gbps device logs in. Upgrades from prior releases
which supported only Modes 0 and 1 will not change the existing setting, but switches reset to
factory defaults with Fabric OS v6.3.1 or later will be configured to Mode 0 by default. The default
setting on new units may vary by vendor.
Modes 2 and 3 are compliant with FC-FS-3 specifications (standards specify the IDLE/ARBF
behavior of Mode 2, which is used by Mode 3 if ARBF/ARBF fails after three attempts). For most
environments, Brocade recommends using Mode 3, as it provides more flexibility and compatibility
with a wide range of devices. In the event that the default setting or Mode 3 does not work with a
particular device, contact your switch vendor for further assistance.

Swapping port area IDs
If a device that uses port binding is connected to a port that fails, you can use port swapping to
make another physical port use the same PID as the failed port. The device can then be plugged
into the new port without the need to reboot the device.
If two ports are changed using the portSwap command, their respective areas and “P” values are
exchanged.
For ports that are numbered above 255, the “P” value is a logical index. The first 256 ports
continue to have an index value equal to the area ID assigned to the port. If a switch is using Core
PID format, and no port swapping has been done, the port index value for all ports is the same as
the physical port numbers. Using portSwap on a pair of ports will exchange those ports’ area ID
and index values.
Port swapping has the following restrictions:

•
•
•
•
•

Shared area ports cannot be swapped.
Ports that are part of a trunk group cannot be swapped.
GbE ports cannot be swapped.
Ports on a faulty blade cannot be swapped.
Swapping ports between different logical switches is not supported. The ports on the source
and destination blades must be in the same logical switch.

• The portSwap command is not supported for ports above 256.
Use the following procedure to swap the port area IDs of two physical switch ports. To swap port
area IDs, the port swap feature must be enabled, and both switch ports must be disabled. The
swapped area IDs for the two ports remain persistent across reboots, power cycles, and failovers.

NOTE

On the Brocade DCX and DCX 8510 Backbone families, you can swap only ports 0 through 15 on the
FC8-48 port blades. You cannot swap ports 16 through 47.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portSwapEnable command to enable the feature.
3. Enter the portDisable command on each of the source and destination ports to be swapped.
switch:admin>portdisable 1
ecp:admin>portdisable 1/2

4. Enter the portSwap command.
switch:admin>portswap 1 2
ecp:admin>portswap 1/1 2/2

Fabric OS Administrator’s Guide
53-1002920-02

91

3

Ports

5. Enter the portSwapShow command to verify that the port area IDs have been swapped.
A table shows the physical port numbers and the logical area IDs for any swapped ports.
6. Enter the portSwapDisable command to disable the port swap feature.

Port activation and deactivation
By default, all licensed ports are enabled. You can disable and re-enable them as necessary. Ports
that you activate with the Ports on Demand license must be enabled explicitly, as described in
“Ports on Demand” on page 535.

CAUTION
The fabric will be reconfigured if the port you are enabling or disabling is connected to another
switch.
The switch with a port that has been disabled will be segmented from the fabric and all traffic
flowing between it and the fabric will be lost.

Enabling a port
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the appropriate command based on the current state of the port and whether it is
necessary to specify a slot number:

• To enable a port that is disabled, enter the portEnable [slot/]port command.
• To enable a port that is persistently disabled, enter the portCfgPersistentEnable [slot/]port
command.
If you change port configurations during a switch failover, the ports may become disabled. To
bring the ports online, re-issue the portEnable command after the failover is complete.

Disabling a port
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the appropriate command based on the current state of the port and whether it is
necessary to specify a slot number:

• To disable a port that is enabled, enter the portDisable [slot/]port command.
• To disable a port that is persistently enabled, enter the portCfgPersistentDisable
[slot/]port command.

Port decommissioning
Port decommissioning is an Fabric OS 7.0.0 and later provides an automated mechanism to
remove an E_Port or E_Port trunk port from use. The port decommissioning feature identifies the
target port and communicates the intention to decommission the port to those systems within the
fabric affected by the action. Each affected system can agree or disagree with the action, and
these responses are automatically collected before a port is decommissioned.

92

Fabric OS Administrator’s Guide
53-1002920-02

Ports

3

Fabric OS 7.1.0 and later provides F_Port decommissioning and recommissioning using Brocade
Network Advisor 12.1.0 and later. Refer to the Brocade Network Advisor User Manual for details.

NOTE

All members of a trunk group must have an equal link cost value in order for any of the members to
be decommissioned. If any member of a trunk group does not have an equal cost, requests to
decommission a trunk member will fail and an error reminding the caller of this requirement is
produced.
The following restrictions apply to port decommissioning:

• The local switch and the remote switch on the other end of the E_Port must both be running
Fabric OS 7.0.0 or later.

• Port decommissioning is not supported on links configured for encryption or compression.
• Port decommissioning is not supported on ports with DWDM, CWDM, or TDM.
• Port decommissioning requires that the lossless feature is enabled on both the local switch
and the remote switch.
Use the portDecom command to begin the decommission process.

Setting port modes
Ports can be set to use one of three link operating modes: full duplex, half duplex, or autonegotiate.
Changing the link operating mode is not supported for all network interfaces or for all Ethernet
network interfaces. On the CP blade in a Brocade DCX, DCX-4S, DCX 8510-4, or DCX 8510-8, the
supported interfaces are eth0 and eth3. On all other platforms, only eth0 is supported.
For dual-CP systems, the ifModeSet command affects only the CP to which you are currently logged
in. Therefore, to set the link operating mode on the active CP, you must issue the ifModeSet
command on the active CP; and to set the mode on the standby CP, you must issue the ifModeSet
command on the standby CP. During failover, the mode is retained separately for each CP because
the physical links may be set to operate in different modes.

ATTENTION
Forcing the link to an operating mode not supported by the network equipment to which it is
attached may result in an inability to communicate with the system through its Ethernet interface. It
is recommended that the ifModeSet command be used only from the serial console port. When used
through an interface other than the serial console port, the command displays a warning message
and prompts for verification before continuing. This warning is not displayed and you are not
prompted when the command is used through the serial console port.
Use the following procedure to set the mode of a port.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the ifModeSet command.
Enter y or yes at the prompts to confirm the active link operating mode values. Enter n or no to
deactivate that mode.
Example of setting the port mode to full autonegotiate

The following example sets the mode for eth3 to autonegotiate, and permits both full and half
duplex modes to be selected at both 10 and 100 Mbps. Note that the caution shown in this
example is not displayed when the command is entered using the serial console port.

Fabric OS Administrator’s Guide
53-1002920-02

93

3

Ports

switch:admin> ifmodeset eth3
Exercise care when using this command. Forcing the link to an operating mode
not supported by the network equipment to which it is attached may result in an
inability to communicate with the system through its ethernet interface.
It is recommended that you only use this command from the serial console port.
Are you sure you really want to do this? (yes, y, no, n): [no] y
Proceed with caution.
Auto-negotiate (yes, y, no, n): [no] y
Advertise 100 Mbps / Full Duplex (yes, y, no, n): [yes] y
Advertise 100 Mbps / Half Duplex (yes, y, no, n): [yes] y
Advertise 10 Mbps / Full Duplex (yes, y, no, n): [yes] y
Advertise 10 Mbps / Half Duplex (yes, y, no, n): [yes] y
Committing configuration...done.

Example of setting the port mode to 10 Mbps half duplex operation

The following example forces the link for the eth0 interface from autonegotiation to 10 Mbps
half-duplex operation:
switch:admin> ifmodeset eth0
Auto-negotiate (yes, y, no, n): [yes] n
Force 100 Mbps / Full Duplex (yes, y, no, n): [no] n
Force 100 Mbps / Half Duplex (yes, y, no, n): [no] n
Force 10 Mbps / Full Duplex (yes, y, no, n): [no] n
Force 10 Mbps / Half Duplex (yes, y, no, n): [no] y
Committing configuration...done.

Setting port speeds
Use the following procedure to set port speeds.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portCfgSpeed command.
The following example sets the speed for port 3 on slot 2 to 4 Gbps:
switch:admin> portcfgspeed 2/3 4

The following example sets the speed for port 3 on slot 2 to autonegotiate:
switch:admin> portcfgspeed 2/3 0

Setting all ports on a switch to the same speed
Use the following procedure to set all ports on a switch to the same speed.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the switchCfgSpeed command.
The following example sets the speed for all ports on the switch to 8 Gbps:
switch:admin> switchcfgspeed 8
Committing configuration...done.

The following example sets the speed for all ports on the switch to autonegotiate:
switch:admin> switchcfgspeed 0
Committing configuration...done.

94

Fabric OS Administrator’s Guide
53-1002920-02

Blade terminology and compatibility

3

Setting port speed for a port octet
You can use the portCfgOctetSpeedCombo command to configure the speed for a port octet. Be
aware that in a Virtual Fabrics environment, this command configures the speed of a port octet
chassis-wide and not only on the logical switch.
Use the following procedure to set the port speed for a port octet.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portCfgOctetSpeedCombo command.
The following example configures the ports in the first octet for combination 3
(support autonegotiated or fixed port speeds of 16 Gbps and 10 Gbps):
switch:admin> portcfgoctetspeedcombo 1 3

NOTE
For information on how encryption and compression can affect port speed, refer to “Port speed on
encryption- or compression-enabled ports” on page 447.

Blade terminology and compatibility
Before configuring a chassis, familiarize yourself with the platform CP blade and port blade
nomenclature, as well as the port blade compatibilities. Table 6 includes core and CP blade
terminology and descriptions. Table 7 on page 96 includes port blade terminology and
descriptions.

TABLE 6

Core and CP blade terminology and platform support
Supported on:

Blade

Blade ID
DCX family
(slotshow)

DCX 8510 family

Definition

CP8

50

Yes

Yes

Brocade DCX and DCX 8510 Backbone family control
processor blade. This CP supports all blades used in the
DCX and DCX 8510 Backbone families.

CORE8

52

Yes
DCX only

No

A 16-port blade that provides 8 Gbps connectivity
between port blades in the Brocade DCX chassis.

CR4S-8

46

Yes
No
DCX-4S only

A 16-port blade that provides 8 Gbps connectivity
between port blades in the Brocade DCX-4S chassis.

CR16-8

98

No

Yes
DCX 8510-8 only

A core blade that has 16x4 QSFPs per blade. It can be
connected to another CR16-8 or a CR16-4 core blade.

CR16-4

99

No

Yes
DCX 8510-4 only

A core blade that has 8x4 QSFPs per blade. It can be
connected to another CR16-4 or a CR16-8 core blade.

Fabric OS Administrator’s Guide
53-1002920-02

95

3
TABLE 7

Blade terminology and compatibility

Port blade terminology, numbering, and platform support
Supported on:

Blade

Blade ID
DCX family
(slotshow)

DCX 8510
family

Ports

Definition

FC8-161

21

Yes

No

16

8-Gbps port blade supporting 1, 2, 4, and 8 Gbps port speeds.
Ports are numbered from 0 through 15 from bottom to top.

FC8-321

55

Yes

No

32

8-Gbps port blade supporting 1, 2, 4, and 8 Gbps port speeds.
Ports are numbered from 0 through 15 from bottom to top on the left set of
ports and 16 through 31 from bottom to top on the right set of ports.

FC8-32E

125

No

Yes

32

8-Gbps port blade supporting 2, 4, and 8 Gbps port speeds.
Ports are numbered from 0 through 15 from bottom to top on the left set of
ports and 16 through 31 from bottom to top on the right set of ports.

FC8-481

51

Yes

No

48

8-Gbps port blade supporting 1, 2, 4, and 8 Gbps port speeds.
Ports are numbered from 0 through 23 from bottom to top on the left set of
ports and 24 through 47 from bottom to top on the right set of ports.

FC8-48E

126

No

Yes

48

8-Gbps port blade supporting 2, 4, and 8 Gbps port speeds.
Ports are numbered from 0 through 23 from bottom to top on the left set of
ports and 24 through 47 from bottom to top on the right set of ports.

FC8-64

77

Yes

Yes

64

8-Gbps port blade supporting 2, 4, and 8 Gbps port speeds. The Brocade
DCX and Brocade DCX 8510 Backbone families support loop devices on
64-port blades in a Virtual Fabrics-enabled environment. The loop devices
can only be attached to ports on a 64-port blade that is not a part of the
default logical switch.
Ports are numbered from 0 through 31 from bottom to top on the left set of
ports and 32 through 63 from bottom to top on the right set of ports.

FC16-32

97

No

Yes

32

A 32-port, 16-Gbps port blade supporting 2, 4, 8, 10, and 16 Gbps port
speeds.
NOTE: 10 Gbps speed for FC16-xx blades requires the 10G license.
Ports are numbered from 0 through 15 from bottom to top on the left set of
ports and 16 through 31 from bottom to top on the right set of ports.

FC16-48

96

No

Yes

48

A 48-port, 16-Gbps port blade supporting 2, 4, 8, 10, and 16 Gbps port
speeds.
NOTE: 10 Gbps speed for FC16-xx blades requires the 10G license.
Ports are numbered from 0 through 23 from bottom to top on the left set of
ports and 24 through 47 from bottom to top on the right set of ports.

FS8-18

96

68

Yes

Yes

16 FC
2 GbE

Brocade Encryption blade that provides high performance 32-port
auto-sensing 8-Gbps Fibre Channel connectivity with data cryptographic
(encryption and decryption) and data compression capabilities.
Ports are numbered from 0 through 15 from bottom to top.
GbE ports are numbered ge0 through ge1 from top to bottom.
Going from top to bottom, the 2 GbE ports appear on the top of the blade
followed by the 16 FC ports.

Fabric OS Administrator’s Guide
53-1002920-02

Blade terminology and compatibility

TABLE 7

3

Port blade terminology, numbering, and platform support (Continued)
Supported on:

Blade

Blade ID
DCX family
(slotshow)

DCX 8510
family

Ports

Definition

FCOE10-24

74

Yes

No

24
10-GbE
DCB ports

An application blade that provides Converged Enhanced Ethernet to bridge
a Fibre Channel and Ethernet SAN.
Ports are numbered from 0 through 11 from bottom to top on the left set of
ports and 12 through 23 from bottom to top on the right set of ports.

FX8-24

75

Yes

Yes

12 FC
10 1-GbE
2 10-GbE

Extension blade with 8-Gbps Fibre Channel, FCIP, and 10-GbE technology.
Port numbering on this blade is as follows.
On the left side of the blade going from bottom to top:
• Six FC ports numbered from 0 through 5
• Two 10-GbE ports numbered xge0 and xge1
• Four 1-GbE ports numbered from ge0 through ge3
On the right side of the blade going from bottom to top:
• Six FC ports numbered from 6 through 11
• Six 1-GbE ports numbered from ge4 through ge9

1.

The Brocade DCX and DCX-4S support loop devices on this blade in a Virtual Fabrics-enabled environment.

CP blades
The control processor (CP) blade provides redundancy and acts as the main controller on the
Brocade Backbone. The Brocade DCX and DCX 8510 Backbone families support the CP8 blades.
The CP blades in the Brocade DCX and DCX 8510 Backbone families are hot-swappable. The CP8
blades are fully interchangeable among Brocade DCX, DCX-4S, DCX 8510-4, and DCX 8510-8
Backbones.
Brocade recommends that each CP (primary and secondary partition) should maintain the same
firmware version.
For more information on maintaining firmware in your Backbone, refer to Chapter 10, “Installing
and Maintaining Firmware”.

Core blades
Core blades provide intra-chassis switching and inter-chassis link (ICL) connectivity between
DCX/DCX-4S platforms and between DCX 8510 platforms.

•
•
•
•

Brocade DCX supports two CORE8 core blades.
Brocade DCX-4S supports two CR4S-8 core blades.
Brocade DCX 8510-8 supports two CR16-8 core blades.
Brocade DCX 8510-4 supports two CR16-4 core blades.

The core blades for each platform are not interchangeable or hot-swappable with the core blades
for any other platform. If you try to interchange the blades, they become faulty.

Fabric OS Administrator’s Guide
53-1002920-02

97

3

Enabling and disabling blades

Port and application blade compatibility
Table 7 on page 96 identifies which port and application blades are supported for each Brocade
Backbone.

NOTE
During power up of a Brocade DCX or DCX-4S Backbone, if an FCOE10-24 is detected first before any
other AP blade, all other AP and FC8-64 blades are faulted. If a non-FCOE10-24 blade is detected
first, then any subsequently-detected FCOE10-24 blades are faulted. Blades are powered up
starting with slot 1.
The maximum number of intelligent blades supported on a Brocade DCX or DCX 8510-8 is eight.
The maximum number of intelligent blades supported on a Brocade DCX-4S or DCX 8510-4 is four.
Table 8 lists the maximum supported limits of each blade for a specific Fabric OS release. Software
functions are not supported across application blades.

TABLE 8

Blade compatibility within Brocade Backbone families

Intelligent blade

Fabric OS v6.3.0

Fabric OS v6.4.0

Fabric OS v7.0.0

DCX

DCX-4S

DCX

DCX-4S

DCX

DCX-4S

DCX 8510-8

DCX 8510-4

FS8-18

4

4

4

4

4

4

4

4

FCOE10-241

2

2

2

2

4

4

0

0

FX8-242

2

4

4

4

4

4

4

4

1.

Not compatible with other application blades or with the FC8-64 in the same chassis.

2.

The hardware limit is enforced by software.

FX8-24 compatibility notes
Follow these guidelines when using an FX8-24 in the Brocade DCX and DCX-4S Backbones:

• Brocade 7500 GbE ports cannot be connected to either the FX8-24 or Brocade 7800 GbE
ports. The ports may come online, but they will not communicate with each other.

• If an FX8-24 blade is replaced by another FX8-24 blade, the previous IP configuration data
would be applied to the new FX8-24.

• The FX8-24 and FS8-18 blades cannot coexist with the FCOE10-24 blade.

Enabling and disabling blades
Port blades are enabled by default. In some cases, you will need to disable a port blade to perform
diagnostics. When diagnostics are executed manually (from the Fabric OS command line), many
commands require the port blade to be disabled. This ensures that diagnostic activity does not
interfere with normal fabric traffic.
If you need to replace an application blade with a different application blade, there may be extra
steps you need to take to ensure that the previous configuration is not interfering with your new
application blade.

98

Fabric OS Administrator’s Guide
53-1002920-02

Blade swapping

3

Enabling blades
Use the following procedure to enable a blade.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the bladeEnable command with the slot number of the port blade you want to enable.
ecp:admin> bladeenable 3
Slot 3 is being enabled

FC8-48, FC8-48E, FC8-64, and FC16-48 port blade enabling exceptions
Because the area IDs are shared with different port IDs, the FC8-48, FC8-48E, FC8-64, and
FC16-48 blades support only F_Ports and E_Ports. They do not support FL_Ports.
Port swapping on an FC8-48, FC8-48E, FC8-64, and FC16-48 is supported only on ports 0 through
15. For the FC8-32, FC8-32E, and FC16-32 port blades, port swapping is supported on all 32 ports.
This means that if you replace a 32-port blade where a port has been swapped on ports 16 through
31 with a 48-port blade, the 48-port blade faults. To correct this, reinsert the 32-port blade and
issue portSwap to restore the original area IDs to ports 16 through 31.

Disabling blades
Use the following procedure to disable a blade.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the bladeDisable command with the slot number of the port blade you want to disable.
ecp:admin> bladedisable 3
Slot 3 is being disabled

Blade swapping
Blade swapping allows you to swap one blade with another of the same type; in this way, you can
replace a FRU with minimal traffic disruption.
The entire operation is accomplished when the bladeSwap command runs on the Fabric OS. Fabric
OS then validates each command before implementing the command on the Backbone. If an error
is encountered, the blade swap quits without disrupting traffic flowing through the blades. If an
unforeseen error does occur during the bladeSwap command, an entry will be made in the RASlog
and all ports that have been swapped as part of the blade swap operation will be swapped back.
On successful completion of the command, the source and destination blades are left in a disabled
state, allowing you to complete the cable move.
Blade swapping is based on port swapping and has the same restrictions:

•
•
•
•
•

Shared area ports cannot be swapped.
Ports that are part of a trunk group cannot be swapped.
GbE ports cannot be swapped.
Faulty blades cannot be swapped.
Swapping ports between different logical switches is not supported. The ports on the source
and destination blades must be in the same logical switch.

Fabric OS Administrator’s Guide
53-1002920-02

99

3

Blade swapping

• Undetermined board types cannot be swapped. For example, a blade swap will fail if the blade
type cannot be identified.

• Blade swapping is not supported when swapping to a different model of blade or a different
port count. For example, you cannot swap an FC8-32 blade with an FC8-48 port blade.

How blades are swapped
The bladeSwap command performs the following operations:
1. Blade selection
The selection process includes selecting the switch and the blades to be affected by the swap
operation. Figure 2 shows the source and destination blades identified to begin the process.

FIGURE 2

Identifying the blades

2. Blade validation
The validation process includes determining the compatibility between the blades selected for
the swap operation:

• Blade technology. Both blades must be of compatible technology types (for example, Fibre
Channel to Fibre Channel, Ethernet to Ethernet, application to application, and so on).

• Port count. Both blades must support the same number of front ports (for example, 16
ports to 16 ports, 32 ports to 32 ports, 48 ports to 48 ports, and so on).

• Availability. The ports on the destination blade must be available for the swap operation
and not attached to any other devices.
3. Port preparation
The process of preparing ports for a swap operation includes basic operations such as
ensuring the source and destination ports are offline, or verifying that none of the destination
ports have failed.

100

Fabric OS Administrator’s Guide
53-1002920-02

Blade swapping

3

The preparation process also includes any special handling of ports associated with logical
switches. For example, Figure 3 shows the source blade has ports in a logical switch or logical
fabric, and the corresponding destination ports must be included in the associated logical
switch or logical fabric of the source ports.

FIGURE 3

Blade swap with Virtual Fabrics during the swap

4. Port swapping
The swap ports action is an iteration of the portSwap command for each port on the source
blade to each corresponding port on the destination blade.
As shown in Figure 4, the blades can be divided into different logical switches as long as they
are divided the same way. If slot 1 and slot 2 ports 0 through 7 are all in the same logical
switch, then blade swapping slot 1 to slot 2 will work. The entire blade does not need to be in
the same partition.

Fabric OS Administrator’s Guide
53-1002920-02

101

3

Disabling switches

FIGURE 4

Blade swap with Virtual Fabrics after the swap

Swapping blades
Use the following procedure to swap blades.
1. Connect to the Backbone and log in using an account with admin permissions.
2. Enter the bladeSwap command.
If no errors are encountered, the blade swap will complete successfully. If errors are
encountered, the command is interrupted and the ports are set back to their original
configurations.
3. Once the command completes successfully, move the cables from the source blade to the
destination blade.
4. Enter the bladeEnable command on the destination blade to enable all user ports.

Disabling switches
Switches are enabled by default. In some cases, you may need to disable a switch to perform
diagnostics. This ensures that diagnostic activity does not interfere with normal fabric traffic.
Use the following procedure to disable a switch.
1. Connect to the Backbone and log in using an account with admin permissions.
2. Enter the switchCfgPersistentDisable --setdisablestate command.
This procedure sets the switch to the disabled state without disabling it. On reset, the switch will be
in a disabled state, and will need to be enabled.

102

Fabric OS Administrator’s Guide
53-1002920-02

Power management

3

Power management
All blades are powered on by default when the switch chassis is powered on. Blades cannot be
powered off when POST or AP initialization is in progress.
To manage power and ensure that more critical components are the least affected by power
changes, you can specify the order in which the components are powered off by using the
powerOffListSet command.
The power monitor compares the available power with the power required to determine if there will
be enough power to operate. If it is predicted to be less power available than required, the
power-off list is processed until there is enough power for operation. By default, the processing
begins with slot 1 and proceeds to the last slot in the chassis. As power becomes available, slots
are powered up in the reverse order. During the initial power up of a chassis, or using the
slotPowerOn command, or the insertion of a blade, the available power is compared to required
power before power is applied to the blade.

NOTE

Some FRUs in the chassis may use significant power, yet cannot be powered off through software.
The powerOffListShow command displays the power-off order.

NOTE
In the Backbones, the core blades and CP blades cannot be powered off from the CLI. You must
manually power off the blades by lowering the slider or removing power from the chassis. If there is
no CP up and running, then physical removal or powering off the chassis is required.

Powering off a port blade
Use the following procedure to power off a port blade.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the slotPowerOff command with the slot number of the port blade you want to power off.
ecp:admin> slotpoweroff 3
Slot 3 is being powered off

Powering on a port blade
Use the following procedure to power on a port blade.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the slotPowerOn command with the slot number of the port blade you want to power on.
ecp:admin> slotpoweron 3
Powering on slot 3

Fabric OS Administrator’s Guide
53-1002920-02

103

3

Equipment status

Equipment status
You can check the status of switch operation, High Availability features, and fabric connectivity.

Checking switch operation
Use the following procedure to check switch operation.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the switchShow command. This command displays a switch summary and a port
summary.
3. Check that the switch and ports are online.
4. Use the switchStatusShow command to further check the status of the switch.

Verifying High Availability features (Backbones only)
High Availability (HA) features provide maximum reliability and nondisruptive management of key
hardware and software modules.
Use the following procedure to verify High Availability features for a Brocade Backbone.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the chassisShow command to verify the model of the DCX and obtain a listing of all
field-replaceable units (FRUs).
3. Enter the haShow command to verify HA is enabled, the heartbeat is up, and that the HA state
is synchronized between the active and standby CP blades.
4. Enter the fanShow command to display the current status and speed of each fan in the system.
Refer to the hardware reference manual of your system to determine the appropriate values.
5. Enter the psShow command to display the current status of the switch power supplies. Refer to
the hardware reference manual of your system to determine the appropriate values.
6. Enter the slotShow -m command to display the inventory and the current status of each slot in
the system.
Example of the slot information displayed for a DCX chassis
DCX:FID128:admin> slotshow -m
Slot
Blade Type
ID
Model Name
Status
-------------------------------------------------1
SW BLADE
55
FC8-32
ENABLED
2
SW BLADE
51
FC8-48
ENABLED
3
SW BLADE
39
FC8-16
ENABLED
4
SW BLADE
51
FC8-48
ENABLED
5
CORE BLADE
52
CORE8
ENABLED
6
CP BLADE
50
CP8
ENABLED
7
CP BLADE
50
CP8
ENABLED
8
CORE BLADE
52
CORE8
ENABLED
9
SW BLADE
37
FC8-16
ENABLED
10
AP BLADE
43
FS8-18
ENABLED
11
SW BLADE
55
FC8-32
ENABLED
12
AP BLADE
24
FS8-18
ENABLED

104

Fabric OS Administrator’s Guide
53-1002920-02

Equipment status

3

Verifying fabric connectivity
Use the following procedure to verify fabric connectivity.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fabricShow command. This command displays a summary of all the switches in the
fabric.
The output of the fabricShow command is discussed in “Domain IDs” on page 75.

Verifying device connectivity
Use the following procedure to verify device connectivity.
1. Connect to the switch and log in using an account with admin permissions.
2. Optional: Enter the switchShow command to verify devices, hosts, and storage are connected.
3. Optional: Enter the nsShow command to verify devices, hosts, and storage have successfully
registered with the name server.
4. Enter the nsAllShow command to display the 24-bit Fibre Channel addresses of all devices in
the fabric.
switch:admin> nsallshow
{
010e00 012fe8 012fef 030500
030b1e 030b1f 040000 050000
050def 051700 061c00 071a00
0a07cb 0a07cc 0a07cd 0a07ce
0a07d5 0a07d6 0a07d9 0a07da
0a0f02 0a0f0f 0a0f10 0a0f1b
0b2fef 0f0000 0f0226 0f0233
211700 211fe8 211fef 2c0000
611600 620800 621026 621036
621500 621700 621a00
75 Nx_Ports in the Fabric }

030b04
050200
073c00
0a07d1
0a07dc
0a0f1d
0f02e4
2c0300
6210e4

030b08
050700
090d00
0a07d2
0a07e0
0b2700
0f02e8
611000
6210e8

030b17
050800
0a0200
0a07d3
0a07e1
0b2e00
0f02ef
6114e8
6210ef

030b18
050de8
0a07ca
0a07d4
0a0f01
0b2fe8
210e00
6114ef
621400

The number of devices listed should reflect the number of devices that are connected.

Viewing the switch status policy threshold values
For switches running Fabric Watch, you can view the switch status policy threshold values using the
switchStatusPolicyShow command
The policy parameter determines the number of failed or inoperable units for each contributor that
triggers a status change in the switch. Each parameter can be adjusted so that a specific threshold
must be reached before that parameter changes the overall status of a switch to MARGINAL or
DOWN. For example, if the FaultyPorts DOWN parameter is set to 3, the status of the switch will
change if three ports fail. Only one policy parameter needs to pass the MARGINAL or DOWN
threshold to change the overall status of the switch.
For more information about setting policy parameters, refer to the Fabric Watch Administrator’s
Guide.

Fabric OS Administrator’s Guide
53-1002920-02

105

3

Equipment status

If the switch is running Fabric Watch, you can use the following procedure to view the switch status
policy threshold values. If the switch is running MAPS, refer to the Monitoring and Alerting Policy
Suite Administrator’s Guide.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the switchStatusPolicyShow command.
Whenever there is a switch change, an error message is logged and an SNMP
connUnitStatusChange trap is sent.
The output is similar to the following:
ecp:admin> switchstatuspolicyshow
switch:admin> switchstatuspolicyshow
The current overall switch status policy parameters:
Down
Marginal
---------------------------------PowerSupplies
0
0
Temperatures
0
0
Fans
1
0
WWN
0
0
CP
0
0
Blade
0
0
CoreBlade
0
0
Flash
0
0
MarginalPorts 0.00%[0]
0.00%[0]
FaultyPorts 0.00%[0]
0.00%[0]
MissingSFPs 0.00%[0]
0.00%[0]
ErrorPorts 0.00%[0]
0.00%[0]
Number of ports: 4

Setting the switch status policy threshold values
For switches running Fabric Watch, you can set the switch status policy threshold values using the
switchStatusPolicySet command
If the switch is running Fabric Watch, you can use the following procedure to set the switch status
policy threshold values. If the switch is running MAPS, refer to the Monitoring and Alerting Policy
Suite Administrator’s Guide.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the switchStatusPolicySet command.
The current switch status policy parameter values are displayed. You are prompted to enter
values for each DOWN and MARGINAL threshold parameter.

NOTE

By setting the DOWN and MARGINAL values for a parameter to 0,0, that parameter is no longer
used in setting the overall status for the switch.
3. Verify the threshold settings you have configured for each parameter.
Enter the switchStatusPolicyShow command to view your current switch status policy
configuration.

106

Fabric OS Administrator’s Guide
53-1002920-02

Audit log configuration

3

Example output from a switch

The following example displays what is typically seen from a Brocade switch, but the quantity and
types vary by platform.
switch:admin> switchstatuspolicyshow
To change the overall switch status policy parameters
The current overall switch status policy parameters:
Down
Marginal
----------------------------------PowerSupplies
2
1
Temperatures
2
1
Fans
2
1
Flash
0
1
MarginalPorts
25.00%[12]
10.00%[5]
FaultyPorts
25.00%[12]
10.00%[5]
MissingSFPs
0.00%[0]
0.00%[0]
ErrorPorts
0.00%[0]
0.00%[0]
Number of ports: 48
Note that the value, 0, for a parameter, means that it is
NOT used in the calculation.
** In addition, if the range of settable values in the prompt is (0..0),
** the policy parameter is NOT applicable to the switch.
** Simply hit the Return key.
The minimum number of
Bad PowerSupplies contributing to DOWN status: (0..2) [2]
Bad PowerSupplies contributing to MARGINAL status: (0..2) [1]
Bad Temperatures contributing to DOWN status: (0..4) [2]1
Bad Temperatures contributing to MARGINAL status: (0..4) [1]2
Bad Fans contributing to DOWN status: (0..2) [2]
Bad Fans contributing to MARGINAL status: (0..2) [1]
(output truncated)

NOTE

On the Brocade Backbones, the command output includes parameters related to CP blades.

Audit log configuration
When managing SANs, you may want to audit certain classes of events to ensure that you can view
and generate an audit log for what is happening on a switch, particularly for security-related event
changes. These events include login failures, zone configuration changes, firmware downloads,
and other configuration changes; in other words, critical changes that have a serious effect on the
operation and security of the switch.
Important information related to event classes is also tracked and made available. For example,
you can track changes from an external source by the user name, IP address, or type of
management interface used to access the switch.

Fabric OS Administrator’s Guide
53-1002920-02

107

3

Audit log configuration

Auditable events are generated by the switch and streamed to an external host through a
configured system message log daemon (syslog). You specify a filter on the output to select the
event classes that are sent through the system message log. The filtered events are streamed
chronologically and sent to the system message log on an external host in the specified audit
message format. This ensures that they can be easily distinguished from other system message log
events that occur in the network. Then, at some regular interval of your choosing, you can review
the audit events to look for unexpected changes.
Before you configure audit event logging, familiarize yourself with the following audit event log
behaviors and limitations:

• By default, all event classes are configured for audit; to create an audit event log for specific
events, you must explicitly set a filter with the class operand and then enable it.

• Audited events are generated specific to a switch and have no negative impact on
performance.

• The last 256 events are persistently stored on the switch and are streamed to a system
message log.

• The audit log depends on the system message log facility and IP network to send messages
from the switch to a remote host. Because the audit event log configuration has no control over
these facilities, audit events can be lost if the system message log and IP network facilities fail.

• If too many events are generated by the switch, the system message log becomes a bottleneck
and audit events are dropped by the Fabric OS.

• If the user name, IP address, or user interface is not transported, None is used instead for
each of the respective fields.

• For High Availability, the audit event logs exist independently on both active and standby CPs.
The configuration changes that occur on the active CP are propagated to the standby CP and
take effect.

• Audit log configuration is also updated through a configuration download.
Before configuring an audit log, you must select the event classes you want audited.

NOTE

Only the active CP can generate audit messages because event classes being audited occur only on
the active CP. Audit messages cannot originate from other blades in a Backbone.
Switch names are logged for switch components and Backbone names for Backbone components.
For example, a Backbone name may be FWDL or RAS and a switch component name may be zone,
name server, or SNMP.
Pushed messages contain the administrative domain of the entity that generated the event. Refer
to the Fabric OS Message Reference for details on event classes and message formats. For more
information on setting up the system error log daemon, refer to the Fabric OS Troubleshooting and
Diagnostics Guide.

NOTE

If an AUDIT message is logged from the CLI, any environment variables will be initialized with proper
values for login, interface, IP and other session information. Refer to the Fabric OS Message
Reference for more information.

108

Fabric OS Administrator’s Guide
53-1002920-02

Audit log configuration

3

Verifying host syslog prior to configuring the audit log
Audit logging assumes that your syslog is operational and running. Before configuring an audit log,
you must perform the following steps to ensure that the host syslog is operational.
1. Set up an external host machine with a system message log daemon running to receive the
audit events that will be generated.
2. On the switch where the audit configuration is enabled, enter the syslogdIpAdd command to
add the IP address of the host machine so that it can receive the audit events.
You can use IPv4, IPv6, or DNS names for the syslogdIpAdd command.
3. Ensure the network is configured with a network connection between the switch and the
remote host.
4. Check the host syslog configuration. If all error levels are not configured, you may not see some
of the audit messages.

Configuring an audit log for specific event classes
1. Connect to the switch from which you want to generate an audit log and log in using an account
with admin permissions.
2. Enter the auditCfg --class command, which defines the specific event classes to be filtered.
switch:admin> auditcfg --class 2,4
Audit filter is configured.

3. Enter the auditCfg --enable command, which enables audit event logging based on the classes
configured in step 2.
switch:admin> auditcfg --enable
Audit filter is enabled.

To disable an audit event configuration, enter the auditCfg --disable command.
4. Enter the auditCfg --show command to view the filter configuration and confirm that the
correct event classes are being audited, and the correct filter state appears (enabled or
disabled).
switch:admin> auditcfg --show
Audit filter is enabled.
2-SECURITY
4-FIRMWARE

5. Enter the auditDump -s command to confirm that the audit messages are being generated.
Example of the syslog (system message log) output for audit logging
Oct 10 08:52:06 10.3.220.7 raslogd: AUDIT, 2008/10/10-08:20:19 (GMT),
[SEC-3020], INFO, SECURITY, admin/admin/10.3.220.13/telnet/CLI,
ad_0/ras007/FID 128, , Event: login, Status: success, Info: Successful login
attempt via REMOTE, IP Addr: 10.3.220.13.
Oct 10 08:52:23 10.3.220.7 raslogd: 2008/10/10-08:20:36, [CONF-1001], 13, WWN
10:00:00:05:1e:34:02:0c | FID 128, INFO, ras007, configUpload completed
successfully. All config parameters are uploaded.

Fabric OS Administrator’s Guide
53-1002920-02

109

3

Duplicate PWWN handling during device login

Oct 10 09:00:04 10.3.220.7 raslogd: AUDIT, 2008/10/10-08:28:16 (GMT),
[SEC-3021], INFO, SECURITY, admin/NONE/10.3.220.13/None/CLI, None/ras007/FID
128, , Event: login, Status: failed, Info: Failed login attempt via REMOTE, IP
Addr: 10.3.220.13.

Duplicate PWWN handling during device login
If a device attempts to log in with the same port WWN (PWWN) as another device on the switch, you
can configure whether the new login or the existing login takes precedence.
You can configure how duplicate PWWNs are handled by selecting an option in the Enforce
FLOGI/FDISC login prompt of the configure command:

• Setting 0: First login takes precedence over second login (default behavior).
• Setting 1: Second login overrides first login.
• Setting 2: The port type determines whether the first or second login takes precedence.

Setting 0, First login precedence
When setting 0 is selected, the first login takes precedence over the second. This is the default
behavior. Table 9 describes the behavior when setting 0 is selected.

TABLE 9

Duplicate PWWN behavior: First login takes precedence over second login

Input port

First port login is NPIV port

First port login is F_Port

FLOGI received

The new login is rejected and the new port is
persistently disabled.

The new login is rejected and the new port is
persistently disabled.

FDISC received The new FDISC is rejected.

The new FDISC is rejected.

Setting 1, Second login precedence
When setting 1 is selected, the second login takes precedence over the first. Table 10 describes
the behavior when setting 1 is selected.

TABLE 10

Duplicate PWWN behavior: Second login overrides first login

Input port

First port login is F_Port

First port login is NPIV port

FLOGI received

New login forces an explicit logout of original
login on the previous F_Port.
The previous F_Port is persistently disabled.

New login forces an explicit logout of original
FDISC on the previous NPIV port.

FDISC received New FDISC forces an explicit logout of original
login on the previous F_Port.
The previous F_Port is persistently disabled.

New FDISC forces an explicit logout of original
FDISC on the previous NPIV port.

Setting 2, Mixed precedence
When setting 2 is selected, the precedence depends on the port type of the first login:

• If the previous port is an F_Port, the first login takes precedence.
• If the previous port is an NPIV port, the second login overrides the first login.

110

Fabric OS Administrator’s Guide
53-1002920-02

Enabling forward error correction

TABLE 11

3

Duplicate PWWN behavior: Port type determines which login takes precedence

Input port

First port login is NPIV port

First port login is F_Port

FLOGI received

New login forces an explicit logout of original
FDISC on the previous NPIV port.

New login is rejected and the new port is
persistently disabled.

FDISC received New FDISC forces an explicit logout of original
FDISC on the previous NPIV port.

New FDISC is rejected.

Setting the behavior for handling duplicate PWWNs
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the switchDisable command to disable the switch.
3. Enter the configure command.
4. Enter y after the F_Port login parameters prompt.
F-Port login parameters (yes, y, no, n): [no] y

5. Enter one of the following options at the Enforce FLOGI/FDISC login prompt to select the
behavior for handling duplicate PWWNs.

• Enter 0 to have the first login take precedence over the second login (default).
• Enter 1 to have the second login override the first login.
• Enter 2 to have the port type determine the behavior.
If a duplicate login is received on an F_Port, the duplicate login is rejected and the old
login is preserved; if a duplicate login is received on an NPIV port, the newer login is
accepted.
Enforce FLOGI/FDISC login: (0..2) [0] 1

6. Respond to the remaining prompts, or press Ctrl + D to accept the other settings and exit.
7.

Enter the switchEnable command to re-enable the switch.

With any of these settings, detection of duplicate PWWNs results in a RASLog. Ports that are
restricted become persistently disabled, marked with the reason “Duplicate Port WWN detected”.

Enabling forward error correction
Forward error correction (FEC) provides a data transmission error control method by including
redundant data (error-correcting code) to ensure error-free transmission on a specified port or port
range. When FEC is enabled, it can correct one burst of up to 11-bit errors in every 2112-bit
transmission, whether the error is in a frame or a primitive.
The following considerations apply to FEC:

• FEC is supported on E_Ports on 16 Gbps-capable switches.
• FEC is supported on the N_Ports and F_Ports of an access gateway using RDY, Normal
(R_RDY), or Virtual Channel (VC_RDY) flow control modes.

• FEC is supported on F_Ports on a switch if the device attached supports FEC by using a
Brocade host bust adaptor (HBA) ).

• FEC is enabled by default.

Fabric OS Administrator’s Guide
53-1002920-02

111

3

Enabling forward error correction

• FEC enables automatically when negotiation with a switch detects FEC capability.
• FEC persists after driver reloads and system reboots.
• FEC functions with features such as QoS, trunking, and BB_Credit recovery.

FEC Limitations
The following limitations apply to FEC:

• FEC is configurable only on 16 Gbps-capable switches (Brocade 6505, 6510, 6520, M6505,
6547, and the Brocade DCX 8510 Backbone family).

• For switch to adaptor connections, FEC is supported only on 1860 and 1867 Fabric Adapter
ports operating in HBA mode connected to 16 Gbps Brocade switches running Fabric OS 7.1
and later.

• FEC is supported only on link speeds of 10 Gbps and 16 Gbps, regardless of whether the
platform is FEC capable.

• FEC is not supported in the following situations:
- When the HBA port is running on a 16 Gbps link. When the HBA port speed changes to
less than this, FEC is disabled.

-

For HBA ports operating in loop mode or in direct-attach configurations.
On ports with some DWDM devices.

Using the portCfgFec command
Use the following procedure to enable FEC.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portCfgFec command, specifying the port or range of ports on which FEC is to be
enabled.
portcfgfec --enable slot/port

To enable the FEC feature on a single port and display the configuration, enter the following
commands:
switch:admin> portcfgfec --enable 1
switch:admin> portcfgfec --show 1
Port: 1
FEC Capable: YES
FEC Configured: ON

Enabling forward error correction
To enable the FEC feature on a port range, enter the portCfgFec --enable command. In this
example, port 1 already has FEC enabled, and so it remains enabled.
switch:admin> portcfgfec --enable 0-8
Same configuration for port 1

112

Fabric OS Administrator’s Guide
53-1002920-02

Enabling forward error correction

3

Disabling forward error correction
To disable the FEC feature on a port range, enter the portCfgFec --disable command.
switch:admin> portcfgfec --disable 0-8

Enabling or disabling FEC for long-distance ports
To enable or disable FEC for long-distance ports, use portCfgLongDistance with the -fecEnable or
-fecDisable parameter as required.
switch:admin> portcfglongdistance 12/6 LS 1 -distance 100 -fecenable

Refer to Chapter 25, “Managing Long-Distance Fabrics” for more details on working with
long-distance ports.

Viewing current FEC settings
Enter the portCfgFec --show command to display the current FEC configuration.
switch:admin> portcfgfec --show 1
Port: 1
FEC Capable: YES
FEC Configured: ON

Fabric OS Administrator’s Guide
53-1002920-02

113

3

114

Enabling forward error correction

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

4

Routing Traffic

In this chapter
• Routing overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Inter-switch links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Gateway links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Routing policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Route selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Frame order delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Lossless Dynamic Load Sharing on ports . . . . . . . . . . . . . . . . . . . . . . . . . .
• Frame Redirection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

115
118
120
122
125
126
129
132

Routing overview
Data moves through a fabric from switch to switch and from storage to server along one or more
paths that make up a route. Routing policies determine the path for each frame of data.
Before the fabric can begin routing traffic, it must discover the route a packet should take to reach
the intended destination. Route tables are lists that indicate the next hop to which packets are
directed to reach a destination. Route tables include network addresses, the next address in the
data path, and a cost to reach the destination network. There are two kinds of routing protocols on
intranet networks, distance vector and link state.

• Distance vector is based on hop count. This is the number of switches that a frame passes
through to get from the source switch to the destination switch.

• Link state is based on a metric value based on a cost. The cost could be based on bandwidth,
line speed, or round-trip time.
With the link state protocol, switches that discover a route identify the networks to which they are
attached, receiving an initial route table from the principal switch. After an initial message is sent
out, the switch only notifies the others when changes occur.
It is recommended that no more than seven hops occur between any two switches. This limit is not
required or enforced by Fabric Shortest Path First (FSPF). Its purpose is to ensure that a frame is
not delivered to a destination after the Resource Allocation TimeOut Value (R_A_TOV) has expired.
Fabric OS supports unicast Class 2 and Class 3 traffic, multicast, and broadcast traffic. Broadcast
and multicast are supported in Class 3 only.

Fabric OS Administrator’s Guide
53-1002920-02

115

4

Routing overview

Paths and route selection
Paths are possible ways to get from one switch to another. Each inter-switch link (ISL) has a metric
cost based on bandwidth. The cumulative cost is based on the sum of all costs of all traversed ISLs.
Route selection is the path that is chosen. Paths that are selected from the routing database are
chosen based on the minimal cost.

FSPF
Fabric Shortest Path First (FSPF) is a link state path selection protocol that directs traffic along the
shortest path between the source and destination based upon the link cost. FSPF is also referred
to as Layer 2 routing. FSPF detects link failures, determines the shortest route for traffic, updates
the routing table, provides fixed routing paths within a fabric, and maintains correct ordering of
frames. FSPF also keeps track of the state of the links on all switches in the fabric and associates a
cost with each link. The protocol computes paths from a switch to all the other switches in the
fabric by adding the cost of all links traversed by the path, and chooses the path that minimizes the
costs. This collection of the link states, including costs, of all the switches in the fabric constitutes
the topology database or link state database.
Once established, FSPF programs the hardware routing tables for all active ports on the switch.
FSPF is not involved in frame switching. FSPF uses several frames to perform its functions.
Because it may run before fabric routing is set up, FSPF does not use the routing tables to
propagate the frames, but floods the frames throughout the fabric hop-by-hop. Frames are first
flooded on all the ISLs; as the protocol progresses, it builds a spanning tree rooted on the principal
switch. Frames are only sent on the principal ISLs that belong to the spanning tree. When there are
multiple ISLs between switches, the first ISL to respond to connection requests becomes the
principal ISL. Only one ISL from each switch is used as the principal ISL. Figure 5 shows the thick
red lines as principal ISLs, and thin green lines as regular ISLs.

FIGURE 5

Principal ISLs

NOTE

FSPF only supports 16 routes in a zone, including Traffic Isolation Zones.

116

Fabric OS Administrator’s Guide
53-1002920-02

Routing overview

4

FSPF makes minimal use of the ISL bandwidth, leaving virtually all of it available for traffic. In a
stable fabric, a switch transmits 64 bytes every 20 seconds in each direction. FSPF frames have
the highest priority in the fabric. This guarantees that a control frame is not delayed by user data
and that FSPF routing decisions occur very quickly during convergence.
FSPF guarantees a routing loop-free topology at all times. It is essential for a fabric to include many
physical loops because, without loops, there would not be multiple paths between switches, and
consequently no redundancy. Without redundancy, if a link goes down, part of the fabric is isolated.
FSPF ensures both that the topology is loop-free and that a frame is never forwarded over the same
ISL more than once.
FSPF calculates paths based on the destination domain ID. The fabric protocol must complete
domain ID assignments before routing can begin. ISLs provide the physical pathway when the
Source ID (SID) address has a frame destined to a port on a remote switch Destination ID (DID).
When an ISL is attached or removed from a switch, FSPF updates the route tables to reflect the
addition or deletion of the new routes.
As each host transmits a frame to the switch, the switch reads the SID and DID in the frame
header. If the domain ID of the destination address is the same as the switch (intra-switch
communications), the frame buffer is copied to the destination port and a credit R_RDY message is
sent to the host. The switch only needs to read word zero and word one of the Fibre Channel frame
to perform what is known as cut-through routing. A frame may begin to emerge from the output
port before it has been entirely received by the input port. The entire frame does not need to be
buffered in the switch.
If the destination domain ID is different from the source domain ID, then the switch consults the
FSPF route table to identify which local E_Port provides Fabric Shortest Path First (FSPF) to the
remote domain.

Fibre Channel NAT
Within an edge fabric or across a backbone fabric, the standard Fibre Channel FSPF protocol
determines how frames are routed from the source Fibre Channel (FC) device to the destination FC
device. The source or destination device can be a proxy device.
Fibre Channel fabrics require that all ports be identified by a unique port identifier (PID). In a single
fabric, FC protocol guarantees that domain IDs are unique, and so a PID formed by a domain ID and
area ID is unique within a fabric. However, the domain IDs and PIDs in one fabric may be duplicated
within another fabric, just as IP addresses that are unique to one private network are likely to be
duplicated within another private network.
In an IP network, a network router can maintain network address translation (NAT) tables to replace
private network addresses with public addresses when a packet is routed out of the private
network, and to replace public addresses with private addresses when a packet is routed from the
public network to the private network. The Fibre Channel routing equivalent to this IP-NAT is Fibre
Channel network address translation (FC-NAT). Using FC-NAT, the proxy devices in a fabric can have
PIDs that are different from the real devices they represent, allowing the proxy devices to have
appropriate PIDs for the address space of their corresponding fabric.

Fabric OS Administrator’s Guide
53-1002920-02

117

4

Inter-switch links

Inter-switch links
An inter-switch link (ISL) is a link between two switches, E_Port-to-E_Port. The ports of the two
switches automatically come online as E_Ports once the login process finishes successfully. For
more information on the login process, refer to Chapter 1, “Understanding Fibre Channel Services”.
You can expand your fabric by connecting new switches to existing switches. Figure 6 shows a new
switch being added into an existing fabric. The thick red line is the newly formed ISL.

FIGURE 6

New switch added to existing fabric

When connecting two switches together, Brocade recommends the best practice that the following
parameters are differentiated:

• Domain ID
• Switch name
• Chassis name
You must also verify the following fabric parameters are identical on each switch for a fabric to
merge:

•
•
•
•
•
•
•

R_A_TOV (Resource Allocation TimeOut Value)
E_D_TOV (Error Detect TimeOut Value)
Data Field Size
Sequence Level Switching
Disable Device Probing
Suppress Class F Traffic
Per-frame Route Priority

There are non-fabric parameters that must match as well, such as zoning. Some fabric services,
such as management server, must match. If the fabric service is enabled in the fabric, then the
switch you are introducing into the fabric must also have it enabled. If you experience a segmented
fabric, refer to the Fabric OS Troubleshooting and Diagnostics Guide to fix the problem.

118

Fabric OS Administrator’s Guide
53-1002920-02

Inter-switch links

4

Buffer credits
In order to prevent the dropping of frames in the fabric, a device can never send frames without the
receiving device being able to receive them, so an end-to-end flow control is used on the switch.
Flow control in Fibre Channel uses buffer-to-buffer credits, which are distributed by the switch.
When all buffer-to-buffer credits are utilized, a device waits for a VC_RDY or an R_RDY primitive
from the destination switch before resuming I/O. The primitive is dependent on whether you have
R_RDYs enabled on your switch using the portCfgISLMode command. When a device logs in to a
fabric, it typically requests from two to sixteen buffer credits from the switch, depending on the
device type, driver version, and configuration. This determines the maximum number of frames the
port can transmit before receiving an acknowledgement from the receiving device.
For more information on how to set the buffer-to-buffer credits on an extended link, refer to Chapter
5, “Buffer-to-Buffer Credits and Credit Recovery”.

Congestion versus over-subscription
Congestion occurs when a channel is bottlenecked and fully utilized. This kind of bottleneck is a
congestion bottleneck. You should be aware that “over-subscription” does not have the same
meaning as “congestion”. Over-subscription refers only to the potential for congestion; an
over-subscribed link may go through a lifetime of normal operation and never be congested. The
term over-subscription is not to be used in place of congestion, which is the actual contention for
bandwidth by devices through an ISL.

Virtual channels
Virtual channels create multiple logical data paths across a single physical link or connection. They are
allocated their own network resources such as queues and buffer-to-buffer credits. Virtual channel
technology is the fundamental building block used to construct Adaptive Networking services. For
more information on Adaptive Networking services, refer to Chapter 14, “Optimizing Fabric Behavior”.
Virtual channels are divided into three priority groups. P1 is the highest priority, which is used for
Class F, F_RJT, and ACK traffic. P2 is the next highest priority, which is used for data frames. The data
virtual channels can be further prioritized to provide higher levels of Quality of Service. P3 is the
lowest priority and is used for broadcast and multicast traffic. This example is illustrated in Figure 7.
Quality of Service (QoS) is a licensed traffic shaping feature available in Fabric OS. QoS allows the
prioritization of data traffic based on the SID and DID of each frame. Through the use of QoS zones, traffic
can be divided into three priorities: high, medium, and low, as shown in Figure 7. The seven data virtual
channels (VC8 through VC14) are used to multiplex data frames based upon QoS zones when congestion
occurs. For more information on QoS zones, refer to Chapter 14, “Optimizing Fabric Behavior”.

Fabric OS Administrator’s Guide
53-1002920-02

119

4

Gateway links

FIGURE 7

Virtual channels on a QoS-enabled ISL

Gateway links
A gateway merges SANs into a single fabric by establishing point-to-point E_Port connectivity
between two Fibre Channel switches that are separated by a network with a protocol such as IP or
SONET.
Except for link initialization, gateways are transparent to switches; the gateway simply provides
E_Port connectivity from one switch to another. Figure 8 shows two separate SANs, A-1 and A-2,
merged together using a gateway.

120

Fabric OS Administrator’s Guide
53-1002920-02

Gateway links

FIGURE 8

4

Gateway link merging SANs

By default, switch ports initialize links using the Exchange Link Parameters (ELP) mode 1. However,
gateways expect initialization with ELP mode 2, also referred to as ISL R_RDY mode. Therefore, to
enable two switches to link through a gateway, the ports on both switches must be set for ELP mode 2.
Any number of E_Ports in a fabric can be configured for gateway links, provided the following
guidelines are followed:

• All switches in the fabric use the core PID format, as described in “Configuring a link through a
gateway” on page 121.

• The switches connected to both sides of the gateway are included when determining
switch-count maximums.

• Extended links (those created using the Extended Fabrics licensed feature) are not supported
through gateway links.

Configuring a link through a gateway
1. Connect to the switch at one end of the gateway and log in using an account assigned to the
admin role.
2. Enter the portCfgIISLMode command.
3. Repeat step 1 and step 2 for any additional ports that are connected to the gateway.
4. Repeat this procedure on the switch at the other end of the gateway.

Fabric OS Administrator’s Guide
53-1002920-02

121

4

Routing policies

Example of enabling a gateway link on slot 2, port 3
ecp:admin> portcfgislmode 2/3, 1
Committing configuration...done.
ISL R_RDY Mode is enabled for port 3. Please make sure the PID
formats are consistent across the entire fabric.

Routing policies
By default, all routing protocols place their routes into a routing table. You can control the routes
that a protocol places into each table and the routes from that table that the protocol advertises by
defining one or more routing policies and then applying them to the specific routing protocol.
The routing policy is responsible for selecting a route based on one of three user-selected routing
policies:

• Port-based routing
• Exchange-based routing
• Device-based routing
Notes
Routing is handled by the FSPF protocol and routing policy.
Each switch can have its own routing policy and different policies can exist in the same fabric.

ATTENTION
For most configurations, the default routing policy is optimal and provides the best performance. You
should change the routing policy only if there is a significant performance issue, or a particular fabric
configuration or application requires it.

Displaying the current routing policy
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aptPolicy command with no parameters.
The current policy is displayed, followed by the supported policies for the switch.
In the following example, the current policy is exchange-based routing (3) with the additional AP
Dedicated Link policy.
switch:admin> aptpolicy
Current Policy: 3
3 : Default Policy
1: Port Based Routing Policy
2: Device Based Routing Policy (FICON support only)
3: Exchange Based Routing Policy
0: AP Shared Link Policy
1: AP Dedicated Link Policy

122

Fabric OS Administrator’s Guide
53-1002920-02

Routing policies

4

Port-based routing
The choice of routing path is based only on the incoming port and the destination domain. To
optimize port-based routing, Dynamic Load Sharing (DLS) can be enabled to balance the load
across the available output ports within a domain.

NOTE
For FC routers only: When an FC router is in port-based routing mode, the backbone traffic is
load-balanced based on SID and DID. When an FC router is in exchange-based routing mode, the
backbone traffic is load-balanced based on SID, DID, and OXID.
Whatever routing policy a switch is using applies to the VE_Ports as well. For more information on
VE_Ports, refer to the Fibre Channel over IP Administrator’s Guide.

Exchange-based routing
The choice of routing path is based on the Source ID (SID), Destination ID (DID), and Fibre Channel
originator exchange ID (OXID) optimizing path utilization for the best performance. Thus, every
exchange can take a different path through the fabric. Exchange-based routing requires the use of
the Dynamic Load Sharing (DLS) feature; when this policy is in effect, you cannot disable the DLS
feature.
Exchange-based routing is also known as Dynamic Path Selection (DPS). For more information on
DPS refer to “Dynamic Path Selection” on page 124.

Device-based routing
Device-based routing optimizes routing path selection and utilization based on the Source ID (SID)
and Destination ID (DID) of the path source and destination ports. As a result, every distinct flow in
the fabric can take a different path through the fabric. Effectively, device-based routing works the
same as exchange-based routing but does not use the OXID field. This helps to ensure that the
exchanges between a pair of devices stay in order.

NOTE
Device-based routing requires the use of Dynamic Load Sharing (DLS); when this policy is in effect,
you cannot disable the DLS feature.
Device-based routing is also a form of Dynamic Path Selection (DPS). For more information on DPS
refer to “Dynamic Path Selection” on page 124.

NOTE

Device-based routing is supported in FICON environments, and in open environments only when
FICON coexists.

Fabric OS Administrator’s Guide
53-1002920-02

123

4

Routing policies

Dynamic Path Selection
DPS assigns communication paths between end devices in a fabric to egress ports in ratios
proportional to the potential bandwidth of the ISL, ICL, or trunk group. When there are multiple
paths to a destination, the input traffic is distributed across the different paths in proportion to the
bandwidth available on each of the paths. This improves utilization of the available paths, thus
reducing possible congestion on the paths. Every time there is a change in the network (which
changes the available paths), the input traffic can be redistributed across the available paths.
This is a very easy and non-disruptive process when the exchange-based routing policy is engaged.

AP route policies
Two additional AP policies are supported under exchange-based routing:

• AP Shared Link policy (default)
• AP Dedicated Link policy
NOTE

AP policies are independent of routing policies. Every routing policy supports both AP policies.

The AP Dedicated Link policy relieves internal congestion in an environment in which:

• There is a large amount of traffic going through both directions at the same time.
• There is a reduction of the effect of slow devices on the overall switch performance.
It is recommended that the default AP Shared Link policy be used for most environments. Also, it is
recommended that you design a SAN that localizes host-to-target traffic by reducing the amount of
traffic through the router.

ATTENTION
Setting either AP route policy is a disruptive process.

Routing in Virtual Fabrics
Virtual Fabrics (VF) supports DPS on all partitions. DPS is limited where multiple paths are
available for a logical fabric frame entering a Virtual Fabrics chassis from a base fabric that is sent
out using one of the dedicated ISLs in a logical switch.
The AP policy affecting the DPS behavior, whether it is exchange-based, device-based, or
port-based, is configured on a per-logical switch basis. In-order delivery (IOD) and DLS settings are
set per logical switch as well. IOD and DLS settings for the base switch affect all traffic going over
the base fabric including any logical fabric traffic that uses the base fabric.

CAUTION
Setting the routing policy is disruptive to the fabric because it requires that you disable the switch
where the routing policy is being changed.

124

Fabric OS Administrator’s Guide
53-1002920-02

Route selection

4

Setting the routing policy
Use the following procedure to set the routing policy.
1. Connect to the VF switch and log in as admin.
2. Enter the setcontext [FID | switchname] command for the correct Fabric ID or switch name.

• The fabricID parameter is the FID of the logical switch you just created.
• The switchname parameter is the name assigned to the logical switch.
• You can only use one parameter at a time.
switch:admin> setcontext 20

3. Enter the switchDisable command to disable the switch.
4. Take the appropriate following action based on the AP route policy you choose to implement:

• If the exchange-based policy is required, enter the aptPolicy 3 command.
• If the port-based policy is required, enter the aptPolicy 1 command.

Setting the AP route policy
The AP route policy can only be set in the base switches that are using Virtual Fabrics.
Use the following procedure to set the AP route policy.
1. Connect to the base switch and log in as admin.
2. Enter the switchDisable command to disable the switch.
3. Take the appropriate following action based on the AP route policy you choose to implement:

• If the AP Shared Link policy (default) is required, enter the aptPolicy -ap 0 command.
• If the AP Dedicated Link policy is required, enter the aptPolicy -ap 1 command.

Route selection
Selection of specific routes can be dynamic, so that the router can constantly adjust to changing
network conditions; or it may be static, so that data packets always follow a predetermined path.

Dynamic Load Sharing
The Fabric OS Dynamic Load Sharing (DLS) feature for dynamic routing path selection is required
by the exchange-based and device-based routing policies. When using these policies, DLS is
enabled by default and cannot be disabled. In other words, you cannot enable or disable DLS when
the exchange-based routing policy is in effect.
When the port-based policy is in force, you can enable DLS to optimize routing. When DLS is
enabled, it shares traffic among multiple equivalent paths between switches. DLS recomputes load
sharing when any of the following occurs:

• A switch boots up
• An E_Port goes offline and online

Fabric OS Administrator’s Guide
53-1002920-02

125

4

Frame order delivery

• An EX_Port goes offline
• A device goes offline

Setting DLS
Use the following procedure to set DLS.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the dlsShow command to view the current DLS setting.
One of the following messages appears:

• “DLS is set” indicates that DLS is turned on.
• “DLS is not set” indicates that DLS is turned off.
• ”DLS is set with Lossless enabled.” DLS is enabled with the Lossless feature. Load sharing
is recomputed with every change in the fabric, and existing routes can be moved to
maintain optimal balance. In Lossless mode, no frames are lost during this operation.

• “DLS is set by default with current routing policy. DLS is set with Lossless enabled.” The
current routing policy (exchange-based) requires DLS to be enabled by default. In addition,
the Lossless option is enabled. Frame loss is prevented during a load sharing
recomputation. If you get this message, you cannot perform step 3, so you are done with
this procedure.
3. Enter the dlsSet command to enable DLS or enter the dlsReset command to disable it.
Example of setting and resetting DLS
switch:admin> dlsshow
DLS is not set
switch:admin> dlsset
switch:admin> dlsshow
DLS is set
switch:admin> dlsreset
switch:admin> dlsshow
DLS is not set

Frame order delivery
The order in which frames are delivered is maintained within a switch and determined by the
routing policy in effect. The frame delivery behaviors for each routing policy are:

• Port-based routing
All frames received on an incoming port destined for a destination domain are guaranteed to
exit the switch in the same order in which they were received.

• Exchange-based routing
All frames received on an incoming port for a given exchange are guaranteed to exit the switch
in the same order in which they were received. Because different paths are chosen for
different exchanges, this policy does not maintain the order of frames across exchanges.

• Device-based routing
All frames received on an incoming port for a given pair of devices are guaranteed to exit the
switch in the same order in which they were received.

126

Fabric OS Administrator’s Guide
53-1002920-02

Frame order delivery

4

If even one switch in the fabric delivers out-of-order exchanges, then exchanges are delivered to the
target out of order, regardless of the policy configured on other switches in the fabric.

NOTE

Some devices do not tolerate out-of-order exchanges; in such cases, use the port-based routing
policy.
In a stable fabric, frames are always delivered in order, even when the traffic between switches is
shared among multiple paths. However, when topology changes occur in the fabric (for example, if
a link goes down), traffic is rerouted around the failure, and some frames could be delivered out of
order. Most destination devices tolerate out-of-order delivery, but some do not.
By default, out-of-order frame-based delivery is allowed to minimize the number of frames dropped.
Enabling in-order delivery (IOD) guarantees that frames are either delivered in order or dropped.
You should only force in-order frame delivery across topology changes if the fabric contains
destination devices that cannot tolerate occasional out-of-order frame delivery.

Forcing in-order frame delivery across topology changes
Use the following procedure to force in-order frame delivery across topology changes.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the iodSet command.

NOTE

The iodSet command can cause a delay in the establishment of a new path when a topology
change occurs; use it with care.
3. Confirm the in-order delivery has been set by entering the iodShow command.

Restoring out-of-order frame delivery across topology changes
Use the following procedure to restore out-of-order frame delivery across topology changes.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the iodReset command.

Using Frame Viewer to understand why frames are dropped
When a frame is unable to reach its destination because of a timeout, it is discarded. You can use
Frame Viewer to find out which flows contained the dropped frames, which in turn can help you
determine which applications might be impacted. Frame Viewer allows you to see the exact time
(within one second) that the frames were dropped.
You can view and filter up to 20 discarded frames per chip per second for 1200 seconds using a
number of fields with the frameLog command.

Fabric OS Administrator’s Guide
53-1002920-02

127

4

Frame order delivery

Use the following procedure to view frames.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter frameLog --show.
EDCX16_114064:root> framelog --show
==================================================================================================
Fri Jul 13 23:47:08 UTC 2012
==================================================================================================
Log
TX
RX
timestamp
port
port
SID
DID
SFID
DFID
Type
Count
===============================================================================================
Jul 13 23:47:07
11/45 11/45
0xfffffd
0x40e580
0
0
timeout
2
Jul 13 23:47:07
11/45 11/45
0xfffffc
0x40e580
0
0
timeout
5
Jul 13 23:47:07
11/45 11/45
0xfffffc
0x40e580
0
0
timeout
3
Jul 13 23:47:07
11/45 11/45
0xfffc40
0x40e580
0
0
timeout
2
Jul 13 23:47:07
11/45 11/45
0xfffc40
0x40e580
0
0
timeout
1

Using the frameLog --show command
The output of --show displays the type of each discard.
The -type option of the frameLog --show command requires an argument, but only timeout is
supported at present. The timeout argument specifies that only timeout discards are shown.

Filtering results by back-end port in Frame Viewer
The Frame Viewer --show command supports specifying that the TX port or RX port of displayed
frames should be a back-end port. To filter by TX port or RX port, use a following command:
framelog --show -txport [slot/]port
or
framelog --show -rxport [slot/]port
or
framelog --show -txport [slot/]port -rxport [slot/]port
The -txport and -rxport options accept the arguments “-1” (for fixed-port switches) or “-1/-1”
(for modular switches). These arguments stand for “any back-end port.” Using this notation, you
can select specifically those discarded frames that have a back-end port in the TX port or RX port
field.

NOTE
Individual back-end ports cannot be specified, only the quality of being a back-end port can be
specified.

128

Fabric OS Administrator’s Guide
53-1002920-02

Lossless Dynamic Load Sharing on ports

4

Lossless Dynamic Load Sharing on ports
Lossless Dynamic Load Sharing (DLS) allows you to rebalance port paths without causing
input/output (I/O) failures. For devices where in-order delivery (IOD) of frames is required, you can
set IOD separately. You can use this feature with the following hardware:

•
•
•
•
•
•
•
•
•
•
•
•
•
•

Brocade 300
Brocade 5100
Brocade 5300
Brocade 6505
Brocade 6510
Brocade 6520
Brocade M6505
Brocade 6547
Brocade VA-40FC
Brocade FC8-16, FC8-32, FC8-48, and FC8-64 port blades
Brocade DCX 8510 Backbone family and supported blades
Brocade FC16-32 and FC16-48 port blades
Brocade FC8-32E and FC8-48E port blades
Brocade FX8-24 application blades in the Brocade DCX and DCX-4S Backbones

On the Brocade 7800 switch and the FX8-24 application blade, Lossless DLS is supported only on
FC-to-FC port flows.

ATTENTION
When you implement Lossless DLS, the switches in the fabric must have either Fabric OS v6.3.0 or
Fabric OS v6.4.0 or later installed to guarantee no frame loss.
Lossless DLS must be implemented along the path between the target and the initiator. You can
use Lossless DLS on ports connecting switches to perform the following functions:

• Eliminate dropped frames and I/O failures by rebalancing the paths going over the ISLs
whenever there is a fabric event that might result in suboptimal utilization of the ISLs.

• Eliminate the frame delay caused by establishing a new path when a topology change occurs.
Lossless mode means no frame loss during a rebalance and only takes effect if DLS is enabled.
Lossless DLS can be enabled on a fabric topology to have zero frame drops during rebalance
operations. If the end device also requires the order of frames to be maintained during the
rebalance operation, then IOD must be enabled. However, this combination of Lossless DLS and
IOD is supported only in specific topologies, such as in a FICON environment.
You can disable or enable IOD when Lossless DLS is enabled. You can also choose between
exchange- or port-based policies with Lossless DLS. The following events cause a rebalance:

•
•
•
•

Adding an E_Port
Adding a slave E_Port
Removing an E_Port (However, frame loss occurs on traffic flows to this port.)
Removing an F_Port (However, frame loss occurs on traffic flows to this port.)

Fabric OS Administrator’s Guide
53-1002920-02

129

4

Lossless Dynamic Load Sharing on ports

Lossless DLS does the following whenever paths need to be rebalanced:
1. Pauses ingress traffic by not returning credits. Frames that are already in transit are not
dropped.
2. Changes the existing path to a more optimal path.
3. If IOD is enabled, waits for sufficient time for frames already received to be transmitted. This is
needed to maintain IOD.
4. Resumes traffic.
Table 12 shows the effect of frames when you have a specific routing policy turned on with IOD.

TABLE 12

Combinations of routing policy and IOD with Lossless DLS enabled

Policy

IOD

Rebalance result with Lossless DLS enabled

Port-based

Disabled

No frame loss, but out-of-order frames may occur.

Port-based

Enabled

No frame loss and no out-of-order frames. Topology restrictions apply. Intended
for FICON environment.

Exchange-based

Disabled

No frame loss, but out-of-order frames may occur.

Exchange-based

Enabled

No frame loss and no out-of-order frames. Topology restrictions apply. Intended
for FICON environment.

Device-based

Disabled

No frame loss, but out-of-order frames may occur.

Device-based

Enabled

No frame loss and no out-of-order frames. Topology restrictions apply. Intended
for FICON environment.

Lossless core
Lossless core works with the default configuration of the Brocade DCX 8510-8 and DCX 8510-4
hardware to prevent frame loss during a core blade removal and insertion. This feature is on by
default and cannot be disabled. Lossless core has the following limitations:

• Only supported with IOD disabled, which means Lossless core cannot guarantee in-order
delivery of exchanges

• ICL limitations
• Traffic flow limitations

ICL limitations
If ICL ports are connected during a core blade removal, it is equivalent to removing external E_Ports
which may cause I/O disruption on the ICL ports that have been removed.
If ICL ports are connected during a core blade insertion, it is equivalent to adding external E_Ports
which may cause I/O disruption because of reroutes. Lossless DLS, if enabled, takes effect to
prevent I/O disruption.

Traffic flow limitations
FA4-18 AP blades, which are supported on the Brocade DCX and DCX-4S devices, may continue to
experience frame drops after core blade removal or insertion. The path between an FA4-18 blade
and an FX8-24 blade, or vice versa, experiences I/O disruption because the FA4-18 blades do not
support this feature.

130

Fabric OS Administrator’s Guide
53-1002920-02

Lossless Dynamic Load Sharing on ports

4

Configuring Lossless Dynamic Load Sharing
You configure Lossless DLS switch- or chassis-wide by using the dlsSet command to specify that no
frames are dropped while rebalancing or rerouting traffic.
Use the following procedure to configure Lossless Dynamic Load Sharing.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the appropriate dlsSet command to enable or disable Lossless Dynamic Load Sharing.
switch:admin> dlsset --enable -lossLess
switch:admin> dlsset --disable -lossLess

Lossless Dynamic Load Sharing in Virtual Fabrics
Enabling Lossless Dynamic Load Sharing is optional on logical switches in Virtual Fabrics. If you
enable this feature, it must be on a per-logical switch basis and can affect other logical switches in
the fabric. XISL use must be disabled for Lossless DLS to be enabled.

How DLS affects other logical switches in the fabric
On a Brocade DCX platform, logical switch 1 consists of ports 0 through 5 in slot 1. Logical switch 2
consists of ports 6 through 10 in slot 1. The Lossless DLS feature is enabled on logical switch 1.
Because ports 0 through 10 in slot 1 belong to a logical switch where Lossless DLS is enabled, the
traffic in logical switch 2 is affected whenever traffic for logical switch 1 is rebalanced.

ATTENTION
Although Lossless DSL is enabled for a specific logical switch, you must have chassis-level
permissions to use this feature.
The effect on logical switch 2 is based on the configuration on logical switch 2:

• If logical switch 2 has IOD enabled (iodSet only), IOD is enforced.
• If logical switch 2 has Lossless DLS enabled, traffic is paused and resumed.
• If logical switch 2 has no IOD (iodReset), traffic is paused and resumed.
To avoid this behavior, it is recommended to define your logical switches as follows:

• Define logical switches that require Lossless DLS at the blade boundary.
• Define logical switches that require Lossless DLS only using supported blades. For example,
do not use blades that support IOD, but do not support Lossless DLS.
For more information on Virtual Fabrics and chassis-level permissions, refer to Chapter 11,
“Managing Virtual Fabrics”.

Fabric OS Administrator’s Guide
53-1002920-02

131

4

Frame Redirection

Frame Redirection
Frame Redirection provides a means to redirect traffic flow between a host and a target that use
virtualization and encryption applications, such as the Brocade SAS blade and Brocade Data
Migration Manager (DMM), so that those applications can perform without having to reconfigure
the host and target. You can use this feature if the hosts and targets are not directly attached.
Frame Redirection depends on the wide distribution of the Defined Zone Database. The Defined
Zone Database on Fabric OS switches is pushed out to all other Fabric OS switches in the fabric
that support Frame Redirection. Redirection zones exist only in the defined configuration and
cannot be added to the effective configuration.

NOTE

Fabric OS v7.2.0 is not supported on the Brocade 7600 or Brocade SAS blade. However, this
hardware can run in a pre-Fabric OS v7.2.0 system and attach to a Fabric OS v7.2.0 fabric.
Frame Redirection uses a combination of special frame redirection zones and name server
changes to spoof the mapping of real device WWNs to virtual PIDs.

FIGURE 9

Single host and target

Figure 9 demonstrates the flow of Frame Redirection traffic. A frame starts at the host with a
destination to the target. The port where the appliance is attached to the host switch acts as the
virtual initiator and the port where the appliance is attached to the target switch is the virtual target.

Creating a frame redirect zone
The first time the zone –-rdcreate command is run, the following zone objects are created by default:

• The base zone object, “red_______base”.
• The redirect (RD) zone configuration, “r_e_d_i_r_c__fg”.
NOTE
Frame redirect zones are not supported with D or I initiator target zones.

ATTENTION
Prior to creating the frame redirect zone, you must create a Layer 2 zone for the Initiator (host) and
Target (storage). This zone must be part of the effective configuration and must be defined using the
port World Wide Name (WWN). Refer to “Creating a zone” on page 350, and “Enabling a zone
configuration” on page 364.
Use the following procedure to create a frame redirect zone.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zone –-rdcreate command.

132

Fabric OS Administrator’s Guide
53-1002920-02

Frame Redirection

4

3. Enter the cfgSave command to save the frame redirect zones to the defined configuration.
The following example creates a redirect zone, given a host (10:10:10:10:10:10:10:10), target
(20:20:20:20:20:20:20:20), virtual initiator (30:30:30:30:30:30:30:30), and virtual target
(40:40:40:40:40:40:40:40):
switch:admin>zone --rdcreate 10:10:10:10:10:10:10:10 20:20:20:20:20:20:20:20 \
30:30:30:30:30:30:30:30 40:40:40:40:40:40:40:40 restartable noFCR

Deleting a frame redirect zone
Use the following procedure to delete a frame redirect zone.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zone --rddelete command to remove the base redirect zone object,
“red_______base”.

NOTE

When the base zone is removed, the redirect zone configuration “r_e_d_i_r_c__fg” is removed
as well.
3. Enter the cfgSave command to save changes to the defined configuration.
Example of deleting a frame redirect zone
switch:admin> zone --rddelete \
red_0917_10_10_10_10_10_10_10_10_20_20_20_20_20_20_20_20

Viewing frame redirect zones
Use the following procedure to view frame redirect zones.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgShow command.

Fabric OS Administrator’s Guide
53-1002920-02

133

4

134

Frame Redirection

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

Buffer-to-Buffer Credits and Credit Recovery

5

In this chapter
• Buffer credit management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
• Buffer credit recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
• Credit loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Buffer credit management
Buffer-to-buffer credit management affects performance over distances; therefore, allocating a
sufficient number of buffer credits for long-distance traffic is essential to performance.
To prevent a target device (either host or storage) from being overwhelmed with frames, the Fibre
Channel architecture provides flow control mechanisms based on a system of credits. Each of
these credits represents the ability of the device to accept additional frames. If a recipient issues
no credits to the sender, no frames can be sent. Pacing the transport of subsequent frames on the
basis of this credit system helps prevent the loss of frames and reduces the frequency of entire
Fibre Channel sequences needing to be retransmitted across the link.
Because the number of buffer credits available for use within each port group is limited,
configuring buffer credits for extended links may affect the performance of the other ports in the
group used for core-to-edge connections. You must balance the number of long-distance ISL
connections and core-to-edge ISL connections within a switch.

NOTE
Configuring long-distance ISLs between core and edge switches is possible, but is not a
recommended practice.
All switch ports provide protection against buffer depletion through buffer limiting. A buffer-limited
port reserves a minimum of eight buffer credits, allowing the port to continue to operate rather
than being disabled because of a lack of buffers.
Buffer-limited operations are supported for the static mode (LS) and dynamic mode (LD) extended
ISL modes only. For LD, distance in kilometers is the smaller of the distance measured during port
initialization versus the desired_distance value. For LS, distance in kilometers is always the
desired_distance value.

Buffer-to-buffer flow control
Buffer-to-buffer (BB) credit flow control is implemented to limit the amount of data that a port may
send, and is based on the number and size of the frames sent from that port. Buffer credits
represent finite physical-port memory. Within a fabric, each port may have a different number of
buffer credits. Within a connection, each side may have a different number of buffer credits.

Fabric OS Administrator’s Guide
53-1002920-02

135

5

Buffer credit management

Buffer-to-buffer flow control is flow control between adjacent ports in the I/O path, for example,
transmission control over individual network links. A separate, independent pool of credits is used
to manage buffer-to-buffer flow control. A sending port uses its available credit supply and waits to
have the credits replenished by the port on the opposite end of the link. These buffer credits are
used by Class 2 and Class 3 services and rely on the Fibre Channel Receiver-Ready (R_RDY) control
word to be sent by the receiving link port to the sender. The rate of frame transmission is regulated
by the receiving port, and is based on the availability of buffers to hold received frames.
If Virtual Channel technology is in use, the VC_RDY or EXT_VC control word is used instead of the
R_RDY control word to manage buffer credits. For Virtual Channels, the buffer credits are managed
for each Virtual Channel, and not for the entire physical link.
The Virtual Channels used in VC_RDY flow-control mode range from VC0 through VC7. When QoS is
enabled, EXT_VC_RDY flow-control mode allocates VC0 through VC14. VC8 through VC14 are
allocated specifically for QoS VCs.
Upon arriving at a receiver, a frame goes through several steps. It is received, deserialized, and
decoded, and is stored in a receive buffer where it is processed by the receiving port. If another
frame arrives while the receiver is processing the first frame, a second receive buffer is needed to
hold this new frame. Unless the receiver is capable of processing frames as fast as the transmitter
is capable of sending them, it is possible for all of the receive buffers to fill up with received frames.
At this point, if the transmitter should send another frame, the receiver will not have a receive
buffer available and the frame is lost. Buffer-to-buffer flow control provides consistent and reliable
frame delivery of information from sender to receiver.

Optimal buffer credit allocation
The optimal number of buffer credits is determined by the distance (frame delivery time), the
processing time at the receiving port, the link signaling rate, and the size of the frames being
transmitted. As the link speed increases, the frame delivery time is reduced and the number of
buffer credits must be increased to obtain full link utilization, even in a short-distance environment.
For each frame that is transferred, the hardware at the other end must acknowledge that the frame
has been received before a successful transmission occurs. This flow requires enough capacity in
the hardware to allow continuous transmission of frames on the link, while waiting for the
acknowledgment to be sent by the receiver at the other end.
As the distance between switches and the link speed increases, additional buffer credits are
required for the ports used for long-distance connections. Distance levels define how buffer credits
are allocated and managed for extended ISLs. Buffer credits are managed from a common pool
available to a group of ports on a switch. The buffer credit can be changed for specific applications
or operating environments, but it must be in agreement among all switches to allow formation of
the fabric.
Smaller frame sizes need more buffer credits. Two commands are available to help you determine
whether you need to allocate more buffer credits to handle the average frame size. The
portBufferShow command calculates the average frame size. The portBufferCalc command uses
the average frame size with the speed and link distance to determine the number of buffer credits
needed.

136

Fabric OS Administrator’s Guide
53-1002920-02

Buffer credit management

5

Considerations for calculating buffer credits
Considerations follow for calculating how many ports can be configured for long distance on all
Fabric OS v7.x-capable switch modules:

• Each port is part of a port group that includes a pool of buffer credits that can be used. This
port group is not the same as the port groups used for ISL Trunking.

• Each user port reserves eight buffer credits when online or offline.
• Any remaining buffers can be reserved by any port in the port group.
• When QoS is enabled and the port is online, additional buffers are allocated to that port. Refer
to “Calculating the number of buffers required based on full-size frames” on page 138 and
“Configuring buffers for a single port directly” on page 141 for more information.

Fibre Channel gigabit values reference definition
Use the following Fibre Channel gigabit values to calculate buffer requirements:
Table 13 shows the Fibre Channel gigabit values used to calculate buffer requirements.

TABLE 13

Fibre Channel gigabit values

Gigabit
value

Buffer requirements

1 Gbps

1.0625

2 Gbps

2.125

4 Gbps

4.25

8 Gbps

8.5

10 Gbps 10.625
16 Gbps 17

Buffer credit allocation based on full-size frames
Assuming that the frame is a full-size frame, one buffer credit allows a device to send one payload
up to 2,112 bytes (2,148 with headers). Assuming that each payload is 2,112, you need one credit
per 1 km of link length at 2 Gbps (smaller payloads require additional buffer credits to maintain
link utilization). Refer to“Allocating buffer credits based on average-size frames” on page 140 for
additional information.

Fibre Channel data frames
The final frame size must be a multiple of 4 bytes. If the data (payload) needs to be segmented, it
will be padded with 1 to 3 “fill-bytes” to achieve an overall 4-byte frame alignment. The standard
frame header size is 24 bytes. If applications require extensive control information, up to 64
additional bytes (for a total of an 88-byte header) can be included. Because the total frame size
cannot exceed the maximum of 2,148 bytes, the additional header bytes will subtract from the data
segment by as much as 64 bytes (per frame). This is why the maximum data (payload) size is 2,112
(because [2,112 – 64] = 2,048, which is 2 kb of data). The final frame, after it is constructed, is
passed through the 8-byte-to-10-byte conversion process.

Fabric OS Administrator’s Guide
53-1002920-02

137

5

Buffer credit management

Table 14 describes Fibre Channel data frames.

TABLE 14

Fibre Channel data frames

Fibre Channel frame fields

Field size

Final frame size

Start of frame

4 bytes

32 bits

Standard frame header

24 bytes

192 bits

Data (payload)

0–2,112 bytes

0–16,896 bits

CRC

4 bytes

32 bits

End of frame

4 bytes

32 bits

Total (number bits/frame)

36–2,148 bytes

288–7,184 bits

Allocating buffer credits based on full-sized frames
You can allocate buffer credits based on distance using the portCfgLongDistance command. The
long-distance link modes allow you to select the dynamic mode (LD) or the static mode (LS) to
calculate the buffer credits.
For LD, the estimated distance in kilometers is the smaller of the distance measured during port
initialization versus the desired_distance parameter, which is required when a port is configured as
an LD or an LS mode link. A best practice is to use LS over LD. The assumption that Fibre Channel
payloads are consistently 2,112 bytes is not realistic in practice. To gain the proper number of
buffer credits with the LS mode, there must be enough buffer credits available in the pool, because
Fabric OS will check before accepting a value.

NOTE
The desired_distance parameter of the portCfgLongDistance command’s is the upper limit of the
link distance and is used to calculate buffer availability for other ports in the same port group. When
the measured distance exceeds the value of desired_distance, this value is used to allocate the
buffers. In this case, the port operates in degraded mode instead of being disabled as a result of
insufficient buffer availability. In LS mode, the actual link distance is not measured; instead, the
desired_distance value is used to allocate the buffers required for the port.
Refer to the data in Table 15 on page 143 and Table 16 on page 144 to get the total ports in a
switch or blade, the number of user ports in a port group, and the unreserved buffer credits
available per port group. The values reflect an estimate, and may differ from the supported values
in Table 16.

Calculating the number of buffers required based on full-size frames
Use the following procedure to calculate the number of buffers required for a long-distance
connection:
1. Determine the desired distance in kilometers of the switch-to-switch connection.
2. Determine the speed that you will use for the long-distance connection.

138

Fabric OS Administrator’s Guide
53-1002920-02

Buffer credit management

5

3. Use one of the following formulas to calculate the reserved buffers for distance:

• If QoS is enabled:
(Reserved Buffer for Distance Y) = (X * LinkSpeed / 2) + 6 + 14

• If QoS is not enabled:
(Reserved Buffer for Distance Y) = (X * LinkSpeed / 2) + 6
The formulas use the following parameters:
X = The distance determined in step 1 (in km).
LinkSpeed = The speed of the link determined in step 2.
6 = The number of buffer credits reserved for fabric services, multicast, and broadcast
traffic. This number is static.
14 = The number of buffer credits reserved for QoS. This number is static.
Using 50 km as the desired distance of the switch-to-switch connection and 2 Gbps as the
speed of the long-distance connection, insert the numbers into the appropriate formula. The
formula should read as follows:
(50 km * 2 Gbps / 2) + 6 = 56 buffers, which is the number of buffers reserved for distance.
The following examples use different speeds, all based on a distance of 50 km. The distances and
speeds are variables that can change depending on how your network is set up.

•
•
•
•
•
•

If you have a distance of 50 km at 1 Gbps, then (50 km * 1 Gbps / 2) + 6 = 31 buffers.
If you have a distance of 50 km at 2 Gbps, then (50 km * 2 Gbps / 2) + 6 = 56 buffers.
If you have a distance of 50 km at 4 Gbps, then (50 km * 4 Gbps / 2) + 6 = 106 buffers.
If you have a distance of 50 km at 8 Gbps, then (50 km * 8 Gbps / 2) + 6 = 206 buffers.
If you have a distance of 50 km at 10 Gbps, then (50 km * 10 Gbps / 2) +6 = 256 buffers.
If you have a distance of 50 km at 16 Gbps, then (50 km * 16 Gbps / 2) + 6 = 406 buffers.

Example

Consider the Brocade 300, which has a single 24-port port group and a total of 676 buffer credits
for that port group. The formulas use the following parameters:
24 = The number of user ports in a port group retrieved from Table 15 on page 143
8 = The number of reserved credits for each user port
676 = The number of buffer credits available in the port group
The maximum remaining number of buffer credits for the port group, after each port reserves its 8
buffer credits, is obtained from the following formula:
676 – (24 * 8) = 484 unreserved buffer credits
492 buffers to a single port (484 + 8 [8 for the reserved buffers already allocated to that user
port]), you can calculate the maximum single-port extended distance supported:
Maximum Distance X (in km) = (BufferCredits + 6) * 2 / LinkSpeed
498 km = (492 + 6 buffers for Fabric Services) * 2 / 2 Gbps
If you have a distance of 50 km at 8 Gbps, then 484 / (206 – 8) = 2 ports.

Fabric OS Administrator’s Guide
53-1002920-02

139

5

Buffer credit management

The following values are used in the example:

• 484 — The total number of unreserved buffer credits
• 206 — Buffer credits needed for 50 km at 8 Gbps
• 8 — The number of reserved buffer credits already allocated to that port
The resulting number is rounded down to the next whole number because fractions of a port are
not allowed.
If you have a distance of 50 km at 1 Gbps, then 484 / (31 – 8) = 21 ports.

Allocating buffer credits based on average-size frames
In cases where the frame size is average, for example 1,024 bytes, you must allocate twice the
buffer credits or configure twice the distance in the long-distance LS configuration mode. Refer to
“Fibre Channel gigabit values reference definition” on page 137 for an approximation of the
calculated number of buffer credits.
1. Use the following formula to calculate the value for the desired_distance parameter needed for
Fabric OS to determine the number of buffer credits to allocate:
desired_distance = roundup [(real_estimated_distance * 2112) / average_payload_size]
The average_payload_size in this equation uses 1024 bytes
If the real estimated distance is 100 km, the desired_distance is 207.
desired_distance = roundup [(100 * 2112) / 1024] = 207
When configuring the LS mode with the portCfgLongDistance command, enter a
desired_distance value of 207 for an actual 100-km link connected to an 8-Gbps E_Port. This
causes Fabric OS to allocate the correct number of buffer credits.
2. Determine the speed you will use for the long-distance connection. This example uses 8 Gbps.
3. Look up the data_rate value for the speed of the connection. Refer to “Fibre Channel gigabit
values reference definition” on page 137 to determine the data_rate value.
For example, the data_rate is 8.5 for a speed of 8 Gbps.
4. Use the following formula to calculate the number of buffer credits to allocate:
buffer credits = [desired_distance * (data_rate / 2.125)]
With the values for desired_distance and data_rate from step 1 and step 3, the value for buffer
credits is calculated as follows:
buffer credits = [207 * (8.5 / 2.125)] = 828

NOTE

This buffer credits formula does not work with LD mode because LD mode checks the distance and
limits the estimated distance to the real value of 100 km. LS mode allows for the necessary
desired_distance value based on the data size entered, regardless of the distance.

140

Fabric OS Administrator’s Guide
53-1002920-02

Buffer credit management

5

If buffer credit recovery is enabled, Fabric OS supports a BB_SC_N range of 1 to 15; therefore, it is
impossible for the desired_distance value to be more than the number of buffer credits available in
the pool as determined by the previous calculations The distance for buffer credit recovery is well
within the range of all possible connections. An estimated distance of 32,768 is considerably
higher than the available buffer credits and only lower values of desired_distance are permitted by
Fabric OS.

Configuring buffers for a single port directly
To configure the number of buffers directly, use the -buffers option of the portCfgLongDistance
command. Fabric OS uses this value to calculate the total number of buffers according to the
following formula:
Total Buffers = Configured Buffers + QOS_VC_Credits + Non-data_VC_Credits
Seven Virtual Channels (VCs) are required for each QoS port. Each VC requires two buffers. Thus,
the total number of QoS buffers required for a port is 14 (7*2). An additional 6 VCs are required for
nondata transmission (for example, control traffic). As a consequence, for a QoS port, 20 buffers
are added. For a non-QoS port, 6 buffers are added.
For example, if the configured number of buffers is 100, then the total number of buffers allocated
for a QoS port is 120, as shown in the following example.
Total Buffers = 100 + 14 + 6 = 120
If the configured number of buffers is 100, the total number of buffers allocated for a non-QoS port
is 106, as shown in the following example.
Total Buffers = 100 + 6 = 106

NOTE

You cannot use the -buffers option with the -distance option or the -frameSize option.
Example
switch:admin> portcfglongdistance 2/35 LS 1 -buffers 400
Reserved Buffers =
420

Configuring buffers using frame size
You can configure the number of buffers by using the -frameSize option of the portCfgLongDistance
command along with the -distance option. Fabric OS calculates the number of buffers from the
-frameSize option value according to the following formula:
buffers_required = (2048/framesize) * data_vc_credits
If you enter the average frame size of 1024, Fabric OS will allocate almost twice as many buffers as
for the maximum frame size of 2048.
The -frameSize option value is persistent across reboots and HA failover.
Example
switch:admin> portcfglongdistance 2/35 LS 1 –distance 100 –framesize 1024

Fabric OS Administrator’s Guide
53-1002920-02

141

5

Buffer credit management

Calculating the number of buffers required given the distance, speed,
and frame size
If you know the distance, speed, and frame size for a given port, you can use the portBufferCalc
command to calculate the number of buffers required. If you omit the distance, speed, or frame
size, the command uses the currently configured values for the port. Given the buffer requirement
and port speed, you can use the same distance and frame size values when using the
portCfgLongDistance command.
To determine the number of buffers required, complete the following steps:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portBufferCalc command and provide values for the distance, port speed, and frame
size.
The following example calculates the number of buffers required for an 8-Gbps port on a 100-km
link with an average frame size of 512 bytes.
switch:admin> portbuffercalc 9/4 -distance 100 -speed 8 -framesize 512
1606 buffers required for 100km at 8G and framesize of 512 bytes

Allocating buffer credits for F_Ports
The default configured F_Port buffer credit is fixed at eight buffers. You can use the
portCfgFPortBuffers command to configure a given port with the specified number of buffers.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgFPortBuffers command.
switch:admin> portcfgfportbuffers --enable 2/44 12

Note that in the sample commands provided in the following procedure, 12 buffers are
configured for an F_Port.
To disable the port buffer configuration and return to the default buffer allocation, use the --disable
option.
switch:admin> portcfgfportbuffers --disable 2/44

NOTE
The configured number of buffers for a given port is stored in the configuration database and is
persistent across reboots. The F_Port buffer feature does not support EX_Port, Port Mirroring,
Long-Distance, L_Port, FastWrite, QoS, and Trunk Area enabled ports. F_Port Buffers are mutually
exclusive to E_Port Credits

Monitoring buffers in a port group
Use the portBufferShow command to monitor the remaining buffers on a long-distance link and to
monitor the average frame size and average buffer usage for a given port.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portBufferShow command.

142

Fabric OS Administrator’s Guide
53-1002920-02

Buffer credit management

5

The average frame size in bytes is shown in parentheses with the average buffer usage for
packet transmission and reception.
switch:admin> portbuffershow 17
User
Port
Port
Type
----------------

Lx
Mode
----

Max/Resv
Buffers
-------

Avg Buffer Usage & FrameSize Buffer Needed
Tx
Rx
Usage Buffers
---------------------------- ------ -------

Link
Remaining
Distance Buffers
---------

64
8
- ( - )
- ( - )
0
65
8
- ( - )
- ( - )
0
66
8
- ( - )
- ( - )
0
67
8
- ( - )
- ( - )
0
68
E
LS
806
197(2012)
201(2044)
206
206
100km
69
E
8
1(2016)
1(2020)
26
26
2km
70
E
8
1(2012)
1(2036)
26
26
2km
71
E
8
1(2008)
2(2052)
26
26
2km
192
8
- ( - )
- ( - )
0
193
8
- ( - )
- ( - )
0
194
8
- ( - )
- ( - )
0
195
8
- ( - )
- ( - )
0
196
8
- ( - )
- ( - )
0
197
8
- ( - )
- ( - )
0
198
8
- ( - )
- ( - )
0
199
8
- ( - )
- ( - )
0
4556
--------------------------------------------------------------------------------------------------

Buffer credits switch or blade model
Table 15 shows the total FC ports in a switch or blade, the number of user ports in a port group,
and the unreserved buffer credits available per port group.

TABLE 15

Total FC ports, ports per port group, and unreserved buffer credits per port group

Switch/blade model

Total FC ports (per switch/blade) User port group size Unreserved buffer credits per port
group

300

24

24

484

5100

40

40

1692

5300

80

16

292

5410

12

12

580

5424

24

24

484

5431

16

16

548

5450

26

26

468

5480

24

24

484

M6505

24

24

7904

6505

24

24

7952

6510

48

48

7760

6520

96

48

4256

6547

48

48

7712

7800

16

16

408

Fabric OS Administrator’s Guide
53-1002920-02

143

5

Buffer credit management

TABLE 15

Total FC ports, ports per port group, and unreserved buffer credits per port group (Continued)

Switch/blade model

Total FC ports (per switch/blade) User port group size Unreserved buffer credits per port
group

VA-40FC

40

40

1692

Brocade Encryption Switch 32

16

1392

FC8-16

16

16

1292/508

FC8-32

32

16

1292/508

FC8-32E

32

16

5456

FC8-48

48

24

1228/716

FC8-48E

48

24

5008

FC8-64

*** Extended Fabrics is not supported on this blade ***

FC16-32

32

16

5456

FC16-48

48

24

5008

FS8-18

16

8

1604

FX8-24

12

12

1060

For the FC8-x port blades, the first number in the “Unreserved buffer credits per port group”
column designates the number of unreserved buffers per port group without buffer optimized
mode; the second number designates the unreserved buffers with buffer optimized mode enabled
on the slot. Use the bufOpMode command to display or change the buffer optimized mode.

Maximum configurable distances for Extended Fabrics
Table 16 shows the maximum supported extended distances (in kilometers) that can be configured
for one port on a specific switch or blade at different speeds.

TABLE 16

Configurable distances for Extended Fabrics
Maximum distances (km) that can be configured (assuming a 2112-byte frame size)

144

Switch/blade model

2 Gbps

4 Gbps

8 Gbps

10 Gbps

16 Gbps

300

486

243

121

N/A

N/A

5100

1694

847

423

N/A

N/A

5300

294

147

73

N/A

N/A

5410

582

291

145.5

N/A

N/A

5424

486

243

121.5

N/A

N/A

5431

550

275

137.5

N/A

N/A

5450

470

235

117.5

N/A

N/A

5480

486

243

121.5

N/A

N/A

M6505

N/A

3953

1976

N/A

988

6505

7426

3713

1856

1485

928

6510

6754

3377

1688

1350

844

6520

4064

2032

1016

812

508

Fabric OS Administrator’s Guide
53-1002920-02

Buffer credit management

TABLE 16

5

Configurable distances for Extended Fabrics (Continued)
Maximum distances (km) that can be configured (assuming a 2112-byte frame size)

Switch/blade model

2 Gbps

4 Gbps

8 Gbps

10 Gbps

16 Gbps

6547

7714

3857

1928

1542

964

7800

410

205

102

N/A

N/A

VA-40FC

1694

847

423

N/A

N/A

Brocade Encryption Switch

1392

696

348

N/A

N/A

FC8-16

1294

647

323

N/A

N/A

FC8-32

1294

647

323

N/A

N/A

FC8-32E

5190

2595

1297

1038

648

FC8-48

1230

615

307

N/A

N/A

FC8-48E

4486

2243

1121

897

560

FC8-64

*** Extended Fabrics is not supported on this blade ***

FC16-32

5190

2595

1297

1038

648

FC16-48

4486

2243

1121

897

560

FS8-18

1604

802

401

N/A

N/A

FX8-24

1062

531

265

N/A

N/A

NOTE
The distances in table 17 assume that QoS is enabled. If QoS is disabled, the maximum supported
distances are higher, because QoS requires an additional 20 buffer credits per active port.
Estimated maximum equally distributed distance = 1-port maximum distance/Number of ports
For example, for three ports running at 2 Gbps on a Brocade 300 switch, the maximum equally
distributed distance is calculated as 486 / 3 = 164 km.

Downgrade considerations
When Fabric OS firmware is downgraded from version 7.1 to an earlier version, the effect depends
on whether the number of buffer credits for the long-distance port is configured with the -framesize
and -distance options or with the -buffers option.

When a port is configured with –framesize and –distance options
In Fabric OS v7.1, if you configure the port by using the -distance option alone, the reserved buffers
are calculated according to the distance. If you configure both the -framesize option and the
-distance option, more buffers will be reserved, depending on the frame size.
With a firmware downgrade, those ports that were configured with more reserved buffers will keep
the reserved buffers as long as the ports remain online. The next time the port is toggled, buffers
will again be reserved on the basis of distance only.

Fabric OS Administrator’s Guide
53-1002920-02

145

5

Buffer credit recovery

When a port is configured with the –buffers option
A firmware downgrade is blocked when a port is configured as a long-distance port by means of the
–buffers option. The following warning message is displayed:
Downgrade to selected version is not allowed because few ports are configured with
Longdistance -buffers option. Please remove the configuration using
portcfglongdistance / L0 CLI or change the configuration with
-distance option on the console.

Configuring credits for a single VC
You can alter the default credit allocation for a normal distance E_Port or EX_Port so that a specific
number of credits is allocated to a port. When you allocate a specific number of credits to an
E_Port or EX_Port, the number of credits specified override the default credit allocation. When this
feature is disabled, the default credit model is restored. Only a normal distance E_Port and EX_Port
can utilize the new credit model, and the allocated credits are reserved only for that port.
When this feature is enabled, the E_Port credit configuration is persistent across system reboots
and High Availability (HA) failover.
This feature is supported on E_Ports and EX_Ports. It does not support ports configured as F_Ports,
Mirror Ports, L_Ports, and Longdistance Ports. If E_Port credits are configured on ports, you cannot
move the ports from one logical switch to another. This feature is not applicable on ICL_ports.

Increasing credits for normal distance E_Ports
Use the following steps to allocate credits to an E_Port.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgEPortCredits --enable command to allocate credits to an E_Port. In the
following example, 14 credits are allocated to an E_Port.
switch:admin> portcfgeportcredits --enable 12/6 14
Success

3. Enter the portCfgEPortCredits --show command to verify that the credits have been allocated
to the E_Port. In the following example, it is verified that 14 credits have been allocated to the
E_Port.
switch:admin> portcfgeportcredits --show 12/6
E-Port Credit Configured : 14
Success.

Buffer credit recovery
Buffer credit recovery allows links to recover after buffer credits are lost when the buffer credit
recovery logic is enabled. The buffer credit recovery feature also maintains performance. If a credit
is lost, a recovery attempt is initiated. During link reset, the frame and credit loss counters are
reset without performance degradation.
Credit recovery is supported on E_Ports, F_Ports, and EX_Ports.

146

Fabric OS Administrator’s Guide
53-1002920-02

Buffer credit recovery

5

Buffer credit recovery is enabled automatically across any long-distance connection for which the
E_Port, F_Port, or EX_Port buffer credit recovery mechanism is supported. For 16-Gbps FC devices
and blades (Brocade 6505, 6510, 6520, M6505, 6547, CR16-4, CR16-8, FC8-32E, FC8-48E,
FC16-32, FC16-48), you can use the portCfgCreditRecovery command to disable or enable buffer
credit recovery on a port.

Buffer credit recovery over an E_Port
To support buffer credit recovery, E_Ports must be connected between devices that support 16
Gbps or between devices that support 8 Gbps.

• Devices that support 16 Gbps:
- Brocade 6505, 6510, 6520, M6505, 6547
- FC8-32E, FC8-48E,FC16-32, FC16-48
• Devices that support 8 Gbps:
- Brocade 300, 5100, 5300, 5410, 5424, 5450, 5480, VA-40FC
- FC8-16, FC8-32, FC8-48
If a device that supports 16 Gbps is connected to a device that supports only 8 Gbps, buffer credit
recovery is disabled, even if both devices are running 8 Gbps.
The buffer credit recovery feature for E_Ports is enabled for the following flow-control modes:

• Normal (R_RDY)
• Virtual Channel (VC_RDY)
• Extended VC (EXT_VC_RDY)

Buffer credit recovery over an F_Port
Buffer credit recovery for F_Ports is supported for F_Port-to-N_Port links between a Brocade switch
and Access Gateway, between a Brocade switch and an adapter, and between an Access Gateway
and an adapter. For an F_Port on a Brocade switch connected to an Access Gateway, the following
conditions must be met:

• Both devices must run Fabric OS v7.1 or later.
• Fabric OS must support buffer credit recovery at either end of the link.
• If both devices support 16 Gbps, the flow-control mode can be either Normal mode (R_RDY) or
VC mode (VC_RDY); otherwise the flow-control mode must be R_RDY.
For an F_Port on a Brocade switch or Access Gateway connected to an adapter, the following
conditions must be met:

•
•
•
•
•

The Brocade switch or Access Gateway must run Fabric OS v7.1 or later.
Fabric OS must support buffer credit recovery at both ends of the link.
The adapter must be running HBA v3.2 firmware or later.
The adapter must operate at maximum speed.
The flow-control mode must be R_RDY.

The feature is enabled automatically during a link reset if the conditions are met. If the conditions
for buffer credit recovery are not met, the link will come up, but buffer credit recovery will not be
enabled.

Fabric OS Administrator’s Guide
53-1002920-02

147

5

Buffer credit recovery

Buffer credit recovery over an EX_Port
Buffer credit recovery is supported on a Fibre Channel router (FCR) EX_Port that connects over an
inter-fabric link (IFL) to an edge fabric E_Port when the following conditions are met:

• The FCR and the switch at the other end of the IFL must both run Fabric OS v7.1 or later.
• The FCR and the switch at either end of the IFL must both support 16 Gbps or 8 Gbps. Buffer
credit recovery is not supported if the EX_Ports do not support the same data rate.

• Either end of the IFL must support buffer credit recovery.
• If the inter-fabric link (IFL) connects devices that support 8 Gbps only, long-distance mode
must also be enabled. Long-distance mode can be enabled or disabled on devices that
support 16 Gbps.

• Virtual Channel flow control (VC_RDY) or Extended VC flow control (EXT_VC_RDY) mode must
be in use. Buffer credit recovery is not supported for EX_Ports when normal (R_RDY) flow
control mode is in use.
The feature is enabled automatically during a link reset if the conditions are met. If the capabilities
at either end of the EX_Port-to-E_Port link are not matched, the link will come up, but refer to the
Fabric OS Command Reference for lists of devices and blades that support 16 Gbps and 8 Gbps.

Enabling and disabling buffer credit recovery
To disable buffer credit recovery on a port, perform the following steps.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgCreditRecovery command and include the -disable option.
The following example disables buffer credit recovery on port 1/20.
switch:admin> portcfgcreditrecovery 1/20 -disable

To enable buffer credit recovery on a port for which it has been disabled, perform the following
steps.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgCreditRecovery command and include the -enable option.
The following example enables buffer credit recovery on port 1/20.
switch:admin> portcfgcreditrecovery 1/20 -enable

148

Fabric OS Administrator’s Guide
53-1002920-02

Credit loss

5

Credit loss
Fabric OS v7.1 and later supports back-end credit loss detection, back-end ports and core blades,
and the Brocade 5300 and 6520 switches, although the support is slightly different on each
device. Refer to the following details on these switches, and the Fabric OS Troubleshooting and
Diagnostics Guide for more general information.

Back-end credit loss detection and recovery support on Brocade 5300
switches
The following credit loss detection methods are supported for Brocade 5300 back-end ports:

• Per-port polling to detect credit loss. If credit loss is detected using this method, the
RASlog C2-1012 message is displayed and recorded.

• On-demand VC credit loss detection. If credit loss is detected using this method, the
RASlog C2-1027 message is displayed and recorded.

• TX timeout trigger automatic VC credit loss detection. If credit loss is detected using this
method, the RASlog C2-1027 message is displayed and recorded.
The following credit loss recovery methods are supported for Brocade 5300 back-end ports:

• For per-port polling and on-demand VC credit loss methods, a link reset will automatically be
performed, assuming that this option was enabled. Refer to “Enabling back-end credit loss
detection and recovery” for details on enabling this feature.

• For the TX timeout trigger automatic VC method, a link reset will be automatically performed if
complete credit loss on a VC is detected.

• A manual link reset option using the creditRecovMode command is also available. Refer
to “Enabling back-end credit loss detection and recovery” for instructions.

NOTE

Whenever a link reset is performed on this switch, the RASlog C2-1014 message is displayed
and recorded.

Back-end credit loss detection and recovery support on Brocade 6520
switches
The following credit loss detection methods are supported for Brocade 6520 back-end ports:

• Per-port polling to detect credit loss. If credit loss is detected using this method, the
RASlog C3-1012 message is displayed and recorded.

• Per-VC credit loss detection. If single-credit loss is detected using this method, it will be
automatically recovered and the RASlog C3-1023 message is displayed and recorded.
If multi-credit loss is detected using this method, the RASlog C3-1013 message is displayed
and recorded. There is no automatic recovery for multi-credit loss.

• Complete VC credit loss detection. If credit loss is detected using this method, the
RASlog C3-1011 message is displayed and recorded.

Fabric OS Administrator’s Guide
53-1002920-02

149

5

Credit loss

The following credit loss recovery methods are supported for Brocade 6520 back-end ports:

• For all the credit loss methods described previously, a link reset will automatically be
performed, assuming that this option was enabled. Refer to “Enabling back-end credit loss
detection and recovery” for details on enabling this feature.

• A manual link reset option using the creditRecovMode command is also available. Refer
to “Enabling back-end credit loss detection and recovery” for instructions.

NOTE

Whenever a link reset is performed on this switch, the RASlog C3-1014 message is displayed
and recorded.

Enabling back-end credit loss detection and recovery
Credit loss detection and recovery is enabled and disabled through the CLI using the
creditRecovMode --cfg command.

• The execution of this command is subject to Virtual Fabrics or Admin Domain restrictions that
may be in place. Refer to the Fabric OS Troubleshooting and Diagnostics Guide for more
information.

• The bottleneck detection commands are supported on F_Ports, FL_Ports, E_Ports, and
EX_Ports.

• The credit recovery commands are supported only on back-end ports of 4 Gbps-, 8 Gbps-, and
16 Gbps-capable FC platforms for blades in the Brocade DCX, DCX-4S, DCX 8510-8, and DCX
8510-4 chassis.
To enable back-end credit loss detection and recovery, perform the following steps.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the creditRecovMode --cfg command to enable credit recovery of back-end ports. In the
following example, back-end port credit loss recovery is enabled with the link reset only option.
switch:admin> creditrecovmode--cfg onLrOnly

3. Enter the creditRecovMode --show command to display information about the back-end port
credit recovery configuration. In the following example, back-end port credit loss recovery is
enabled with the link reset only option.
switch:admin> creditrecovmode --show
Internal port credit recovery is Enabled with LrOnly
C2 FE Complete Credit Loss Detection is Enabled

150

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

6

Managing User Accounts

In this chapter
• User accounts overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Local database user accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Local user account database distribution . . . . . . . . . . . . . . . . . . . . . . . . . .
• Password policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• The boot PROM password . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Remote authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

151
155
158
159
163
167

User accounts overview
In addition to the default permissions assigned to the roles of root, factory, admin, and user,
Fabric OS supports up to 252 additional user accounts on the chassis. These accounts expand
your ability to track account access and audit administrative activities.
Each user account is associated with the following:

• Admin Domain list — Specifies the Administrative Domains to which a user account is allowed
to log in.

• Home Admin Domain — Specifies the Admin Domain that the user is logged in to by default.
The home Admin Domain must be a member of the user’s Admin Domain list.

• Permissions — Associate roles with each user account to determine the functional access
levels within the bounds of the user’s current Admin Domain.

• Virtual Fabric list — Specifies the Virtual Fabric a user account is allowed to log in to.
• Home Virtual Fabric — Specifies the Virtual Fabric that the user is logged in to, if available. The
home Virtual Fabric must be a member of the user’s Virtual Fabric list. If the fabric ID is not
available, the next-lower valid fabric ID is used.

• LF Permission List — Determines functional access levels within the bounds of the user’s
Virtual Fabrics.

• Chassis role — Similar to switch-level roles, but applies to a different subset of commands.
NOTE
Admin Domains are mutually exclusive from Virtual Fabrics permissions when you set up user
accounts. You will need to set up different user accounts for each feature.
You cannot have Admin Domain mode and Virtual Fabrics mode enabled at the same time.
For more information about Admin Domains, refer to Chapter 20, “Managing Administrative
Domains”.
For more information about Virtual Fabrics, refer to Chapter 11, “Managing Virtual Fabrics”.

Fabric OS Administrator’s Guide
53-1002920-02

151

6

User accounts overview

Fabric OS provides four options for authenticating users: remote RADIUS service, remote LDAP
service, remote TACACS+ service, and the local-switch user database. All options allow users to be
managed centrally by means of the following methods:

• Remote RADIUS service: Users are managed in a remote RADIUS server. All switches in the
fabric can be configured to authenticate against the centralized remote database.

• Remote LDAP service: Users are managed in a remote LDAP server. All switches in the fabric
can be configured to authenticate against the centralized remote database. The remote LDAP
server can run Microsoft Active Directory or OpenLDAP.

• Remote TACACS+ service: Users are managed in a remote TACACS+ server. All switches in the
fabric can be configured to authenticate against the centralized remote database.

• Local user database: Users are managed by means of the local user database. The local user
database is manually synchronized by means of the distribute command to push a copy of the
switch’s local user database to all other switches in the fabric running Fabric OS v5.3.0 and
later, but the distribute command is blocked if users with user-defined roles exist on the
sending switch or on any remote, receiving switch.

Role-Based Access Control
Role-Based Access Control (RBAC) specifies the permissions that a user account has on the basis
of the role the account has been assigned. For each role, a set of predefined permissions
determines the jobs and tasks that can be performed on a fabric and its associated fabric
elements. Fabric OS uses RBAC to determine which commands a user is allowed to access.
When you log in to a switch, your user account is associated with a predefined role or a
user-defined role. The role that your account is associated with determines the level of access you
have on that switch and in the fabric. The chassis role can also be associated with user-defined
roles; it has permissions for RBAC classes of commands that are configured when user-defined
roles are created. The chassis role is similar to a switch-level role, except that it affects a different
subset of commands. You can use the userConfig command to add this permission to a user
account.
Table 17 outlines the Fabric OS predefined (default) roles.

TABLE 17

152

Default Fabric OS roles

Role name

Duties

Description

Admin

All administration

All administrative commands

BasicSwitchAdmin

Restricted switch administration

Mostly monitoring with limited switch (local) commands

FabricAdmin

Fabric and switch administration

All switch and fabric commands, excluding user
management and Admin Domains commands

Operator

General switch administration

Routine switch-maintenance commands.

SecurityAdmin

Security administration

All switch security and user management functions

SwitchAdmin

Local switch administration

Most switch (local) commands, excluding security, user
management, and zoning commands

User

Monitoring only

Nonadministrative use, such as monitoring system
activity

ZoneAdmin

Zone administration

Zone management commands only

Fabric OS Administrator’s Guide
53-1002920-02

User accounts overview

6

Admin Domain considerations
Legacy users with no Admin Domain specified and whose current role is admin will have access to
AD0 through AD255 (physical fabric admin); otherwise, they will have access to AD0 only.
If some Admin Domains have been defined for the user and all of them are inactive, the user will
not be allowed to log in to any switch in the fabric. If no home domain is specified for a user, the
system provides a default home domain.
The default home domain for the predefined account is AD0. For user-defined accounts, the default
home domain is the Admin Domain in the user’s Admin Domain list with the lowest ID.

Role permissions
Table 18 describes the types of permissions that are assigned to roles.

TABLE 18

Permission types

Abbreviation

Definition

Description

O

Observe

The user can run commands by using options that display information only, such as
running userConfig --show -a to show all users on a switch.

M

Modify

The user can run commands by using options that create, change, and delete
objects on the system, such as running userConfig --change username -r rolename
to change a user’s role.

OM

Observe and
Modify

The user can run commands by using both observe and modify options; if a role has
modify permissions, it almost always has observe permissions.

N

None

The user is not allowed to run commands in a given category.

To view the permission type for categories of commands, use the classConfig command.

• Enter the classConfig --show -classlist command to list all command categories.
• Enter the classConfig --showroles command with the command category of interest as the
argument.
This command shows the permissions that apply to all commands in a specific category.
> classconfig --showroles authentication
Roles that have access to the RBAC Class ‘authentication’ are:
Role name
--------Admin
Factory
Root
Security Admin

Permission
---------OM
OM
OM
OM

You can also use the classConfig --showcli command to show the permissions that apply to a
specific command.

Fabric OS Administrator’s Guide
53-1002920-02

153

6

User accounts overview

Management channel
The management channel is the communication established between the management
workstation and the switch. Table 19 shows the number of simultaneous login sessions allowed for
each role when authenticated locally. The roles are displayed in alphabetic order, which does not
reflect their importance. When LDAP, RADIUS, or TACACS+ are used for authentication, the total
number of sessions on a switch may not exceed 32.

TABLE 19

Maximum number of simultaneous sessions

Role name

Maximum sessions

Admin

2

BasicSwitchAdmin

4

FabricAdmin

4

Operator

4

SecurityAdmin

4

SwitchAdmin

4

User

4

ZoneAdmin

4

Managing user-defined roles
Fabric OS provides an extensive toolset for managing user-defined roles:

• The roleConfig command is available for defining new roles, deleting created roles, or viewing
information about user-defined roles.

• The classConfig command is available for displaying RBAC information about each category or
class of commands, and includes an option to show all roles associated with a given RBAC
command category.

• The userConfig command can be used to assign a user-defined role to a user account.

Creating a user-defined role
You can define a role as long as it has a unique name that is not the same as any of the Fabric OS
default roles, any other user-defined role, or any existing user account name.
The following conditions also apply:

• A role name is case-insensitive and contains only letters.
• The role name should have a minimum of 4 letters and can be up to 16 letters long.
• The maximum number of user-defined roles that are allowed on a chassis is 256.
The roleConfig command can be used to define unique roles. You must have chassis-level access
and permissions to execute this command. The following example creates a user-defined role
called mysecurityrole. The RBAC class Security is added to the role, and the Observe permission is
assigned:
> roleconfig --add mysecurityrole -class security -perm O
Role added successfully

154

Fabric OS Administrator’s Guide
53-1002920-02

Local database user accounts

6

The assigned permissions can be no higher than the admin role permission assigned to the class.
The admin role permission for the Security class is Observe/Modify. Therefore, the Observe
permission is valid.
The roleConfig --show command is available to view the permissions assigned to a user-defined
role. You can also use the classConfig --showroles command to see that the role was indeed
added with Observe permission for the security commands.
> classConfig --showroles security
Roles that have access to RBAC Class ‘security’ are:
Role Name
--------User
Admin
Factory
Root
SwitchAdmin
FabricAdmin
BasicSwitchAdmin
SecurityAdmin
mysecurityrole

Permissions
----------O
OM
OM
OM
O
OM
O
OM
O

To delete a user-defined role, use the roleConfig --delete command.

Assigning a user-defined role to a user
You can assign a user-defined role to a user by using one of the following options of the userConfig
command:

• userConfig --add with the -r option to create a new user account and assign a role.
• userConfig --change with the -r option to add or change a user-defined role for an existing user
account.

• userConfig --add with the -c option to create a new user account and assign a chassis role.
• userConfig --change with the -c option to add a chassis role to an account.
The following example assigns the mysecurityrole role to the existing anewuser account and adds
the admin chassis role:
> userConfig --change anewuser -r mysecurityrole -c admin

Local database user accounts
User add, change, and delete operations are subject to the subset rule: an admin with ADlist 0–10
or LFlist 1–10 cannot perform operations on an admin, user, or any role with ADlist 11–25 or LFlist
11–128. The user account being changed must have an ADlist or LFlist that is a subset of the
account that is making the change.
In addition to the default administrative and user accounts, Fabric OS supports up to 252
user-defined accounts in each switch (domain). These accounts expand your ability to track
account access and audit administrative activities.

Fabric OS Administrator’s Guide
53-1002920-02

155

6

Local database user accounts

Default accounts
Table 20 lists the predefined accounts offered by Fabric OS that are available in the local-switch
user database. The password for all default accounts should be changed during the initial
installation and configuration of each switch.

TABLE 20

Default local user accounts

Account name

Role

Admin Domain

Logical Fabric

Description

admin

Admin

AD0–255
home: 0

LF1–128
home: 128

Most commands have
Observe/Modify permission.

factory

Factory

AD0–255
home: 0

LF1–128
home: 128

Reserved

root

Root

AD0-255
home: 0

LF1-128
home: 128

Reserved.

user

User

AD0
home: 0

LF-128
home: 128

Most commands have observe-only
permission.

Admin Domain and Virtual Fabrics considerations: Administrators can act on other accounts only if
that account has an Admin Domain or Logical Fabric list that is a subset of the administrator.

Displaying account information
1. Connect to the switch and log in using an account with admin permissions, or an account
associated with a user-defined role with permissions for the UserManagement class of
commands.
2. Enter the appropriate show operands for the account information you want to display:

• userConfig --show -a to show all account information for a switch
• userConfig --show username to show account information for the specified account
• userConfig --showad -a adminDomain_ID to show all accounts permitted to select the
specified adminDomain_ID

• userConfig --showlf -l logicalFabric_ID for each LF in an LF_ID_list, displays a list of users
that include that LF in their LF permissions.

Creating an account
1. Connect to the switch and log in using an account with admin permissions, or an account
associated with a user-defined role with permissions for the UserManagement class of
commands.
2. Enter the userConfig --add command.
> userconfig --add metoo -l 1-128 -h 128 -r admin -c admin

This example creates a user account for the user metoo with the following properties:

•
•
•
•

156

Access to Virtual Fabrics 1 through 128
Default home logical switch to 128
Admin role permissions
Admin chassis role permissions

Fabric OS Administrator’s Guide
53-1002920-02

Local database user accounts

6

3. In response to the prompt, enter a password for the account.
The password is not displayed when you enter it on the command line.

Deleting an account
This procedure can be performed on local user accounts.
1. Connect to the switch and log in using an account with admin permissions, or an account
associated with a user-defined role with permissions for the UserManagement class of
commands.
2. Enter the userConfig --delete command.
You cannot delete the default accounts. An account cannot delete itself. All active CLI sessions
for the deleted account are logged out.
3. At the prompt for confirmation, enter y.

Changing account parameters
This procedure can be performed on local user accounts.
When changing account parameters, if you change the ADlist for the user account, all of the
currently active sessions for that account will be logged out. For more information about changing
the Admin Domain on an account, refer to Chapter 20, “Managing Administrative Domains”.
1. Connect to the switch and log in using an account with admin permissions, or an account
associated with a user-defined role with permissions for the UserManagement class of
commands.
2. Enter the userConfig --change command.

Local account passwords
The following rules apply to changing passwords:

• Users can change their own passwords.
• To change the password for another account requires admin permissions or an account
associated with a user-defined role with Modify permissions for the LocalUserEnvironment
RBAC class of commands. When changing an admin account password, you must provide the
current password.

• An admin with ADlist 0–10 or LFlist 1–10 cannot change the password on an account with
admin, user, or any permission with an ADlist 11–25 or LFlist 11–128. The user account being
changed must have an ADlist that is a subset of the account that is making the change.

• A new password must have at least one character different from the previous password.
• You cannot change passwords by using SNMP.

Changing the password for the current login account
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the passwd command.
3. Enter the requested information at the prompts.

Fabric OS Administrator’s Guide
53-1002920-02

157

6

Local user account database distribution

Changing the password for a different account
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the passwd command specifying the name of the account for which the password is
being changed.
3. Enter the requested information at the prompts.

Local user account database distribution
Fabric OS allows you to distribute the user database and passwords to other switches in the fabric.
When the switch accepts a distributed user database, it replaces the local user database with the
user database it receives.
By default, switches accept the user databases and passwords distributed from other switches.
The “Locked” status of a user account is not distributed as part of local user database distribution.
When the user database is distributed, it may be rejected by a switch for one of the following
reasons:

•
•
•
•

One of the target switches does not support local account database distribution.
One of the target switch’s user databases is protected.
One of the remote switches has logical switches defined.
Either the local switch or one of the remote switches has user accounts associated with
user-defined roles.

Distributing the local user database
When the local user database is distributed, all user-defined accounts residing in the receiving
switches are logged out of any active sessions.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the distribute -p PWD -d command.

NOTE

If Virtual Fabrics mode is enabled and there are logical switches defined other than the default
logical switch, then distributing the password database to switches is not supported.
Distributing the password database to switches is not allowed if there are users associated with
user-defined roles in either the sending switch or the remote switch

Accepting distributed user databases on the local switch
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fddCfg --localaccept PWD command.

158

Fabric OS Administrator’s Guide
53-1002920-02

Password policies

6

Rejecting distributed user databases on the local switch
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fddCfg --localreject PWD command.

Password policies
The password policies described in this section apply to the local-switch user database only.
Configured password policies (and all user account attribute and password state information) are
synchronized across CPs and remain unchanged after an HA failover. Password policies can also be
manually distributed across the fabric (refer to “Local user account database distribution” on
page 158).
All password policies are enforced during logins to the standby CP. However, you may observe that
the password enforcement behavior on the standby CP is inconsistent with prior login activity; this
is because password state information from the active CP is automatically synchronized with the
standby CP, thereby overwriting any password state information that was previously stored there.
Also, password changes are not permitted on the standby CP.
Password authentication policies configured using the passwdCfg command are not enforced
during initial prompts to change default passwords.

Password strength policy
The password strength policy is enforced across all user accounts, and enforces a set of format
rules to which new passwords must adhere. The password strength policy is enforced only when a
new password is defined. The total of the other password strength policy parameters (lowercase,
uppercase, digits, and punctuation) must be less than or equal to the value of the MinLength
parameter.
Use the following attributes to the passwdCfg command to set the password strength policy:

• Lowercase
Specifies the minimum number of lowercase alphabetic characters that must appear in the
password. The default value is zero. The maximum value must be less than or equal to the
MinLength value.

• Uppercase
Specifies the minimum number of uppercase alphabetic characters that must appear in the
password. The default value is zero. The maximum value must be less than or equal to the
MinLength value.

• Digits
Specifies the minimum number of numeric digits that must appear in the password. The
default value is zero. The maximum value must be less than or equal to the MinLength value.

• Punctuation
Specifies the minimum number of punctuation characters that must appear in the password.
All printable, non-alphanumeric punctuation characters except the colon ( : ) are allowed. The
default value is zero. The maximum value must be less than or equal to the MinLength value.

Fabric OS Administrator’s Guide
53-1002920-02

159

6

Password policies

• MinLength
Specifies the minimum length of the password. The minimum can be from 8 through 40
characters. New passwords must be between the minimum length specified and 40
characters. The default value is 8. The maximum value must be greater than or equal to the
MinLength value.

• Repeat
Specifies the length of repeated character sequences that will be disallowed. For example, if
the “repeat” value is set to 3, a password “passAAAword” is disallowed because it contains the
repeated sequence “AAA”. A password of “passAAword” would be allowed because no repeated
character sequence exceeds two characters. The range of allowed values is from 1 through 40.
The default value is 1.

• Sequence
Specifies the length of sequential character sequences that will be disallowed. A sequential
character sequence is defined as a character sequence in which the ASCII value of each
contiguous character differs by one. The ASCII value for the characters in the sequence must
all be increasing or decreasing. For example, if the “sequence” value is set to 3, a password
“passABCword” is disallowed because it contains the sequence “ABC”. A password of
“passABword” would be allowed because it contains no sequential character sequence
exceeding two characters. The range of allowed values is from 1 through 40. The default
value is 1. When set to 1, sequential characters are not enforced.

• Reverse
Activates or deactivates the validation check to determine whether the password is an exact
reverse string of the user name. This option is disabled by default.
Example of a password strength policy

The following example shows a password strength policy that requires passwords to contain at
least 3 uppercase characters, 4 lowercase characters, and 2 numeric digits; the minimum
length of the password is 9 characters. The password cannot be an exact reverse string of the
username.
> passwdcfg --set -uppercase 3 -lowercase 4 -digits 2 -minlength 9 -reverse 1

Password history policy
The password history policy prevents users from recycling recently used passwords, and is enforced
across all user accounts when users are setting their own passwords. The password history policy is
enforced only when a new password is defined.
Specify the number of past password values that are disallowed when setting a new password. Allowable
password history values range from 0 through 24. If the value is set to 0, the new password cannot be set
to the current password, but can be set to the most recent password. The default value is 1, which
means the current and one previous password cannot be reused. The value 2 indicates that the current
and the two previous passwords cannot be used (and so on, up to 24 passwords).
This policy does not verify that a new password meets a minimal standard of difference from prior
passwords; rather, it only determines whether or not a newly specified password is identical to one
of the specified number (1–24) of previously used passwords.
The password history policy is not enforced when an administrator sets a password for another
user; instead, the user’s password history is preserved and the password set by the administrator
is recorded in the user’s password history.

160

Fabric OS Administrator’s Guide
53-1002920-02

Password policies

6

Password expiration policy
The password expiration policy forces the expiration of a password after a configurable period of
time. The expiration policy can be enforced across all user accounts or on specified users only.
A warning that password expiration is approaching is displayed when the user logs in. When a
password expires, the user must change the password to complete the authentication process and
open a user session. You can specify the number of days prior to password expiration during which
warnings will commence. Password expiration does not disable or lock out the account.
Use the following attributes to the passwdCfg command to set the password expiration policy:

• MinPasswordAge
Specifies the minimum number of days that must elapse before a user can change a
password. MinPasswordAge values range from 0 through 999. The default value is zero.
Setting this parameter to a nonzero value discourages users from rapidly changing a password
in order to circumvent the password history setting to select a recently used password. The
MinPasswordAge policy is not enforced when an administrator changes the password for
another user.

• MaxPasswordAge
Specifies the maximum number of days that can elapse before a password must be changed,
and is also known as the password expiration period. MaxPasswordAge values range from 0
through 999. The default value is zero. Setting this parameter to zero disables password
expiration.

• Warning
Specifies the number of days prior to password expiration that a warning about password
expiration is displayed. Warning values range from 0 through 999. The default value is 0 days.

NOTE

When MaxPasswordAge is set to a nonzero value, MinPasswordAge and Warning must be set
to a value that is less than or equal to MaxPasswordAge.
Example password expiration policies

The following example configures a password expiration policy for the metoo user account. This
user must change the password within 90 days of setting the current password and no sooner than
10 days after setting the current password. The user will start to receive warning messages 3 days
before the 90-day limit, if the password is not already changed.
> passwdcfg --setuser metoo -minpasswordage 10 -maxpasswordage 90 -warning 3

The following example configures a password expiration policy for all users.
> passwdcfg --set -minpasswordage 5 -maxpasswordage 30 -warning 5

Account lockout policy
The account lockout policy disables a user account when that user exceeds a specified number of
failed login attempts, and is enforced across all user accounts. You can configure this policy to
keep the account locked until explicit administrative action is taken to unlock it, or the locked
account can be automatically unlocked after a specified period. Administrators can unlock a locked
account at any time.

Fabric OS Administrator’s Guide
53-1002920-02

161

6

Password policies

A failed login attempt counter is maintained for each user on each switch instance. The counters
for all user accounts are reset to zero when the account lockout policy is enabled. The counter for
an individual account is reset to zero when the account is unlocked after a lockout duration period
expires, or when the account user logs in successfully.
The admin account can also have the lockout policy enabled on it. The admin account lockout
policy is disabled by default and uses the same lockout threshold as the other permissions. It can
be automatically unlocked after the lockout duration passes or when it is manually unlocked by
either a user account that has a securityAdmin or other admin permissions.
Virtual Fabrics considerations: The home logical fabric context is used to validate user enforcement
for the account lockout policy.
Note that the account-locked state is distinct from the account-disabled state.
Use the following attributes to set the account lockout policy:

• LockoutThreshold
Specifies the number of times a user can attempt to log in using an incorrect password before
the account is locked. The number of failed login attempts is counted from the last successful
login. LockoutThreshold values range from 0 through 999, and the default value is 0. Setting
the value to 0 disables the lockout mechanism.

• LockoutDuration
Specifies the time, in minutes, after which a previously locked account is automatically
unlocked. LockoutDuration values range from 0 through 99999, and the default value is 30.
Setting the value to 0 disables lockout duration, and requires a user to seek administrative
action to unlock the account. The lockout duration begins with the first login attempt after the
LockoutThreshold has been reached. Subsequent failed login attempts do not extend the
lockout period.

Enabling the admin lockout policy
1. Log in to the switch using an account that has admin or securityAdmin permissions.
2. Enter the passwdCfg --enableadminlockout command.

Unlocking an account
1. Log in to the switch using an account that has admin or securityAdmin permissions.
2. Enter the userConfig --change account_name -u command specifying the name of the user
account that is locked out.

Disabling the admin lockout policy
1. Log in to the switch using an account that has admin or securityAdmin permissions.
2. Enter the passwdCfg --disableadminlockout command.

162

Fabric OS Administrator’s Guide
53-1002920-02

The boot PROM password

6

Denial of service implications
The account lockout mechanism may be used to create a denial of service condition when a user
repeatedly attempts to log in to an account by using an incorrect password. Selected privileged
accounts are exempted from the account lockout policy to prevent users from being locked out
from a denial of service attack. However, these privileged accounts may then become the target of
password-guessing attacks. Audit logs should be examined to monitor if such attacks are
attempted.

The boot PROM password
The boot PROM password provides an additional layer of security by protecting the boot PROM from
unauthorized use. Setting a recovery string for the boot PROM password enables you to recover a
lost boot PROM password by contacting your switch service provider. Without the recovery string, a
lost boot PROM password cannot be recovered.
Although you can set the boot PROM password without also setting the recovery string, it is strongly
recommended that you set both the password and the recovery string. If your site procedures
dictate that you set the boot PROM password without the recovery string, refer to “Setting the boot
PROM password for a switch without a recovery string” on page 165.
To set the boot PROM password with or without a recovery string, refer to the section that applies to
your switch or Backbone model.

CAUTION
Setting the boot PROM password requires accessing the boot prompt, which stops traffic flow
through the switch until the switch is rebooted. Perform this procedure during a planned
downtime.

Setting the boot PROM password for a switch with a recovery string
This procedure applies to the fixed-port switch models. The password recovery instructions
provided within this section are only for the switches listed in “Supported hardware and software”
on page 35. If your switch is not listed, contact your switch support provider for instructions.
1. Connect to the serial port interface as described in “Connecting to Fabric OS through the serial
port” on page 58.
2. Reboot the switch.
3. Press Esc within four seconds after the message “Press escape within 4 seconds...” is
displayed.
The following options are available:
Option
1
2
3

Fabric OS Administrator’s Guide
53-1002920-02

Description
Start system.
Recovery password.
Enter command shell.

Continues the system boot process.
Lets you set the recovery string and the boot PROM password.
Provides access to boot parameters.

163

6

The boot PROM password

4. Enter 2.

• If no password was previously set, the following message is displayed:
Recovery password is NOT set. Please set it now.

• If a password was previously set, the following messages is displayed:
Send the following string to Customer Support for password recovery:
afHTpyLsDo1Pz0Pk5GzhIw==
Enter the supplied recovery password.
Recovery Password:

5. Enter the recovery password (string).
The recovery string must be from 8 through 40 alphanumeric characters in length. A random
string that is 15 characters or longer is recommended for higher security. The firmware
prompts for this password only once. It is not necessary to remember the recovery string,
because it is displayed the next time you enter the command shell.
The following prompt is displayed:
New password:

6. Enter the boot PROM password, and then re-enter it when prompted. The password must be
eight alphanumeric characters long (any additional characters are not recorded). Record this
password for future use.
The new password is automatically saved.
7.

Reboot the switch by entering the reset command at the prompt.

Setting the boot PROM password for a Backbone with a recovery string
This procedure applies to the Brocade DCX, DCX-4S, DCX 8510-4, and DCX 8510-8 Backbones. The
boot PROM and recovery passwords must be set for each CP blade.
1. Connect to the serial port interface on the standby CP blade, as described in “Connecting to
Fabric OS through the serial port” on page 58.
2. Connect to the active CP blade over a serial or Telnet connection and enter the haDisable
command to prevent failover during the remaining steps.
3. Reboot the standby CP blade by sliding the On/Off switch on the ejector handle of the standby
CP blade to Off, and then back to On.
4. Press Esc within four seconds after the message “Press escape within 4 seconds...” is
displayed.
The following options are available:

164

Option

Description

1
2
3

Continues the system boot process.
Lets you set the recovery string and the boot PROM password.
Provides access to boot parameters.

Start system.
Recovery password.
Enter command shell.

Fabric OS Administrator’s Guide
53-1002920-02

The boot PROM password

6

5. Enter 2. Take the following appropriate action based on whether you find the password was
previously set:

• If no password was previously set, the following message is displayed:
Recovery password is NOT set. Please set it now.

• If a password was previously set, the following messages are displayed:
Send the following string to Customer Support for password recovery:
afHTpyLsDo1Pz0Pk5GzhIw==
Enter the supplied recovery password.
Recovery Password:

6. Enter the recovery password (string).
The recovery string must be from 8 through 40 alphanumeric characters in length. A random
string that is 15 characters or longer is recommended for higher security. The firmware only
prompts for this password once. It is not necessary to remember the recovery string because it
is displayed the next time you enter the command shell.
The following prompt is displayed:
New password:

7.

Enter the boot PROM password, and then re-enter it when prompted. The password must be
eight alphanumeric characters long any additional characters are not recorded). Record this
password for future use.
The new password is automatically saved (the saveEnv command is not required).

8. Connect to the active CP blade over a serial or Telnet connection and enter the haEnable
command to restore high availability, and then fail over the active CP blade by entering the
haFailover command.
Traffic flow through the active CP blade resumes when the failover is complete.
9. Connect the serial cable to the serial port on the new standby CP blade (previously the active
CP blade).
10. Repeat step 2 through step 7 for the new standby CP blade (each CP blade has a separate
boot PROM password).
11. Connect to the active CP blade over a serial or Telnet connection and enter the haEnable
command to restore high availability.
Although you can set the boot PROM password without also setting the recovery string, it is strongly
recommended that you set both the password and the string as described in “Setting the boot
PROM password for a switch with a recovery string” on page 163. If your site procedures dictate
that you must set the boot PROM password without the string, follow the procedure that applies to
your switch model.

Setting the boot PROM password for a switch without a recovery string
This procedure applies to the fixed-port switch models.
The password recovery instructions provided within this section are only for the switches listed in
“Supported hardware and software” on page 35. If your switch is not listed, contact your switch
support provider for instructions.

Fabric OS Administrator’s Guide
53-1002920-02

165

6

The boot PROM password

1. Create a serial connection to the switch as described in “Connecting to Fabric OS through the
serial port” on page 58.
2. Reboot the switch by entering the reboot command.
3. Press Esc within four seconds after the message “Press escape within 4 seconds...” is
displayed.
The following options are available:
Option

Description

1
2
3

Continues the system boot process.
Lets you set the recovery string and the boot PROM password.
Provides access to boot parameters.

Start system.
Recovery password.
Enter command shell.

4. Enter 3.
5. At the shell prompt, enter the passwd command.
The passwd command only applies to the boot PROM password when it is entered from the
boot interface.
6. Enter the boot PROM password at the prompt, and then re-enter it when prompted. The
password must be eight alphanumeric characters long (any additional characters are not
recorded). Record this password for future use.
7.

Enter the saveEnv command to save the new password.

8. Reboot the switch by entering the reset command.

Setting the boot PROM password for a Backbone without a recovery
string
This procedure applies to the Brocade DCX, DCX-4S, DCX 8510-4, and DCX 8510-8 Backbones.
On the Brocade DCX Backbone, set the password on the standby CP blade, fail over, and then set
the password on the previously active (now standby) CP blade to minimize disruption to the fabric.
1. Determine the active CP blade by opening a Telnet session to either CP blade, connecting as
admin, and entering the haShow command.
2. Connect to the active CP blade over a serial or Telnet connection and enter the haDisable
command to prevent failover during the remaining steps.
3. Create a serial connection to the standby CP blade as described in “Connecting to Fabric OS
through the serial port” on page 58.
4. Reboot the standby CP blade by sliding the On/Off switch on the ejector handle of the standby
CP blade to Off, and then back to On.
This causes the blade to reset.
5. Press Esc within four seconds after the message “Press escape within 4 seconds...”is
displayed.

166

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

The following options are available:
Option

Description

1
2
3

Continues the system boot process.
Lets you set the recovery string and the boot PROM password.
Provides access to boot parameters.

Start system.
Recovery password.
Enter command shell.

6. Enter 3.
7.

Enter the passwd command at the shell prompt.
The passwd command applies only to the boot PROM password when it is entered from the
boot interface.

8. Enter the boot PROM password at the prompt, and then re-enter it when prompted.
The password must be eight alphanumeric characters long (any additional characters are not
recorded). Record this password for future use.
9. Enter the saveEnv command to save the new password.
10. Reboot the standby CP blade by entering the reset command.
11. Connect to the active CP blade over a serial or Telnet connection and enter the haEnable
command to restore high availability, and then fail over the active CP blade by entering the
haFailover command.
Traffic resumes flowing through the newly active CP blade after it has completed rebooting.
12. Connect the serial cable to the serial port on the new standby CP blade (previously the active
CP blade).
13. Repeat step 3 through step 10 for the new standby CP blade.
14. Connect to the active CP blade over a serial or Telnet connection and enter the haEnable
command to restore high availability.

NOTE

To recover lost passwords, refer to the Fabric OS Troubleshooting and Diagnostics Guide.

Remote authentication
Fabric OS supports user authentication through the local user database or one of the following
external authentication services:

• Remote authentication dial-in user service (RADIUS)
• Lightweight Directory Access Protocol (LDAP) using Microsoft Active Directory in Windows or
OpenLDAP in Linux.

• Terminal Access Controller Access-Control System Plus (TACACS+)

Remote authentication configuration
A switch can be configured to try one of the supported remote authentication services (RADIUS,
LDAP, or TACACS+) and local switch authentication. The switch can also be configured to use only a
remote authentication service, or only local switch authentication.

Fabric OS Administrator’s Guide
53-1002920-02

167

6

Remote authentication

Client/server model
When configured to use one of the supported remote authentication services, the switch acts as a
Network Access Server (NAS) and RADIUS, LDAP, or TACACS+ client. The switch sends all
authentication, authorization, and accounting (AAA) service requests to the authentication server.
The authentication server receives the request, validates the request, and sends its response back
to the switch.
The supported management access channels that integrate with RADIUS, LDAP, and TACACS+
include serial port, Telnet, SSH, Web Tools, and API. All these access channels require the switch IP
address or name to connect. RADIUS, LDAP, and TACACS+ servers accept both IPv4 and IPv6
address formats. For accessing both the active and standby CP blades, and for the purpose of HA
failover, both CP IP addresses of a Backbone should be included in the authentication server
configuration.

NOTE

For systems such as the Brocade DCX Backbone, the switch IP addresses are aliases of the physical
Ethernet interfaces on the CP blades. When specifying client IP addresses for the logical switches in
such systems, make sure that the CP IP addresses are used.

Authentication server data
When configured for remote authentication, a switch becomes a RADIUS, LDAP, or TACACS+ client.
In any of these configurations, authentication records are stored in the authentication host server
database. Login and logout account name, assigned permissions, and time-accounting records are
also stored on the authentication server for each user.

Switch configuration
By default, the remote authentication services are disabled, so AAA services default to the switch’s
local database.
To enable remote authentication, it is strongly recommended that you access the CLI through an
SSH connection so that the shared secret is protected. Multiple login sessions can configure
simultaneously, and the last session to apply a change leaves its configuration in effect. After a
configuration is applied, it persists after a reboot or an HA failover.
To enable the secure LDAP service, you must install a certificate from the Microsoft Active Directory
server or the OpenLDAP server. By default, the LDAP service does not require certificates.
The configuration applies to all switches. On a Backbone, the configuration replicates itself on a
standby CP blade if one is present. It is saved in a configuration upload and applied in a
configuration download.
Brocade recommends configuring at least two authentication servers, so that if one fails, the other
will assume service. Up to five servers are supported.
You can set the configuration with any one of the supported authentication services and local
authentication enabled, so that if the authentication servers do not respond because of a power
failure or network problems, the switch uses local authentication.

168

Fabric OS Administrator’s Guide
53-1002920-02

6

Remote authentication

Consider the effects of the use of a remote authentication service on other Fabric OS features. For
example, when a remote authentication service is enabled, all account passwords must be
managed on the authentication server. The Fabric OS mechanisms for changing switch passwords
remain functional; however, such changes affect only the involved switches locally. They do not
propagate to the authentication server, nor do they affect any account on the authentication server.
Authentication servers also support notifying users of expiring passwords.
When RADIUS, LDAP, or TACACS+ is set up for a fabric that contains a mix of switches with and
without RADIUS, LDAP, and TACACS+ support, the way a switch authenticates users depends on
whether a RADIUS, LDAP, or TACACS+ server is set up for that switch. For a switch with remote
authentication support and configuration, authentication bypasses the local password database.
For a switch without remote authentication support or configuration, authentication uses the
switch’s local account names and passwords.

Supported LDAP options
Table 21 summarizes the various LDAP options and Brocade support for each.

TABLE 21

LDAP options

Protocol

Description

Channel type Default port

URL

Brocade
supported?

LDAPv3

LDAP over TCP

Unsecured

389

ldap://

No

LDAPv3 with TLS
extension

LDAPv3 over TLS

Secured

389

ldap://

Yes

LDAPv3 with TLS
and Certificate

LDAPv3 over TLS channel and
authenticated using a certificate

Secured

389

ldap://

Yes

LDAPv2 with SSL1

LDAPv2 over SSL. Port 636 is used for
SSL. Port 389 is for connecting to
LDAP.

Secured

636 and 389

ldaps://

No

1.

This protocol was deprecated in 2003 when LDAPv3 was standardized.

Command options
Table 22 outlines the aaaConfig command options used to set the authentication mode.

TABLE 22

Authentication configuration options

aaaConfig options

Description

Equivalent setting in
Fabric OS v5.1.0 and
earlier
--radius

--switchdb1

--authspec “local”

Default setting. Authenticates management
connections against the local database only.
If the password does not match or the user is
not defined, the login fails.

Off

On

--authspec “radius”

Authenticates management connections
against any RADIUS databases only.
If the RADIUS service is not available or the
credentials do not match, the login fails.

On

Off

Fabric OS Administrator’s Guide
53-1002920-02

169

6

Remote authentication

TABLE 22

Authentication configuration options (Continued)

aaaConfig options

Description

Equivalent setting in
Fabric OS v5.1.0 and
earlier
--radius

--authspec “radius;local”

Authenticates management connections
against any RADIUS databases first.
If RADIUS fails for any reason, authenticates
against the local user database.

not
not
supported supported

--authspec “radius;local” --backup

Authenticates management connections
against any RADIUS databases. If RADIUS fails
because the service is not available, it then
authenticates against the local user database.
The --backup option directs the service to try
the secondary authentication database only if
the primary authentication database is not
available.

On

On

--authspec “ldap”

Authenticates management connections
against any LDAP databases only. If LDAP
service is not available or the credentials do
not match, the login fails.

n/a

n/a

--authspec “ldap; local”

Authenticates management connections
against any LDAP databases first. If LDAP fails
for any reason, it then authenticates against
the local user database.

n/a

On

--authspec “ldap; local” --backup

Authenticates management connections
against any LDAP databases first. If LDAP fails
for any reason, it then authenticates against
the local user database. The --backup option
states to try the secondary authentication
database only if the primary authentication
database is not available.

n/a

On

--authspec “tacacs+”

Authenticates management connections
against any TACACS+ databases only. If
TACACS+ service is not available or the
credentials do not match, the login fails.

not
not
supported supported

--authspec “tacacs+; local”

Authenticates management connections
against any TACACS+ databases first. If
TACACS+ fails for any reason, it then
authenticates against the local user database.

not
not
supported supported

--authspec “tacacs+; local” --backup

Authenticates management connections
against any TACACS+ databases first. If
TACACS+ fails for any reason, it then
authenticates against the local user database.
The --backup option states to try the secondary
authentication database only if the primary
authentication database is not available.

not
not
supported supported

--authspec -nologout

Prevents users from being logged out when you
change authentication. Default behavior is to
log out users when you change authentication.

n/a

1.

170

--switchdb1

n/a

Fabric OS v5.1.0 and earlier aaaConfig --switchdb  setting.

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

Setting the switch authentication mode
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aaaConfig --authspec command.

Fabric OS user accounts
RADIUS, LDAP, and TACACS+ servers allow you to set up user accounts by their true network-wide
identities rather than by the account names created on a Fabric OS switch. With each account
name, assign the appropriate switch access permissions. For LDAP servers, you can use the
ldapCfg -–maprole ldap_role name switch_role command to map LDAP server permissions.
RADIUS, LDAP, and TACACS+ support all the defined RBAC roles described in Table 17 on page 152.
Users must enter their assigned RADIUS, LDAP, or TACACS+ account name and password when
logging in to a switch that has been configured with remote authentication. After the remote
authentication (RADIUS, LDAP, or TACACS+) server authenticates a user, it responds with the
assigned switch role in a Brocade Vendor-Specific Attribute (VSA). If the response does not have a
VSA permissions assignment, the user role is assigned. If no Administrative Domain is assigned,
then the user is assigned to the default Admin Domain AD0.
You can set a user password expiration date and add a warning for RADIUS login and TACACS+
login. The password expiry date must be specified in UTC and in MM/DD/YYYY format. The
password warning specifies the number of days prior to the password expiration that a warning of
password expiration notifies the user. You either specify both attributes or none. If you specify a
single attribute or there is a syntax error in the attributes, the password expiration warning will not
be issued. If your RADIUS server maintains its own password expiration attributes, you must set the
exact date twice to use this feature, once on your RADIUS server and once in the VSA. If the dates
do not match, then the RADIUS server authentication fails.
Table 23 describes the syntax used for assigning VSA-based account switch roles on a RADIUS
server.

TABLE 23

Syntax for VSA-based account roles

Item

Value

Description

Type

26

1 octet

Length

7 or higher

1 octet, calculated by the server

Vendor ID

1588

4 octet, Brocade SMI Private Enterprise Code

Fabric OS Administrator’s Guide
53-1002920-02

171

6

Remote authentication

TABLE 23

Syntax for VSA-based account roles (Continued)

Item

Value

Description

Vendor type

1

1 octet, Brocade-Auth-Role; valid attributes for the Brocade-Auth-Role are:
Admin
BasicSwitchAdmin
FabricAdmin
Operator
SecurityAdmin
SwitchAdmin
User
ZoneAdmin

2

Optional: Specifies the Admin Domain or Virtual Fabric member list. For
more information on Admin Domains or Virtual Fabrics, refer to “RADIUS
configuration with Admin Domains or Virtual Fabrics” on page 173.
Brocade-AVPairs1

3

Brocade-AVPairs2

4

Brocade-AVPairs3

5

Brocade-AVPairs4

6

Brocade Password ExpiryDate

7

Brocade Password ExpiryWarning

Vendor length

2 or higher

1 octet, calculated by server, including vendor-type and vendor-length

Attribute-specific data

ASCII string

Multiple octet, maximum 253, indicating the name of the assigned role and
other supported attribute values such as Admin Domain member list.

Fabric OS users on the RADIUS server
All existing Fabric OS mechanisms for managing local-switch user accounts and passwords remain
functional when the switch is configured to use RADIUS. Changes made to the local switch
database do not propagate to the RADIUS server, nor do the changes affect any account on the
RADIUS server.

Windows 2000 IAS
To configure a Windows 2000 Internet authentication service (IAS) server to use VSA to pass the
admin role to the switch in the dial-in profile, the configuration specifies the Vendor code (1588),
Vendor-assigned attribute number (1), and attribute value (admin), as shown in Figure 10.

172

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

FIGURE 10

6

Windows 2000 VSA configuration

Linux FreeRADIUS server
For the configuration on a Linux FreeRADIUS server, define the values outlined in Table 24 in a
vendor dictionary file called dictionary.brocade.

TABLE 24

Entries in dictionary.brocade file

Include

Key

Value

VENDOR

Brocade

1588

ATTRIBUTE

Brocade-Auth-Role

1 string Brocade

Brocade-AVPairs1, 2, 3, 4

2, 3, 4, 5 string
Admin Domain or Virtual Fabric member list

Brocade-Passwd-ExpiryDate

6 string MM/DD/YYYY in UTC

Brocade-Passwd-WarnPeriod

7 integer in days

After you have completed the dictionary file, define the permissions for the user in a configuration
file. For example, to grant the user admin permissions, you would add the following statement to
the configuration file:
swladmin

Auth-Type := Local, User-Password == "myPassword"
Brocade-Auth-Role = "admin",
Brocade-AVPairs1 = "HomeLF=70",
Brocade-AVPairs2 =
"LFRoleList=admin:2,4-8,70,80,128;ChassisRole=admin",
Brocade-Passwd-ExpiryDate = "11/10/2011",
Brocade-Passwd-WarnPeriod = "30"

RADIUS configuration with Admin Domains or Virtual Fabrics
When configuring users with Admin Domains or Virtual Fabrics, you must also include the Admin
Domain or Virtual Fabric member list. This section describes the way that you configure attribute
types for this configuration.

Fabric OS Administrator’s Guide
53-1002920-02

173

6

Remote authentication

The values for these attribute types use the syntax key=val[;key=val], where key is a text description
of attributes, val is the attribute value for the given key, the equal sign (=) is the separator between
key and value, and the semicolon (;) is an optional separator for multiple key-value pairs.
Multiple key-value pairs can appear for one Vendor-Type code. Key-value pairs with the same key
name may be concatenated across multiple Vendor-Type codes. You can use any combination of
the Vendor-Type codes to specify key-value pairs. Note that a switch always parses these attributes
from Vendor-Type code 2 to Vendor-Type code 4.
Only the following keys are accepted; all other keys are ignored.

• HomeAD is the designated home Admin Domain for the account. The valid range of values is
from 0 through 255. The first valid HomeAD key-value pair is accepted by the switch, and any
additional HomeAD key-value pairs are ignored.

• ADList is a comma-separated list of Administrative Domain numbers of which this account is a
member. Valid numbers range from 0 through 255. A dash between two numbers specifies a
range. Multiple ADlist key-value pairs within the same or across the different Vendor-Type
codes are concatenated. Multiple occurrences of the same Admin Domain number are
ignored.

• HomeLF is the designated home Virtual Fabric for the account. The valid values are from
1 through 128 and chassis context. The first valid HomeLF key-value pair is accepted by the
switch; additional HomeLF key-value pairs are ignored.

• LFRoleList is a comma-separated list of Virtual Fabric ID numbers of which this account is a
member. Valid numbers range from 1 through 128. A dash between two numbers specifies a
range. Multiple Virtual Fabric list key-value pairs within the same or across different
Vendor-Type codes are concatenated. Multiple occurrences of the same Virtual Fabric ID
number are ignored.

• ChassisRole is the account access permission at the chassis level. The chassis role allows the
user to execute chassis-related commands in a Virtual Fabrics-enabled environment. Valid
chassis roles include the default roles and any of the user-defined roles.
RADIUS authentication requires that the account have valid permissions through the attribute type
Brocade-Auth-Role. The additional attribute values ADList, HomeAD, HomeLF, and LFRoleList are
optional. If they are unspecified, the account can log in with AD0 as its member list and home
Admin Domain or VF128 as its member list and home Virtual Fabric. If there is an error in the
ADlist, HomeAD, LFRoleList, or HomeLF specification, the account cannot log in until the AD list or
Virtual Fabric list is corrected; an error message is displayed.
For example, on a Linux FreeRADIUS Server, the user (user-za) with the following settings takes the
“zoneAdmin” permissions, with AD member list: 1, 2, 4, 5, 6, 7, 8, 9, 12; the Home Admin Domain
will be 1.
user-za Auth-Type := Local, User-Password == "password"
Brocade-Auth-Role = "ZoneAdmin",
Brocade-AVPairs1 = "ADList=1,2,6,"
Brocade-AVPairs2 = "ADList=4-8;ADList=7,9,12"

In the next example, on a Linux FreeRADIUS Server, the user has the “operator” permissions, with
ADList 1, 2, 4, 5, 6, 7, 8, 9, 12, 20 and HomeAD 2.
user-opr Auth-Type := Local, User-Password == "password"
Brocade-Auth-Role = "operator",
Brocade-AVPairs1 = "ADList=1,2;HomeAD=2",
Brocade-AVPairs2 = "ADList=-4-8,20;ADList=7,9,12"

174

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

In the next example, on a Linux FreeRADIUS Server, the user has the “zoneAdmin” permissions,
with VFlist 2, 4, 5, 6, 7, 8, 10, 11, 12, 13, 15 17, 19, 22, 23, 24, 25, 29, 31 and HomeLF 1.
user300 Auth-Type := Local, User-Password == "password"
Brocade-Auth-Role = "zoneadmin",
Brocade-AVPairs1 = "HomeLF=1;LFRoleList=securityadmin:2,4-8,10”
Brocade-AVPairs2 = "LFRoleList=admin:11-13, 15, 17, 19;user:22-25,29,31"
Brocade-AVPairs3 = "ChassisRole=switchadmin"

Setting up a RADIUS server
NOTE
To set up the RADIUS server, you must know the switch IP address, in either IPv4 or IPv6 notation,
or the name to connect to switches. Use the ipAddrShow command to display a switch IP address.
For Brocade Backbones, the switch IP addresses are aliases of the physical Ethernet interfaces on
the CP blades. When specifying client IP addresses for the logical switches in these systems, make
sure the CP blade IP addresses are used. For accessing both the active and standby CP blades,
and for the purpose of HA failover, both of the CP blade IP addresses must be included in the
RADIUS server configuration.
User accounts should be set up by their true network-wide identities rather than by the account
names created on a Fabric OS switch. Along with each account name, the administrator must
assign appropriate switch access permissions. To manage a fabric, one can set these permissions
to user, admin, and securityAdmin.

Configuring RADIUS server support with Linux
The following procedures work for FreeRADIUS on Solaris and Red Hat Linux. FreeRADIUS is a
freeware RADIUS server that you can find at the following website:
http://www.freeradius.org
Follow the installation instructions at the website. FreeRADIUS runs on Linux (all versions),
FreeBSD, NetBSD, and Solaris. If you make a change to any of the files used in this configuration,
you must stop the server and restart it for the changes to take effect.
FreeRADIUS installation places the configuration files in $PREFIX/etc/raddb. By default, the
PREFIX is /usr/local.
Configuring RADIUS service on Linux consists of the following tasks:

• Adding the Brocade attributes to the server
• Creating the user
• Enabling clients
Adding the Brocade attributes to the server
1. Create and save the file $PREFIX/etc/raddb/dictionary.brocade with the following information:
# dictionary.brocade
#
VENDOR Brocade 1588
#
# attributes
#

Fabric OS Administrator’s Guide
53-1002920-02

175

6

Remote authentication

ATTRIBUTE
ATTRIBUTE
ATTRIBUTE
ATTRIBUTE
ATTRIBUTE
ATTRIBUTE
ATTRIBUTE

Brocade-Auth-Role
Brocade-AVPairs1
Brocade-AVPairs2
Brocade-AVPairs3
Brocade-AVPairs4
Brocade-Passwd-ExpiryDate
Brocade-Passwd-WarnPeriod

1
2
3
4
5
6
7

string
string
string
string
string
string
string

Brocade
Brocade
Brocade
Brocade
Brocade
Brocade
Brocade

This information defines the Brocade vendor ID as 1588, Brocade attribute 1 as
Brocade-Auth-Role, Brocade attribute 6 as Brocade-Passwd-ExpiryDate, and Brocade attribute
7 as Brocade-Passwd-WarnPeriod.
2. Open the file $PREFIX/etc/raddb/dictionary in a text editor and add the line:
$INCLUDE dictionary.brocade

As a result, the file dictionary.brocade is located in the RADIUS configuration directory and
loaded for use by the RADIUS server.
Creating the user
1. Open the $PREFIX/etc/raddb/user file in a text editor.
2. Add the user names and their permissions for users accessing the switch and authenticating
through RADIUS.
The user logs in using the permissions specified with Brocade-Auth-Role. The valid permissions
include root, admin, switchAdmin, zoneAdmin, securityAdmin, basic SwitchAdmin, fabricAdmin,
operator, and user. You must use quotation marks around “password” and “role”.
Example of adding a user name to the RADIUS authentication

For example, to set up an account called JohnDoe with admin permissions with a password
expiry date of May 28, 2008 and a warning period of 30 days:
JohnDoe Auth-Type := Local
User-Password == "johnPassword",
Brocade-Auth-Role = "admin",
Brocade-Passwd-ExpiryDate = "05/28/08",
Brocade-Passwd-WarnPeriod = "30"

Example of using the local system password to authenticate users

The following example uses the local system password file to authenticate users.
Auth-Type := System
Brocade-Auth-Role = "admin",
Brocade-AVPairs1 = "HomeLF=70",
Brocade-AVPairs2 = "LFRoleList=admin:2,4-8,70,80,128",
Brocade-AVPairs3 = "ChassisRole=switchadmin",
Brocade-Passwd-ExpiryDate = "11/10/2008",
Brocade-Passwd-WarnPeriod = "30"

When you use network information service (NIS) for authentication, the only way to enable
authentication with the password file is to force the Brocade switch to authenticate using Password
Authentication Protocol (PAP); this requires the -a pap option with the aaaConfig command.
Enabling clients
Clients are the switches that will use the RADIUS server; each client must be defined. By default, all
IP addresses are blocked.

176

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

The Brocade Backbones send their RADIUS requests using the IP address of the active CP. When
adding clients, add both the active and standby CP IP addresses so that, in the event of a failover,
users can still log in to the switch.
1. Open the $PREFIX/etc/raddb/client.config file in a text editor and add the switches that are to
be configured as RADIUS clients.
For example, to configure the switch at IP address 10.32.170.59 as a client:
client 10.32.170.59
secret
= Secret
shortname
= Testing Switch
nastype
= other

In this example, shortname is an alias used to easily identify the client. Secret is the shared
secret between the client and server. Make sure the shared secret matches that configured on
the switch (refer to “Adding an authentication server to the switch configuration” on page 193).
2. Save the file $PREFIX/etc/raddb/client.config, and then start the RADIUS server as follows:
$PREFIX/sbin/radiusd

Configuring RADIUS server support with Windows 2000
The instructions for setting up RADIUS on a Windows 2000 server are listed here for your
convenience but are not guaranteed to be accurate for your network environment. Always check
with your system administrator before proceeding with setup.

NOTE
All instructions involving Microsoft Windows 2000 can be obtained from www.microsoft.com or your
Microsoft documentation. Confer with your system or network administrator prior to configuration
for any special needs your network environment may have.
Configuring RADIUS service on Windows 2000 consists of the following steps:
1. Installing Internet Authentication Service (IAS)
For more information and instructions on installing IAS, refer to the Microsoft website.
2. Enabling the Challenge Handshake Authentication Protocol (CHAP)
If CHAP authentication is required, then Windows must be configured to store passwords with
reversible encryption. Reverse password encryption is not the default behavior; it must be
enabled.

NOTE

If a user is configured prior to enabling reverse password encryption, then the user’s password
is stored and cannot utilize CHAP. To use CHAP, the password must be re-entered after
encryption is enabled. If the password is not re-entered, then CHAP authentication will not work
and the user will be unable to authenticate from the switch.
Alternatives to using CHAP are Password Authentication Protocol (PAP), or PEAP-MSCHAPv2.

Fabric OS Administrator’s Guide
53-1002920-02

177

6

Remote authentication

3. Configuring a user
IAS is the Microsoft implementation of a RADIUS server and proxy. IAS uses the Windows
native user database to verify user login credentials; it does not list specific users, but instead
lists user groups. Each user group should be associated with a specific switch role. For
example, you should configure a user group for root, admin, factory, switchAdmin, and user,
and then add any users whose logins you want to associate to the appropriate group.
4. Configuring the server
For more information and instructions on configuring the server, refer to the Microsoft website.
You will need the following information to configure the RADIUS server for a Brocade switch. A
client is the device that uses the RADIUS server; in this case, it is the switch.
a.

For the Add RADIUS Client window, provide the following:
Client address (IP or DNS) — Enter the IP address of the switch.
Client-Vendor — Select RADIUS Standard.
Shared secret — Provide a password. Shared secret is a password used between the client
device and server to prevent IP address spoofing by unwanted clients. Keep your shared
secret password in a safe place. You will need to enter this password in the switch
configuration.
After clicking Finish, add a new client for all switches on which RADIUS authentication will
be used.

b.

In the Internet Authentication Service window, right-click the Remote Access Policies
folder, and then select New Remote Access Policy from the pop-up window.
A remote access policy must be created for each group of Brocade login permissions (root,
admin, factory, switchAdmin, and user) for which you want to use RADIUS. Apply this policy
to the user groups that you already created.

c.

In the Vendor-Specific Attribute Information window, enter the vendor code value 1588.
Click the Yes. It conforms option, and then click Configure Attribute.

d.

In the Configure VSA (RFC compliant) window, enter the following values, and then click
OK.
Vendor-assigned attribute number — Enter the value 1.
Attribute format — Enter String.
Attribute value — Enter the login role (root, admin, switchAdmin, user, and so on) that the
user group must use to log in to the switch.

e.

After returning to the Internet Authentication Service window, add additional policies for all
Brocade login types for which you want to use the RADIUS server. After this is done, you
can configure the switch.

NOTE

Windows 2008 RADIUS (NPS) support is also available.

178

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

RSA RADIUS server
Traditional password-based authentication methods are based on one-factor authentication, where
you confirm your identity using a memorized password. Two-factor authentication increases the
security by using a second factor to corroborate identification. The first factor is either a PIN or
password and the second factor is the RSA SecurID token.
RSA SecurID with an RSA RADIUS server is used for user authentication. The Brocade switch does
not communicate directly with the RSA Authentication Manager, so the RSA RADIUS server is used
in conjunction with the switch to facilitate communication.
To learn more about how RSA SecurID works, visit www.rsa.com for more information.
Setting up the RSA RADIUS server
For more information on how to install and configure the RSA Authentication Manager and the RSA
RADIUS server, refer to your documentation or visit www.rsa.com.
1. Create user records in the RSA Authentication Manager.
2. Configure the RSA Authentication Manager by adding an agent host.
3. Configure the RSA RADIUS server.
Setting up the RSA RADIUS server involves adding RADIUS clients, users, and vendor-specific
attributes to the RSA RADIUS server.
a.

Add the following data to the vendor.ini file:
vendor-product = Brocade
dictionary = brocade
ignore-ports = no
port-number-usage = per-port-type
help-id = 2000

b.

Create a brocade.dct file that must be added into the dictiona.dcm file located in the
following path:
C:\Program Files\RSA Security\RSA RADIUS\Service
Figure 11 on page 180 shows what the brocade.dct file should look like and Figure 12 on
page 180 shows what needs to be modified in the dictiona.dcm file.

NOTE

The dictionary files for the RSA RADIUS server must remain in the installation directory. Do
not move the files to other locations on your computer.

c.

Add Brocade-VSA macro and define the attributes as follows:

• vid (Vendor-ID): 1588
• type1 (Vendor-Type): 1
• len1 (Vendor-Length): >=2

Fabric OS Administrator’s Guide
53-1002920-02

179

6

Remote authentication

#######################################################################
# brocade.dct -- Brocade Dictionary
#
# (See readme.dct for more details on the format of this file)
#######################################################################
#
# Use the Radius specification attributes in lieu of the Brocade one:
#
@radius.dct

MACRO Brocade-VSA(t,s) 26 [vid=1588 type1=%t% len1=+2 data=%s%]
ATTRIBUTE
ATTRIBUTE
ATTRIBUTE

Brocade-Auth-Role
Brocade-Passwd-ExpiryDate
Brocade-Passwd-WarnPeriod

Brocade-VSA(1,string) r
Brocade-VSA(6,string) r
Brocade-VSA(7,integer) r

#######################################################################
# brocade.dct -- Brocade Dictionary
#######################################################################

FIGURE 11

Example of a brocade.dct file

#######################################################################
# dictiona.dcm
#######################################################################
# Generic Radius
@radius.dct
#
# Specific Implementations (vendor specific)
#
@3comsw.dct
@aat.dct
@acc.dct
@accessbd.dct
@agere.dct
@agns.dct
@airespace.dct
@alcatel.dct
@altiga.dct
@annex.dct
@aptis.dct
@ascend.dct
@ascndvsa.dct
@axc.dct
@bandwagn.dct
@brocade.dct <-------

FIGURE 12
d.

180

Example of the dictiona.dcm file

When selecting items from the Add Return List Attribute, select Brocade-Auth-Role and
type the string Admin. The string you type equals the role on the switch.

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

e.

Add the Brocade profile.

f.

In RSA Authentication Manager, edit the user records that will be authenticated using RSA
SecurID.

LDAP configuration and Microsoft Active Directory
LDAP provides user authentication and authorization using the Microsoft Active Directory service or
using OpenLDAP in conjunction with LDAP on the switch. This section discusses authentication and
authorization using Microsoft Active Directory. For information about authentication and
authorization using OpenLDAP, refer to “LDAP configuration and OpenLDAP” on page 184.
Two operational modes exist in LDAP authentication, FIPS mode and non-FIPS mode. This section
discusses LDAP authentication in non-FIPS mode. For more information on LDAP in FIPS mode,
refer to Chapter 8, “Configuring Security Policies”. The following are restrictions when using LDAP in
non-FIPS mode:

• There is no password change through Active Directory.
• There is no automatic migration of newly created users from the local switch database to
Active Directory. This is a manual process explained later.

• Only IPv4 is supported for LDAP on Windows 2000 and LDAP on Windows Server 2003.
For LDAP on Windows Server 2008, both IPv4 and IPv6 are supported.

• LDAP authentication is used on the local switch only and not for the entire fabric.
• You can use the User-Principal-Name and not the Common-Name for AD LDAP authentication.
To provide backward compatibility, authentication based on the Common Name is still
supported for Active Directory LDAP 2000 and 2003. Common Name-based authentication is
not recommended for new installations.

• A user can belong to multiple groups as long as one of the groups is the primary group. The
primary group in the AD server should not be set to the group corresponding to the switch role.
You can choose any other group.

• A user can be part of any Organizational Unit (OU).
• Active Directory LDAP 2000, 2003, and 2008 are supported.
When authentication is performed by User-Principal-Name, in Fabric OS 7.1.0 and later releases,
the suffix part of the name (the @domain-name part) can be omitted when the user logs in. If the
suffix part of the User-Principal-Name name is omitted, the domain name configured for the LDAP
server (in the aaaConfig --add server -d domain command) is added and used for authentication
purposes.
Roles for Brocade-specific users can be added through the Microsoft Management Console.
Groups created in Active Directory must correspond directly to the RBAC user roles on the switch.
Role assignments can be achieved by including the user in the respective group. A user can be
assigned to multiple groups such as Switch Admin and Security Admin. For LDAP servers, you can
use the ldapCfg -–maprole ldap_role_name switch_role command to map LDAP server
permissions to one of the default roles available on a switch. For more information on RBAC roles,
refer to “Role-Based Access Control” on page 152.

NOTE
All instructions involving Microsoft Active Directory can be obtained from www.microsoft.com or your
Microsoft documentation. Confer with your system or network administrator prior to configuration
for any special needs your network environment may have.

Fabric OS Administrator’s Guide
53-1002920-02

181

6

Remote authentication

Configuring Microsoft Active Directory LDAP service
The following is an overview of the process used to set up LDAP.
1. If your Windows Active Directory server for LDAP needs to be verified by the LDAP client (that is,
the Brocade switch), then you must install a Certificate Authority (CA) certificate on the
Windows Active Directory server for LDAP.
Follow Microsoft instructions for generating and installing CA certificates on a Windows server.
2. Create a user in Microsoft Active Directory server.
For instructions on how to create a user, refer to www.microsoft.com or Microsoft
documentation to create a user in your Active Directory.
3. Create a group name that uses the switch’s role name so that the Active Directory group’s
name is the same as the switch’s role name.
or
Use the ldapCfg -–maprole ldap_role_name switch_role command to map an LDAP server role
to one of the default roles available on the switch.
4. Associate the user to the group by adding the user to the group.
5. Add the user’s Administrative Domains or Virtual Fabrics to the CN_list by either editing the
adminDescription value or adding the brcdAdVfData attribute to the existing Active Directory
schema.
This action maps the Admin Domains or Virtual Fabrics to the user name. Multiple Admin
Domains can be added as a string value separated by the underscore character ( _ ). Virtual
Fabrics are added as a string value separate by a comma ( , ) and entered as a range.

Creating a user
To create a user in Active Directory, refer to www.microsoft.com or Microsoft documentation. There
are no special attributes to set. You can use a fully qualified name for logging in; for example, you
can log in as "user@domain.com".

Creating a group
To create a group in Active Directory, refer to www.microsoft.com or Microsoft documentation. You
must verify that the group has the following attributes:

•
•
•
•

The name of the group must match the RBAC role.
The Group Type must be Security.
The Group Scope must be Global.
The primary group in the AD server should not be set to the group corresponding to the switch
role. You can choose any other group.

• If the user you created is not a member of the Users OU, then the User Principal Name, in the
format of "user@domain", is required to log in.

182

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

Assigning the group (role) to the user
To assign the user to a group in Active Directory, refer to www.microsoft.com or Microsoft
documentation. If you have a user-defined group, use the ldapCfg -–maprole ldap_role_name
switch_role command to map LDAP server permissions to one of the default roles available on a
switch. Alternatively, update the memberOf field with the login permissions (root, admin,
switchAdmin, user, and so on) that the user must use to log in to the switch.

Adding an Admin Domain or Virtual Fabric list
1. From the Windows Start menu, select Programs> Administrative Tools> ADSI.msc.
ADSI is a Microsoft Windows Resource Utility. This utility must be installed to proceed with the
rest of the setup. For Windows 2003, this utility comes with Service Pack 1 or you can
download this utility from the Microsoft website.
2. Go to CN=Users.
3. Select Properties. Click the Attribute Editor tab.
4. Double-click the adminDescription attribute.
The String Attribute Editor dialog box displays.

NOTE

The attribute can be added to user objects only.
5. Perform the appropriate action based on whether you are using Administrative Domains or
Virtual Fabrics:

• If you are using Administrative Domains, enter the values of the Admin Domains separated
by an underscore ( _ ) into the Value field.
Example for adding Admin Domains

adlist_0_10_200_endAd
Home Admin Domain (homeAD) for the user will be the first value in the adlist (Admin
Domain list). If a user has no values assigned in the adlist attribute, then the homeAD "0"
will be the default administrative domain for the user.

• If you are using Virtual Fabrics, enter the values of the logical fabrics separated by a
semi-colon ( ; ) into the Value field.
Example for adding Virtual Fabrics

HomeLF=10;LFRoleList=admin:128,10;ChassisRole=admin
In this example, the logical switch that would be logged in to by default is 10. If 10 is not
available, then the lowest FID available will be chosen. You would have permission to enter
logical switch 128 and 10 in an admin role and you would also have the chassis role
permission of admin.

NOTE

You can perform batch operations using the Ldifde.exe utility. For more information on
importing and exporting schemas, refer to your Microsoft documentation or visit
www.microsoft.com.

Fabric OS Administrator’s Guide
53-1002920-02

183

6

Remote authentication

Adding attributes to the Active Directory schema
To create a group in Active Directory, refer to www.microsoft.com or Microsoft documentation. You
must:

• Add a new attribute brcdAdVfData as Unicode String.
• Add brcdAdVfData to the person’s properties.

LDAP configuration and OpenLDAP
Fabric OS provides user authentication and authorization by means of OpenLDAP or the Microsoft
Active Directory service in conjunction with LDAP on the switch. This section discusses
authentication and authorization using OpenLDAP. For information about authentication and
authorization using Microsoft Active Directory, refer to “LDAP configuration and Microsoft Active
Directory” on page 181.
Two operational modes exist in LDAP authentication: FIPS mode and non-FIPS mode. This section
discusses LDAP authentication in non-FIPS mode. For information on LDAP in FIPS mode, refer to
Chapter 8, “Configuring Security Policies”. When using OpenLDAP in non-FIPS mode, you must use
the Common-Name for OpenLDAP authentication. User-Principal-Name is not supported in
OpenLDAP. OpenLDAP 2.4.23 is supported.
When a user is authenticated, the role of the user is obtained from the memberOf attribute, which
determines group membership. This feature is supported in OpenLDAP through the memberOf
overlay. You must use this overlay on the OpenLDAP server to assign membership information.
For OpenLDAP servers, you can use the ldapCfg -–maprole ldap_role_name switch_role command
to map LDAP server permissions to one of the default roles available on a switch. For more
information on RBAC roles, see “Role-Based Access Control” on page 152.

OpenLDAP server configuration overview
For complete details about how to install and configure an OpenLDAP server, refer to the OpenLDAP
user documentation at http://www.openldap.org/doc/. A few key steps for the Brocade
environment are outlined here.
1. If your OpenLDAP server needs to be verified by the LDAP client (that is, the Brocade switch),
then you must install a Certificate Authority (CA) certificate on the OpenLDAP server.
Follow OpenLDAP instructions for generating and installing CA certificates on an OpenLDAP
server.
2. Enable group membership through the memberOf mechanism by including the memberOf
overlay in the slapd.conf file.
3. Create entries (users) in the OpenLDAP Directory.
4. Assign users to groups by using the member attribute.
5. Use the ldapCfg -–maprole ldap_role_name switch_role command to map an LDAP server role
to one of the default roles available on the switch.
6. Add the user’s Admin Domains or Virtual Fabrics to the user entry.

184

a.

Add the brcdAdVfData attribute to the existing OpenLDAP schema,

b.

Add the brcdAdVfData attribute to the user entry in the LDAP directory with a value that
identifies the Administrative Domains or Virtual Fabrics with which to associate the user.

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

Enabling group membership
Group membership in OpenLDAP is specified by an overlay called memberOf. Overlays are helpful
in customizing the back-end behavior without requiring changes to the back-end code. The
memberOf overlay updates the memberOf attribute whenever changes occur to the membership
attribute of entries of the groupOfNames objectClass. To include this overlay, add “overlay
memberof” to the slapd.conf file, as shown in the following example.
overlay memberof
Example file:
include
include
include

/usr/local/etc/openldap/schema/core.schema
/usr/local/etc/openldap/schema/cosine.schema
/usr/local/etc/openldap/schema/local.schema

###############################################
TLSCACertificateFile /root/sachin/ldapcert/cacert.pem
TLSCertificateFile
/root/sachin/ldapcert/serverCert.pem
TLSCertificateKeyFile /root/sachin/ldapcert/serverKey.pem
TLSVerifyClient never

pidfile
argsfile

/usr/local/var/run/slapd.pid
/usr/local/var/run/slapd.args

database
suffix
rootdn
rootpw

bdb
"dc=mybrocade,dc=com"
"cn=Manager,dc=mybrocade,dc=com"
{SSHA}HL8uT5hPaWyIdcP6yAheMT8n0GoWubr3

# The database directory MUST exist prior to running slapd AND
# should only be accessible by the slapd and slap tools.
# Mode 700 recommended.
directory
/usr/local/var/openldap-data
# Indices to maintain
index
objectClass
eq
overlay memberof

Adding entries to the directory
To add entries in the OpenLDAP directory, perform the following steps.
1. Using a text editor of your choice, create a .ldif file and enter the information for the entry.
The following example defines an organizational role for the Directory Manager in a .ldif file for
an organization with the domain name mybrocade.com.
# Organization for mybrocade Corporation
dn: dc=mybrocade,dc=com
objectClass: dcObject
objectClass: organization
dc: mybrocade
o: Mybrocade Corporation
description: Mybrocade Corporation

############################################################################
# Organizational Role for Directory Manager

Fabric OS Administrator’s Guide
53-1002920-02

185

6

Remote authentication

dn: cn=Manager,dc=mybrocade,dc=com
objectClass: organizationalRole
cn: Manager
description: Directory Manager

2. Enter the ldapadd command to add the contents of the .ldif file to the Directory, where test.ldif
is the file you created in step 1.
> ldapadd -D cn=Manager,dc=mybrocade,dc=com -x -w secret -f test.ldif

Assigning a user to a group
Before you can assign a user to a group, the memberOf overlay must be added to the slapd.conf
file. Refer to “Enabling group membership” on page 185 for details.
1. In a .ldif file, create a “groupOfNames” objectClass entry with the name of the group, for
example, “admin,” to create a group.
2. Set a “member” attribute for the group instance to identify the member, as in this example:
“cn=Sachin,cn=Users,dc=mybrocade,dc=com”
Automatically, the “memberOf” attribute of the entry Sachin will have the value
“cn=admin,ou=groups,dc=mybrocade,dc=com”, which assigns Sachin to the admin group.
3. Enter the ldapadd command.
For example, the .ldif file might contain information similar to the following:
#Groups in organization
dn: ou=groups,dc=mybrocade,dc=com
objectclass:organizationalunit
ou: groups
description: generic groups branch
dn: cn=admin,ou=groups,dc=mybrocade,dc=com
objectclass: groupofnames
cn: admin
description: Members having admin permission
#Add members for admin group
member: cn=sachin,cn=Users,dc=mybrocade,dc=com

Assigning the LDAP role to a switch role
Use the ldapCfg -–maprole ldap_role_name switch_role command to map LDAP server
permissions to one of the default roles available on a switch.

Modifying an entry
To modify a directory entry, perform the following steps.
1. Create a .ldif file containing the information to be modified.
2. Enter the ldapmodify -f filename command, where filename is the .ldif file you created in
step 1.
Example to delete a user attribute

1. Create or edit a .ldif file with an entry similar to the following.
#########Deleting an attr
#dn: cn=Sachin,cn=Users,dc=mybrocade,dc=com

186

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

#changetype: modify
#delete: memberof

2. Enter the following ldapmodify command, where test.ldif is the name of the file you edited in
step 1.
> ldapmodify -D cn=Sachin,dc=mybrocade,dc=com –x -w secret -f test.ldif

Example to add a group member

1. Create or edit a .ldif file with an entry similar to the following.
##########Adding an attr value
dn: cn=admin,ou=groups,dc=mybrocade,dc=com
changetype: modify
add: member
member: cn=test1,cn=Users,dc=mybrocade,dc=com

2. Enter the following ldapmodify command, where test1.ldif is the name of the file you edited in
step 1.
> ldapmodify -D cn=admin,dc=mybrocade,dc=com –x -w secret -f test1.ldif

Example to delete a group member

1. Create or edit a .ldif file with contents similar to the following.
########Deleting an attr value
#dn: cn=admin,ou=groups,dc=mybrocade,dc=com
#changetype: modify
#delete: member
#member: cn=Sachin,cn=Users,dc=mybrocade,dc=com

2. Enter the following ldapmodify command, where test2.ldif is the name of the file you edited in
step 1.
> ldapmodify -D cn=admin,dc=mybrocade,dc=com –x -w secret -f test2.ldif

Example to change the value of an attribute

1. Create or edit a .ldif file with contents similar to the following.
#######Replacing an attribute value
dn: cn=test,cn=Users,dc=mybrocade,dc=com
changetype: modify
replace: uid
uid: test

2. Enter the following ldapmodify command, where test3.ldif is the name of the file you edited in
step 1.
> ldapmodify -D cn=admin,dc=mybrocade,dc=com –x -w secret -f test3.ldif

The value of the uid attribute is changed to “test”.

Adding an Admin Domain or Virtual Fabric list
If your network uses Admin Domains, you can specify a list of Admin Domain numbers to which the
user has access.
Use the brcdAdVfData attribute to map a role to a Virtual Fabric or Admin Domain. To perform this
operation, you must modify the schema to include the definition of the brcdAdVfData attribute and
the definition of a user class that can use this attribute. You can then add this attribute to user
entries in the LDAP directory.

Fabric OS Administrator’s Guide
53-1002920-02

187

6

Remote authentication

1. In a schema file, assign the brcdAdVfData attribute to a user class.
The following sample schema file defines a new objectClass named “user” with optional
attributes “brcdAdVfData” and “description”.
#New attr brcdAdVfData
attributetype ( 1.3.6.1.4.1.8412.100
NAME ( 'brcdAdVfData' )
DESC 'Brocade specific data for LDAP authentication'
EQUALITY caseIgnoreIA5Match
SUBSTR caseIgnoreIA5SubstringsMatch
SYNTAX 1.3.6.1.4.1.1466.115.121.1.26{256} )
objectclass ( 1.3.6.1.4.1.8412.110 NAME 'user'
DESC 'Brocade switch specific person'
SUP top AUXILIARY
MAY ( brcdAdVfData $ description ) )

2. Include the schema file in the slapd.conf file.
The following example slapd.conf line assumes that local.schema contains the attribute
definition provided in step 1.
include

/usr/local/etc/openldap/schema/local.schema

3. Include the brcdAdVfData attribute in a user entry in the LDAP directory.

• If you are using Administrative Domains, enter the value of each Admin Domain separated by
an underscore ( _ ). Each number represents the number of the Admin Domain to which the
user has access. The first such number represents the user’s Home domain.
Example for adding Admin Domains

In the following example, the user is granted access to Admin Domains 0, 10, and 200. Admin
Domain 0 is the domain that the user initially logs in to.
brcdAdVfData: adlist_0_10_200_endAd

• If you are using Virtual Fabrics, enter the value of the logical fabrics to which the user has
access. Up to three value fields can be specified, separated by an semicolons ( ; ):

-

The HomeLF field specifies the user’s home Logical Fabric.

-

The ChassisRole field designates the permissions that apply to the ChassisRole subset of
commands.

The LFRole list field specifies the additional Logical Fabrics to which the user has access
and the user’s access permissions for those Logical Fabrics. Logical Fabric numbers are
separated by commas ( , ). A hyphen ( - ) indicates a range.

Example for adding Virtual Fabrics

In the following example, the logical switch that would be logged in to by default is 10. If 10 is
not available, then the lowest FID available will be chosen.The user is given permission to enter
logical switches 1 through 128 in an admin role and is also given the chassis role permission
of admin.
brcdAdVfData: HomeLF=10;LFRoleList=admin:1-128;ChassisRole=admin

The following fragment from a file named test4.ldif provides an entry for a user with Virtual Fabric
access roles.
# Organizational Role for Users
dn: cn=Users,dc=mybrocade,dc=com

188

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

objectClass: organizationalRole
cn: Users
description: User
# User entries
dn: cn=Sachin,cn=Users,dc=mybrocade,dc=com
objectClass: user
objectClass: person
objectClass: uidObject
cn: Sachin
sn: Mishra
description: First user
brcdAdVfData: HomeLF=30;LFRoleList=admin:1-128;ChassisRole=admin
userPassword: pass
uid: mishras@mybrocade.com

The following command adds the user to the LDAP directory.
> ldapadd -D cn=Sachin,dc=mybrocade,dc=com -x -w secret -f test4.ldif

TACACS+ service
Fabric OS can authenticate users with a remote server using the Terminal Access Controller
Access-Control System Plus (TACACS+) protocol. TACACS+ is a protocol used in AAA server
environments consisting of a centralized authentication server and multiple Network Access
Servers or clients. Once configured to use TACACS+, a Brocade switch becomes a Network Access
Server (NAS).
The following authentication protocols are supported by the TACACS+ server for user
authentication:

• Password Authentication Protocol (PAP)
• Challenge Handshake Authentication Protocol (CHAP)
TACACS+ is not a FIPS-supported protocol, so you cannot configure TACACS+ in FIPS mode. To
enable FIPS, any TACACS+ configuration must be removed.
The TACACS+ server can be a Microsoft Windows server or a Linux server. For Linux servers, use
TACACS+ 4.0.4 or later from Cisco. For Microsoft Windows servers, use any TACACS+ freeware that
uses TACACS+ protocol v1.78 or later.

TACACS+ configuration overview
Configuration is required on both the TACACS+ server and the Brocade switch. On the TACACS+
server, you should assign a role for each user and, if Admin Domains or Virtual Fabrics are in use,
provide lists of Admin Domains or Virtual Fabrics to which the user should have access. For details,
refer to “The tac_plus.cfg file” on page 190.
On the Brocade switch, use the aaaConfig command to configure the switch to use TACACS+ for
authentication. The aaaConfig command also allows you to specify up to five TACACS+ servers.
When a list of servers is configured, failover from one server to another server happens only if a
TACACS+ server fails to respond. It does not happen when user authentication fails.
Failover to another TACACS+ server is achieved by means of a timeout. You can configure a timeout
value for each TACACS+ server, so that the next server can be used in case the first server is
unreachable. The default timeout value is 5 seconds.

Fabric OS Administrator’s Guide
53-1002920-02

189

6

Remote authentication

Retry, the number of attempts to authenticate with a TACAS+ server, is also allowed. The default
value is 5 attempts. If authentication is rejected or times out, Fabric OS will try again. The retry
value can also be customized for each user.
Refer to “Remote authentication configuration on the switch” on page 192 for details about
configuring the Brocade switch for authenticating users with a TACACS+ server.

Configuring the TACACS+ server on Linux
Fabric OS software supports TACACS+ authentication on a Linux server running the Open Source
TACACS+ LINUX package v4.0.4 from Cisco. To install and configure this software, perform the
following steps.
1. Download the TACACS+ software from http://www.cisco.com and install it.
2. Configure the TACACS+ server by editing the tac_plus.cfg file.
Refer to “The tac_plus.cfg file” for details.
3. Run the tac_plus daemon to start and enable the TACACS+ service on the server.
> tac_plus -d 16 /usr/local/etc/mavis/sample/tac_plus.cfg

The tac_plus.cfg file
The TACACS+ server is configured in the tac_plus.cfg file. Open the file by using the editor of your
choice and customize the file as needed.
You must add users into this file and provide some attributes specific to the Brocade
implementation. Table 25 lists and defines attributes specific to Brocade.

TABLE 25

Brocade custom TACACS+ attributes

Attribute

Purpose

brcd-role

Role assigned to the user account

brcd-AV-Pair1

The Admin Domain or Virtual Fabric member list, and chassis role

brcd-AV-Pair2

The Admin Domain or Virtual Fabric member list, and chassis role

brcd-passwd-expiryDate

The date on which the password expires

brcd-passwd-warnPeriod

The time before expiration for the user to receive a warning message

Adding a user and assigning a role
When adding a user to the tac_plus.cfg file, you should at least provide the brcd-role attribute. The
value assigned to this attribute should match a role defined for the switch. When a login is
authenticated, the role specified by the brcd-role attribute represents the permissions granted to
the account. If no role is specified, or if the specified role does not exist on the switch, the account
is granted user role permissions only.
Refer to “Role-Based Access Control” on page 152 for details about roles.
The following fragment from a tac_plus.cfg file adds a user named fosuser1 and assigns the
securityAdmin role to the account.
user = fosuser1 {
chap = cleartext "my$chap$pswrd"
pap = cleartext "pap-password"
service = exec {

190

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

brcd-role = securityAdmin;
}
}

Configuring Admin Domain lists
If your network uses Admin Domains, you should create Admin Domain lists for each user to
identify the Admin Domains to which the user has access.
Assign the following key-value pairs to the brcd-AV--Pair1 and, optionally, brcd-AV-Pair2 attributes to
grant the account access to the Admin Domains:

• HomeAD is the designated home Admin Domain for the account. The valid range of values is
from 0 through 255. The first valid HomeAD key-value pair is accepted by the switch, and any
additional HomeAD key-value pairs are ignored.

• ADList is a comma-separated list of Administrative Domain numbers of which this account is a
member. Valid numbers range from 0 through 255. A - between two numbers specifies a
range.
The following example sets the home Admin Domain for the fosuser4 account to 255 and grants
the account access to Admin Domains 1, 2, 3, and 200 through 255.
user = fosuser4 {
pap = clear "password"
chap = clear "password"
password = clear "password"
service = shell {
set brcd-role = securityAdmin
set brcd-AV-Pair1 = "homeAD=255;ADList=1,2,3";
set brcd-AV-Pair2 = “ADList=200-255”;
}
}

Configuring Virtual Fabric lists
If your network uses Virtual Fabrics, you should create Virtual Fabric lists for each user to identify
the Virtual Fabrics to which the account has access.
Assign the following key-value pairs to the brcd-AV--Pair1 and, optionally, brcd-AV-Pair2 attributes to
grant the account access to the Virtual Fabrics:

• HomeLF is the designated home Virtual Fabric for the account. The valid values are from 1
through 128 and chassis context. The first valid HomeLF key-value pair is accepted by the
switch. Additional HomeLF key-value pairs are ignored.

• LFRoleList is a comma-separated list of Virtual Fabric ID numbers to which this account is a
member, and specifies the role the account has on those Virtual Fabrics. Valid numbers range
from 1 through 128. A - between two numbers specifies a range.
The following example sets the home Virtual Fabric for the userVF account to 30 and allows the
account admin role access to Virtual Fabrics 1, 3, and 4 and securityAdmin access to Virtual
Fabrics 5 and 6.
user = userVF {
pap = clear “password”
service = shell {
set brcd-role = zoneAdmin
set brcd-AV-Pair1 = “homeLF=30;LFRoleList=admin:1,3,4;securityAdmin:5,6”
set brcd-AV-Pair2 = “chassisRole=admin”
}
}

Fabric OS Administrator’s Guide
53-1002920-02

191

6

Remote authentication

Configuring the password expiration date
FabricOS allows you to configure a password expiration date for each user account and to configure
a warning period for notifying the user that the account password is about to expire. To configure
these values, set the following attributes:

• brcd-passwd-expiryDate sets the password expiration date in mm/dd/yyyy format.
• brcd-passwd-warnPeriod sets the warning period as a number of days.
The following example sets the password expiration date for the fosuser5 account. It also specifies
that a warning be sent to the user 30 days before the password is due to expire.
user = fosuser5 {
pap = clear "password"
chap = clear "password"
password = clear "password"
service = shell {
set brcd-role = securityAdmin
set brcd-passwd-expiryDate = 03/21/2014;
set brcd-passwd-warnPeriod = 30;
}
}

Configuring a Windows TACACS+ server
Fabric OS is compatible with any TACACS+ freeware for Microsoft Windows that uses TACACS+
protocol version v1.78. Refer to the vendor documentation for configuration details.

Remote authentication configuration on the switch
At least one RADIUS, LDAP, or TACACS+ server must be configured before you can enable a remote
authentication service. You can configure the remote authentication service even if it is disabled on
the switch. You can configure up to five RADIUS, LDAP, or TACACS+ servers. You must be logged in
as admin or switchAdmin to configure the RADIUS service.

NOTE

On dual-CP Backbones (Brocade DCX, DCX-4S, DCX 8510-4, and DCX 8510-8 devices), the switch
sends its RADIUS, LDAP, or TACACS+ request using the IP address of the active CP. When adding
clients, add both the active and standby CP IP addresses so that users can still log in to the switch
in the event of a failover.
RADIUS, LDAP, or TACACS+ configuration is chassis-based configuration data. On platforms
containing multiple switch instances, the configuration applies to all instances. The configuration is
persistent across reboots and firmware downloads. On a chassis-based system, the command
must replicate the configuration to the standby CP.
Multiple login sessions can invoke the aaaConfig command simultaneously. The last session that
applies the change is the one whose configuration is in effect. This configuration is persistent after
an HA failover.
The authentication servers are contacted in the order they are listed, starting from the top of the
list and moving to the bottom.

192

Fabric OS Administrator’s Guide
53-1002920-02

Remote authentication

6

Adding an authentication server to the switch configuration
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aaaConfig --add command.
At least one authentication server must be configured before you can enable the RADIUS, LDAP, or
TACACS+ service.
If no RADIUS, LDAP, or TACACS+ configuration exists, turning on the authentication mode triggers
an error message. When the command succeeds, the event log indicates that the configuration is
enabled or disabled.

Enabling and disabling remote authentication
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aaaConfig --authspec command to enable or disable RADIUS, LDAP, or TACACS+.
You must specify the type of service as one of RADIUS, LDAP, or TACACS+. Local is used for
local authentication if the user authentication fails on the authentication server.
Example enabling RADIUS
switch:admin> aaaconfig --authspec "radius;local" -backup

Example enabling LDAP
switch:admin> aaaconfig --authspec "ldap;local" -backup

Example enabling TACACS+
switch:admin> aaaconfig --authspec "tacacs+;local" -backup

Deleting an authentication server from the configuration
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aaaConfig --remove command.
When the command succeeds, the event log indicates that the server is removed.

Changing an authentication server configuration
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aaaConfig --change command.

Changing the order in which authentication servers are contacted for service
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aaaConfig --move command.
When the command succeeds, the event log indicates that a server configuration is changed.

Fabric OS Administrator’s Guide
53-1002920-02

193

6

Remote authentication

Displaying the current authentication configuration
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aaaConfig --show command.
If a configuration exists, its parameters are displayed. If the RADIUS, LDAP, or TACACS+ service
is not configured, only the parameter heading line is displayed. Parameters include:
Position
Server
Port
Secret
Timeouts
Authentication

The order in which servers are contacted to provide service.
The server names or IPv4 or IPv6 addresses. IPv6 is not supported when using PEAP
authentication.
The server ports.
The shared secrets.
The length of time servers have to respond before the next server is contacted.
The type of authentication being used on servers.

Configuring local authentication as backup
It is useful to enable local authentication, so that the switch can take over authentication locally if
the RADIUS or LDAP servers fail to respond because of power outage or network problems.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aaaConfig --authspec command to enable or disable RADIUS, LDAP, or TACACS+
with local authentication as a backup authentication mechanism.
You must specify the type of service as one of RADIUS, LDAP, or TACACS+. Local is used for
local authentication if the user authentication fails on the authentication server.of enabling
local authentication as a backup for RADIUS.
switch:admin> aaaconfig --authspec "radius;local" -backup

Example for LDAP
switch:admin> aaaconfig --authspec "ldap;local" -backup

Example for TACACS+
switch:admin> aaaconfig --authspec "tacacs+;local" -backup

For details about the aaaConfig command refer toTable 22 on page 169.
When local authentication is enabled and the authentication servers fail to respond, you can log in
to the default switch accounts (admin and user) or any user-defined account. You must know the
passwords of these accounts.
When the aaaConfig command succeeds, the event log indicates that local database
authentication is disabled or enabled.

194

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

7

Configuring Protocols

In this chapter
• Security protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Secure Copy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Secure Shell protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Secure Sockets Layer protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Simple Network Management Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Telnet protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Listener applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Ports and applications used by switches. . . . . . . . . . . . . . . . . . . . . . . . . . .

195
196
197
200
206
226
228
229

Security protocols
Security protocols provide endpoint authentication and communications privacy using
cryptography. Typically, you are authenticated to the switch while the switch remains
unauthenticated to you. This means that you can be sure with what you are communicating. The
next level of security, in which both ends of the conversation are sure with whom they are
communicating, is known as two-factor authentication. Two-factor authentication requires public
key infrastructure (PKI) deployment to clients.
Fabric OS supports the secure protocols shown in Table 26.

TABLE 26

Secure protocol support

Protocol Description
HTTPS

HTTPS is a Uniform Resource Identifier scheme used to indicate a secure HTTP connection. Web Tools
supports the use of Hypertext Transfer Protocol over SSL (HTTPS).

IPsec

Internet Protocol Security (IPsec) is a framework of open standards for providing confidentiality,
authentication and integrity for IP data transmitted over untrusted links or networks.

LDAP

Lightweight Directory Access Protocol with TLS uses a certificate authority (CA). By default, LDAP traffic is
transmitted unsecured. With the import of signed certificates, you can make LDAP traffic confidential
and secure by using Secure Sockets Layer (SSL) / Transport Layer Security (TLS) technology in
conjunction with LDAP.

SCP

Secure Copy (SCP) is a means of securely transferring computer files between a local and a remote host
or between two remote hosts, using the Secure Shell (SSH) protocol. Configuration upload and download
support the use of SCP.

SNMP

Simple Network Management Protocol (SNMP) is used in network management systems to monitor
network-attached devices for conditions that warrant administrative attention. Supports SNMPv1, v2,
and v3.

Fabric OS Administrator’s Guide
53-1002920-02

195

7

Secure Copy

TABLE 26

Secure protocol support (Continued)

Protocol Description
SSH

Secure Shell (SSH) is a network protocol that allows data to be exchanged over a secure channel
between two computers. Encryption provides confidentiality and integrity of data. SSH uses public-key
cryptography to authenticate the remote computer and allow the remote computer to authenticate the
user, if necessary.

SSL

Fabric OS uses Secure Socket Layer (SSL) to support HTTPS. A certificate must be generated and
installed on each switch to enable SSL. Supports SSLv3, 128-bit encryption by default.

Table 27 describes additional software or certificates that you must obtain to deploy secure
protocols.

TABLE 27

Items needed to deploy secure protocols

Protocol

Host side

Switch side

SSHv2

Secure shell client

None

HTTPS

No requirement on host side except a browser
that supports HTTPS

Switch IP certificate for SSL

SCP

SSH daemon, SCP server

None

SNMPv1, SNMPv2, SNMPv3

None

None

The security protocols are designed with the four main use cases described in Table 28.

TABLE 28

Main security scenarios

Fabric

Management interfaces Comments

Nonsecure

Nonsecure

No special setup is needed to use Telnet or HTTP.

Nonsecure

Secure

Secure protocols may be used. An SSL switch certificate must be installed if
HTTPS is used.

Secure

Secure

Switches running earlier Fabric OS versions can be part of the secure fabric,
but they do not support secure management.
Secure management protocols must be configured for each participating
switch. Nonsecure protocols may be disabled on nonparticipating switches.
If SSL is used, then certificates must be installed. For more information on
installing certificates, refer to “Installing a switch certificate” on page 203.

Secure

Nonsecure

You must use SSH because Telnet is not allowed with some features.

Secure Copy
The Secure Copy protocol (SCP) runs on port 22. It encrypts data during transfer, thereby avoiding
packet sniffers that attempt to extract useful information during data transfer. SCP relies on SSH to
provide authentication and security.

196

Fabric OS Administrator’s Guide
53-1002920-02

Secure Shell protocol

7

Setting up SCP for configuration uploads and downloads
Use the following procedure to configure SCP for configuration uploads and downloads.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter one of the following commands:

• If Virtual Fabrics is enabled, enter the configurechassis command.
• If Virtual Fabrics is not enabled, enter the configure command.
3. Enter y or yes at the cfgload attributes prompt.
4. Enter y or yes at the Enforce secure configUpload/Download prompt.
Example of setting up SCP for configUpload/download
switch:admin# configure
Not all options will be available on an enabled switch.
To disable the switch, use the "switchDisable" command.
Configure...
System services (yes, y, no, n): [no] n
ssl attributes (yes, y, no, n): [no] n
http attributes (yes, y, no, n): [no] n
snmp attributes (yes, y, no, n): [no] n
rpcd attributes (yes, y, no, n): [no] n
cfgload attributes (yes, y, no, n): [no] y
Enforce secure config Upload/Download (yes, y, no, n): [no]# y
Enforce signature validation for firmware (yes, y, no, n): [no]#

Secure Shell protocol
To ensure security, Fabric OS supports Secure Shell (SSH) encrypted sessions. SSH encrypts all
messages, including the client transmission of the password during login. The SSH package
contains a daemon (sshd), which runs on the switch. The daemon supports a wide variety of
encryption algorithms, such as Blowfish-Cipher block chaining (CBC) and Advanced Encryption
Standard (AES).

NOTE

To maintain a secure network, you should avoid using Telnet or any other unprotected application
when you are working on the switch.
Commands that require a secure login channel must originate from an SSH session. If you start an
SSH session, and then use the login command to start a nested SSH session, commands that
require a secure channel will be rejected.
Fabric OS supports OpenSSH protocol v2.0 (ssh2) version 5.2p1. For more information on SSH,
refer to the SSH IETF website:
http://www.ietf.org/ids.by.wg/secsh.html
You can also refer to SSH, The Secure Shell: The Definitive Guide by Daniel J. Barrett, Ph. D.,
Richard E. Silverman, and Robert G. Byrnes.

Fabric OS Administrator’s Guide
53-1002920-02

197

7

Secure Shell protocol

SSH public key authentication
OpenSSH public key authentication provides password-less logins, known as SSH authentication,
that uses public and private key pairs for incoming and outgoing authentication. This feature allows
only one allowed-user to be configured to utilize outgoing OpenSSH public key authentication.Any
admin user can perform incoming Open SSH public key authentication. Using OpenSSH RSA and
DSA, the authentication protocols are based on a pair of specially generated cryptographic keys,
called the private key and the public key. The advantage of using these key-based authentication
systems is that in many cases, it is possible to establish secure connections without having to
depend on passwords for security. RSA asynchronous algorithms are FIPS-compliant.
Incoming authentication is used when the remote host needs to authenticate to the switch.
Outgoing authentication is used when the switch needs to authenticate to a server or remote host,
such as when running the configUpload or configDownload commands, or performing firmware
download. Both password and public key authentication can coexist on the switch.

Allowed-user
For outgoing authentication, the default admin user must set up the allowed-user with admin
permissions. By default, the admin is the configured allowed-user. While creating the key pair, the
configured allowed-user can choose a passphrase with which the private key is encrypted. Then the
passphrase must always be entered when authenticating to the switch. The allowed-user must
have admin permissions to perform OpenSSH public key authentication, import and export keys,
generate a key pair for an outgoing connection, and delete public and private keys.

Configuring incoming SSH authentication
1. Log in to your remote host.
2. Generate a key pair for host-to-switch (incoming) authentication by verifying that SSH v2 is
installed and working (refer to your host’s documentation as necessary) by entering the
following command:
ssh-keygen -t rsa

Example of RSA/DSA key pair generation
anyuser@mymachine: ssh-keygen -t rsa
Generating public/private rsa key pair.
Enter file in which to save the key (/users/anyuser/.ssh/id_rsa):
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /users/anyuser/.ssh/id_rsa.
Your public key has been saved in /users/anyuser/.ssh/id_rsa.pub.
The key fingerprint is:
32:9f:ae:b6:7f:7e:56:e4:b5:7a:21:f0:95:42:5c:d1 anyuser@mymachine

3. Import the public key to the switch by logging in to the switch as any user with the admin role
and entering the sshUtil importpubkey command to import the key.
Example of adding the public key to the switch
switch:anyuser> sshutil importpubkey
Enter user name for whom key is imported: aswitchuser
Enter IP address:192.168.38.244
Enter remote directory:~auser/.ssh
Enter public key name(must have .pub suffix):id_rsa.pub

198

Fabric OS Administrator’s Guide
53-1002920-02

Secure Shell protocol

7

Enter login name:auser
Password:
Public key is imported successfully.

4. Test the setup by logging in to the switch from a remote device, or by running a command
remotely using SSH.

Configuring outgoing SSH authentication
After the allowed-user is configured, the remaining setup steps must be completed by the
allowed-user.
Use the following procedure to configure outgoing SSH authentication:
1. Log in to the switch as the default admin.
2. Change the allowed-user’s permissions to admin, if applicable.
switch:admin> userconfig --change username -r admin

where the username variable is the name of the user who can perform SSH public key
authentication, and who can import, export, and delete keys.
3. Set up the allowed-user by typing the following command:
switch:admin> sshutil allowuser username

where the username variable is the name of the user who can perform SSH public key
authentication, and who can import, export, and delete keys.
4. Generate a key pair for switch-to-host (outgoing) authentication by logging in to the switch as
the allowed user and entering the sshUtil genkey command.
You may enter a passphrase for additional security.
Example of generating a key pair on the switch
switch:alloweduser> sshutil genkey
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Key pair generated successfully.

5. Export the public key to the host by logging in to the switch as the allowed-user and entering
the sshUtil exportpubkey command to export the key.
Example of exporting a public key from the switch
switch:alloweduser> sshutil exportpubkey
Enter IP address:192.168.38.244
Enter remote directory:~auser/.ssh
Enter login name:auser
Password:
public key out_going.pub is exported successfully.

6. Append the public key to a remote host by logging in to the remote host, locating the directory
where authorized keys are stored, and appending the public key to the file.
You may need to refer to the host’s documentation to locate where the authorized keys are
stored.
7.

Test the setup by using a command that uses SCP and authentication, such as
firmwareDownload or configUpload.

Fabric OS Administrator’s Guide
53-1002920-02

199

7

Secure Sockets Layer protocol

Deleting public keys on the switch
Use the following procedure to delete public keys from the switch.
1. Connect to the switch and log in using an account with admin permissions.
2. Use the sshUtil delpubkeys command to delete public keys.
You will be prompted to enter the name of the user whose the public keys you want to delete.
Enter all to delete public keys for all users.
For more information on IP filter policies, refer to Chapter 8, “Configuring Security Policies”.

Deleting private keys on the switch
Use the following procedure to delete private keys from the switch.
1. Log in to the switch as the allowed-user.
2. Use the sshUtil delprivkey command to delete the private key.
For more information on IP filter policies, refer to Chapter 8, “Configuring Security Policies”.

Secure Sockets Layer protocol
Secure Sockets Layer (SSL) protocol provides secure access to a fabric through web-based
management tools such as Web Tools. SSL support is a standard Fabric OS feature.
Switches configured for SSL grant access to management tools through Hypertext Transfer Protocol
over SSL links (which begin with https://) instead of standard links (which begin with http://).
SSL uses public key infrastructure (PKI) encryption to protect data transferred over SSL
connections. PKI is based on digital certificates obtained from an Internet Certificate Authority (CA)
that acts as the trusted key agent.
Certificates are based on the switch IP address or fully qualified domain name (FQDN), depending
on the issuing CA. If you change a switch IP address or FQDN after activating an associated
certificate, you may have to obtain and install a new certificate. Check with the CA to verify this
possibility, and plan these types of changes accordingly.

Browser and Java support
Fabric OS supports the following web browsers for SSL connections:

• Internet Explorer v7.0 or later (Microsoft Windows)
• Mozilla Firefox v2.0 or later (Solaris and Red Hat Linux)
NOTE
Review the release notes for the latest information and to verify if your platform and browser are
supported.
In countries that allow the use of 128-bit encryption, you should use the latest version of your
browser. For example, Internet Explorer 7.0 and later supports 128-bit encryption by default. You
can display the encryption support (called “cipher strength”) using the Internet Explorer Help:About
menu option. If you are running an earlier version of Internet Explorer, you may be able to download
an encryption patch from the Microsoft website at http://www.microsoft.com.

200

Fabric OS Administrator’s Guide
53-1002920-02

Secure Sockets Layer protocol

7

You should upgrade to the Java 1.6.0 plug-in on your management workstation. To find the Java
version that is currently running, open the Java console and look at the first line of the window.
For more details on levels of browser and Java support, refer to the Web Tools Administrator’s
Guide.

SSL configuration overview
You configure SSL access for a switch by obtaining, installing, and activating digital certificates.
Certificates are required on all switches that are to be accessed through SSL.
Also, you must install a certificate in the Java plug-in on the management workstation, and you may
need to add a certificate to your web browser.
Configuring for SSL involves these main steps, which are shown in detail in the next sections.
1. Choose a certificate authority (CA).
2. Generate the following items on each switch:
a.

A public and private key by using the secCertUtil genkey command.

b.

A certificate signing request (CSR) by using the secCertUtil gencsr command.

3. Store the CSR on a file server by using the secCertUtil export command.
4. Obtain the certificates from the CA.
You can request a certificate from a CA through a web browser. After you request a certificate,
the CA either sends certificate files by e-mail (public) or gives access to them on a remote host
(private).
5. On each switch, install the certificate. Once the certificate is loaded on the switch, HTTPS
starts automatically.
6. If necessary, install the root certificate to the browser on the management workstation.
7.

Add the root certificate to the Java plug-in keystore on the management workstation.

Certificate authorities
To ease maintenance and allow secure out-of-band communication between switches, consider
using one certificate authority (CA) to sign all management certificates for a fabric. If you use
different CAs, management services operate correctly, but the Web Tools Fabric Events button is
unable to retrieve events for the entire fabric.
Each CA (for example, Verisign or GeoTrust) has slightly different requirements; for example, some
generate certificates based on IP address, while others require an FQDN, and most require a 1024-bit
public/private key pair while some may accept a 2048-bit key. Consider your fabric configuration,
check CA websites for requirements, and gather all the information that the CA requires.

Generating a public-private key pair
Use the following procedure to generate a public-private key pair.

NOTE
You must perform this procedure on each switch.

Fabric OS Administrator’s Guide
53-1002920-02

201

7

Secure Sockets Layer protocol

1. Connect to the switch and log in using an account with admin permissions.
2. Enter the secCertUtil genkey command to generate a public/private key pair.
The system reports that this process will disable secure protocols, delete any existing CSR, and
delete any existing certificates.
3. Respond to the prompts to continue and select the key size.
Example of generating a key
Continue (yes, y, no, n): [no] y
Select key size [1024 or 2048]: 1024
Generating new rsa public/private key pair
Done.

Generating and storing a Certificate Signing Request
After generating a public/private key pair, you must generate and store a certificate signing request
(CSR).
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the secCertUtil gencsr command.
3. Enter the requested information.
Example of generating a CSR
Country Name (2 letter code, eg, US):US
State or Province Name (full name, eg, California):California
Locality Name (eg, city name):San Jose
Organization Name (eg, company name):Brocade
Organizational Unit Name (eg, department name):Eng
Common Name (Fully qualified Domain Name, or IP address): 192.1.2.3
Generating CSR, file name is: 192.1.2.3.csr
Done.

Your CA may require specific codes for Country, State or Province, Locality, Organization, and
Organizational Unit names. Make sure that your spelling is correct and matches the CA
requirements. If the CA requires that the Common Name be specified as an FQDN, make sure
that the fully qualified domain name is set on the domain name switch/director. The IP
address or FQDN is the switch where the certificate gets installed.
4. Enter the secCertUtil export command to store the CSR.
5. Enter the requested information. You can use either FTP or SCP.
Example of exporting a CSR
Select protocol [ftp or scp]: ftp
Enter IP address: 192.1.2.3
Enter remote directory: path_to_remote_directory
Enter Login Name: your account
Enter Password: your password
Success: exported CSR.

If you are set up for Secure Copy Protocol (SCP), you can select it; otherwise, select FTP. Enter
the IP address of the switch on which you generated the CSR. Enter the remote directory name
of the FTP server to which the CSR is to be sent. Enter your account name and password on the
server.

202

Fabric OS Administrator’s Guide
53-1002920-02

Secure Sockets Layer protocol

7

Obtaining certificates
Once you have generated a CSR, you will need to follow the instructions on the website of the
certificate issuing authority that you want to use; and then obtain the certificate.
Fabric OS and HTTPS support the following types of files from the Certificate Authority(CA):

• .cer (binary)
• .crt (binary)
• .pem (text)
Typically, the CA provides the certificate files listed in Table 29.

TABLE 29

SSL certificate files

Certificate file

Description

name.pem

The switch certificate.

nameRoot.pem

The root certificate. Typically, this certificate is already installed in the browser, but if not, you
must install it.

nameCA.pem

The CA certificate. It must be installed in the browser to verify the validity of the server certificate
or server validation fails.

NOTE

You must perform this procedure for each switch.
Use the following procedure to obtain a security certificate.
1. Generate and store the CSR as described in “Generating and storing a Certificate Signing
Request” on page 202.
2. Open a web browser window on the management workstation and go to the CA website. Follow
the instructions to request a certificate. Locate the area in the request form into which you are
to paste the CSR.
3. Through a Telnet window, connect to the switch and log in using an account with admin
permissions.
4. Enter the secCertUtil showcsr command. The contents of the CSR are displayed.
5. Locate the section that begins with “BEGIN CERTIFICATE REQUEST” and ends with “END
CERTIFICATE REQUEST”.
6. Copy and paste this section (including the BEGIN and END lines) into the area provided in the
request form; then, follow the instructions to complete and send the request.
It may take several days to receive the certificates. If the certificates arrive by e-mail, save them to
an FTP server. If the CA provides access to the certificates on an FTP server, make note of the path
name and make sure you have a login name and password on the server.

Installing a switch certificate
Use the following procedure to install a security certificate on a switch.

NOTE

You must perform this procedure on each switch.

Fabric OS Administrator’s Guide
53-1002920-02

203

7

Secure Sockets Layer protocol

1. Connect to the switch and log in using an account with admin permissions.
2. Enter the secCertUtil import command.
3. Select a protocol, enter the IP address of the host on which the switch certificate is saved, and
enter your login name and password.
Example of installing a switch certificate in interactive mode
switch:admin> seccertutil import -config swcert -enable https
Select protocol [ftp or scp]: ftp
Enter IP address: 192.10.11.12
Enter remote directory: path_to_remote_directory
Enter certificate name (must have ".crt", ".cer", \
".pem" or “.psk” suffix): 192.1.2.3.crt
Enter Login Name: your_account
Enter Password: *****
Success: imported certificate [192.1.2.3.crt].

Example of installing a switch certificate in noninteractive mode
switch:admin> seccertutil import -config swcert -enable https \
-protocol ftp -ipaddr 192.10.11.12 -remotedir path_to_remote_directory \
-certname 192.1.2.3.crt -login your_account -password passwd
Success: imported certificate [192.1.2.3.crt].
Certificate file in configuration has been updated.
Secure http has been enabled.

Example of installing a common certificate in non-interactive mode
switch:admin> seccertutil import -commonswcert -config swcert -enable https
-protocol scp -ipaddr 192.10.11.12 -remotedir path_to_remote_directory -login
cert -certname 192.1.2.3.pem

Important Notes
• Certificate Authorities may provide their certificates in different encodings and different
extensions. Be sure to save the certificate with the applicable file extension before you import
the certificate to the switch:
For example, certificates that contain lines similar to the following are usually .pem encoded:
“----BEGIN REQUEST----” and “----END REQUEST---- (and may include the
strings "x509" or "certificate")

• For Certificate Authorities that request information regarding the type of web server, Fabric OS
uses the Apache web server running on Linux.

The browser
The root certificate may already be installed on your browser, if not, you must install it. To see
whether it is already installed, check the certificate store on your browser.
The next procedures are guides for installing root certificates to Internet Explorer and Mozilla
Firefox browsers. For more detailed instructions, refer to the documentation that came with the
certificate.

204

Fabric OS Administrator’s Guide
53-1002920-02

Secure Sockets Layer protocol

7

Checking and installing root certificates on Internet Explorer
Use the following procedure to check and install a root security certificate on a switch using IE:
1. Select Tools > Internet Options.
2. Click the Content tab.
3. Click Certificates.
4. Click the Intermediate or Trusted Root tab and scroll the list to see if the root certificate is
listed. Take the appropriate following action based on whether you find the certificate:

• If the certificate is listed, you do not need to install it. You can skip the rest of this
procedure.

• If the certificate is not listed, click Import.
5. Follow the instructions in the Certificate Import wizard to import the certificate.

Checking and installing root certificates on Mozilla Firefox
Use the following procedure to check and install a root security certificate on a switch using Firefox:
1. Select Tools > Options.
2. Click Advanced.
3. Click the Encryption tab.
4. Click View Certificates > Authorities tab and scroll the list to see if the root certificate is listed.
For example, its name may have the form nameRoot.crt. Take the appropriate following action
based on whether you find the certificate:

• If the certificate is listed, you do not need to install it. You can skip the rest of this
procedure.

• If the certificate is not listed, click Import.
5. Browse to the certificate location and select the certificate. For example, select nameRoot.crt.
6. Click Open and follow the instructions to import the certificate.

Root certificates for the Java plugin
For information on Java requirements, refer to “Browser and Java support” on page 200.
This procedure is a guide for installing a root certificate to the Java plugin on the management
workstation. If the root certificate is not already installed to the plugin, you should install it.
For more detailed instructions, refer to the documentation that came with the certificate and to the
Sun Microsystems website (www.sun.com).

Installing a root certificate to the Java plugin
Use the following procedure to install a root certificate to the Java plugin.
1. Copy the root certificate file from its location on the FTP server to the Java plugin bin directory.
For example, the bin location may be:
C: \program files\java\j2re1.6.0\bin

Fabric OS Administrator’s Guide
53-1002920-02

205

7

Simple Network Management Protocol

2. Open a Command Prompt window and change the directory to the Java plugin bin directory.
3. Enter the keyTool command and respond to the prompts.
Example of installing a root certificate
C:\Program Files\Java\j2re1.6.0\bin> keytool -import -alias RootCert -file
RootCert.crt -keystore ..\lib\security\RootCerts
Enter keystore password: changeit
Owner: CN=Brocade, OU=Software, O=Brocade Communications, L=San Jose,
ST=California, C=US
Issuer: CN=Brocade, OU=Software, O=Brocade Communications, L=San Jose,
ST=California, C=US
Serial number: 0
Valid from: Thu Jan 15 16:27:03 PST 2007 until: Sat Feb 14 16:27:03 PST 2007
Certificate fingerprints:
MD5: 71:E9:27:44:01:30:48:CC:09:4D:11:80:9D:DE:A5:E3
SHA1: 06:46:C5:A5:C8:6C:93:9C:FE:6A:C0:EC:66:E9:51:C2:DB:E6:4F:A1
Trust this certificate? [no]: yes
Certificate was added to keystore

In the example, changeit is the default password and RootCert is an example root certificate name.

Simple Network Management Protocol
Simple Network Management Protocol (SNMP) is a set of protocols for managing complex
networks. SNMP protocols are application layer protocols. Using SNMP, devices within a network
send messages, called protocol data units (PDUs), to different parts of a network. Network
management using SNMP requires three components:

• SNMP Manager
• SNMP Agent
• Management Information Base (MIB)

SNMP Manager
The SNMP Manager can communicate to the devices within a network using the SNMP protocol.
Typically, SNMP Managers are network management systems (NMS) that manage networks by
monitoring the network parameters, and optionally, setting parameters in managed devices.
Normally, the SNMP Manager sends read requests to the devices that host the SNMP Agent, to
which the SNMP Agent responds with the requested data. In some cases, the managed devices
can initiate the communication, and send data to the SNMP Manager using asynchronous events
called traps.

SNMP Agent
The SNMP agent is a software that resides in the managed devices in the network, and collects
data from these devices. Each device hosts an SNMP Agent. The SNMP Agent stores the data, and
sends these when requested by an SNMP Manager. In addition, the Agent can asynchronously alert
the SNMP Manager about events, by using special PDUs called traps.

206

Fabric OS Administrator’s Guide
53-1002920-02

7

Simple Network Management Protocol

Management Information Base (MIB)
SNMP Agents in the managed devices store the data about these devices in a database called
Management Information Base (MIB). The MIB is a hierarchical database, which is structured on
the standard specified in the RFC 2578 (Structure of Management Information Version 2 (SMIv2)).
The MIB is a database of objects that can be used by a network management system to manage
and monitor devices on the network. The MIB can be retrieved by a network management system
that uses SNMP. The MIB structure determines the scope of management access allowed by a
device. By using SNMP, a manager application can issue read or write operations within the scope
of the MIB.

Basic SNMP operation
Every Brocade device carries an agent and management information base (MIB), as shown in
Figure 13. The agent accesses information about a device and makes it available to an SNMP
network management station.
SNMP

FIGURE 13

MIB

Agent

Management Station

SNMP structure

When active, the management station can get information or set information when it queries an
agent. SNMP commands, such as get, set, getnext, and getresponse, are sent from the
management station, and the agent replies once the value is obtained or modified (Figure 14).
Agents use variables to report such data as the number of bytes and packets in and out of the
device, or the number of broadcast messages sent and received. These variables are also known
as managed objects. All managed objects are contained in the MIB.
get, getnext, set
Management Station

FIGURE 14

reply

Agent

SNMP query

The management station can also receive traps, unsolicited messages from the switch agent if an
unusual event occurs (Figure 15). For more information, refer to “Traps” on page 209.
Management Station

FIGURE 15

TRAP

Agent

SNMP trap

The agent can receive queries from one or more management stations and can send traps to up to
six management stations.

Fabric OS Administrator’s Guide
53-1002920-02

207

7

Simple Network Management Protocol

Understanding MIBs
The management information base (MIB) is a database of monitored and managed information on
a device, in this case a Brocade switch. The MIB structure can be represented by a tree hierarchy.
The root splits into three main branches: International Organization for Standardization (ISO),
Consultative Committee for International Telegraph and Telephone (CCITT), and joint ISO/CCITT.
These branches have short text strings and integers (OIDs) to identify them. Text strings describe
object names, while integers allow software to create compact, encoded representations of the
names.
Each MIB variable is assigned an object identifier (OID). The OID is the sequence of numeric labels
on the nodes along a path from the root to the object. For example, as shown in Figure 16, the
Brocade SW.MIB OID is:
1.3.6.1.4.1.1588.2.1.1.1

The corresponding name is:
iso.org.dod.internet.private.enterprise.bcsi.commDev.fibreChannel.fcSwitch.sw

The other branches are part of the standard MIBs, and the portions relevant to configuring SNMP
on a Brocade switch are referenced in the remainder of this reference.
iso (1)

org (3)

dod (6)

internet (1)

directory (1)

mgmt (2)

experimental (3)

private (4)

mib-2 (1)

fcmgmt (94)

enterprise (1)

connSet (1)

bcsi (1588)

system (1)

sysDescr (1)

interface (2)

sysObjectID (2)
Brocade SW MIB
1.3.6.1.4.1.1588.2.1.1.1

FIGURE 16

Brocade MIB tree location

Access to MIB variables
Use a MIB browser to access the MIB variables: all MIB browsers perform queries and load MIBs.

208

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

7

Once loaded, the MAX-ACCESS provides access levels between the agent and management station.
The access levels are as follows:

• not accessible
You cannot read or write to this variable.

• read create
Specifies a tabular object that can be read, modified, or created as a new row in a table.

• read only - Public
You can only monitor information.

• read-write - Private
You can read or modify this variable.

• accessible-to-notify
You can read this information only through traps.

SNMP support
In addition to the standard MIBs that Brocade devices support, these devices also support
Brocade-specific MIBs. Since different vendors vary the information in their private enterprise
MIBs, it is necessary to verify their information. The Fibre Channel MIB standards dictate that
certain information be included in all MIBs: it is the vendors' responsibility to follow the standards.
The standards are as follows:

• FibreAlliance (FA) MIB: Brocade supports v4.4.0 and later releases.
• Fabric Element (FE) MIB: accepted by the Internet Engineering Task Force (IETF).
Brocade supports FE_RFC2837.mib under the MIB-II branch in Fabric OS v7.1.0, v7.0.0,
v6.4.1_fcoe, v6.4.0, v6.3.0, v6.2.0, v6.1.2_CEE, v6.1.0, and v6.0.0. The latest version of the FE
MIB references the FRAMEWORK.MIB and, based on the MIB browser, it is necessary to load this
MIB before the FE.MIB. For more information, refer to “Loading Brocade MIBs” on page 212.

Traps
An unsolicited message that comes to the management station from the SNMP agent on the
device is called a trap. Brocade switches send traps out on UDP port 162 and to any configured
port. In order to receive traps, the management station IP address and severity level must be
configured on the switch. You can define a different message severity level for each recipient so
that some recipients receive all trap messages and others receive only the most critical.

NOTE

Due to design limitation, IP address validation cannot be done for trap recipients.
There are two main MIB trap choices:

• FibreAlliance MIB trap - Associated with the Fibre Alliance MIB (FA-MIB), this MIB manages SAN
switches and devices from any company that complies with Fibre Alliance specifications.

• Brocade-specific MIB trap - Associated with the Brocade-specific Brocade MIB (SW-MIB),
manages Brocade switches only.

Fabric OS Administrator’s Guide
53-1002920-02

209

7

Simple Network Management Protocol

There is some overlap in the functionality of these MIBs. If you enable both SW-MIB and FA-MIB
traps, you could receive duplicate messages for the switch events that trigger the trap.
You can also use these additional MIBs and their associated traps: HA-MIB; FICON-MIB; and
SWEXTTRAP. In Fabric OS v6.4.0, you can use the snmpConfig--set mibCapability command to
enable or disable all the MIBs.
An event trap (swEventTrap, connUnitEventTrap, or swFabricWatchTrap) is basically an error
message (errShow output) that is SNMP-formatted and delivered.

FA traps
Consider enabling the FA traps if you want to use SNMP to monitor multiple connectivity units,
including Brocade switches.
The switchStatusPolicySet command determines the FA-TRAP switch status-related outputs:

• connUnitStatusChange
This trap is generated by Fabric watch such that only the swUnitsStatusChange is controlled by
switchStatusPolicySet command.

• connUnitSensorStatusChange
This trap is generated by any sensor event.

• connUnitPortStatusChange
This trap sends the instance of connUnitPortName as part of the trap; the instance string is
NULL, if the port name is not defined for the specified port.

• connUnitEventTrap
All the external traps gets converted into swEventTrap except for AN-1006, AUTH-3001 to
AUTH-3008, FW-3001, SEC-3001 to SEC-3034, and SEC-3044 to SEC-3048 RASlog messages.
Events in the Error Log of a severity at or above the configured threshold will generate SNMP traps.
The Fibre Alliance Trap (FA-TRAP) can be configured to send traps using the snmpConfig command.
For more information on this command, refer to the Fabric OS Command Reference.

HA traps
Consider enabling these traps to monitor field-replaceable unit (FRU) status and control processor
(CP) status when you have a Brocade director in your environment:

• fruStatusChanged
This trap is generated by a FRU status change, such as a switch reboot or disabling or enabling
a FRU component such as fandisable, fanenable and so on.

• cpStatusChanged
This trap is generated by a change in the status of a CP, including a reboot or firmware
download.

• fruHistoryTrap
This trap is generated when a FRU is added or removed. It is not generated when standby CP is
removed.

210

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

7

The high availability trap (HA-TRAP) can be configured to send traps using the snmpConfig
command. For more information on this command, refer to the Fabric OS Command Reference.

SW traps
There are fourteen specific traps defined in Brocade SW-TRAP.

• swfault (no longer supported)
• swSensorScn (no longer supported)
• swFCPortScn
This trap is generated by a port state change.

• swEventTrap
This trap is generated by any switch event reported to the system error log.
The desired severity level is introduced to filter a swEvent trap based on the severity level.

• swFabricWatchTrap
This trap is generated when any Fabric Watch threshold is reached.
The desired severity level is introduced to filter a swFabricWatchTrap based on the severity
level.

• swTrackChangesTrap
This trap is generated by a login or a logout.

• swIPv6ChangeTrap
This trap is generated when an IPv6 address status change event occurs. It is generated only
when IPv6 stateless state changes to the deprecation state and not for address change
notification.

• swPmgrEventTrap
This trap is generated when any partition manager change happens.

• swFabricReconfigTrap
The trap to be sent for tracking fabric reconfiguration.

• swFabricSegmentTrap
The trap to be sent for tracking segmentation.

• swExtTrap
The trap adds the SSN binding to the traps if it is enabled.

• swStateChangeTrap
This trap is sent when the switch changes its state to online or offline.

• swPortMoveTrap
This trap is sent when the virtual ports are moved from one switch to another.

• swBrcdGenericTrap
This trap is sent for one of the events, such as fabric change, device change, FAPWWN change,
and FDMI events. This trap is for Brocade use.

Fabric OS Administrator’s Guide
53-1002920-02

211

7

Simple Network Management Protocol

• swDeviceStatusTrap
This trap is sent whenever a device logs in or logs out.
The Brocade trap (SW-TRAP) can be configured to send traps using the snmpConfig command.

FICON traps
• linkRNIDDeviceRegistration
A device registered with the switch.

• linkRNIDDeviceDeRegistration
A device de-registered with the switch.

• linkLIRRListenerAdded
A listener for link failure incident is added.

• linkLIRRListenerRemoved
A listener for link failure incident is removed.

• linkRLIRFailureIncident
A link failure incident has occurred.

IF traps
• linkDown
A linkDown trap signifies that the SNMPv2 entity, acting in an agent role has detected that the
ifOperStatus object for one of its communication links is about to transition into the down
state.

• linkUp
A linkUp trap signifies that the SNMPv2 entity, acting in an agent role has detected that the
ifOperStatus object for one of its communication links has transitioned out of the down state."

BD traps
• bdTrap
Traps to be sent for bottleneck detection.

• bdClearTrap
Traps to be sent for bottleneck clearance.

Loading Brocade MIBs
The Brocade MIB is a set of variables that are private extensions to the Internet standard MIB-II.
The Brocade agents support many other Internet-standard MIBs. These standard MIBs are defined
in RFC publications. To find specific MIB information, examine the Brocade proprietary MIB
structure and the standard RFC MIBs supported by Brocade.

212

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

7

Brocade MIB files
The Brocade MIB files are as follows:

•
•
•
•
•
•
•
•
•
•
•
•
•
•

bd.mib
bcCustomOperation.mib
BRCD_REG.mib
BRCD_TC.mib
brcdfcip.mib
CPQ_HOST.mib
CPQ_RACK.mib
FA.mib
faext.mib
FICON.mib
fod.mib
HA.mib
IbmBladeCenter.mib
SW.mib

Standard MIBs
Distribution of standard MIBs has been stopped from Fabric OS v6.4.0. Download the following
MIBs from the http://www.oidview.com/ website:

•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•

BRIDGE-MIB
ENTITY-MIB
FC-MGMT-MIB
FCIP-MGMT-MIB
FIBRE-CHANNEL-FE-MIB
IANAifType-MIB
IEEE 802.1x PAE MIB
IEEE 802.3 LAG MIB
IF-MIB
INET-ADDRESS-MIB
IP MIB
LLDP MIB
LLDP-EXT-DOT1-MIB
LLDP-EXT-DOT3-MIB
P-BRIDGE MIB
Q-BRIDGE MIB
RFC1155-SMI
RFC1158-MIB
RFC-1212

Fabric OS Administrator’s Guide
53-1002920-02

213

7

Simple Network Management Protocol

•
•
•
•
•
•
•
•
•
•
•
•
•
•

RFC1213-MIB
RFC-1215
RMON-MIB
RSTP-MIB
SNMP-COMMUNITY-MIB
SNMP-FRAMEWORK-MIB
SNMPv2-CONF
SNMPv2-MIB
SNMPv2-PARTY-MIB
SNMPv2-SMI-MIB
SNMPv2-TC
SNMP-VIEW-BASED-ACM-MIB
SNMP-USER-BASED-SM-MIB
SNMP-TARGET-MIB

MIB loading order
Many MIBs use definitions that are defined in other MIBs. These definitions are listed in the
IMPORTS section near the top of the MIB. When loading the Brocade MIBs, refer to Table 30 to
ensure any MIB dependencies are loading in the correct order.

NOTE

Before loading Brocade MIB files, ensure that you have the correct version of SNMP for your
Fabric OS version. All versions of Fabric OS support SNMPv1 and SNMPv3. SNMPv2 is not
supported on Fabric OS v6.1.2_CEE and later versions.

TABLE 30

214

Brocade SNMP MIB dependencies

MIB Name

Dependencies

BRCD_REG.mib

RFC1155-SMI

BRCD_TC.mib

Brocade-REG-MIB
SNMPv2-TC
SNMPv2-SMI

FC-MGMT-MIB

SNMPv2-SMI
SNMPv2-CONF
SNMPv2-MIB
IANAifType-MIB
SNMPv2-TC
IF-MIB
SNMP-FRAMEWORK-MIB

FA.mib

RFC1155-SMI
RFC1158-MIB
RFC-1212
RFC1213-MIB
RFC-1215

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

TABLE 30

Brocade SNMP MIB dependencies

MIB Name

Dependencies

FIBRE-CHANNEL-FE-MIB

SNMPv2-SMI
SNMPv2-TC
SNMP-FRAMEWORK-MIB
SNMPv2-CONF

FCIP-MGMT-MIB

SNMPv2-SMI
SNMPv2-TC
INET-ADDRESS-MIB
FC-MGMT-MIB
IF-MIB
SNMPv2-CONF
SNMP-FRAMEWORK-MIB

ENTITY-MIB

SNMPv2-SMI
SNMPv2-TC
SNMPv2-CONF
SNMP-FRAMEWORK-MIB

SW.mib

SNMPv2-TC
SNMPv2-SMI
Brocade-TC
Brocade-REG-MIB
FCMGMT-MIB

bd.mib

SNMPv2-TC
SNMPv2-SMI
Brocade-TC
Brocade-REG-MIB
SW-MIB

brcdfcip.mib

SNMPv2-SMI
Brocade-REG-MIB
SNMPv2-TC
INET-ADDRESS-MIB
IF-MIB
SNMPv2-CONF

faext.mib

SNMPv2-TC
SNMPv2-SMI
SW-MIB
FCMGMT-MIB

FICON.mib

SNMPv2-SMI
SNMPv2-TC
Brocade-REG-MIB

HA.mib

SNMPv2-SMI
Brocade-REG-MIB
SW-MIB
ENTITY-MIB
SNMPv2-TC

Fabric OS Administrator’s Guide
53-1002920-02

7

215

7

Simple Network Management Protocol

Access Gateway and Brocade MIBs
Table 31 shows the MIBs supported by Brocade Access Gateway.

TABLE 31
.

Access Gateway MIB support

MIB name

Description

MIB-2

Supported in v5.2.1 and later releases.

Entity-MIB

Supported.

HA-MIB

Supported.

SW-MIB

Disabled in Access Gateway because the conventions are specific to fabric switches.
In Fabric OS v6.4.0, swConnUnitPortExtensionTable is supported in Access Gateway mode.
In Fabric OS v7.0.0, SNMP allows you to access the following tables to support the Advanced
Performance Monitoring feature on Access Gateway, even if the SW-MIB is disabled:
• “swBlmPerfEEMntTable”
• “swBlmPerfFltMntTable”

FA-MIB

The connUnitSnsTable is not supported because a switch in Access Gateway does support
name server services.

CPQ-Rack MIB

Supported on embedded switches only.

IF-MIB

Supported.

BD-MIB

Supported for F-ports.

FA-Ext

Supported.

SNMPv2 MIB

Supported.

Firmware upgrades and enabled traps
You can turn on and off traps individually within a trap group. By default the individual traps are
turned off even if the corresponding trap group was enabled before upgrading. You must use the
snmpconfig command to turn on the individual traps within each trap group.

Support for Administrative Domains
Administrative Domains are supported in Fabric OS v5.3.0 and later releases. An Administrative
Domain (AD) is a domain within a fabric. Administrative domains can be used to limit administrator
access within a fabric, and to provide service providers with a means to assign portions of a fabric
to individual consumers. An AD may contain switches, devices, and ports. An AD may also limit
access to a configured set of users.
The following example shows how the AD:xxx field is used in the snmpwalk command. This
command is executed on the host and it walks the entire MIB tree specified (.1).
switch# snmpwalk -u admin -v 3 -n AD:4 10.168.176.181.1

Support for Role-Based Access Control
Role-Based Access Control (RBAC) is supported in Fabric OS v5.3.0 and later releases. RBAC
applies a fixed set of roles that address the access control needs of a majority of customers. Each
role is a set of permissions that can be applied to a user that controls the kinds of jobs and tasks
the user can perform on a fabric or fabric element.

216

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

7

Support for IPv6 addressing
IPv6 addressing is supported in Fabric OS v5.3.0 and later releases.

Support for Virtual Fabric
Virtual Fabric is supported in Fabric OS v6.2.0 and later releases.
When an SNMPv3 request arrives with a particular user name, it executes in the home Virtual
Fabric. From the SNMP manager, all SNMPv3 requests must have a home Virtual Fabric that is
specified in the contextName field. When the home Virtual Fabric is specified, it will be converted to
the corresponding switch ID and the home Virtual Fabric will be set. If the user does not have
permission for the specified home Virtual Fabric, this request fails with an error code of noAccess.
For an SNMPv3 user to have a home Virtual Fabric, a list of allowed Virtual Fabrics, an RBAC role,
and the name of the SNMPv3 user should match that of the Fabric OS user in the local switch
database. SNMPv3 users whose names do not match with any of the existing Fabric OS local users
have a default RBAC role of admin with the SNMPv3 user access control of read/write. Their
SNMPv3 user logs in with an access control of read-only. Both user types will have the default
switch as their home Virtual Fabrics.
The contextName field should have the format “VF:xxx”, where xxx is the actual VF_ID, for example
“VF:1”. If the contextName field is empty, then the home Virtual Fabric of the local Fabric OS user
with the same name is used. As Virtual Fabrics and Admin Domains are mutually exclusive, this
field is considered as Virtual Fabrics context when Virtual Fabrics is enabled. You cannot specify
chassis context in the contextName field.
The following example shows how the VF:xxx field is used in the snmpwalk command. This
command is executed on the host and it walks the entire MIB tree specified (.1).
switch# snmpwalk -u admin -v 3 -n VF:4 10.168.176.181.1

Filtering ports
Each port can belong to only one Virtual Fabric at any time. An SNMP request coming to one Virtual
Fabric can only view the port information of the ports belonging to that Virtual Fabric. All port
attributes are filtered to allow SNMP to obtain the port information only from within the current
Virtual Fabrics context.
Switch and chassis context enforcement
All attributes are classified into one of two categories:

• Chassis-level attributes
• Switch-level attributes
Attributes that are specific to each logical switch belong to the switch category. These attributes are
available in the Virtual Fabrics context and not available in the Chassis context.
Attributes that are common across the logical switches belong to the chassis level. These attributes
are accessible to users having the chassis-role permission. When a chassis table is queried, the
context is set to chassis context, if the user has the chassis-role permission. The context is
switched back to the original context after the operation is performed.

Fabric OS Administrator’s Guide
53-1002920-02

217

7

Simple Network Management Protocol

Configuring SNMP using CLI
For information about Fabric OS commands for configuring SNMP, refer to the Fabric OS Command
Reference.

Configuring SNMP security level
The following example sets the SNMP security level to 1 (authentication only). This setting allows all
SNMPv1 users to perform GET and SET operations on MIBs, but creates an exception for SNMPv3
users that do not have authentication and privacy privileges (noAuthnoPriv).
switch:admin> snmpconfig --set seclevel
Select SNMP Security Level
(0 = No security, 1 = Authentication only, 2 = Authentication and Privacy, 3 =
sxNo Access): (0..3) [0]
Select SNMP SET Security Level
(0 = No security, 1 = Authentication only, 2 = Authentication and Privacy, 3 =
No Access): (0..3) [0]

Table 32 shows the security level options.

TABLE 32

Security level options

Security level

Protocol

Query behavior

Traps

No security [0]
(noAuthnoPriv)

SNMPv1
SNMPv3

Allowed.
Allowed.

Sent.
Sent.

Authentication only [1]
(authNoPriv)

SNMPv1
SNMPv3

Allowed.
All SNMPv3 users allowed except
noAuthNoPriv users.

Sent.
Sent for all SNMPv3 users
except noAuthNoPriv users.

Authentication and
Privacy [2]
(authPriv)

SNMPv1
SNMPv3

Not allowed.
Only SNMPv3 users with authPriv
privilege are allowed.

Not Sent.
Sent only for authPriv users.

No Access [3]

SNMPv1
SNMPv3

Not allowed.

Not Sent.

Configuring SNMPv3 user/traps
The following examples list how to configure SNMPv3 users/traps.
1. Create a user on the switch in non-VF Context using CLI userconfig, with the required role.
switch:admin> userconfig --add fa_adm -r fabricadmin -h0 -a 0-255
Setting initial password for fa_adm
Enter new password:********
Re-type new password:********
Account fa_adm has been successfully added.
switch:admin>

Create a user on the switch in VF Context using CLI userconfig, with the required role.
switch:admin> userconfig --add sa_user -r switchadmin -l 1-128 -h1 -c admin
Setting initial password for sa_user
Enter new password:********
Re-type new password:********
Account sa_user has been successfully added.
switch:admin>

218

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

7

2. Create the SNMPv3 user.
switch:root> snmpconfig --set snmpv3
SNMP Informs Enabled (true, t, false, f): [false] t
SNMPv3 user configuration(snmp user not configured in FOS user database will
have physical AD and admin role as the default):
User (rw): [snmpadmin1]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
Engine ID: [0:0:0:0:0:0:0:0:0] 80:00:05:23:01:0A:23:34:21
User (rw): [snmpadmin2]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3] 1
New Auth Passwd:
Verify Auth Passwd:
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
New Priv Passwd:
Verify Priv Passwd:
Engine ID: [0:0:0:0:0:0:0:0:0] 80:00:05:23:01:0A:23:34:1B
User (rw): [snmpadmin3]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
Engine ID: [0:0:0:0:0:0:0:0:0]
User (ro): [snmpuser1]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
Engine ID: [0:0:0:0:0:0:0:0:0]
User (ro): [snmpuser2]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
Engine ID: [0:0:0:0:0:0:0:0:0]
User (ro): [snmpuser3]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
Engine ID: [0:0:0:0:0:0:0:0:0]
SNMPv3 trap recipient configuration:
Trap Recipient's IP address : [0.0.0.0] 10.35.52.33
UserIndex: (1..6) [1]
Trap recipient Severity level : (0..5) [0] 4
Trap recipient Port : (0..65535) [162]
Trap Recipient's IP address : [0.0.0.0] 10.35.52.27
UserIndex: (1..6) [2]
Trap recipient Severity level : (0..5) [0] 4
Trap recipient Port : (0..65535) [162]
Trap Recipient's IP address : [0.0.0.0]
Trap Recipient's IP address : [0.0.0.0]
Trap Recipient's IP address : [0.0.0.0]
Trap Recipient's IP address : [0.0.0.0]
Committing configuration.....done.
switch:root> snmpconfig --show snmpv3
SNMP Informs = 1 (ON)
SNMPv3 USM configuration:
User 1 (rw): snmpadmin1
Auth Protocol: noAuth

Fabric OS Administrator’s Guide
53-1002920-02

219

7

Simple Network Management Protocol

User 2

User 3

User 4

User 5

User 6

Priv Protocol: noPriv
Engine ID: 80:00:05:23:01:0a:23:34:21
(rw): snmpadmin2
Auth Protocol: MD5
Priv Protocol: DES
Engine ID: 80:00:05:23:01:0a:23:34:1b
(rw): snmpadmin3
Auth Protocol: noAuth
Priv Protocol: noPriv
Engine ID: 00:00:00:00:00:00:00:00:00
(ro): snmpuser1
Auth Protocol: noAuth
Priv Protocol: noPriv
Engine ID: 00:00:00:00:00:00:00:00:00
(ro): snmpuser2
Auth Protocol: noAuth
Priv Protocol: noPriv
Engine ID: 00:00:00:00:00:00:00:00:00
(ro): snmpuser3
Auth Protocol: noAuth
Priv Protocol: noPriv
Engine ID: 00:00:00:00:00:00:00:00:00

SNMPv3 Trap configuration:
Trap Entry 1:
10.35.52.33
Trap Port: 162
Trap User: snmpadmin1
Trap recipient Severity level: 4
Trap Entry 2:
10.35.52.27
Trap Port: 162
Trap User: snmpadmin2
Trap recipient Severity level: 4
Trap Entry 3:
No trap recipient configured
Trap Entry 4:
No trap recipient configured
Trap Entry 5:
No trap recipient configured
Trap Entry 6:
No trap recipient configured

yet
yet
yet
yet

An example of the SNMPv3 user trap recipients configured with DNS names and IPv6
addresses:
switch:admin> snmpconfig --set snmpv3
SNMP Informs Enabled (true, t, false, f): [false]
SNMPv3 user configuration(snmp user not configured in FOS user database will
have physical AD and admin role as the default):
User (rw): [snmpadmin1]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
User (rw): [snmpadmin2]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3] 1
New Auth Passwd:
Verify Auth Passwd:
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
User (rw): [snmpadmin3]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3] 1
New Auth Passwd:
Verify Auth Passwd:
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
New Priv Passwd:

220

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

7

Verify Priv Passwd:
User (ro): [snmpuser1]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
User (ro): [snmpuser2]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
User (ro): [snmpuser3]
Auth Protocol [MD5(1)/SHA(2)/noAuth(3)]: (1..3) [3]
Priv Protocol [DES(1)/noPriv(2)/AES128(3)]): (1..3) [2]
SNMPv3 trap recipient configuration:
Trap Recipient's IP address : [0.0.0.0] 172.26.4.102
UserIndex: (1..6) [1]
Trap recipient Severity level : (0..5) [0] 1
Trap recipient Port : (0..65535) [162]
Trap Recipient's IP address : [0.0.0.0] fe80::224:1dff:fef6:3f98
UserIndex: (1..6) [2]
Trap recipient Severity level : (0..5) [0] 2
Trap recipient Port : (0..65535) [162]
Trap Recipient's IP address : [0.0.0.0]
UserIndex: (1..6) [3]
Trap recipient Severity level : (0..5) [0] 5
Trap recipient Port : (0..65535) [162]
Trap Recipient's IP address : [0.0.0.0]
Trap Recipient's IP address : [0.0.0.0]
Trap Recipient's IP address : [0.0.0.0]
Committing configuration.....done.
DCX_128:FID128:admin>
switch:admin> snmpconfig --show snmpv3
SNMP Informs = 0 (OFF)
SNMPv3 USM configuration:
User 1 (rw): snmpadmin1
Auth Protocol:
Priv Protocol:
User 2 (rw): snmpadmin2
Auth Protocol:
Priv Protocol:
User 3 (rw): snmpadmin3
Auth Protocol:
Priv Protocol:
User 4 (ro): snmpuser1
Auth Protocol:
Priv Protocol:
User 5 (ro): snmpuser2
Auth Protocol:
Priv Protocol:
User 6 (ro): snmpuser3
Auth Protocol:
Priv Protocol:

noAuth
noPriv
MD5
noPriv
MD5
DES
noAuth
noPriv
noAuth
noPriv
noAuth
noPriv

SNMPv3 Trap configuration:
Trap Entry 1:
172.26.4.102
Trap Port: 162
Trap User: snmpadmin1
Trap recipient Severity level: 1
Trap Entry 2:
fe80::224:1dff:fef6:3f98

Fabric OS Administrator’s Guide
53-1002920-02

221

7

Simple Network Management Protocol

Trap Port: 162
Trap User: snmpadmin2
Trap recipient Severity level: 2
Trap Entry 3:
HCL0389U.corp.brocade.com
Trap Port: 162
Trap User: snmpadmin3
Trap recipient Severity level: 5
Trap Entry 4:
No trap recipient configured yet
Trap Entry 5:
No trap recipient configured yet
Trap Entry 6:
No trap recipient configured yet

To display the traps and MIBs supported in Fabric OS:
switch:root> snmpTraps --show
# |Mib Name
|Supported Traps
---|----------------|-------------------------------001|SW-MIB
|sw-track-changes-trap
|
|sw-fabric-watch-trap
|
|sw-fc-port-scn
|
|ip-v6-change-trap
|
|sw-pmgr-event-trap
|
|sw-event-trap
|
|sw-fabric-reconfig-trap
|
|sw-fabric-segment-trap
|
|sw-state-trap
|
|sw-port-move-trap
|
|sw-brcd-genric-trap
|
|sw-device-status-trap
002|FICON-MIB
|link-rnid-device-registration
|
|link-rnid-device-deregistration
|
|link-lirr-listerner-added
|
|link-lirr-listerner-removed
|
|link-rlir-failure-incident
003|FA-MIB
|conn-unit-status-change
|
|conn-unit-sensor-status-change
|
|conn-unit-port-status-change
|
|conn-unit-event-trap
004|RFC1157
|cold-restart-trap
|
|warm-restart-trap
|
|if-link-up-trap
|
|if-link-down-trap
|
|snmp-authetication-trap
005|HA-MIB
|fru-status-change-trap
|
|fru-history-trap
|
|cp-status-change-trap
006|BD-MIB
|bd-trap
|
|bd-clear-trap

To send all traps to the configured recipients:
switch:root> snmpTraps --send
Number of traps sent : 30

To send all traps to the recipient 10.35.52.33:
switch:root> snmpTraps --send -ip_address 10.35.52.33
Number of traps sent : 30

222

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

7

To send the sw-fc-port-scn trap to the configured recipients:
switch:root> snmpTraps --send -trap_name sw-fc-port-scn
Number of traps sent : 1

To send the sw-fc-port-scn trap to the recipient 10.35.52.33:
switch:root> snmpTraps --send -trap_name sw-fc-port-scn -ip_address
10.35.52.33
Number of traps sent : 1

To unblock port traps on all the ports or on a specific port:
switch:admin> snmptraps --unblock -ports ALL
switch:admin> snmptraps -unblock -port 1/10

To block port traps on slot 1 and port 10:
Switch:admin> snmptraps -block -port

1/10

Example of accessControl configuration:
switch:admin> snmpconfig --set accessControl
SNMP access list configuration:
Access host subnet area in dot notation: [0.0.0.0] 192.168.0.0
Read/Write? (true, t, false, f): [true]
Access host subnet area in dot notation: [0.0.0.0] 10.32.148.0
Read/Write? (true, t, false, f): [true] f
Access host subnet area in dot notation: [0.0.0.0]
Read/Write? (true, t, false, f): [true]
Access host subnet area in dot notation: [0.0.0.0] 10.33.0.0
Read/Write? (true, t, false, f): [true] f
Access host subnet area in dot notation: [0.0.0.0]
Read/Write? (true, t, false, f): [true]
Access host subnet area in dot notation: [0.0.0.0]
Read/Write? (true, t, false, f): [true]
Committing configuration...done.

Example of mibCapability configuration:
To enable the swFabricWatchTrap non-interactively:
switch:admin> snmpconfig --enable mibCapability -mib_name SW-MIB -trap_name
swFabricWatchTrap
Operation succeeded

To enable the swEventTrap of the SW-MIB category only (this operation disables all other SNMP
traps in this MIB category):
switch:admin> snmpconfig --set mibCapability -mib_name SW-MIB -bitmask 0x10
Operation succeeded
switch:admin> snmpconfig --show mibCapability
[...]
SW-MIB: NO
swFault: NO
swSensorScn: NO
swFCPortScn: NO
swEventTrap: YES

Fabric OS Administrator’s Guide
53-1002920-02

223

7

Simple Network Management Protocol

DesiredSeverity:None
swFabricWatchTrap: NO
DesiredSeverity:None
swTrackChangesTrap: NO
swIPv6ChangeTrap: NO
swPmgrEventTrap: NO
swFabricReconfigTrap: NO
swFabricSegmentTrap: NO
swExtTrap: NO
[...]

To enable the SW-MIB MIB only without changing the current trap configuration:
switch:admin> snmpconfig --enable mibCapability -mib_name SW-MIB
Operation succeeded
switch:admin> snmpconfig --show mibCapability
[...]
SW-MIB: YES
swFault: NO
swSensorScn: NO
swFCPortScn: NO
swEventTrap: YES
DesiredSeverity:None
swFabricWatchTrap: YES
DesiredSeverity:None
swTrackChangesTrap: NO
swIPv6ChangeTrap: NO
swPmgrEventTrap: NO
swFabricReconfigTrap: NO
swFabricSegmentTrap: NO
swExtTrap: NO
[...]

To re-enable all traps under the SW-MIB category after they were disabled:
switch:admin> snmpconfig --set mibCapability -mib_name SW-MIB -bitmask 0xFFF
Operation succeeded
switch:admin> snmpconfig --show mibCapability
[...]
SW-MIB: YES
swFault: YES
swSensorScn: YES
swFCPortScn: YES
swEventTrap: YES
DesiredSeverity:None
swFabricWatchTrap: YES
DesiredSeverity:None
swTrackChangesTrap: YES
swIPv6ChangeTrap: YES
swPmgrEventTrap: YES
swFabricReconfigTrap: Yes
swFabricSegmentTrap: Yes
swExtTrap: Yes
[...]

To display the configuration for all MIBs and associated traps:
switch:admin> snmpconfig --show mibcapability
FE-MIB: YES

224

Fabric OS Administrator’s Guide
53-1002920-02

Simple Network Management Protocol

7

SW-MIB: YES
FA-MIB: YES
FICON-MIB: YES
HA-MIB: YES
FCIP-MIB: YES
ISCSI-MIB: YES
IF-MIB: YES
BD-MIB: YES
SW-TRAP: YES
swFault: YES
swSensorScn: YES
swFCPortScn: YES
swEventTrap: YES
DesiredSeverity:None
swFabricWatchTrap: YES
DesiredSeverity:None
swTrackChangesTrap: YES
swIPv6ChangeTrap: YES
swPmgrEventTrap: YES
swFabricReconfigTrap: YES
swFabricSegmentTrap: YES
swExtTrap: YES
FA-TRAP: YES
connUnitStatusChange: YES
connUnitDeletedTrap: YES
connUnitEventTrap: YES
connUnitSensorStatusChange: YES
connUnitPortStatusChange: YES
FICON-TRAP: YES
linkRNIDDeviceRegistration: YES
linkRNIDDeviceDeRegistration: YES
linkLIRRListenerAdded: YES
linkLIRRListenerRemoved: YES
linkRLIRFailureIncident: YES
HA-TRAP: YES
fruStatusChanged: YES
cpStatusChanged: YES
fruHistoryTrap: YES
ISCSI-TRAP: YES
iscsiTgtLoginFailure: YES
iscsiIntrLoginFailure: YES
iscsiInstSessionFailure: YES
IF-TRAP: YES
linkDown: YES
linkUp: YES
BD-TRAP: YES
bdTrap: YES
bdClearTrap: YES

To set the system group:
DCX_128:FID128:admin> snmpconfig --set systemgroup
Example of systemGroup configuration (default)
switch:admin> snmpconfig --default systemGroup
*****
This command will reset the agent's system group configuration back to
factory default
*****

Fabric OS Administrator’s Guide
53-1002920-02

225

7

Telnet protocol

sysDescr = Fibre Channel Switch
sysLocation = End User Premise
sysContact = Field Support
authTraps = 0 (OFF)
*****
Are you sure? (yes, y, no, n): [no] y

3. Set the security level.
switch:admin> snmpconfig --set secLevel
Select SNMP GET Security Level
(0 = No security, 1 = Authentication only, 2
No Access): (0..3) [0] 2
Select SNMP SET Security Level
(0 = No security, 1 = Authentication only, 2
No Access): (2..3) [2] 2
switch:admin> snmpconfig --show secLevel
GET security level = 2, SET level = 2
SNMP GET Security Level: Authentication and
SNMP SET Security Level: Authentication and

= Authentication and Privacy, 3 =

= Authentication and Privacy, 3 =

Privacy
Privacy

To set the security level to default:
DCX_128:FID128:admin> snmpconfig --default seclevel
GET security level = 0, SET level = 0
SNMP GET Security Level: No security
SNMP SET Security Level: No security
SNMP GET Security Level will be set to 'No Security'
SNMP SET Security Level will be set to 'No Security'
Do you want to continue? (yes, y, no, n): [no] y
DCX_128:FID128:admin>
DCX_128:FID128:admin> snmpconfig --show seclevel
GET security level = 0, SET level = 0
SNMP GET Security Level: No security
SNMP SET Security Level: No security
DCX_128:FID128:admin

4. In the Manager (SNMP Browser), create a user snmpadmin1 with Authentication protocol as
noAuth, Privacy protocol as noPriv, set the password and set the trap port as 162. (Same
values are set as in the switch SNMPv3 configuration.)

NOTE

SNMPv3 supports AES-128 and DES protocols.

Telnet protocol
Telnet is enabled by default. To prevent passing clear text passwords over the network when
connecting to the switch, you can block the Telnet protocol using an IP filter policy. For more
information on IP filter policies, refer to “IP Filter policy” on page 253.

ATTENTION
Before blocking Telnet, make sure you have an alternate method of establishing a connection with
the switch.

226

Fabric OS Administrator’s Guide
53-1002920-02

Telnet protocol

7

Blocking Telnet
If you create a new policy using commands with just one rule, all the missing rules have an implicit
deny and you lose all IP access to the switch, including Telnet, SSH, and management ports.
Use the following procedure to block Telnet access.
1. Connect to the switch and log in using an account with admin permissions.
2. Clone the default policy by typing the ipFilter --clone command.
switch:admin> ipfilter --clone BlockTelnet -from default_ipv4

3. Save the new policy by typing the ipFilter --save command.
switch:admin> ipfilter --save BlockTelnet

4. Verify the new policy exists by typing the ipFilter --show command.
switch:admin> ipfilter --show

5. Add a rule to the policy, by typing the ipFilter --addrule command.
switch:admin> ipfilter --addrule BlockTelnet -rule 1 -sip any -dp 23 -proto
tcp -act deny

ATTENTION
The rule number assigned must precede the default rule number for this protocol. For
example, in the defined policy, the Telnet rule number is 2. Therefore, to effectively block
Telnet, the rule number to assign must be 1.
If you choose not to use 1, you must delete the Telnet rule number 2 after adding this rule.
Refer to “Deleting a rule from an IP Filter policy” on page 259 for more information on deleting
IP filter rules.
6. Save the new IP filter policy by typing the ipfilter --save command.
7.

Verify the new policy is correct by typing the ipFilter --show command.

8. Activate the new IP filter policy by typing the ipfilter --activate command.
switch:admin> ipfilter --activate BlockTelnet

9. Verify the new policy is active (the default_ipv4 policy should be displayed as defined).
switch:admin> ipfilter --show
Name: default_ipv4, Type: ipv4, State: defined
Rule
Source IP
Protocol
Dest Port
1
any
tcp
22
2
any
tcp
23
3
any
tcp
80
4
any
tcp
443
5
any
udp
161
6
any
udp
123
7
any
tcp
600 - 1023
8
any
udp
600 - 1023

Action
permit
permit
permit
permit
permit
permit
permit
permit

Name: default_ipv6, Type: ipv6, State: defined
Rule
Source IP
Protocol
Dest Port

Action

Fabric OS Administrator’s Guide
53-1002920-02

227

7

Listener applications

1
2
3
4
5
6
7
8

any
any
any
any
any
any
any
any

tcp
tcp
tcp
tcp
udp
udp
tcp
udp

22
23
80
443
161
123
600 - 1023
600 - 1023

permit
permit
permit
permit
permit
permit
permit
permit

Unblocking Telnet
Use the following procedure to unblock Telnet access.
1. Connect to the switch through a serial port or SSH and log in as admin.
2. Enter the ipfilter --delete command.
Refer to “Deleting a rule from an IP Filter policy” on page 259 for more information on deleting
IP filter rules.
3. To permanently delete the policy, type the ipfilter --save command.

ATTENTION
If you deleted the rule to permit Telnet, you must add a rule to permit Telnet.

Listener applications
Brocade switches block Linux subsystem listener applications that are not used to implement
supported features and capabilities.
Table 33 lists the listener applications that Brocade switches either block or do not start. Note that
RPC ports are blocked.

TABLE 33

228

Blocked listener applications

Listener application

Brocade DCX and DCX 8510 Backbone families

Brocade switches

chargen

Disabled

Disabled

daytime

Disabled

Disabled

discard

Disabled

Disabled

echo

Disabled

Disabled

ftp

Disabled

Disabled

rexec

Block with packet filter

Disabled

rlogin

Block with packet filter

Disabled

rsh

Block with packet filter

Disabled

rstats

Disabled

Disabled

rusers

Disabled

Disabled

time

Block with packet filter

Disabled

Fabric OS Administrator’s Guide
53-1002920-02

Ports and applications used by switches

7

Ports and applications used by switches
If you are using the FC-FC Routing Service, be aware that the secModeEnable command is not
supported.
Table 34 lists the defaults for accessing hosts, devices, switches, and zones.

TABLE 34

Access defaults
Access default

Hosts

Any host can access the fabric by SNMP.
Any host can Telnet to any switch in the fabric.
Any host can establish an HTTP connection to any switch in the fabric.
Any host can establish an API connection to any switch in the fabric.

Devices

All devices can access the management server.
Any device can connect to any FC port in the fabric.

Switch access

Any switch can join the fabric.
All switches in the fabric can be accessed through a serial port.

Zoning

No zoning is enabled.

Port configuration
Table 35 provides information on ports that the switch uses. When configuring the switch for
various policies, take into consideration firewalls and other devices that may sit between switches
in the fabric and your network or between the managers and the switch.

TABLE 35

Port information

Port

Type

Common use

22

TCP

SSH, SCP

23

TCP

Telnet

Use the ipfilter command to block the port.

80

TCP

HTTP

Use the ipfilter command to block the port.

123

UDP

NTP

161

UDP

SNMP

Disable the SNMP service on the remote host if you do not use it, or filter
incoming UDP packets going to this port.

443

TCP

HTTPS

Use the ipfilter command to block the port.

512

TCP

exec

513

TCP

login

514

TCP

shell

Fabric OS Administrator’s Guide
53-1002920-02

Comment

229

7

230

Ports and applications used by switches

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

8

Configuring Security Policies

In this chapter
• ACL policies overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• ACL policy management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• FCS policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Device Connection Control policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• SCC Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Authentication policy for fabric elements . . . . . . . . . . . . . . . . . . . . . . . . . .
• IP Filter policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Policy database distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Management interface security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

231
232
235
238
242
243
253
260
266

ACL policies overview
Each supported Access Control List (ACL) policy listed below is identified by a specific name, and
only one policy of each type can exist, except for DCC policies. Policy names are case-sensitive and
must be entered in all uppercase. Fabric OS provides the following policies:

• Fabric configuration server (FCS) policy — Used to restrict which switches can change the
configuration of the fabric.

• Device connection control (DCC) policies — Used to restrict which Fibre Channel device ports
can connect to which Fibre Channel switch ports.

• Switch connection control (SCC) policy — Used to restrict which switches can join with a switch.
NOTE

Run all commands in this chapter by logging in to Administrative Domain (AD) 255 with the
suggested permissions. If Administrative Domains have not been implemented, log in to AD0.

How the ACL policies are stored
The policies are stored in a local database. The database contains the ACL policy types of FCS,
DCC, SCC, and IPFilter. The number of policies that may be defined is limited by the size of the
database. FCS, SCC and DCC policies are all stored in the same database.
The limit for security policy database size is set to 1Mb. The policies are grouped by state and type.
A policy can be in either of the following states:

• Active, which means the policy is being enforced by the switch.
• Defined, which means the policy has been set up but is not enforced.

Fabric OS Administrator’s Guide
53-1002920-02

231

8

ACL policy management

Policies with the same state are grouped together in a Policy Set. Each switch has the following two
sets:

• Active policy set, which contains ACL policies being enforced by the switch.
• Defined policy set, which contains a copy of all ACL policies on the switch.
When a policy is activated, the defined policy either replaces the policy with the same name in the
active set or becomes a new active policy. If a policy appears in the defined set but not in the active
set, the policy was saved but has not been activated. If a policy with the same name appears in
both the defined and active sets but they have different values, then the policy has been modified
but the changes have not been activated.
Admin Domain considerations: ACL management can be done on AD255 and in AD0 only if there
are no user-defined Admin Domains. Both AD0 (when no other user-defined Admin Domains exist)
and AD255 provide an unfiltered view of the fabric.
Virtual Fabric considerations: ACL policies such as DCC, SCC, and FCS can be configured on each
logical switch. The limit for security policy database size is set to 1Mb per logical switch.

Policy members
The FCS, DCC and SCC policy members are specified by device port WWN, switch WWN, domain
IDs, or switch names, depending on the policy. The valid methods for specifying policy members
are listed in Table 36.

TABLE 36

Valid methods for specifying policy members

Policy name

Device port WWN or Fabric port WWN

Switch WWN

Domain ID

Switch name

FCS_POLICY

No

Yes

Yes

Yes

DCC_POLICY_nnn

Yes

Yes

Yes

Yes

SCC_POLICY

No

Yes

Yes

Yes

ACL policy management
All policy modifications are temporarily stored in volatile memory until those changes are saved or
activated. You can create multiple sessions to the switch from one or more hosts. It is
recommended you make changes from one switch only to prevent multiple transactions from
occurring. Each logical switch will have its own access control list.
The FCS, SCC and DCC policies in Secure Fabric OS are not interchangeable with Fabric OS FCS,
SCC and DCC policies. Uploading and saving a copy of the Fabric OS configuration after creating
policies is strongly recommended. For more information on configuration uploads, see Chapter 9,
“Maintaining the Switch Configuration File”.

NOTE

All changes, including the creation of new policies, are saved and activated on the local switch only—
unless the switch is in a fabric that has a strict or tolerant fabric-wide consistency policy for the ACL
policy type for SCC or DCC. See “Policy database distribution” on page 260 for more information on
the database settings and fabric-wide consistency policy.

232

Fabric OS Administrator’s Guide
53-1002920-02

ACL policy management

8

Displaying ACL policies
You can view the active and defined policy sets at any time. Additionally, in a defined policy set,
policies created in the same login session also appear but these policies are automatically deleted
if the you log out without saving them.
1. Connect to the switch and log in using an account with admin permissions, or an account with
O permission for the Security RBAC class of commands.
2. Type the secPolicyShow command.
switch:admin> secPolicyShow
____________________________________________________
ACTIVE POLICY SET
____________________________________________________
DEFINED POLICY SET

Saving changes without activating the policies
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Enter the secPolicySave command.

Activating ACL policy changes
You can implement changes to the ACL policies using the secPolicyActivate command. This saves
the changes to the active policy set and activates all policy changes since the last time the
command was issued. You cannot activate policies on an individual basis; all changes to the entire
policy set are activated by the command. Until a secPolicySave or secPolicyActivate command is
issued, all policy changes are in volatile memory only and are lost upon rebooting.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Type the secPolicyActivate command.
Example of activating policy changes
switch:admin> secpolicyactivate
About to overwrite the current Active data.
ARE YOU SURE (yes, y, no, n): [no] y

Deleting an ACL policy
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Type secPolicyDelete “policy_name”.
where policy_name is the name of the ACL policy.
3. Save and activate the policy deletion by entering the secPolicyActivate command.

Fabric OS Administrator’s Guide
53-1002920-02

233

8

ACL policy management

Example of deleting an ACL policy
switch:admin> secpolicydelete "DCC_POLICY_010"
About to delete policy Finance_Policy.
Are you sure (yes, y, no, n):[no] y
Finance_Policy has been deleted.

Adding a member to an existing ACL policy
As soon as a policy has been activated, the aspect of the fabric managed by that policy is enforced.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Enter the secPolicyAdd command.
3. To implement the change immediately, enter the secPolicyActivate command.
Example of adding to an ACL policy

For example, to add a member to the SCC_POLICY using the switch WWN:
switch:admin> secpolicyadd "SCC_POLICY", "12:24:45:10:0a:67:00:40"
Member(s) have been added to SCC_POLICY.

Example of adding members to the DCC policy

To add two devices to the DCC policy, and to attach domain 3 ports 1 and 3 (WWNs of devices
are 11:22:33:44:55:66:77:aa and 11:22:33:44:55:66:77:bb):
switch:admin> secpolicyadd "DCC_POLICY_abc",
"11:22:33:44:55:66:77:aa;11:22:33:44:55:66:77:bb;3(1,3)"

Removing a member from an ACL policy
As soon as a policy has been activated, the aspect of the fabric managed by that policy is enforced.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Enter the secPolicyRemove command.
3. To implement the change immediately, enter the secPolicyActivate command.
Example of removing a member

For example, to remove a member that has a WWN of 12:24:45:10:0a:67:00:40 from the
SCC_POLICY:
switch:admin> secpolicyremove "SCC_POLICY", "12:24:45:10:0a:67:00:40"
Member(s) have been removed from SCC_POLICY.

Abandoning unsaved ACL policy changes
You can abandon all ACL policy changes that have not yet been saved.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Enter the secPolicyAbort command.

234

Fabric OS Administrator’s Guide
53-1002920-02

FCS policies

8

Example of aborting unsaved changes
switch:admin> secpolicyabort
Unsaved data has been aborted.

All changes since the last time the secPolicySave or secPolicyActivate commands were entered are
aborted.

FCS policies
Fabric configuration server (FCS) policy in base Fabric OS may be performed on a local switch basis
and may be performed on any switch in the fabric.
The FCS policy is not present by default, but must be created. When the FCS policy is created, the
WWN of the local switch is automatically included in the FCS list. Additional switches can be
included in the FCS list. The first switch in the list becomes the Primary FCS switch.
Switches in the fabric are designated as either a Primary FCS, backup FCS, or non-FCS switch. Only
the Primary FCS switch is allowed to modify and distribute the database within the fabric.
Automatic distribution is supported and you can either configure the switches in your fabric to
accept the FCS policy or manually distribute the FCS policy. Changes made to the FCS policy are
saved to permanent memory only after the changes have been saved or activated; they can be
aborted later if you have set your fabric to distribute the changes manually.

TABLE 37

FCS policy states

Policy state

Characteristics

No active policy

Any switch can perform fabric-wide configuration changes.

Active policy with one entry

A Primary FCS switch is designated (local switch), but there are no backup
FCS switches. If the Primary FCS switch becomes unavailable for any reason,
the fabric is left without an FCS switch.

Active policy with multiple entries

A Primary FCS switch and one or more backup FCS switches are designated. If
the Primary FCS switch becomes unavailable, the next switch in the list
becomes the Primary FCS switch.

FCS policy restrictions
The backup FCS switches normally cannot modify the policy. However, if the Primary FCS switch in
the policy list is not reachable, then a backup FCS switch is allowed to modify the policy.
Once an FCS policy is configured and distributed across the fabric, only the Primary FCS switch can
perform certain operations. Operations that affect fabric-wide configuration are allowed only from
the Primary FCS switch. Backup and non-FCS switches cannot perform security, zoning and AD
operations that affect the fabric configuration. The following error message is returned if a backup
or non-FCS switch tries to perform these operations:
Can only execute this command on the Primary FCS switch.

Operations that do not affect the fabric configuration, such as show or local switch commands, are
allowed on backup and non-FCS switches.
FCS enforcement applies only for user-initiated fabric-wide operations. Internal fabric data
propagation because of a fabric merge is not blocked. Consequently, a new switch that joins the
FCS-enabled fabric could still propagate the AD and zone database.

Fabric OS Administrator’s Guide
53-1002920-02

235

8

FCS policies

Table 38 shows the commands for switch operations for Primary FCS enforcement.

TABLE 38

FCS switch operations

Allowed on FCS switches

Allowed on all switches

secPolicyAdd (Allowed on all switches for SCC and DCC
policies as long as it is not fabric-wide)

secPolicyShow

secPolicyCreate (Allowed on all switches for SCC and
DCC policies as long as it is not fabric-wide)

fddCfg –-localaccept or fddCfg --localreject

secPolicyDelete (Allowed on all switches for SCC and
DCC policies as long as its not fabric-wide)

userconfig, Passwd, Passwdcfg (Fabric-wide distribution
is not allowed from a backup or non-FCS switch.)

secPolicyRemove (Allowed on all switches for SCC and
DCC policies as long as its not fabric-wide)

secPolicyActivate

fddCfg –-fabwideset

secPolicySave

Any fabric-wide commands

secPolicyAbort

All zoning commands except the show commands

SNMP commands

All AD commands

configupload
Any local-switch commands
Any AD command that does not affect fabric-wide
configuration

In Fabric OS v7.1.0 and later, to avoid segmentation of ports due to a member-list order mismatch,
security policy members are sorted based on WWN. By default, DCC and SCC policy members are
sorted based on WWN. Switches running earlier Fabric OS versions will have the member list in the
unsorted manner. Any older-version switch with a policy already created in unsorted order will have
port segmentation due to order mismatch when attempting to join any switch with Fabric OS v7.1.0
or later. To overcome the order mismatch, you can modify the member list in the switch by using the
-legacy option in the secPolicyAdd and secPolicyCreate commands.

Ensuring fabric domains share policies
Whether your intention is to create new FCS policies or manage your current FCS policies, you must
follow certain steps to ensure the domains throughout your fabric have the same policy.
The local-switch WWN cannot be deleted from the FCS policy.
1. Create the FCS policy using the secPolicyCreate command.
2. Activate the policy using the secPolicyActivate command.
If the command is not entered, the changes are lost when the session is logged out.
3. To distribute the policies, enter the distribute -p policy_list -d switch_list command to either
send the policies to intended domains, or enter the distribute -p policy_list -d wild_card (*)
command to send the policies to all switches.

Creating an FCS policy
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Enter the secPolicyCreate “FCS_POLICY” command.

236

Fabric OS Administrator’s Guide
53-1002920-02

FCS policies

8

Example of creating an FCS policy

The following example creates an FCS policy that allows a switch with domain ID 2 to become a
primary FCS and domain ID 4 to become a backup FCS:
switch:admin> secpolicycreate "FCS_POLICY", "2;4"
FCS_POLICY has been created

3. To save or activate the new policy, enter either the secPolicySave or the secPolicyActivate
command. Once the policy has been activated you can distribute the policy.

NOTE
FCS policy must be consistent across the fabric. If the policy is inconsistent in the fabric, then you
will not be able to perform any fabric-wide configurations from the primary FCS.

Modifying the order of FCS switches
1. Log in to the Primary FCS switch using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Type secPolicyShow “Defined”, “FCS_POLICY”.
This displays the WWNs of the current Primary FCS switch and backup FCS switches.
3. Type secPolicyFCSMove; then provide the current position of the switch in the list and the
desired position at the prompts.
Alternatively, enter secPolicyFCSMove [From, To] command. From is the current position in the
list of the FCS switch and To is the desired position in the list for this switch.
Example of moving an FCS policy

The following example moves a backup FCS switch from position 2 to position 3 in the FCS list,
using interactive mode:
primaryfcs:admin> secpolicyfcsmove
Pos
Primary WWN
DId swName.
=================================================
1
Yes
10:00:00:60:69:10:02:18
1 switch5.
2
No
10:00:00:60:69:00:00:5a
2 switch60.
3
No
10:00:00:60:69:00:00:13
3 switch73.
Please enter position you’d like to move from : (1..3) [1] 2
Please enter position you’d like to move to : (1..3) [1] 3
____________________________________________________
DEFINED POLICY SET
FCS_POLICY
Pos
Primary WWN
DId swName
__________________________________________________
1
Yes
10:00:00:60:69:10:02:18
1 switch5.
2
No
10:00:00:60:69:00:00:13
3 switch73.
3
No
10:00:00:60:69:00:00:5a
2 switch60.
____________________________________________________

4. Type the secPolicyActivate command to activate and save the new order.

Fabric OS Administrator’s Guide
53-1002920-02

237

8

Device Connection Control policies

FCS policy distribution
The FCS policy can be automatically distributed using the fddCfg --fabwideset command or it can
be manually distributed to the switches using the distribute -p command. Each switch that receives
the FCS policy must be configured to receive the policy. To configure the switch to accept
distribution of the FCS policy, refer to “Database distribution settings” on page 261.
Database distributions may be initiated from only the Primary FCS switch. FCS policy configuration
and management is performed using the command line or a manageability interface.
Only the Primary FCS switch is allowed to distribute the database. The FCS policy can be manually
distributed across the fabric using the distribute -p command. Since this policy is distributed
manually, the command fddCfg –-fabwideset is used to distribute a fabric-wide consistency policy
for FCS policy in an environment consisting of only Fabric OS v6.2.0 and later switches.
FCS enforcement for the distribute command is handled differently for FCS and other databases in
an FCS fabric:

• For an FCS database, the enforcement allows any switch to initiate the distribution. This is to
support FCS policy creation specifying a remote switch as Primary.

• For other database distributions, only the Primary FCS switch can initiate the distribution.
The FCS policy distribution is allowed to be distributed from a switch in the FCS list. However, if
none of the FCS switches in the existing FCS list are reachable, receiving switches accept
distribution from any switch in the fabric. To learn more about how to distribute policies, refer to
“ACL policy distribution to other switches” on page 262.
Local switch configuration parameters are needed to control whether a switch accepts or rejects
distributions of FCS policy and whether the switch is allowed to initiate distribution of an FCS policy.
A configuration parameter controls whether the distribution of the policy is accepted or rejected on
the local switch. Setting the configuration parameter to accept indicates distribution of the policy
will be accepted and distribution may be initiated using the distribute -p command. Setting the
configuration parameter to reject indicates the policy distribution is rejected and the switch may
not distribute the policy.
The default value for the distribution configuration parameter is accept, which means the switch
accepts all database distributions and is able to initiate a distribute operation for all databases.

TABLE 39

Distribution policy states

Fabric OS

State

v6.2.0 and later configured to accept Target switch accepts distribution and fabric state change occurs.
v6.2.0 and later configured to reject

Target switch explicitly rejects the distribution and the operation fails.
The entire transaction is aborted and no fabric state change occurs.

Device Connection Control policies
Multiple Device Connection Control (DCC) policies can be used to restrict which device ports can
connect to which switch ports. The devices can be initiators, targets, or intermediate devices such
as SCSI routers and loop hubs. By default, all device ports are allowed to connect to all switch
ports; no DCC policies exist until they are created. For information regarding DCC policies and
F_Port trunking, refer to the Access Gateway Administrator’s Guide.

238

Fabric OS Administrator’s Guide
53-1002920-02

Device Connection Control policies

8

Each device port can be bound to one or more switch ports; the same device ports and switch
ports may be listed in multiple DCC policies. After a switch port is specified in a DCC policy, it
permits connections only from designated device ports. Device ports that are not specified in any
DCC policies are allowed to connect only to switch ports that are not specified in any DCC policies.
When a DCC violation occurs, the related port is automatically disabled and must be re-enabled
using the portEnable command.
Table 40 shows the possible DCC policy states.

TABLE 40

DCC policy states

Policy state

Characteristics

No policy

Any device can connect to any switch port in the fabric.

Policy with no
entries

Any device can connect to any switch port in the fabric. An empty policy is the same as no
policy.

Policy with entries

If a device WWN or Fabric port WWN is specified in a DCC policy, that device is only allowed
access to the switch if connected by a switch port listed in the same policy.
If a switch port is specified in a DCC policy, it only permits connections from devices that are
listed in the policy.
Devices with WWNs that are not specified in a DCC policy are allowed to connect to the
switch at any switch ports that are not specified in a DCC policy.
Switch ports and device WWNs may exist in multiple DCC policies.
Proxy devices are always granted full access and can connect to any switch port in the fabric.

Virtual Fabrics considerations
The DCC policies that have entries for the ports that are being moved from one logical switch to
another will be considered stale and will not be enforced. You can choose to keep stale policies in
the current logical switch or delete the stale policies after the port movements. Use the
secPolicyDelete command to delete stale DCC policies.

DCC policy restrictions
The following restrictions apply when using DCC policies:

• Some older private-loop host bus adaptors (HBAs) do not respond to port login from the switch
and are not enforced by the DCC policy. This does not create a security problem because these
HBAs cannot contact any device outside of their immediate loop.

• DCC policies cannot manage or restrict iSCSI connections, that is, an FC Initiator connection
from an iSCSI gateway.

• You cannot manage proxy devices with DCC policies. Proxy devices are always granted full
access, even if the DCC policy has an entry that restricts or limits access of a proxy device.

Creating a DCC policy
DCC policies must follow the naming convention “DCC_POLICY_nnn,” where nnn represents a
unique string. The maximum length is 30 characters, including the prefix DCC_POLICY_.
Device ports must be specified by port WWN. Switch ports can be identified by the switch WWN,
domain ID, or switch name followed by the port or area number. To specify an allowed connection,
enter the device port WWN, a semicolon, and the switch port identification.

Fabric OS Administrator’s Guide
53-1002920-02

239

8

Device Connection Control policies

The following methods of specifying an allowed connection are possible:

• deviceportWWN;switchWWN (port or area number)
• deviceportWWN;domainID (port or area number)
• deviceportWWN;switchname (port or area number)
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Enter the secPolicyCreate “DCC_POLICY_nnn” command.
DCC_POLICY_nnn is the name of the DCC policy; nnn is a string consisting of up to 19
alphanumeric or underscore characters to differentiate it from any other DCC policies.
3. To save or activate the new policy, enter the appropriate command:

• To save the policy, enter the secPolicySave command.
• To save and activate the policy, enter the secPolicyActivate command.
If neither of these commands is entered, the changes are lost when the session is logged out.
Example of creating DCC policies

To create the DCC policy “DCC_POLICY_server” that includes device 11:22:33:44:55:66:77:aa and
port 1 and port 3 of switch domain 1:
switch:admin> secpolicycreate
"DCC_POLICY_server","11:22:33:44:55:66:77:aa;1(1,3)"
DCC_POLICY_server has been created

To create the DCC policy “DCC_POLICY_storage” that includes device port WWN
22:33:44:55:66:77:11:bb, all ports of switch domain 2, and all currently connected devices of
switch domain 2:
switch:admin> secpolicycreate "DCC_POLICY_storage",
"22:33:44:55:66:77:11:bb;2[*]"
DCC_POLICY_storage has been created

To create the DCC policy “DCC_POLICY_abc” that includes device 33:44:55:66:77:11:22:cc and
ports 1 through 6 and port 9 of switch domain 3:
switch:admin> secpolicycreate "DCC_POLICY_abc",
"33:44:55:66:77:11:22:cc;3(1-6,9)"
DCC_POLICY_abc has been created

To create the DCC policy “DCC_POLICY_example” that includes devices 44:55:66:77:22:33:44:dd
and 33:44:55:66:77:11:22:cc, ports 1 through 4 of switch domain 4, and all devices currently
connected to ports 1 through 4 of switch domain 4:
switch:admin> secpolicycreate "DCC_POLICY_example",
"44:55:66:77:22:33:44:dd;33:44:55:66:77:11:22:cc;4[1-4]"
DCC_POLICY_example has been created

Deleting a DCC policy
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Enter the secPolicyDelete command.

240

Fabric OS Administrator’s Guide
53-1002920-02

Device Connection Control policies

8

Example of deleting stale DCC policies
switch:admin> secpolicydelete ALL_STALE_DCC_POLICY
About to clear all STALE DCC policies
ARE YOU SURE (yes, y, no, n): [no] y

DCC policy behavior with Fabric-Assigned PWWNs
A DCC policy check is always performed for the physical port WWN of a device when the HBA has
established that the device is attempting a normal FLOGI and has both a fabric-assigned port WWN
(FA-PWWN) and a physical port WWN.
DCC policies created with FA-PWWNs will result in the disabling of FA-PWWN assigned ports on
subsequent FLOGI. It is therefore recommended to create policies with the physical PWWN
DCC policies created with the lock down feature result in DCC policies with FA-PWWNs. It is
therefore recommended to avoid using the lock down feature in fabrics that are using FA-PWWNs.
A DCC policy created with a device WWN for a specific port allows the device to log in only on the
same port. The same device will not be allowed to log in on a different port. For devices that log in
across an AG, the policy should be created with all the NPIV ports, so even if failover occurs the
device will be allowed to log in on a different NPIV port.
Table 41 lists the behavior of the DCC policy with FA-PWWNs in the fabric when the DCC policy is
created using lockdown support.

TABLE 41

DCC policy behavior with FA-PWWN when created using lockdown support

Configuration

WWN seen on
DCC policy list

Behavior when DCC policy
activates

Behavior on portDisable
and portEnable

•
•

FA-PWWN has logged into the switch
DCC policy creation with lock down
(uses FA-PWWN).
DCC policy activation.

FA-PWWN

Traffic will not be
disrupted.1

Ports will be disabled
for security violation.2

DCC policy creation with lockdown
(uses physical PWWN).
FA-PWWN has logged into the switch
DCC policy activation.

Physical
PWWN

Traffic will not be
disrupted.

Ports will come up
without security
issues.

DCC policy creation with lockdown
(uses physical PWWN)
DCC policy activation
FA-PWWN has logged into the switch

Physical
PWWN

Traffic will not be
disrupted.

Ports will come up
without any security
issues.

•
•
•
•
•
•
•

1. Indicates a security concern, because devices that are logged in with FA-PWWNs will not be disabled after
activation of DCC policies that are created with FA-PWWNs. This is done to avoid disturbing any existing
management.
2. Any disruption in the port will disable the port for a security violation. As the traffic is already disrupted for this
port, you must enforce the DCC policy for a physical device WWN; otherwise, the device will not be allowed to login
again.

Table 42 shows the behavior of a DCC policy created manually with the physical PWWN of a device.
The configurations shown in this table are the recommended configurations when an FA-PWWN is
logged into the switch.

Fabric OS Administrator’s Guide
53-1002920-02

241

8

SCC Policies

TABLE 42

DCC policy behavior when created manually with PWWN

Configuration

WWN seen on
DCC policy list

Behavior when DCC policy
activates

Behavior on portDisable
and portEnable

•
•

FA-PWWN has logged into the switch.
DCC policy creation manually with
physical PWWN of device.
DCC policy activation.

PWWN

Traffic will not be
disrupted.

Ports will come up
without security
issues.

DCC policy creation. manually with
physical PWWN
FA-PWWN has logged into the switch.
DCC policy activation.

PWWN

Traffic will not be
disrupted.

Ports will come up
without security
issues.

DCC policy creation manually with
physical PWWN,
DCC policy activation.
FA-PWWN has logged into the switch.

Physical PWWN Traffic will not be
disrupted.

•
•
•
•
•
•
•

Ports will come up
without any security
issues.

SCC Policies
The switch connection control (SCC) policy is used to restrict which switches can join the fabric.
Switches are checked against the policy each time an E_Port-to-E_Port connection is made. The
policy is named SCC_POLICY and accepts members listed as WWNs, domain IDs, or switch names.
Only one SCC policy can be created.
By default, any switch is allowed to join the fabric; the SCC policy does not exist until it is created.
When connecting a Fibre Channel router to a fabric or switch that has an active SCC policy, the
front domain of the Fibre Channel router must be included in the SCC policy.
SCC policy states are shown in Table 43.

TABLE 43

SCC policy states

Policy state

SCC policy enforcement

No active policy

All switches can connect to the switch with the specified policy.

Active policy that has no members

All neighboring switches are segmented.

Active policy that has members

The neighboring switches not specified in the SCC policy are segmented.

Virtual Fabrics considerations: In a logical fabric environment the SCC policy enforcement is not
done on the logical ISL. For a logical ISL-based switch, the SCC policy enforcement is considered as
the reference and the logical ISL is formed if the SCC enforcement passes on the extended ISL. The
following changes:

• A logical switch supports an SCC policy. You can configure and distribute an SCC policy on a
logical switch.

• SCC enforcement is performed on a ISL based on the SCC policy present on the logical switch.
For more information on Virtual Fabrics, refer to Chapter 11, “Managing Virtual Fabrics”.

242

Fabric OS Administrator’s Guide
53-1002920-02

Authentication policy for fabric elements

8

Creating an SCC policy
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Security RBAC class of commands.
2. Enter the secPolicyCreate “SCC_POLICY” command.
3. Save or activate the new policy by entering either the secPolicySave or the secPolicyActivate
command.
If neither of these commands is entered, the changes are lost when the session is logged out.
Example of creating an SCC policy

For example, to create an SCC policy that allows switches that have domain IDs 2 and 4 to join
the fabric:
switch:admin> secpolicycreate "SCC_POLICY", "2;4"
SCC_POLICY has been created
switch:admin> secpolicysave

Authentication policy for fabric elements
By default, Fabric OS v6.2.0 and later use Diffie Hellman - Challenge Handshake Authentication
Protocol) (DH-CHAP) or Fibre Channel Authentication Protocol (FCAP) for authentication.
These protocols use shared secrets and digital certificates, based on switch WWN and public key
infrastructure (PKI) technology, to authenticate switches. Authentication automatically defaults to
FCAP if both switches are configured to accept FCAP protocol in authentication, unless ports are
configured for in-flight encryption, in which case authentication defaults to DH-CHAP if both
switches are configured to accept the DH-CHAP protocol in authentication. To use FCAP on both
switches, PKI certificates have to be installed.
The DH-CHAP and FCAP authentication protocols used by Brocade switches are FC-SP2 standard
compliant.

NOTE
The fabric authentication feature is available in base Fabric OS. No license is required.
FCAP requires the exchange of certificates between two or more switches to authenticate to each
other before they form or join a fabric. Beginning with Fabric OS v7.0.0, these certificates are no
longer issued by Brocade, but by a third-party which is now the root CA for all of the issued
certificates. You can use Brocade and third-party certificates between switches that are Fabric OS
v6.4.0, but only Brocade-issued certificates (where Brocade is the root CA) for Fabric OS versions
earlier than v6.4.0. The certificates must be in PEM (Privacy Enhanced Mail) encoded format for
both root and peer certificates. The switch certificates issued from the third-party vendors can be
directly issued from the root CA or from an intermediate CA authority.
When you configure DH-CHAP authentication, you also must define a pair of shared secrets known
to both switches as a secret key pair. Figure 17 illustrates how the secrets are configured. A secret
key pair consists of a local secret and a peer secret. The local secret uniquely identifies the local
switch. The peer secret uniquely identifies the entity to which the local switch authenticates. Every
switch can share a secret key pair with any other switch or host in a fabric.
To use DH-CHAP authentication, a secret key pair has to be configured on both switches. For more
information on setting up secret key pairs, refer to “Setting a secret key pair” on page 250.

Fabric OS Administrator’s Guide
53-1002920-02

243

8

Authentication policy for fabric elements

When configured, the secret key pair is used for authentication. Authentication occurs whenever
there is a state change for the switch or port. The state change can be due to a switch reboot, a
switch or port disable and enable, or the activation of a policy.

Key database on switch
Local secret A
Peer secret B

Switch A
FIGURE 17

Key database on switch
Local secret B
Peer secret A

Switch B

DH-CHAP authentication

If you use DH-CHAP authentication, then a secret key pair must be installed only in connected
fabric elements. However, as connections are changed, new secret key pairs must be installed
between newly connected elements. Alternatively, a secret key pair for all possible connections
may be initially installed, enabling links to be arbitrarily changed while still maintaining a valid
secret key pair for any new connection.
The switch authentication (AUTH) policy initiates DH-CHAP/FCAP authentication on all E_Ports. This
policy is persistent across reboots, which means authentication will be initiated automatically on
ports or switches brought online if the policy is set to activate authentication. The AUTH policy is
distributed by command; automatic distribution of the AUTH policy is not supported.
The default configuration directs the switch to attempt FCAP authentication first, DH-CHAP second.
The switch may be configured to negotiate FCAP, DH-CHAP, or both.
The DH group is used in the DH-CHAP protocol only. The FCAP protocol exchanges the DH group
information, but does not use it.
Virtual Fabrics considerations
If Virtual Fabrics is enabled, all AUTH module parameters such as shared secrets, and shared
switch and device policies, are logical switch-wide. That means you must configure shared secrets
and policies separately on each logical switch and the shared secrets and policies must be set on
each switch prior to authentication. On logical switch creation, authentication takes default values
for policies and other parameters. FCAP certificates are installed on a chassis, but are configured
on each logical switch.

E_Port authentication
The authentication (AUTH) policy allows you to configure DH-CHAP authentication on switches with
Fabric OS v5.3.0 and later. By default the policy is set to PASSIVE and you can change the policy. All
changes to the AUTH policy take effect during the next authentication request. This includes
starting authentication on all E_Ports on the local switch if the policy is changed to ON or ACTIVE,
and clearing the authentication if the policy is changed to OFF. The authentication configurations
will be effective only on subsequent E_ and F_Port initialization.

244

Fabric OS Administrator’s Guide
53-1002920-02

Authentication policy for fabric elements

8

ATTENTION
A secret key pair has to be installed prior to changing the policy. For more information on setting up
secret key pairs, refer to “Setting a secret key pair” on page 250.
If you must disable authentication on a port that has in-flight encryption or compression
configured, you must first disable in-flight encryption or compression on the port, and then disable
authentication. Refer to Chapter 16, “In-flight Encryption and Compression,” for details.
Virtual Fabrics considerations
The switch authentication policy applies to all E_Ports in a logical switch. This includes ISLs and
extended ISLs. Authentication of extended ISLs between two base switches is considered
peer-chassis authentication. Authentication between two physical entities is required, so the
extended ISL which connects the two chassis needs to be authenticated. The corresponding
extended ISL for a logical ISL authenticates the peer-chassis, therefore the logical ISL
authentication is not required. Because the logical ISLs do not carry actual traffic, they do not need
to be authenticated. Authentication on re-individualization is also blocked on logical ISLs. The
following error message is printed on the console when you execute the authUtil –-authinit
command on logical-ISLs, “Failed to initiate authentication. Authentication is not supported on
logical ports ”. For more information on Virtual Fabrics, refer to Chapter 11, “Managing
Virtual Fabrics”.

Configuring E_Port authentication
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Authentication RBAC class of commands.
2. Enter the authUtil command to set the switch policy mode.
Example of configuring E_Port authentication

The following example shows how to enable Virtual Fabrics and configure the E_Ports to perform
authentication using the AUTH policies authUtil command.
switch:admin> fosconfig -enable vf
WARNING: This is a disruptive operation that requires a reboot to take
effect.
All EX ports will be disabled upon reboot.
Would you like to continue [Y/N] y
switch:admin> authutil --authinit 2,3,4

CAUTION
If data input has not been completed and a failover occurs, the command is terminated without
completion and your entire input is lost.
If data input has completed, the enter key pressed, and a failover occurs, data may or may not be
replicated to the other CP depending on the timing of the failover. Log in to the other CP after the
failover is complete and verify the data was saved. If data was not saved, run the command
again.
Example of setting the policy to active mode
switch:admin> authutil --policy -sw active
Warning: Activating the authentication policy requires

Fabric OS Administrator’s Guide
53-1002920-02

245

8

Authentication policy for fabric elements

either DH-CHAP secrets or PKI certificates depending
on the protocol selected. Otherwise, ISLs will be
segmented during next E-port bring-up.
ARE YOU SURE (yes, y, no, n): [no] y
Auth Policy is set to ACTIVE

NOTE

This authentication-policy change will not affect online EX_Ports.

Re-authenticating E_Ports
Use the authUtil --authinit command to re-initiate the authentication on selected ports. It provides
flexibility to initiate authentication for specified E_Ports, a set of E_Ports, or all E_Ports on the
switch. This command does not work on loop, NPIV and FICON devices, or on ports configured for
in-flight encryption. The command authUtil can re-initiate authentication only if the device was
previously authenticated. If the authentication fails because shared secrets do not match, the port
is disabled.
This command works independently of the authentication policy; this means you can initiate the
authentication even if the switch is in PASSIVE mode. This command is used to restart
authentication after changing the DH-CHAP group, hash type, or shared secret between a pair of
switches.

ATTENTION
This command may bring down E_Ports if the DH-CHAP shared secrets are not installed correctly.
1. Log in to the switch using an account with admin permissions, or an account with OM
permissions for the Authentication RBAC class of commands.
2. Enter the authUtil –-authinit command.
Example for specific ports on the switch
switch:admin> authutil –-authinit 2,3,4

Example for all E_Ports on the switch
switch:admin> authutil –-authinit allE

Example for Backbones using the slot/port format
switch:admin> authutil –-authinit 1/1, 1/2

Device authentication policy
Device authentication policy can also be categorized as an F_Port, node port, or an HBA
authentication policy. Fabric-wide distribution of the device authentication policy is not supported
because the device authentication requires manual interaction in setting the HBA shared secrets
and switch shared secrets, and most of the HBAs do not support the defined DH groups for use in
the DH-CHAP protocol.

NOTE

Authentication is supported from Brocade fabric switches in native mode to Access Gateway
switches and from Access Gateway switches to HBAs. For more information, refer to the Access
Gateway Administrator’s Guide.

246

Fabric OS Administrator’s Guide
53-1002920-02

Authentication policy for fabric elements

8

By default the devicepolicy is in the OFF state, which means the switch clears the security bit in the
FLOGI (fabric login). The authUtil command provides an option to change the device policy mode to
select PASSIVE policy, which means the switch responds to authentication from any device and
does not initiate authentication to devices.
When the policy is set to ON, the switch expects a FLOGI with the FC-SP bit set. If not, the switch
rejects the FLOGI with reason LS_LOGICAL_ERROR (0x03), explanation “Authentication
Required”(0x48), and disables the port. Regardless of the policy, the F_Port is disabled if the
DH-CHAP protocol fails to authenticate.
If the HBA sets the FC-SP bit during FLOGI and the switch sends a FLOGI accept with the FC-SP bit
set, then the switch expects the HBA to start the AUTH_NEGOTIATE. From this point on until the
AUTH_NEGOTIATE is completed, all ELS and CT frames, except the AUTH_NEGOTIATE ELS frame,
are blocked by the switch. During this time, the Fibre Channel driver rejects all other ELS frames.
The F_Port does not form until the AUTH_NEGOTIATE is completed. It is the HBA's responsibility to
send an Authentication Negotiation ELS frame after receiving the FLOGI accept frame with the
FC-SP bit set.
Virtual Fabrics considerations
Because the device authentication policy has switch and logical switch-based parameters, each
logical switch is set when Virtual Fabrics is enabled. Authentication is enforced based on each
logical switch’s policy settings.

Configuring device authentication
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the Authentication RBAC class of commands.
2. Enter the authUtil command to set the device policy mode.
Example of setting the Device policy to passive mode:
switch:admin> authutil --policy -dev passive
Warning: Activating the authentication policy requires
DH-CHAP secrets on both switch and device. Otherwise,
the F-port will be disabled during next F-port
bring-up.
ARE YOU SURE (yes, y, no, n): [no] y
Device authentication is set to PASSIVE

AUTH policy restrictions
All fabric element authentication configurations are performed on a local switch basis.
Device authentication policy supports devices that are connected to the switch in point-to-point
manner and is visible to the entire fabric. The following are not supported:

•
•
•
•
•
•

Public loop devices
Single private devices
Private loop devices
Mixed public and private devices in loop
NPIV devices
FICON channels

Fabric OS Administrator’s Guide
53-1002920-02

247

8

Authentication policy for fabric elements

• Configupload and download will not be supported for the following AUTH attributes: auth type,
hash type, group type.

NOTE

For information about how to use authentication with Access Gateway, refer to the Access Gateway
Administrator’s Guide.

Authentication protocols
Use the authUtil command to perform the following tasks:

• Display the current authentication parameters.
• Select the authentication protocol used between switches.
• Select the DH (Diffie-Hellman) group for a switch.
Run the authUtil command on the switch you want to view or change. Below are the different
options to specify which DH group you want to use.

•
•
•
•
•

00 – DH Null option
01 – 1024 bit key
02 – 1280 bit key
03 – 1536 bit key
04 – 2048 bit key

Viewing the current authentication parameter settings for a switch
1. Log in to the switch using an account with admin permissions, or an account with the O
permission for the Authentication RBAC class of commands.
2. Enter the authUtil --show.
Example of output from the authUtil --show command
AUTH TYPE
HASH TYPE
GROUP TYPE
-------------------------------------fcap,dhchap
sha1,md5
0, 1, 2, 3, 4
Switch Authentication Policy: PASSIVE
Device Authentication Policy: OFF

Setting the authentication protocol
1. Log in to the switch using an account with admin permissions, or an account with OM
permissions for the Authentication RBAC class of commands.
2. Enter the authUtil --set -a command specifying fcap, dhchap, or all.
Example of setting the DH-CHAP authentication protocol
switch:admin> authutil --set -a dhchap
Authentication is set to dhchap.

When using DH-CHAP, make sure that you configure the switches at both ends of a link.

248

Fabric OS Administrator’s Guide
53-1002920-02

Authentication policy for fabric elements

8

NOTE

If you set the authentication protocol to DH-CHAP or FCAP, have not configured shared secrets
or certificates, and authentication is checked (for example, you enable the switch), then switch
authentication will fail.
If the E_Port is to carry in-flight encrypted traffic, the authentication protocol must be set to
DH-CHAP. You must also use the -g option to set the DH group value to group 4 or all groups.
See Chapter 16, “In-flight Encryption and Compression,” for details about in-flight encryption.

Secret key pairs for DH-CHAP
When you configure the switches at both ends of a link to use DH-CHAP for authentication, you
must also define a secret key pair—one for each end of the link. Use the secAuthSecret command
to perform the following tasks:

• View the WWN of switches with a secret key pair.
• Set the secret key pair for switches.
• Remove the secret key pair for one or more switches.
Characteristics of a secret key pair
• The secret key pair must be set up locally on every switch. The secret key pair is not distributed
fabric-wide.

• If a secret key pair is not set up for a link, authentication fails. The “Authentication Failed”
(reason code 05h) error will be reported and logged.

• The minimum length of a shared secret is 8 characters and the maximum length is 40
characters. If the E_Port is to carry in-flight encrypted traffic, a shared secret or at least 32
characters is recommended. See Chapter 16, “In-flight Encryption and Compression” for
details about in-flight encryption.

NOTE

When setting a secret key pair, note that you are entering the shared secrets in plain text. Use a
secure channel (for example, SSH or the serial console) to connect to the switch on which you are
setting the secrets.

Viewing the list of secret key pairs in the current switch database
1. Log in to the switch using an account with admin permissions, or an account with the O
permission for the Authentication RBAC class of commands.
2. Enter the secAuthSecret --show command.
The output displays the WWN, domain ID, and name (if known) of the switches with defined
shared secrets:
WWN
DId
Name
----------------------------------------------10:00:00:60:69:80:07:52
Unknown
10:00:00:60:69:80:07:5c
1
switchA

Fabric OS Administrator’s Guide
53-1002920-02

249

8

Authentication policy for fabric elements

Note about Access Gateway switches
Because Domain ID and name are not supported for Access Gateway, secAuthSecret --show output
for Access Gateway appears as follows:
WWN
DId
Name
----------------------------------------------10:00:8C:7C:FF:03:9E:00
-1
Unknown
10:00:8C:7C:FF:03:9E:01
-1
Unknown
10:00:8C:7C:FF:0D:AF:01
-1
Unknown

When setting and removing the secret for a switch or device on Access Gateway, only the WWN can
be used.

Setting a secret key pair
1. Log in to the switch using an account with admin permissions, or an account with OM
permissions for the Authentication RBAC class of commands.
2. Enter the secAuthSecret --set command.
The command enters interactive mode. The command returns a description of itself and
needed input; then it loops through a sequence of switch specification, peer secret entry, and
local secret entry.
To exit the loop, press Enter for the switch name; then type y.
Example of setting a secret key pair
switchA:admin> secauthsecret --set
This command is used to set up secret keys for the DH-CHAP authentication.
The minimum length of a secret key is 8 characters and maximum 40
characters. Setting up secret keys does not initiate DH-CHAP
authentication. If switch is configured to do DH-CHAP, it is performed
whenever a port or a switch is enabled.
Warning: Please use a secure channel for setting secrets. Using
an insecure channel is not safe and may compromise secrets.
Following inputs should be specified for each entry.
1. WWN for which secret is being set up.
2. Peer secret: The secret of the peer that authenticates to peer.
3. Local secret: The local secret that authenticates peer.
Press Enter to start setting up shared secrets > 
Enter WWN, Domain, or switch name (Leave blank when done):
10:20:30:40:50:60:70:80
Enter peer secret: 
Re-enter peer secret: 
Enter local secret: 
Re-enter local secret: 
Enter WWN, Domain, or switch name (Leave blank when done):
10:20:30:40:50:60:70:81
Enter peer secret: 

250

Fabric OS Administrator’s Guide
53-1002920-02

Authentication policy for fabric elements

8

Re-enter peer secret: 
Enter local secret: 
Re-enter local secret: 
Enter WWN, Domain, or switch name (Leave blank when done): 
Are you done? (yes, y, no, n): [no] y
Saving data to key store… Done.

3. Disable and enable the ports on a peer switch using the portDisable and portEnable
commands.

FCAP configuration overview
Beginning with Fabric OS release 7.0.0, you must configure the switch to use third-party certificates
for authentication with the peer switch.
To perform authentication with FCAP protocol with certificates issued from third party, the user has
to perform following steps:
1. Choose a certificate authority (CA).
2. Generate a public, private key, passphrase and a CSR on each switch.
3. Store the CSR from each switch on a file server.
4. Obtain the certificates from the CA.
You can request a certificate from a CA through a Web browser. After you request a certificate,
the CA either sends certificate files by e-mail (public) or gives access to them on a remote host
(private). Typically, the CA provides the certificate files listed in Table 44.

ATTENTION
Only the .pem file is supported for FCAP authentication.

TABLE 44

FCAP certificate files

Certificate file

Description

nameCA.pem

The CA certificate. It must be installed on the remote and local switch to verify the
validity of the switch certificate or switch validation fails.

name.pem

The switch certificate.

5. On each switch, install the CA certificate before installing switch certificate.
6. After the CA certificate is installed, install the switch certificate on each switch.
7.

Update the switch database for peer switches to use third-party certificates.

8. Use the newly installed certificates by starting the authentication process.

Generating the key and CSR for FCAP
The public/private key and CSR has to be generated for the local and remote switches that will
participate in the authentication. In FCAP, one command is used to generate the public/private key
the CSR, and the passphrase.

Fabric OS Administrator’s Guide
53-1002920-02

251

8

Authentication policy for fabric elements

1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having OM permissions for the PKI RBAC class of commands.
2. Enter the secCertUtil generate -fcapall -keysize command on the local switch.
switch:admin> seccertutil generate -fcapall -keysize 1024
WARNING!!!
About to create FCAP:
ARE YOU SURE (yes, y, no, n): [no] y
Installing Private Key and Csr...
Switch key pair and CSR generated...

3. Repeat step 2 on the remote switch.

Exporting the CSR for FCAP
You will need to export the CSR file created in “Generating the key and CSR for FCAP” section and
send to a Certificate Authority (CA). The CA will in turn provide two files as outlined in “FCAP
configuration overview” on page 251.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having OM permissions for the PKI RBAC class of commands.
2. Enter the secCertUtil export –fcapswcsr command.
switch:admin> seccertutil export -fcapswcsr
Select protocol [ftp or scp]: scp
Enter IP address: 10.1.2.3
Enter remote directory: /myHome/jdoe/OPENSSL
Enter Login Name: jdoe
jdoe@10.1.2.3's password: 
Success: exported FCAP CA certificate

Importing CA for FCAP
Once you receive the files back from the Certificate Authority, you will need to install or import them
onto the local and remote switches.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having OM permissions for the PKI RBAC class of commands.
2. Enter the secCertUtil import –fcapcacert command and verify the CA certificates are
consistent on both local and remote switches.
switch:admin> seccertutil import -fcapcacert
Select protocol [ftp or scp]: scp
Enter IP address: 10.1.2.3
Enter remote directory: /myHome/jdoe/OPENSSL
Enter certificate name (must have a ".pem" suffix):CACert.pem
Enter Login Name: jdoe
jdoe@10.1.2.3's password: 
Success: imported certificate [CACert.pem].

Importing the FCAP switch certificate
ATTENTION
The CA certificates must be installed prior to installing the switch certificate.

252

Fabric OS Administrator’s Guide
53-1002920-02

IP Filter policy

8

1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having OM permissions for the PKI RBAC class of commands.
2. Enter the secCertUtil import –fcapswcert command.
switch:admin> seccertutil import -fcapswcert
Select protocol [ftp or scp]: scp
Enter IP address: 10.1.2.3
Enter remote directory: /myHome/jdoe/OPENSSL
Enter certificate name (must have ".crt" or ".cer" ".pem" or ".psk"
suffix):01.pem
Enter Login Name: jdoe
jdoe@10.1.2.3's password: 
Success: imported certificate [01.pem].

Starting FCAP authentication
1. Log in to the switch using an account with admin permissions, or an account with OM
permissions for the Authentication RBAC class of commands.
2. Enter the authUtil --authinit command to start the authentication using the newly imported
certificates. (This command is not supported in Access Gateway mode.)
3. Enter the authUtil --policy -sw command and select active or on, the default is passive. This
makes the changes permanent and forces the switch to request authentication. (For Access
Gateway mode, the defaults for sw policy and dev policy are off, and there is no passive option
for sw policy.)

NOTE
This authentication-policy change does not affect online EX_Ports.

Fabric-wide distribution of the authorization policy
The AUTH policy can be manually distributed to the fabric by command; there is no support for
automatic distribution. To distribute the AUTH policy, see “Distributing the local ACL policies” on
page 263 for instructions.
Local Switch configuration parameters are needed to control whether a switch accepts or rejects
distributions of the AUTH policy using the distribute command and whether the switch may initiate
distribution of the policy. To set the local switch configuration parameter, refer to “Policy database
distribution” on page 260.

NOTE

This is not supported for Access Gateway mode.

IP Filter policy
The IP Filter policy is a set of rules applied to the IP management interfaces as a packet filtering
firewall. The firewall permits or denies the traffic to go through the IP management interfaces
according to the policy rules.

Fabric OS Administrator’s Guide
53-1002920-02

253

8

IP Filter policy

Fabric OS supports multiple IP Filter policies to be defined at the same time. Each IP Filter policy is
identified by a name and has an associated type. Two IP Filter policy types, IPv4 and IPv6, exist to
provide separate packet filtering for IPv4 and IPv6. It is not allowed to specify an IPv6 address in
the IPv4 filter, or specify an IPv4 address in the IPv6 filter. There can be up to six different IP Filter
policies defined for both types. Only one IP Filter policy for each IP type can be activated on the
affected management IP interfaces.
Audit messages will be generated for any changes to the IP Filter policies.
The rules in the IP Filter policy are examined one at a time until the end of the list of rules. For
performance reasons, the most commonly used rules should be specified at the top.
On a chassis system, changes to persistent IP Filter policies are automatically synchronized to the
standby CP when the changes are saved persistently on the active CP. The standby CP will enforce
the filter policies to its management interface after policies are synchronized with the active CP.
Virtual Fabrics considerations: Each logical switch cannot have its own different IP Filter policies. IP
Filter policies are treated as a chassis-wide configuration and are common for all the logical
switches in the chassis.

Creating an IP Filter policy
You can create an IP Filter policy specifying any name and using type IPv4 or IPv6. The policy
created is stored in a temporary buffer, and is lost if the current command session logs out. The
policy name is a unique string composed of a maximum of 20 alpha, numeric, and underscore
characters. The names default_ipv4 and default_ipv6 are reserved for default IP filter policies. The
policy name is case-insensitive and always stored as lowercase. The policy type identifies the policy
as an IPv4 or IPv6 filter. There can be a maximum of six IP Filter policies.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having OM permissions for the IPfilter RBAC class of commands.
2. Enter in the ipFilter --create command.

Cloning an IP Filter policy
You can create an IP Filter policy as an exact copy of an existing policy. The policy created is stored
in a temporary buffer and has the same type and rules as the existing defined or active policy.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having OM permissions for the IPfilter RBAC class of commands.
2. Enter the ipFilter --clone command.

Displaying an IP Filter policy
You can display the IP Filter policy content for the specified policy name, or all IP Filter policies if a
policy name is not specified.
For each IP Filter policy, the policy name, type, persistent state and policy rules are displayed. The
policy rules are listed by the rule number in ascending order. There is no pagination stop for
multiple screens of information. Pipe the output to the |more command to achieve this.
If a temporary buffer exists for an IP Filter policy, the --show subcommand displays the content in
the temporary buffer, with the persistent state set to no.

254

Fabric OS Administrator’s Guide
53-1002920-02

IP Filter policy

8

1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having the O permission for the IPfilter RBAC class of commands.
2. Enter the ipFilter –-show command.

Saving an IP Filter policy
You can save one or all IP Filter policies persistently in the defined configuration.
Only the CLI session that owns the updated temporary buffer may run this command. Modification
to an active policy cannot be saved without being applied. Hence, the --save subcommand is
blocked for the active policies. Use --activate instead.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having the OM permissions for the IPfilter RBAC class of commands.
2. Enter the ipFilter –-save command.

Activating an IP Filter policy
IP Filter policies are not enforced until they are activated. Only one IP Filter policy per IPv4 and IPv6
type can be active. If there is a temporary buffer for the policy, the policy is saved to the defined
configuration and activated at the same time. If there is no temporary buffer for the policy, the
policy existing in the defined configuration becomes active. The activated policy continues to
remain in the defined configuration. The policy to be activated replaces the existing active policy of
the same type. Activating the default IP Filter policies returns the IP management interface to its
default state. An IP Filter policy without any rule cannot be activated. This subcommand prompts
for a user confirmation before proceeding.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having OM permissions for the IPfilter RBAC class of commands.
2. Enter the ipFilter –-activate command.

Deleting an IP Filter policy
You can delete a specified IP Filter policy. Deleting an IP Filter policy removes it from the temporary
buffer. To permanently delete the policy from the persistent database, run ipfilter --save. An active
IP Filter policy cannot be deleted.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having the OM permissions for the IPfilter RBAC class of commands.
2. Enter the ipFilter --delete command.
3. To permanently delete the policy, enter the ipfilter --save command.

IP Filter policy rules
An IP Filter policy consists of a set of rules. Each rule has an index number identifying the rule.
There can be a maximum of 256 rules within an IP Filter policy.
Each rule contains the following elements:

• Source Address:

Fabric OS Administrator’s Guide
53-1002920-02

A source IP address or a group prefix.

255

8

IP Filter policy

• Destination Port: The destination port number or name, such as: Telnet, SSH, HTTP, HTTPS.
• Protocol:
The protocol type. Supported types are TCP or UDP.
• Action:
The filtering action taken by this rule, either Permit or Deny.
A traffic type and destination IP can also be specified

Source address
For an IPv4 filter policy, the source address has to be a 32-bit IPv4 address in dot decimal notation.
The group prefix has to be a CIDR block prefix representation. For example, 208.130.32.0/24
represents a 24-bit IPv4 prefix starting from the most significant bit. The special prefix 0.0.0.0/0
matches any IPv4 address. In addition, the keyword any is supported to represent any IPv4
address.
For an IPv6 filter policy, the source address has to be a 128-bit IPv6 address, in a format
acceptable in RFC 3513. The group prefix has to be a CIDR block prefix representation. For
example, 12AB:0:0:CD30::/64 represents a 64-bit IPv6 prefix starting from the most significant bit.
In addition, the keyword any is supported to represent any IPv6 address.

Destination port
For the destination port, a single port number or a port number range can be specified. According
to IANA (http://www.iana.org), ports 0 to 1023 are well-known port numbers, ports 1024 to 49151
are registered port numbers, and ports 49152 to 65535 are dynamic or private port numbers.
Well-known and registered ports are normally used by servers to accept connections, while
dynamic port numbers are used by clients.
For an IP Filter policy rule, you can only select port numbers in the well-known port number range,
between 0 and 1023, inclusive. This means that you have the ability to control how to expose the
management services hosted on a switch, but not the ability to affect the management traffic that
is initiated from a switch. A valid port number range is represented by a dash, for example 7-30.
Alternatively, service names can also be used instead of port number. Table 45 lists the supported
service names and their corresponding port numbers.

TABLE 45

Supported services

Service name

Port number

echo

7

discard

256

systat

11

daytime

13

netstat

15

chargen

19

ftp data

20

ftp

21

fsp

21

ssh

22

telnet

23

smtp

25

Fabric OS Administrator’s Guide
53-1002920-02

IP Filter policy

TABLE 45

8

Supported services (Continued)

Service name

Port number

time

27

name

42

whois

43

domain

53

bootps

67

bootpc

68

tftp

69

http

80

kerberos

88

hostnames

101

sftp

115

ntp

123

snmp

161

snmp trap

162

https

443

ssmtp

465

exec

512

login

513

shell

514

uucp

540

biff

512

who

513

syslog

514

route

520

timed

525

kerberos4

750

Protocol
TCP and UDP protocols are valid protocol selections. Fabric OS v6.2.0 and later do not support
configuration to filter other protocols. Implicitly, ICMP type 0 and type 8 packets are always allowed
to support ICMP echo request and reply on commands like ping and traceroute.

Action
For the action, only “permit” and “deny” are valid.

Fabric OS Administrator’s Guide
53-1002920-02

257

8

IP Filter policy

Traffic type and destination IP
The traffic type and destination IP elements allow an IP policy rule to specify filter enforcement for
IP forwarding. The INPUT traffic type is the default and restricts rules to manage traffic on IP
management interfaces,
The FORWARD traffic type allows management of bidirectional traffic between the external
management interface and the inband management interface. In this case, the destination IP
element should also be specified.

Implicit filter rules
For every IP Filter policy, the two rules listed in Table 46 are always assumed to be appended
implicitly to the end of the policy. This ensures that TCP and UDP traffic to dynamic port ranges is
allowed, so that management IP traffic initiated from a switch, such as syslog, radius and ftp, is not
affected.

TABLE 46

Implicit IP Filter rules

Source address

Destination port

Protocol

Action

Any

1024-65535

TCP

Permit

Any

1024-65535

UDP

Permit

Default policy rules
Switches have a default IP Filter policy for IPv4 and IPv6. The default IP Filter policy cannot be
deleted or changed. When an alternative IP Filter policy is activated, the default IP Filter policy
becomes deactivated. Table 47 lists the rules of the default IP Filter policy.

TABLE 47

Default IP policy rules

Rule number

Source address

Destination port

Protocol

Action

1

Any

22

TCP

Permit

2

Any

23

TCP

Permit

6

Any

80

TCP

Permit

7

Any

443

TCP

Permit

8

Any

161

UDP

Permit

10

Any

123

UDP

Permit

1

Any

600-1023

TCP

Permit

1

Any

600-1023

UDP

Permit

11

12
1.

None of the RPC ports are configurable, even though the action shows “Permit”.

IP Filter policy enforcement
An active IP Filter policy is a filter applied to the IP packets through the management interface. IPv4
management traffic passes through the active IPv4 filter policy, and IPv6 management traffic
passes through the active IPv6 filter policy. The IP Filter policy applies to the incoming (ingress)
management traffic only. When a packet arrives, it is compared against each rule, starting from the

258

Fabric OS Administrator’s Guide
53-1002920-02

IP Filter policy

8

first rule. If a match is found for the source address, destination port, and protocol, the
corresponding action for this rule is taken, and the subsequent rules in this policy are ignored. If
there is no match, then it is compared to the next rule in the policy. This process continues until the
incoming packet is compared to all rules in the active policy.
If none of the rules in the policy matches the incoming packet, the two implicit rules are matched to
the incoming packet. If the rules still do not match the packet, the default action, which is to deny,
is taken.
When the IPv4 or IPv6 address for the management interface of a switch is changed through the
ipAddrSet command or manageability tools, the active IP Filter policies automatically become
enforced on the management IP interface with the changed IP address.

NOTE

If a switch is part of a LAN behind a Network Address Translation (NAT) server, depending on the NAT
server configuration, the source address in an IP Filter rule may have to be the NAT server address.

Adding a rule to an IP Filter policy
There can be a maximum of 256 rules created for an IP Filter policy. The change to the specified IP
Filter policy is not saved to the persistent configuration until a save or activate subcommand is run.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having the OM permissions for the IPfilter RBAC class of commands.
2. Enter the ipFilter --addrule command.

Deleting a rule from an IP Filter policy
Deleting a rule in the specified IP Filter policy causes the rules following the deleted rule to shift up
in rule order. The change to the specified IP Filter policy is not saved to persistent configuration
until a save or activate subcommand is run.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having the OM permissions for the IPfilter RBAC class of commands.
2. Enter the ipFilter --delrule command.

Aborting an IP Filter transaction
A transaction is associated with a command line or manageability session. It is opened implicitly
when the --create, --addrule, --delrule, --clone, and --delete subcommands are run. The
--transabort, --save, or --activate subcommands explicitly end the transaction owned by the
current command line or manageability session. If a transaction is not ended, other command line
or manageability sessions are blocked on the subcommands that would open a new transaction.
1. Log in to the switch using an account with admin permissions, or an account associated with
the chassis role and having the OM permissions for the IPfilter RBAC class of commands.
2. Enter the ipFilter –-transabort command.

Fabric OS Administrator’s Guide
53-1002920-02

259

8

Policy database distribution

IP Filter policy distribution
The IP Filter policy is manually distributed by command. The distribution includes both active and
defined IP Filter policies. All policies are combined as a single entity to be distributed and cannot be
selectively distributed. However, you may choose the time at which to implement the policy for
optimization purposes. If a distribution includes an active IP Filter policy, the receiving switches
activate the same IP Filter policy automatically. When a switch receives IP Filter policies, all
uncommitted changes left in its local transaction buffer are lost, and the transaction is aborted.
The IP Filter policy can be manually distributed to the fabric by command; there is no support for
automatic distribution. To distribute the IPFilter policy, see “Distributing the local ACL policies” on
page 263 for instructions.
You can accept or deny IP Filter policy distribution through the commands fddCfg --localaccept or
fddCfg --localreject. See “Policy database distribution” on page 260 for more information on
distributing the IP Filter policy.

NOTE
Any RPC ports that were allowed in Fabric OS versions earlier than 7.2.0 are removed and ignored
in Fabric OS 7.2.0 and later.
Virtual Fabrics considerations: To distribute the IPFilter policy in a logical fabric, use the
chassisDistribute command.

Policy database distribution
Fabric OS lets you manage and enforce the ACL policy database on either a per-switch or
fabric-wide basis. The local switch distribution setting and the fabric-wide consistency policy affect
the switch ACL policy database and related distribution behavior.
The ACL policy database is managed as follows:

• Switch database distribution setting — Controls whether or not the switch accepts or rejects
databases distributed from other switches in the fabric. The distribute command sends the
database from one switch to another, overwriting the target switch database with the
distributed one. To send or receive a database the setting must be accept. For configuration
instructions, see “Database distribution settings” on page 261.
Virtual Fabric considerations: FCS, DCC, SCC, and AUTH databases can be distributed using
the -distribute command, but the PWD and IPFILTER databases are blocked from distribution.

• Manually distribute an ACL policy database — Run the distribute command to push the local
database of the specified policy type to target switches. “ACL policy distribution to other
switches” on page 262.

• Fabric-wide consistency policy — Use to ensure that switches in the fabric enforce the same
policies. Set a strict or tolerant fabric-wide consistency policy for each ACL policy type to
automatically distribute that database when a policy change is activated. If a fabric-wide
consistency policy is not set, then the policies are managed on a per switch basis. For
configuration instructions, see “Fabric-wide enforcement” on page 263.
Virtual Fabric considerations: Fabric-wide consistency policies are configured on a per logical
switch-basis and are applied to the fabrics connected to the logical switches. Automatic policy
distribution behavior for DCC, SCC and FCS is the same as that of pre-v6.2.0 releases and are
configured on a per logical switch basis.

260

Fabric OS Administrator’s Guide
53-1002920-02

Policy database distribution

8

Table 48 on page 261 explains how the local database distribution settings and the fabric-wide
consistency policy affect the local database when the switch is the target of a distribution
command.

TABLE 48

Interaction between fabric-wide consistency policy and distribution settings

Distribution
setting

Fabric-wide consistency policy
Absent (default)

Tolerant

Strict

Reject

Database is protected, it
cannot be overwritten.
May not match other
databases in the fabric.

Invalid configuration.1

Invalid configuration.1

Accept (default)

Database is not protected,
the database can be
overwritten.
If the switch initiating a
distribute command has a
strict or tolerant fabric-wide
consistency policy, the
fabric-wide policy is also
overwritten.
May not match other
databases in the fabric.

Database is not protected.
Automatically distributes
activated changes to other
v6.2.0 or later switches in the
fabric.
May not match other
databases in the fabric.

Database is not protected.
Automatically distributes
activated changes to all
switches in the fabric.
Fabric can only contain
switches running Fabric OS
v6.2.0 or later.
Active database is the same for
all switches in the fabric.

1. An error is returned indicating that the distribution setting must be accept before you can set the fabric-wide
consistency policy.

Database distribution settings
The distribution settings control whether a switch accepts or rejects distributions of databases
from other switches and whether the switch may initiate a distribution. Configure the distribution
setting to reject when maintaining the database on a per-switch basis.
Table 49 lists the databases supported in Fabric OS v6.2.0 and later switches.

TABLE 49

Supported policy databases

Database type

Database identifier (ID)

Authentication policy database

AUTH

DCC policy database

DCC

FCS policy database

FCS

IP Filter policy database

IPFILTER

Password database

PWD

SCC policy database

SCC

Use the chassisDistribute command to distribute IP filter policies. To distribute other security
policies, use the distribute command.

Fabric OS Administrator’s Guide
53-1002920-02

261

8

Policy database distribution

Displaying the database distribution settings
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the FabricDistribution RBAC class of commands.
2. Enter the fddCfg --showall command.
Example shows the database distribution settings
switch:admin> fddcfg --showall
Local Switch Configuration for all Databases:DATABASE - Accept/Reject
--------------------------------SCC accept
DCC accept
PWD accept
FCS accept
AUTH accept
IPFILTER accept
Fabric Wide Consistency Policy:- ""

Enabling local switch protection
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the FabricDistribution RBAC class of commands.
2. Enter the fddCfg --localreject command.

Disabling local switch protection
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the FabricDistribution RBAC class of commands.
2. Enter the fddCfg --localaccept command.

ACL policy distribution to other switches
This section explains how to manually distribute local ACL policy databases. The distribute
command has the following dependencies:

• All target switches must be running Fabric OS v6.2.0 or later.
• All target switches must accept the database distribution (see “Database distribution settings”
on page 261).

• The fabric must have a tolerant or no (absent) fabric-wide consistency policy (see “Fabric-wide
enforcement” on page 263).
If the fabric-wide consistency policy for a database is strict, the database cannot be manually
distributed. When you set a strict fabric-wide consistency policy for a database, the distribution
mechanism is automatically invoked whenever the database changes.

• The local distribution setting must be accepted. To be able to initiate the distribute command,
set the local distribution to accept.

262

Fabric OS Administrator’s Guide
53-1002920-02

Policy database distribution

8

Distributing the local ACL policies
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the FabricDistribution RBAC class of commands.
2. Enter the distribute -p command.

Fabric-wide enforcement
The fabric-wide consistency policy enforcement setting determines the distribution behavior when
changes to a policy are activated. Using the tolerant or strict fabric-wide consistency policy ensures
that changes to local ACL policy databases are automatically distributed to other switches in the
fabric.

NOTE
To completely remove all fabric-wide policy enforcement from a fabric enter the fddCfg --fabwideset
"” command.
When you set the fabric-wide consistency policy using the fddCfg command with the
--fabwideset database_id option, both the fabric-wide consistency policy and specified database
are distributed to the fabric.The active policies of the specified databases overwrite the
corresponding active and defined policies on the target switches.
Policy changes that are saved but not activated are stored locally until a policy database change is
activated. Activating a policy automatically distributes the Active policy set for that policy type (SCC,
DCC, FCS, or any combination of the three) to the other switches in the fabric.

NOTE

FC routers cannot join a fabric with a strict fabric-wide consistency policy. FC routers do not support
the fabric-wide consistency policies.
Table 50 describes the fabric-wide consistency settings.

TABLE 50

Fabric-wide consistency policy settings

Setting

Value

When a policy is activated

Absent

null

Database is not automatically distributed to other switches in the fabric.

Tolerant

database_id

All updated and new policies of the type specified (SCC, DCC, FCS, or any
combination) are distributed to all Fabric v6.2.0 and later switches in the fabric.

Strict

database_id:S

All updated and new policies of the type specified (SCC, DCC, FCS, or any
combination) are distributed to all switches in the fabric.

Displaying the fabric-wide consistency policy
1. Connect to the switch and log in using an account with admin permissions, or an account with
O permission for the FabricDistribution RBAC class of commands.
2. Enter the fddCfg --showall command.
Example shows policies for a fabric where no consistency policy is defined.
switch:admin> fddcfg --showall
Local Switch Configuration for all Databases:DATABASE - Accept/Reject
---------------------------------

Fabric OS Administrator’s Guide
53-1002920-02

263

8

Policy database distribution

SCC
DCC
PWD
FCS
AUTH
IPFILTER

-

accept
accept
accept
accept
accept
accept

Fabric Wide Consistency Policy:- ""

Setting the fabric-wide consistency policy
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the FabricDistribution RBAC class of commands.
2. Enter the fddCfg --fabwideset command.
Example shows how to set a strict SCC and tolerant DCC fabric-wide consistency policy.
switch:admin> fddcfg --fabwideset "SCC:S;DCC"
switch:admin> fddcfg --showall
Local Switch Configuration for all Databases:DATABASE - Accept/Reject
--------------------------------SCC accept
DCC accept
PWD accept
FCS accept
AUTH accept
IPFILTER accept
Fabric Wide Consistency Policy:- "SCC:S;DCC"

Notes on joining a switch to the fabric
When a switch is joined to a fabric with a tolerant SCC, DCC, or FCS fabric-wide consistency policy,
the joining switch must have a matching tolerant SCC, DCC, or FCS fabric-wide consistency policy. If
the tolerant SCC, DCC, or FCS fabric-wide consistency policies do not match, the switch can join the
fabric, but an error message flags the mismatch. If the tolerant SCC, DCC, and FCS fabric-wide
consistency policies match, the corresponding SCC, DCC, and FCS ACL policies are compared.
The enforcement of fabric-wide consistency policy involves comparison of the Active policy set. If
the ACL polices match, the switch joins the fabric successfully. If the ACL policies are absent either
on the switch or on the fabric, the switch joins the fabric successfully, and the ACL policies are
copied automatically from where they are present to where they are absent. The Active policy set
where it is present overwrites the Active and Defined policy set where it is absent. If the ACL
policies do not match, the switch cannot join the fabric and the neighboring E_Ports are disabled.
Use the fddCfg –-fabwideset command on either this switch or the fabric to set a matching strict
SCC, DCC, or FCS fabric-wide consistency policy. Use ACL policy commands to delete the conflicting
ACL policy from one side to resolve ACL policy conflict. If neither the fabric nor the joining switch is
configured with a fabric-wide consistency policy, there are no ACL merge checks required. Under
both conflicting conditions, secPolicyActivate is blocked in the merged fabric. Use the distribute
command to explicitly resolve conflicting ACL policies.
The descriptions above also apply to joining two fabrics. In this context, the joining switch becomes
a joining fabric.

264

Fabric OS Administrator’s Guide
53-1002920-02

Policy database distribution

8

Matching fabric-wide consistency policies
This section describes the interaction between the databases with active SCC and DCC policies
and combinations of fabric-wide consistency policy settings when fabrics are merged.
For example: Fabric A with SCC:S;DCC (strict SCC and tolerant DCC) joins Fabric B with SCC:S;DCC
(strict SCC and tolerant DCC), the fabrics can merge as long as the SCC policies match, including
the order SCC:S;DCC and if both are set to strict.
Table 51 describes the impact of merging fabrics with the same fabric-wide consistency policy that
have SCC, DCC, or both policies.

TABLE 51

Merging fabrics with matching fabric-wide consistency policies

Fabric-wide
Fabric A
consistency policy ACL policies

Fabric B
ACL policies

Merge
results

Database copied

None

None

None

Succeeds

No ACL policies copied.

None

SCC/DCC

Succeeds

No ACL policies copied.

None

None

Succeeds

No ACL policies copied.

None

SCC/DCC

Succeeds

ACL policies are copied from B to A.

SCC/DCC

SCC/DCC

Succeeds

If A and B policies do not match, a
warning displays and policy
commands are disabled1.

None

None

Succeeds

No ACL policies copied.

None

SCC/DCC

Succeeds

ACL policies are copied from B to A.

Matching SCC/DCC

Matching SCC/DCC

Succeeds

No ACL policies copied.

Different SCC/DCC
policies

Different SCC/DCC
policies

Fails

Ports are disabled.

Tolerant

Strict

1. To resolve the policy conflict, manually distribute the database you want to use to the switch with the mismatched
database. Until the conflict is resolved, commands such as fddCfg --fabwideset and secPolicyActivate are
blocked.

Non-matching fabric-wide consistency policies
You may encounter one of the following two scenarios described in Table 52 and Table 53 where
you are merging a fabric with a strict policy to a fabric with an absent, tolerant, or non-matching
strict policy and the merge fails and the ports are disabled.
Table 52 on page 266 shows merges that are not supported.

Fabric OS Administrator’s Guide
53-1002920-02

265

8

Management interface security

TABLE 52

Examples of strict fabric merges

Fabric-wide consistency policy setting

Strict/Tolerant

Strict/Absent

Expected behavior

Fabric A

Fabric B

SCC:S;DCC:S

SCC;DCC:S

SCC;DCC:S

SCC:S;DCC

SCC:S;DCC

SCC:S

Ports connecting switches are disabled.

SCC:S;DCC:S
SCC:S
DCC:S

Strict/Strict

SCC:S

DCC:S

Table 53 has a matrix of merging fabrics with tolerant and absent policies.

TABLE 53

Fabric merges with tolerant and absent combinations

Fabric-wide consistency policy setting
Fabric A
Tolerant/Absent

Expected behavior
Fabric B

SCC;DCC
DCC
SCC;DCC

SCC

DCC

SCC

Error message logged.
Run fddCfg --fabwideset “policy_ID” from any
switch with the desired configuration to fix the
conflict. The secPolicyActivate command is blocked
until conflict is resolved.

Management interface security
You can secure an Ethernet management interface between two Brocade switches or Backbones
by implementing IPsec and IKE policies to create a tunnel that protects traffic flows. While the
tunnel must have a Brocade switch or Backbone at each end, there may be routers, gateways, and
firewalls in between the two ends.

ATTENTION
Enabling secure IPsec tunnels does not provide IPsec protection for traffic flows on the external
management interfaces of intelligent blades in a chassis, nor does it support protection of traffic
flows on FCIP interfaces.
Internet Protocol security (IPsec) is a framework of open standards that ensures private and secure
communications over Internet Protocol (IP) networks through the use of cryptographic security
services. The goal of IPsec is to provide the following capabilities:

• Authentication — Ensures that the sending and receiving end-users and devices are known and
trusted by one another.

• Data Integrity — Confirms that the data received was in fact the data transmitted.
• Data Confidentiality — Protects the user data being transmitted, such as utilizing encryption to
avoid sending data in clear text.

• Replay Protection — Prevents replay attack in which an attacker resends previously-intercepted
packets in an effort to fraudulently authenticate or otherwise masquerade as a valid user.

266

Fabric OS Administrator’s Guide
53-1002920-02

Management interface security

8

• Automated Key Management—Automates the process, as well as manages the periodic
exchange and generation of new keys.
Using the ipSecConfig command, you must configure multiple security policies for traffic flows on
the Ethernet management interfaces based on IPv4 or IPv6 addresses, a range of IPv4 or IPv6
addresses, the type of application, port numbers, and protocols used (UDP/TCP/ICMP). You must
specify the transforms and processing choices for the traffic flow (drop, protect or bypass). Also,
you must select and configure the key management protocol using an automatic or manual key.
For more information on IPv4 and IPv6 addressing, refer to Chapter 2, “Performing Basic
Configuration Tasks”.

Configuration examples
Below are several examples of various configurations you can use to implement an IPsec tunnel
between two devices. You can configure other scenarios as nested combinations of these
configurations.

Endpoint-to-endpoint transport or tunnel
In this scenario, both endpoints of the IP connection implement IPsec, as required of hosts in
RFC4301. Transport mode encrypts only the payload while tunnel mode encrypts the entire packet.
A single pair of addresses will be negotiated for packets protected by this SA.
It is possible in this scenario that one or both of the protected endpoints will be behind a network
address translation (NAT) node, in which case tunneled packets will have to be UDP-encapsulated
so that port numbers in the UDP headers can be used to identify individual endpoints behind the
NAT.

FIGURE 18

Protected endpoints configuration

A possible drawback of end-to-end security is that various applications that require the ability to
inspect or modify a transient packet will fail when end-to-end confidentiality is employed. Various
QoS solutions, traffic shaping, and firewalling applications will be unable to determine what type of
packet is being transmitted and will be unable to make the decisions that they are supposed to
make.

Fabric OS Administrator’s Guide
53-1002920-02

267

8

Management interface security

Gateway-to-gateway tunnel
In this scenario, neither endpoint of the IP connection implements IPsec, but the network nodes
between them protect traffic for part of the way. Protection is transparent to the endpoints, and
depends on ordinary routing to send packets through the tunnel endpoints for processing. Each
endpoint would announce the set of addresses behind it, and packets would be sent in tunnel
mode where the inner IP header would contain the IP addresses of the actual endpoints.

FIGURE 19

Gateway tunnel configuration

Endpoint-to-gateway tunnel
In this scenario, a protected endpoint (typically a portable computer) connects back to its corporate
network through an IPsec-protected tunnel. It might use this tunnel only to access information on
the corporate network, or it might tunnel all of its traffic back through the corporate network in
order to take advantage of protection provided by a corporate firewall against Internet-based
attacks. In either case, the protected endpoint will want an IP address associated with the security
gateway so that packets returned to it will go to the security gateway and be tunneled back.

FIGURE 20

Endpoint-to-gateway tunnel configuration

RoadWarrior configuration
In endpoint-to-endpoint security, packets are encrypted and decrypted by the host which produces
or consumes the traffic. In the gateway-to-gateway example, a router on the network encrypts and
decrypts the packets on behalf of the hosts on a protected network. A combination of the two is
referred to as a RoadWarrior configuration where a host on the Internet requires access to a
network through a security gateway that is protecting the network.

268

Fabric OS Administrator’s Guide
53-1002920-02

Management interface security

8

IPsec protocols
IPsec ensures confidentiality, integrity, and authentication using the following protocols:

•
•

Authentication Header (AH)
Encapsulating Security Payload (ESP)

IPsec protocols protect IP datagram integrity using hash message authentication codes (HMAC).
Using hash algorithms with the contents of the IP datagram and a secret key, the IPsec protocols
generate this HMAC and add it to the protocol header. The receiver must have access to the secret
key in order to decode the hash.
IPsec protocols use a sliding window to assist in flow control, The IPsec protocols also use this
sliding window to provide protection against replay attacks in which an attacker attempts a denial
of service attack by replaying an old sequence of packets. IPsec protocols assign a sequence
number to each packet. The recipient accepts each packet only if its sequence number is within
the window. It discards older packets.

Security associations
A security association (SA) is the collection of security parameters and authenticated keys that are
negotiated between IPsec peers to protect the IP datagram. A security association database (SADB)
is used to store these SAs. Information in these SAs—IP addresses, secret keys, algorithms, and so
on—is used by peers to encapsulate and decapsulate the IPsec packets
An IPsec security association is a construct that specifies security properties that are recognized by
communicating hosts. The properties of the SA are the security protocol (AH or ESP), destination IP
address, and Security Parameter Index (SPI) number. SPI is an arbitrary 32-bit value contained in
IPsec protocol headers (AH or ESP) and an IPsec SA is unidirectional. Because most
communication is peer-to-peer or client-to-server, two SAs must be present to secure traffic in both
directions. An SA specifies the IPsec protocol (AH or ESP), the algorithms used for encryption and
authentication, and the expiration definitions used in security associations of the traffic. IKE uses
these values in negotiations to create IPsec SAs. You must create an SA prior to creating an
SA-proposal. You cannot modify an SA once it is created. Use the ipSecConfig --flush manual-sa
command to remove all SA entries from the kernel SADB and re-create the SA. For more
information on the ipSecConfig command, refer to the Fabric OS Command Reference.

IPsec proposal
The IPsec sa-proposal defines an SA or an SA bundle. An SA is a set of parameters that define how
the traffic is protected using IPsec. These are the IPsec protocols to use for an SA, either AH or ESP,
and the encryption and authentication algorithms to use to protect the traffic. For SA bundles,
[AH, ESP] is the supported combination.

Authentication and encryption algorithms
IPsec uses different protocols to ensure the authentication, integrity, and confidentiality of the
communication. Encapsulating Security Payload (ESP) provides confidentiality, data integrity and
data source authentication of IP packets, and protection against replay attacks. Authentication
Header (AH) provides data integrity, data source authentication, and protection against replay
attacks, but unlike ESP, AH does not provide confidentiality.

Fabric OS Administrator’s Guide
53-1002920-02

269

8

Management interface security

In AH and ESP, hmac_md5 and hmac_sha1 are used as authentication algorithms. Only in ESP,
3des_cbc, blowfish_cbc, aes256_cbc and null_enc are used as encryption algorithms. Use
Table 54 when configuring the authentication algorithm.

TABLE 54

Algorithms and associated authentication policies

Algorithm

Encryption Level Policy

Description

hmac_md5

128-bit

AH, ESP

hmac_sha1

160-bit

AH, ESP

A stronger MAC because it is a keyed hash inside a keyed hash. When
MD5 or SHA-1 is used in the calculation of an HMAC; the resulting MAC
algorithm is termed HMAC-MD5 or HMAC-SHA-1 accordingly.
NOTE: The MD5 hash algorithm is blocked when FIPS mode is
enabled

3des_cbc

168-bit

ESP

Triple DES is a more secure variant of DES. It uses three different
56-bit keys to encrypt blocks of 64-bit plain text. The algorithm is
FIPS-approved for use by Federal agencies.

blowfish_cbc

64-bit

ESP

Blowfish is a 32-bit to 448-bit keyed, symmetric block cipher.

aes128_cbc

128-bit

ESP

aes256_cbc

256-bit

ESP

Advanced Encryption Standard is a 128- or 256-bit fixed block size
cipher.

null_enc

n/a

ESP

A form of plaintext encryption.

IPsec policies
An IPsec policy determines the security services afforded to a packet and the treatment of a packet
in the network. An IPsec policy allows classifying IP packets into different traffic flows and specifies
the actions or transformations performed on IP packets on each of the traffic flows. The main
components of an IPsec policy are: IP packet filter and selector (IP address, protocol, and port
information) and transform set.

IPsec traffic selector
The traffic selector is a traffic filter that defines and identifies the traffic flow between two systems
that have IPsec protection. IP addresses, the direction of traffic flow (inbound, outbound) and the
upper layer protocol are used to define a filter for traffic (IP datagrams) that is protected using
IPsec.

IPsec transform
A transform set is a combination of IPsec protocols and cryptographic algorithms that are applied
on the packet after it is matched to a selector. The transform set specifies the IPsec protocol, IPsec
mode and action to be performed on the IP packet. It specifies the key management policy that is
needed for the IPsec connection and the encryption and authentication algorithms to be used in
security associations when IKE is used as the key management protocol.
IPsec can protect either the entire IP datagram or only the upper-layer protocols using tunnel mode
or transport mode. Tunnel mode uses the IPsec protocol to encapsulate the entire IP datagram.
Transport mode handles only the IP datagram payload.

270

Fabric OS Administrator’s Guide
53-1002920-02

Management interface security

8

IKE policies
When IKE is used as the key management protocol, IKE policy defines the parameters used in IKE
negotiations needed to establish IKE SA and parameters used in negotiations to establish IPsec
SAs. These include the authentication and encryption algorithms, and the primary authentication
method, such as preshared keys, or a certificate-based method, such as RSA signatures.

Key management
The IPsec key management supports Internet Key Exchange or Manual key/SA entry. The Internet
Key Exchange (IKE) protocol handles key management automatically. SAs require keying material
for authentication and encryption. The managing of keying material that SAs require is called key
management.
The IKE protocol secures communication by authenticating peers and exchanging keys. It also
creates the SAs and stores them in the SADB.
The manual key/SA entry requires the keys to be generated and managed manually. For the
selected authentication or encryption algorithms, the correct keys must be generated using a third
party utility on your LINUX system. The key length is determined by the algorithm selected.
Linux IPsec-tools 0.7 provides tools for manual key entry (MKE) and automatic keyed connections.
The LINUX setKey command can be used for manually keyed connections, which means that all
parameters needed for the setup of the connection are provided by you. Based on which protocol,
algorithm, and key used for the creation of the security associations, the switch populates the
security association database (SAD) accordingly.

Pre-shared keys
A pre-shared key has the .psk extension and is one of the available methods IKE can be configured
to use for primary authentication. You can specify the pre-shared keys used in IKE policies; add and
delete pre-shared keys (in local database) corresponding to the identity of the IKE peer or group of
peers.
The ipSecConfig command does not support manipulating pre-shared keys corresponding to the
identity of the IKE peer or group of peers. Use the secCertUtil command to import, delete, or display
the pre-shared keys in the local switch database. For more information on this procedure, refer to
Chapter 7, “Configuring Protocols”.

Security certificates
A certificate is one of the available methods IKE can be configured to use for primary
authentication. You can specify the local public key and private key (in X.509 PEM format) and peer
public key (in X.509 format) to be used in a particular IKE policy.
Use the secCertUtil import command to import public key, private key and peer-public key (in X.509
PEM format) into the switch database. For more information on this procedure, refer to Chapter 7,
“Configuring Protocols”.

ATTENTION
The CA certificate name must have the IPSECCA.pem name.

Fabric OS Administrator’s Guide
53-1002920-02

271

8

Management interface security

Static Security Associations
Manual Key Entry (MKE) provides the ability to manually add, delete and flush SA entries in the
SADB. Manual SA entries may not have an associated IPsec policy in the local policy database.
Manual SA entries are persistent across system reboots.

Creating the tunnel
Each side of the tunnel must be configured in order for the tunnel to come up. Once you are logged
into the switch, do not log off as each step requires that you be logged in to the switch. IPsec
configuration changes take effect upon execution and are persistent across reboots. Configure the
following on each side of the tunnel:
1. Determine the authentication protocol and algorithm to be used on the tunnel.
Refer to Table 54 on page 270 to determine which algorithm to use in conjunction with a
specific authentication protocol.
2. Determine the type of keys to be used on the tunnel.
If you are using CA signed keys, you must generate them prior to setting up your tunnels.
3. Enable IPsec.
a.

Connect to the switch and log in using an account with admin permissions, or an account
associated with the chassis role and having OM permissions for the IPsec RBAC class of
commands.

b.

Enter the ipSecConfig --enable command to enable IPsec on the switch.

4. Create an IPsec SA policy on each side of the tunnel using the ipSecConfig --add command.
Example of creating an IPsec SA policy

This example creates an IPsec SA policy named AH01, which uses AH protection with MD5. You
would run this command on each switch; on each side of the tunnel so that both sides have
the same IPsec SA policy.
switch:admin> ipsecconfig --add policy ips sa -t AH01 -p ah -auth hmac_md5

5. Create an IPsec proposal on each side of the tunnel using the ipSecConfig --add command.
Example of creating an IPsec proposal

This example creates an IPsec proposal IPSEC-AH to use AH01 as SA.
switch:admin> ipsecconfig --add policy ips sa-proposal -t IPSEC-AH –sa AH01

6. Import the pre-shared key file.
Refer to Chapter 7, “Configuring Protocols” for information on how to set up pre-shared keys
and certificates.
7.

Configure the IKE policy using the ipSecConfig --add command.
Example of creating an IKE policy

This example creates an IKE policy for the remote peer.
switch:admin> ipsecconfig --add policy ike –t IKE01 -remote 10.33.74.13
-id 10.33.69.132 -remoteid 10.33.74.13 -enc 3des_cbc -hash hmac_md5
-prf hmac_md5 –auth psk -dh modp1024 -psk ipseckey.psk

272

Fabric OS Administrator’s Guide
53-1002920-02

Management interface security

8

8. Create an IPsec transform on each switch using the ipSecConfig --add command.
Example of creating an IPsec transform

This example creates an IPsec transform TRANSFORM01 to use the transport mode to protect
traffic identified for IPsec protection and use IKE01 as key management policy.
switch:admin> ipsecconfig --add policy ips transform –t TRANSFORM01
-mode transport -sa-proposal IPSEC-AH -action protect –ike IKE01

9. Create a traffic selector on each switch using the ipSecConfig --add command.
Example of creating a traffic selector

This example creates a traffic selector to select outbound and inbound traffic that needs to be
protected.
switch:admin> ipsecconfig --add policy ips selector –t SELECTOR-OUT -d out
-l 10.33.69.132 -r 10.33.74.13 –transform TRANSFORM01
switch:admin> ipsecconfig --add policy ips selector –t SELECTOR-IN -d in
-l 10.33.74.13 -r 10.33.69.132 –t transform TRANSFORM01

Inbound and outbound selectors use opposite values for local and remote IP addresses. In this
example, notice that the local ("-l") address of SELECTOR-OUT is the same as the remote ("-r")
address or SELECTOR-IN, Similarly, the local ("-l") address of SELECTOR-IN is the same as the
remote ("-r") address or SELECTOR-OUT. That is, “local” refers to the source IP address of the
packet, and “remote” is the destination IP address. Hence inbound packets have opposite
source and destination addresses than outbound packets.
10. Verify traffic is protected.
a.

Initiate a telnet, SSH, or ping session from the two switches.

b.

Verify that IP traffic is encapsulated.

c.

Monitor IPsec SAs created using IKE for above traffic flow

• Use the ipSecConfig -–show manual-sa –a command with the operands specified to
display the outbound and inbound SAs in kernel SADB.

• Use the ipSecConfig –-show policy ips sa -a command with the specified operands to
display all IPsec SA policies.

• Use the ipSecConfig –-show policy ips sa-proposal –a command with the specified
operands to display IPsec proposals.

• Use the ipSecConfig –-show policy ips transform –a command with the specified
operands to display IPsec transforms.

• Use the ipSecConfig –-show policy ips selector –a command with the specified
operands to display IPsec traffic selectors.

• Use the ipSecConfig –-show policy ike –a command with the specified operands to
display IKE policies.

• Use the ipSecConfig –-flush manual-sa command with the specified operands to flush
the created SAs in the kernel SADB.

Fabric OS Administrator’s Guide
53-1002920-02

273

8

Management interface security

Example of an end-to-end transport tunnel mode
This example illustrates securing traffic between two systems using AH protection with MD5 and
configure IKE with pre-shared keys. The two systems are a switch, BROCADE300 (IPv4 address
10.33.74.13), and an external host (10.33.69.132).
1. On the system console, log in to the switch as Admin.
2. Enable IPsec.
a.

Connect to the switch and log in using an account with admin permissions, or an account
with OM permissions for the IPsec RBAC class of commands.

b.

Enter the ipSecConfig --enable command to enable IPsec on the switch.

3. Create an IPsec SA policy named AH01, which uses AH protection with MD5.
switch:admin> ipsecconfig --add policy ips sa -t AH01 -p ah -auth hmac_md5

4. Create an IPsec proposal IPSEC-AH to use AH01 as SA.
switch:admin> ipsecconfig --add policy ips sa-proposal -t IPSEC-AH -sa AH01

5. Configure the SA proposal's lifetime in time units. The maximum lifetime is 86400, or one day.
switch:admin> ipsecconfig --add policy ips sa-proposal -t IPSEC-AH
-lttime 86400 -sa AH01

6. Import the pre-shared key file using the secCertUtil command. The file name should have a
.psk extension.
For more information on importing the pre-shared key file, refer to “Installing a switch
certificate” on page 203.
7.

Configure an IKE policy for the remote peer.
switch:admin> ipsecconfig --add policy ike -t IKE01 -remote 10.33.69.132
-id 10.33.74.13 -remoteid 10.33.69.132 -enc 3des_cbc -hash hmac_md5
-prf hmac_md5 -auth psk -dh modp1024 -psk ipseckey.psk

NOTE
IKE version (‘-v’ option) needs to be set to 1 (IKEv1) if remote peer is a Windows XP or 2000 Host as
Windows XP and 2000 do not support IKEv2.
8. Create an IPsec transform named TRANSFORM01 to use transport mode to protect traffic
identified for IPsec protection and use IKE01 as key management policy.
switch:admin> ipsecconfig --add policy ips transform -t TRANSFORM01
-mode transport -sa-proposal IPSEC-AH -action protect -ike IKE01

9. Create traffic selectors to select the outbound and inbound traffic that needs to be protected.
switch:admin> ipsecconfig --add policy ips selector -t SELECTOR-OUT -d out
-l 10.33.74.13 -r 10.33.69.132 -transform TRANSFORM01
switch:admin> ipsecconfig --add policy ips selector -t SELECTOR-IN -d in
-l 10.33.69.132 -r 10.33.74.13 -transform TRANSFORM01

10. Verify the IPsec SAs created with IKE using the ipSecConfig --show manual-sa –a command.

274

Fabric OS Administrator’s Guide
53-1002920-02

Management interface security

8

11. Perform the equivalent steps on the remote peer to complete the IPsec configuration. Refer to
your server administration guide for instructions.
12. Generate IP traffic and verify that it is protected using defined policies.
a.

Initiate Telnet or SSH or ping session from BRCD300 to Remote Host.

b.

Verify that the IP traffic is encapsulated.

c.

Monitor IPsec SAs created using IKE for the above traffic flow.

• Use the ipSecConfig -–show manual-sa –a command with the operands specified to
display the outbound and inbound SAs in the kernel SADB.

• Use the ipSecConfig –-show policy ips sa -a command with the specified operands to
display all IPsec SA policies.

• Use the ipSecConfig –-show policy ips sa-proposal –a command with the specified
operands to display IPsec proposals.

• Use the ipSecConfig –-show policy ips transform –a command with the specified
operands to display IPsec transforms.

• Use the ipSecConfig –-show policy ips selector –a command with the specified
operands to display IPsec traffic selectors.

• Use the ipSecConfig –-show policy ike –a command with the specified operands to
display IKE policies.

• Use the ipSecConfig –-flush manual-sa command with the specified operands to flush
the created SAs in the kernel SADB.

CAUTION
Flushing SAs requires IPsec to be disabled and re-enabled. This operation is disruptive to traffic
using the tunnel.

Notes
• As of Fabric OS 7.0.0, IPsec no longer supports null encryption (null_enc) for IKE policies.
• IPv6 policies cannot tunnel IMCP traffic.

Fabric OS Administrator’s Guide
53-1002920-02

275

8

276

Management interface security

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

9

Maintaining the Switch Configuration File

In this chapter
• Configuration settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Configuration file backup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Configuration file restoration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Configurations across a fabric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Configuration management for Virtual Fabrics . . . . . . . . . . . . . . . . . . . . . .
• Brocade configuration form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

277
279
280
284
285
287

Configuration settings
It is important to maintain consistent configuration settings on all switches in the same fabric
because inconsistent parameters, such as inconsistent PID formats, can cause fabric
segmentation. As part of standard configuration maintenance procedures, Brocade recommends
that you back up all important configuration data for every switch on a host computer server as a
safety measure.

NOTE
For information about AD-enabled switches, refer to Chapter 20, “Managing Administrative
Domains”.
For more information about troubleshooting configuration file uploads and downloads, refer to the
Fabric OS Troubleshooting and Diagnostics Guide.
There are two ways to view configuration settings for a switch in a Brocade fabric:

• Issue the configShow -all command.
To display configuration settings, connect to the switch, log in as admin, and enter the
configShow -all command. The configuration settings vary depending on switch model and
configuration. This command does not show as much configuration information as the text file
created from the configUpload command.

• Issue the configUpload -all command to upload an ASCII text file from the switch or switch
module.
You can open the text file with any text editor to view the configuration information of the
switch.

CAUTION
Editing of the uploaded file is unsupported and can result in system errors if an edited file is
subsequently downloaded.

Fabric OS Administrator’s Guide
53-1002920-02

277

9

Configuration settings

If your user account has chassis account permissions, you can use any of the following options
when uploading or downloading a configuration file:
-fid

To upload the specified FID configuration.

-all

To upload all of the system configuration, including the chassis section
and all switch sections for all logical switches.
NOTE: Use this parameter when obtaining a complete capture of the switch
configuration in a switch that has Virtual Fabrics mode disabled.

-chassis

To upload only the chassis section of the system configuration file.

-switch

To upload the switch configuration only, if Virtual Fabrics mode is disabled.

Configuration file format
The configuration file is divided into three areas: the header, the chassis section, and one or more
logical-switch sections.

Chassis section
There is only one chassis section within a configuration. It defines configuration data for chassis
components that affect the entire system, not just one individual logical switch. The chassis
section is included in non-Virtual Fabric modes only if you use the configUpload -all command.
The chassis section specifies characteristics for the following software components:

•
•
•
•
•
•
•
•
•
•
•
•
•
•

FC Routing – Fibre Channel Routing
Chassis configuration – Chassis configuration
FCOE_CH_CONF – FCoE chassis configuration
UDROLE_CONF – User-defined role configuration
LicensesDB – License Database (slot-based)
DMM_WWN – Data migration manager World Wide Name configuration
Licenses – (Feature-based) Licenses configuration
AGWWN_MAPPING_CONF – Access Gateway WWN mapping configuration
LicensesLservc – Sentinel License configuration
GE blade mode – GigE Mode configuration
FWD CHASSIS CFG – Fabric Watch configuration
FRAME LOG – Frame log configuration (enable/disable)
DMM_TB – Data migration manager configuration
MOTD – Message of the day

Switch section
There is always at least one switch section for the default switch or a switch that has Virtual Fabrics
mode disabled, and there are additional sections corresponding to each additionally defined
logical switch instance on a switch with Virtual Fabrics mode enabled. This data is switch-specific
and affects only that logical switch behavior.

278

Fabric OS Administrator’s Guide
53-1002920-02

Configuration file backup

9

The switch section of the configuration file contains information for all of the following:

•
•
•
•
•
•
•
•
•
•
•
•
•
•

Boot parameters
Configuration
Bottleneck configuration
Flow Vision configuration
FCoE software configuration
Zoning
Defined security policies
Active security policies
iSCSI
CryptoDev
FICU saved files
VS_SW_CONF
MAPS configuration
Banner

Configuration file backup
Brocade recommends keeping a backup configuration file. You should keep individual backup files
for all switches in the fabric and avoid copying configurations from one switch to another. The
configUpload command, by default, only uploads the switch context configuration for the logical
switch context in which the command is executed.
In non-Virtual Fabric mode, you must use the configUpload -all command to include both the
switch and the chassis information. In Virtual Fabric mode, the configUpload -all command can be
selected to upload all logical switches and the chassis configuration. Only administrators with
chassis permissions are allowed to upload other FIDs or the chassis configuration.
The following information is not saved in a backup:

• dnsConfig command information
• Passwords
Before you upload a configuration file, verify that you can reach the FTP server from the switch.
Using a Telnet connection, save a backup copy of the configuration file from a logical switch to a
host computer.
Secure File Transfer Protocol (SFTP) is now an option when uploading a configuration file. SFTP is
analogous to Secure Copy Protocol (SCP). SFTP can be used for the configupload/download,
supportsave, and auto FFDC/trace upload (supportftp) commands.

Uploading a configuration file in interactive mode
1. Verify that the FTP, SFTP, or SCP service is running on the host computer.
2. Connect to the switch and log in using an account with admin permissions.
3. Enter the configUpload command. The command becomes interactive and you are prompted
for the required information.

Fabric OS Administrator’s Guide
53-1002920-02

279

9

Configuration file restoration

4. Store a soft copy of the switch configuration information in a safe place for future reference.
Example of configUpload on a switch without Admin Domains
switch:admin> configupload
Protocol (scp, ftp, sftp, local) [ftp]: sftp
Server Name or IP Address [host]: 10.1.2.3
User Name [user]: UserFoo
Path/Filename [/config.txt]: switchConfig.txt
Section (all|chassis|FID# [all]): chassis
username@10.1.2.3's password:
Password: 
configUpload complete

Example of configUpload on a switch with Admin Domains

NOTE

Administrative domains other than AD255 upload a subset of information. If you want a
complete switch configuration, you must use the configUpload command while logged in to
AD255.
switch:AD5:admin> ad --select 5
switch:AD5:admin> configUpload
Protocol (scp or ftp) [ftp]:
Server Name or IP Address [host]: 10.1.2.3
User Name [user]: UserFoo
Path/Filename [/config.txt]: /pub/configurations/config.txt
Password: 
configUpload complete: Only zoning parameters are uploaded from ad5.

Configuration file restoration
When you restore a configuration file, you overwrite the existing configuration with a previously
saved backup configuration file.
CAUTION
Make sure that the configuration file you are downloading is compatible with your switch model.
Downloading a configuration file from a different switch model or from a different firmware could
cause your switch to fail.

CAUTION
If you have Virtual Fabrics enabled, you must follow the procedure in “Configuration management
for Virtual Fabrics” on page 285 to restore the logical switches.
If a configDownload command is issued on a non-FC router, any FC router parameters may be
viewed in the downloaded data. This is harmless to the switch and can be ignored.
MAPS configuration is downloaded onto the switch only if MAPS is enabled on the local switch.
While it is possible to transfer a Fabric OS 6.4.1 configuration file to a Fabric OS 7.0.0 or later
switch, it is not possible to transfer a Fabric OS 7.0.0 or later configuration file to a Fabric OS 6.4.1
switch.

280

Fabric OS Administrator’s Guide
53-1002920-02

Configuration file restoration

9

Restrictions
This section lists restrictions for some of the options of the configDownload command.
-chassis

The number of switches defined in the downloaded configuration file must
match the number of switches currently defined on the switch.

-fid FID

The FID must be defined in both the downloaded configuration file and the
current system.

NOTE
Brocade recommends you disable a switch before downloading a configuration
file. If you plan to download a configuration file while the switch is enabled, refer
to “Configuration download without disabling a switch” on page 282.
-fid FID -sfid FID The FID must be defined on the switch and the source FID must be defined in the
downloaded configuration file.
-all

The number of switches or FIDs defined in the downloaded configuration file
must match the number of switches or FIDs currently defined on the switch.
The switches must be disabled first. If they are not, the configDownload
command will download the configuration for as many switches as possible until
a non-disabled switch is found. If a non-disabled switch is found, the
downloading process stops. Before running the configDownload command, verify
if any switches must be disabled.
If you are performing a configuration download due to a configuration error, it is
highly recommended to run the configDefault command before running the
configDownload command. Refer to “Configuration download without disabling a
switch” on page 282 for more information on non-disruptive configuration
downloads.
If Virtual Fabrics mode is enabled, the chassisDisable and chassisEnable
commands are used to disable all logical switches on the affected switch. This
process bypasses the need to disable and enable each switch individually once
the configuration download has completed.
Non-Virtual Fabric configuration files downloaded to a Virtual Fabric system have
a configuration applied only to the default switch. If there are multiple logical
switches created in a Virtual Fabric-enabled system, there may be problems if
there are ports that belong to the default switch in a Virtual Fabric-disabled
system, but are now assigned to logical switches in a Virtual Fabric-enabled
system. Only configurations related to ports within the default switch are applied.

If you must set up your switch again, run the commands listed in Table 55 and save the output in a
file format. Store the files in a safe place for emergency reference.

TABLE 55

CLI commands to display or modify switch configuration information

Command

Displays

configShow

System configuration parameters, settings, and license information.

fcLunQuery

LUN IDs and LUNs for all accessible targets.

fcrRouterPortCost FC Router route information.

Fabric OS Administrator’s Guide
53-1002920-02

281

9

Configuration file restoration

TABLE 55

CLI commands to display or modify switch configuration information (Continued)

Command

Displays

fcrXlateConfig

Translate (xlate) domain's domain ID for both EX_Port-attached fabric and backbone fabric.

fosConfig

Fabric OS features.

ipAddrShow

IP address.

isnscCfg

Configuration state of the iSNS client operation.

licenseShow

License keys installed with more detail than the license information from the configShow
command.

portCfgEXPort

EX_Port configuration parameters.

portCfgVEXPort

VEX_Port configuration parameters.

CAUTION
Though the switch itself has advanced error checking, the configdownload feature within
Fabric OS was not designed for users to edit, and is limited in its ability. Edited files can become
corrupted and this corruption can lead to switch failures.

Configuration download without disabling a switch
You can download configuration files to a switch while the switch is enabled; that is, you do not
need to disable the switch for changes in SNMP, MAPS, Fabric Watch, or ACL parameters. However,
if there is any changed parameter that does not belong to SNMP, MAPS, Fabric Watch, or ACL, then
you must disable the switch. When you use the configDownload command, you are prompted to
disable the switch only when necessary.

ATTENTION
The configuration download process can restore only logical switches that already exist and that use
the same FIDs. It cannot be used to clone or repair the current switch because the configDownload
command cannot create logical switches if they do not exist.

Restoring a configuration
CAUTION
Using the SFID parameter erases all configuration information on the logical switch. Use the SFID
parameter only when the logical switch has no configuration information you want to save.
1. Verify that the FTP service is running on the server where the backup configuration file is
located.
2. Connect to the switch and log in using an account with admin permissions, and if necessary,
with chassis permissions.
3. If there are any changed parameters in the configuration file that do not belong to SNMP,
Fabric Watch, or ACL, disable the switch by entering the switchDisable command.

282

Fabric OS Administrator’s Guide
53-1002920-02

Configuration file restoration

9

4. Enter the configDownload command.
The command becomes interactive and you are prompted for the required information.
5. At the “Do you want to continue [y/n]” prompt, enter y.
Wait for the configuration to be restored.
6. If you disabled the switch, enter the switchEnable command when the process is finished.

NOTE

Always perform a reboot after you download a configuration file. On dual-CP platforms, you must
reboot both CPs simultaneously.
Example of configDownload without Admin Domains
switch:admin> configdownload
Protocol (scp, ftp, local) [ftp]:
Server Name or IP Address [host]: 10.1.2.3
User Name [user]: UserFoo
Path/Filename [/config.txt]:
Section (all|chassis|FID# [all]): all
*** CAUTION ***
This command is used to download a backed-up configuration
for a specific switch. If using a file from a different
switch, this file's configuration settings will override
any current switch settings.
Downloading a configuration
file, which was uploaded from a different type of switch,
may cause this switch to fail.
A switch reboot is required for the changes to take effect.
Please make sure all the switches are disabled by
using "chassisdisable" command. Downloading configuration
to an online switch may result in some configuration not
being downloaded to that switch.
configDownload operation may take several minutes
to complete for large files.
Do you want to continue [y/n]:y
Password: 
configDownload complete.

Example of configDownload with Admin Domains
switch:AD5:admin>configdownload
Protocol (scp or ftp) [ftp]:
Server Name or IP Address [host]: 10.1.2.3
User Name [user]: UserFoo
Path/Filename [/config.txt]: /pub/configurations/config.txt
*** CAUTION ***
This command is used to download a backed-up configuration
for a specific switch. If using a file from a different
switch, this file's configuration settings will override
any current switch settings.
Downloading a configuration
file, which was uploaded from a different type of switch,

Fabric OS Administrator’s Guide
53-1002920-02

283

9

Configurations across a fabric

may cause this switch to fail.
A switch reboot is required for the changes to take effect.
Please make sure all the switches are disabled by
using "chassisdisable" command. Downloading configuration
to an online switch may result in some configuration not
being downloaded to that switch.
configDownload operation may take several minutes
to complete for large files.
Do you want to continue [y/n]:y
Password: 
Activating configDownload: Switch is disabled
configDownload complete: Only zoning parameters are downloaded to ad5.

Example of a non-interactive download of all configurations (chassis and switches)
configdownload -a -ftp
10.1.2.3,UserFoo,/pub/configurations/config.txt,password

Configurations across a fabric
To save time when configuring fabric parameters and software features, you can save a
configuration file from one switch and download it to other switches of the same model type.
Do not download a configuration file from one switch to another switch that is a different model or
runs a different firmware version, because it can cause the switch to fail. If you need to reset
affected switches, issue the configDefault command after download is completed but before the
switch is enabled. If a switch is enabled with a duplicate domain ID, the switch becomes
segmented.

Downloading a configuration file from one switch to another switch of
the same model
1. Configure one switch.
2. Use the configUpload command to save the configuration information. Refer to “Configuration
file backup” on page 279 for more information.
3. Run configDefault on each of the target switches, and then use the configDownload command
to download the configuration file to each of the target switches. Refer to “Configuration file
restoration” on page 280 for more information.

Security considerations
Security parameters and the switch identity cannot be changed by the configDownload command.
Parameters such as the switch name and IP address (lines in the configuration file that begin with
“boot”) are ignored. Security parameters (lines in the configuration file that begin with “sec”), such
as secure mode setting and version stamp, are ignored. For more detailed information on security,
refer to Chapter 7, “Configuring Protocols”.

284

Fabric OS Administrator’s Guide
53-1002920-02

Configuration management for Virtual Fabrics

9

Configuration management for Virtual Fabrics
You can use the configUpload -vf or configDownload -vf command to restore configurations to a
logical switch. The -vf option only restores the Virtual Fabrics configuration information on to a
switch of the same model and same release. For example, a Virtual Fabrics configuration file for
Fabric OS 7.2.x cannot be used on a Fabric OS 7.1.x switch and vice versa.
The Virtual Fabrics configuration on the switch defines all of the logical switches allowed and
configured for a particular platform.

Uploading a configuration file from a switch with Virtual Fabrics
enabled
The configUpload command with the -vf option specifies that configuration upload will upload the
Virtual Fabrics configuration instead of the non-Virtual Fabrics configuration information.
You must specify a file name with the configUpload -vf command. It is recommended not to use
config.txt for a file name as this name can be confused with a normal uploaded configuration file.
Example of configUpload on a switch with Virtual Fabrics
Sprint5100:FID128:admin> configupload
Protocol (scp, ftp, sftp, local) [ftp]:
Server Name or IP Address [host]: 10.1.2.3
User Name [user]: UserFoo
Path/Filename [/config.txt]: 5100.txt
Potentially remote file may get overwritten
Section (all|chassis|FID# [all]):
Password: 
configUpload complete: All selected config parameters are uploaded

Example of configUpload on a logical switch configuration
DCX_80:FID128:admin> configupload -vf
Protocol (scp, ftp, sftp, local) [ftp]:
Server Name or IP Address [host]: 10.1.2.3
User Name [user]: anonymous
Path/Filename [/config.txt]: 5100_vf.txt
configUpload complete: VF config parameters are uploaded
2009/07/20-09:13:40, [LOG-1000], 225, SLOT 7 | CHASSIS, INFO, BrocadeDCX,
Previous message repeated 7 time(s)
2009/07/20-10:27:14, [CONF-1001], 226, SLOT 7 | FID 128, INFO, DCX_80,
configUpload completed successfully for VF config parameters.

Restoring a logical switch configuration using configDownload
The configDownload -vf command specifies that the Virtual Fabrics configuration download file is
downloaded instead of the regular configuration. After the Virtual Fabrics configuration file is
downloaded, the switch is automatically rebooted.
On dual-CP platforms, if CPs are incompatible (HA not in sync), the Virtual Fabrics configuration file is
not propagated to the standby CP. If CPs are compatible, the active CP attempts to remain active after
the reboot, and the new Virtual Fabrics configuration file is then propagated to the standby CP.

Fabric OS Administrator’s Guide
53-1002920-02

285

9

Configuration management for Virtual Fabrics

CAUTION
You must issue the configDownload command on the switch after restoring the Virtual Fabrics
configuration to fully restore your switch or chassis configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the configDownload -vf command.
3. Respond to the prompts.
Wait for the configuration file to download on to the switch. You may need to reconnect to the
switch.
4. Enter the configDownload command.
5. Respond to the prompts.
Wait for the configuration file to download to the switch.
6. Verify the LISL ports are set up correctly.
Example of a non-interactive download from a switch with FID = 8 and SFID =10
configdownload -fid 8 -sfid 10 -ftp 10.1.2.3,UserFoo,config.txt,password

Example of configDownload on a switch
5100:FID128:admin> configdownload -vf
Protocol (scp, ftp, sftp, local) [ftp]:
Server Name or IP Address [host]: 10.1.2.3
User Name [user]: UserFoo
Path/Filename [/config.txt]: 5100_FID89.txt
*** CAUTION ***
This command is used to download the VF configuration to the
switch. Afterwards, the switch will be automatically rebooted
and the new VF settings will be used. You will then need to
run configdownload again to install the configuration(s) for
any logical switch(s) that are setup in the new VF configuration.
Do you want to continue [y/n]: y
(output truncated)

Restrictions
The following restrictions apply when using the configUpload or configDownload commands when
Virtual Fabrics mode is enabled:

• The -vf option is incompatible with the –fid, –sfid, or –all options. Any attempt to combine it
with any of the other three will cause the configuration upload or download operation to fail.

• You are not allowed to modify the Virtual Fabrics configuration file after it has been uploaded.
Only minimal verification is done by the configDownload command to ensure it is compatible,
much like the normal downloaded configuration file.

• After the configDownload -vf command completes and reboots your switch, you must then
download the matching regular configuration using the configDownload -all command. This
ensures proper behavior of the system and logical switches.

286

Fabric OS Administrator’s Guide
53-1002920-02

Brocade configuration form

9

All of the attributes of the Virtual Fabrics configuration file will be downloaded to the system and
take effect. This includes, but is not limited to, logical switch definitions, whether Virtual Fabrics is
enabled or disabled, and the F_Port trunking ports, except the LISL ports. The LISL ports on the
system are not affected by the Virtual Fabrics configuration file download.
You can restore Virtual Fabrics configurations only to a switch of the same model and same
release. For example, a Virtual Fabrics configuration file for Fabric OS 7.2.x cannot be used on a
Fabric OS 7.1.x switch and vice versa.

Brocade configuration form
Use the form in Table 56 as a hard copy reference for your configuration information. In the
hardware reference manuals for the Brocade DCX and DCX-4S Backbones, there is a guide for FC
port-setting.

TABLE 56

Brocade configuration and connection form

Brocade configuration settings
IP address
Gateway address
Chassis configuration option

Management connections
Serial cable tag
Ethernet cable tag

Configuration information
Domain ID
Switch name
Ethernet IP address
Ethernet subnet mask
Total number of local devices (nsShow)
Total number of devices in fabric (nsAllShow)
Total number of switches in the fabric (fabricShow)

Fabric OS Administrator’s Guide
53-1002920-02

287

9

288

Brocade configuration form

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

10

Installing and Maintaining Firmware

In this chapter
• Firmware download process overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Preparing for a firmware download . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Firmware download on switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Firmware download on a Backbone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Firmware download from a USB device . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• FIPS support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Testing and restoring firmware on switches . . . . . . . . . . . . . . . . . . . . . . . .
• Testing and restoring firmware on Backbones . . . . . . . . . . . . . . . . . . . . . .
• Validating a firmware download . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

289
292
294
296
299
300
302
304
306

Firmware download process overview
Fabric OS v7.2.0 provides nondisruptive firmware installation.
This chapter refers to the following specific types of blades inserted into the Brocade DCX and
DCX 8510 Backbone families:

• FC blades or port blades that contain only Fibre Channel ports; the Brocade FC8-16, FC8-32,
FC8-48, and FC8-64; and the Brocade FC16-32 and FC16-48 blades for 16-Gbps-capable FC
blades.

• AP blades contain extra processors and specialized ports: FCOE10-24, FX8-24, and FS8-18
encryption blade.

• CP blades have a control processor (CP) used to control the entire switch; CP blades can be
inserted only into slots 6 and 7 on the Brocade DCX or DCX 8510-8, and slots 4 and 5 on the
Brocade DCX-4S or DCX 8510-4.

• CR8 and CR4S-8 core blades provide ICL functionality between two Brocade DCX Backbones.
CR8 blades can be inserted only into slots 5 and 8 on the Brocade DCX. CR4S-8 blades can be
inserted only into slots 3 and 6 on the Brocade DCX-4S.

• CR8 and CR4S-8 core blades provide ICL functionality between two Brocade DCX 8510-8
Backbones. CR4S-8 blades can be inserted only into slots 3 and 6 on the Brocade DCX
8510-4.

NOTE
For more information on troubleshooting a firmware download, refer to the Fabric OS
Troubleshooting and Diagnostics Guide.

Fabric OS Administrator’s Guide
53-1002920-02

289

10

Firmware download process overview

You can download Fabric OS to a Backbone, which is a chassis; and to a nonchassis-based system,
also referred to as a fixed-port switch. The difference in the download process is that Backbones
have two CPs and fixed-port switches have one CP. Use the firmwareDownload command to
download the firmware from either an FTP or SSH server by using FTP, SFTP, or SCP to the switch.
Or you can use a Brocade-branded USB device.
New firmware consists of multiple files in the form of RPM packages listed in a .plist file. The .plist
file contains specific firmware information (time stamp, platform code, version, and so forth) and
the names of packages of the firmware to be downloaded. These packages are made available
periodically to add features or to remedy defects. Contact your switch support provider to obtain
information about available firmware versions.
All systems maintain two partitions (a primary and a secondary) of nonvolatile storage areas to
store firmware images. The firmware download process always loads the new image into the
secondary partition. It then swaps the secondary partition to be the primary and High Availability
(HA) reboots (which is nondisruptive) the system. After the system boots up, the new firmware is
activated. The firmware download process then copies the new image from the primary partition to
the secondary partition.
In dual-CP systems, the firmware download process, by default, sequentially upgrades the firmware
image on both CPs using HA failover to prevent disruption to traffic flowing through the Backbone.
This operation depends on the HA status on the Backbone. If the platform does not support HA, you
can still upgrade the CPs one at a time.
If you are using a Brocade DCX or DCX 8510 Backbone family platform, with one or more
AP blades: Fabric OS automatically detects mismatches between the active CP firmware and the
blade’s firmware and triggers the autoleveling process. This autoleveling process automatically
updates the blade firmware to match the active CP. At the end of the autoleveling process, the
active CP and the blade run the same version of the firmware.
If the firmware download process is interrupted by an unexpected reboot, the system automatically
repairs and recovers the secondary partition. You must wait for the recovery to complete before
issuing another firmwareDownload command.
The command supports both non-interactive and interactive modes. If the firmwareDownload
command is issued without any operands, or if there is any syntax error in the parameters, the
command enters an interactive mode, in which you are prompted for input.

ATTENTION
For each switch in your fabric, complete all firmware download changes on the current switch before
issuing the firmwareDownload command on the next switch. This process ensures nondisruption of
traffic between switches in your fabric.
To verify the firmware download process is complete, enter the firmwareDownloadStatus command
on the switch, verify the process is complete, and then move to the next switch.

290

Fabric OS Administrator’s Guide
53-1002920-02

Firmware download process overview

10

Upgrading and downgrading firmware
Upgrading means installing a newer version of firmware. Downgrading means installing an older
version of firmware.
In most cases, you will be upgrading firmware; that is, installing a newer firmware version than the
one you are currently running. However, some circumstances may require installing an older
version; that is, downgrading the firmware. The procedures in this section assume that you are
upgrading firmware, but they work for downgrading as well, provided the old and new firmware
versions are compatible. Always reference the latest release notes for updates that may exist
regarding downgrades under particular circumstances.
For details on Administrative Domains and the firmware download process, refer to Chapter 20,
“Managing Administrative Domains” for more information.
For details about testing and restoring firmware, refer to “Testing and restoring firmware on
Backbones” on page 304.

Passwordless firmware download
You can download firmware without a password using the sshutil command for public key
authentication when SSH is selected. The switch must be configured to install the private key, and
then you must export the public key to the remote host. Before running the firmwareDownload
command, you must first configure the SSH protocol to permit passwordless logins for outgoing
authentication as described in “Configuring outgoing SSH authentication” on page 199.

Considerations for FICON CUP environments
To prevent channel errors during nondisruptive firmware installation, the switch CUP port must be
taken offline from all host systems.

HA sync state
High Availability (HA) synchronization occurs when two CPs in a Backbone are synchronized. This
state provides redundancy and a nondisruptive firmware download. In order for a firmware
download to successfully occur, the two CPs in a Backbone must be in sync.
If the CPs have mixed versions when you enter the firmwareDownload command, the CPs may not
be in HA sync. In this case, you must enter the firmwareDownload –s command first to upgrade or
downgrade the standby CP to the same level as the active CP, and then upgrade the CPs to the
desired version of firmware.

NOTE
Do not run mixed firmware levels on CPs.
Table 57 shows the sync state of a Backbone that has different Fabric OS versions installed on the
active and standby CPs. Use the table to determine if you need to use the firmwareDownload -s
command.

Fabric OS Administrator’s Guide
53-1002920-02

291

10

Preparing for a firmware download

TABLE 57

Backbone HA sync states

Active CP Fabric OS
version

Standby CP Fabric OS
version

HA sync state

Remedy

v6.4.0

v6.4.0

inSync

Run firmwareDownload -s on the
standby CP to upgrade it to v7.0.0

v6.4.0

v7.1.0

Not inSync

N/A

v7.0.0

v6.4.0

inSync

Run firmwareDownload -s on the
standby CP to upgrade it to v7.0.0

v7.0.0

v7.0.0

inSync

N/A

v7.0.0

v7.1.0

inSync

N/A

v7.0.0

v7.2.0

Not inSync

N/A

v7.1.0

v6.4.0

Not inSync

N/A

v7.1.0

v7.0.0

inSync

N/A

v7.1.0

v7.1.0

inSync

N/A

v7.1.0

v7.2.0

InSync

N/A

v7.2.0

v7.1.0

InSync

N/A

v7.2.0

v7.2.0

InSync

N/A

Preparing for a firmware download
Before executing a firmware download, it is recommended that you perform the tasks listed in this
section. In the unlikely event of a failure or timeout, these preparatory tasks enable you to provide
your switch support provider the information required to troubleshoot the firmware download.
It is recommended that you use the configUpload command to back up the current configuration
before you download firmware to a switch. Refer to “Configuration file backup” on page 279 for
details.
1. Read the release notes for the new firmware to find out if there are any updates related to the
firmware download process.
2. Connect to the switch and log in using an account with admin permissions. Enter the
firmwareShow command to verify the current version of Fabric OS.
Brocade does not support non-disruptive upgrades from more than one previous release.
Non-disruptive upgrade is supported only from Fabric OS 7.1.x to Fabric OS 7.2.x.
Disruptive upgrade from Fabric OS 7.0.x to Fabric OS 7.2.x is supported.
3. Use the configUpload command prior to the firmware download. Save the configuration file on
your FTP or SSH server or USB memory device on supported platforms.
4. Optional: For additional support, connect the switch to a computer with a serial console cable.
Ensure that all serial consoles (both CPs for Backbones) and any open network connection
sessions, such as Telnet, are logged and included with any trouble reports.

292

Fabric OS Administrator’s Guide
53-1002920-02

Preparing for a firmware download

10

5. Connect to the switch and log in using an account with admin permissions. Enter the
supportSave command to retrieve all current core files prior to executing the firmware
download. This information helps to troubleshoot the firmware download process if a problem
is encountered.
6. Optional: Enter the errClear command to erase all existing messages in addition to internal
messages.

Obtaining and decompressing firmware
Firmware upgrades are available for customers with support service contracts and for partners on
the Brocade website at http://www.brocade.com.
You must decompress the firmware before you can use the firmwareDownload command to update
the firmware on your equipment. Use the UNIX tar command for .tar files, the gunzip command for
all .gz files, or a Windows unzip program for all .zip files
When you unpack the downloaded firmware, it expands into a directory that is named according to
the version of Fabric OS it contains. For example, when you download and unzip v7.2.0.zip, it
expands into a directory called v7.2.0. When you issue the firmwareDownload command, there is
an automatic search for the correct package file type associated with the switch. Specify only the
path up to and including the v7.2.0 directory.

Connected switches
Before you upgrade the firmware on your switch, you must check the connected switches to ensure
compatibility and that any older versions are supported. Refer to the Fabric OS Compatibility
section of the Brocade Fabric OS Release Notes for the recommended firmware version.
If fixed-port switches are adjacent and you start firmware downloads on them at the same time,
there may be traffic disruption.
To determine if you need to upgrade switches connected to the switch you are upgrading, use the
following procedure on each connected switch to display firmware information and build dates.

Finding the switch firmware version
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the version command.
The following information is displayed:

•
•
•
•
•

Fabric OS Administrator’s Guide
53-1002920-02

Kernel displays the version of the switch kernel operating system.
Fabric OS displays the version of the switch Fabric OS.
Made on displays the build date of the firmware running on the switch.
Flash displays the install date of firmware stored in nonvolatile memory.
BootProm displays the version of the firmware stored in the boot PROM.

293

10

Firmware download on switches

Firmware download on switches
Brocade fixed-port switches maintain primary and secondary partitions for firmware. The
firmwareDownload command defaults to an autocommit option that automatically copies the
firmware from one partition to the other.

NOTE

This section only applies when upgrading from Fabric OS v7.1.x to v7.2.0, downgrading from v7.2.0
to v7.1.x, or going from v7.2.x to v7.2.x
If you are upgrading from Fabric OS v7.0.x to v7.2.0 or downgrading from v7.2.0 to v7.0.x or earlier,
you must enter the firmwareDownload –s command as described in “Testing and restoring firmware
on switches” on page 302.
You cannot download firmware if you are going from v7.2.0 to v6.4 (or earlier) or from v6.4 (or earlier)
to v7.2.0.
Do not override the autocommit option under normal circumstances; use the default. Refer to
“Testing and restoring firmware on Backbones” on page 304 for details about overriding the
autocommit option.

Switch firmware download process overview
The following list describes the default behavior after you enter the firmwareDownload command
(without options) on Brocade fixed-port switches:

• The Fabric OS downloads the firmware to the secondary partition.
• The system performs a high availability reboot (haReboot). After the haReboot, the former
secondary partition is the primary partition.

• The system replicates the firmware from the primary to the secondary partition.
The upgrade process first downloads and then commits the firmware to the switch. While the
upgrade is proceeding, you can start a session on the switch and use the firmwareDownloadStatus
command to observe the upgrade progress.

CAUTION
After you start the process, do not enter any disruptive commands (such as reboot) that interrupt
the process. The entire firmware download and commit process takes approximately 17 minutes.
If there is a problem, wait for the timeout (30 minutes for network problems) before issuing the
firmwareDownload command again. Disrupting the process can render the switch inoperable and
require you to seek help from your switch service provider.
Do not disconnect the switch from power during the process. The switch could be inoperable
when rebooted.

294

Fabric OS Administrator’s Guide
53-1002920-02

Firmware download on switches

10

Upgrading firmware for Brocade fixed-port switches
1. Take the following appropriate action based on what service you are using:

• If you are using FTP, SFTP, or SCP, verify that the FTP or SSH server is running on the host
server and that you have a valid user ID and password on that server.

• If your platform supports a USB memory device, verify that it is connected and running.
2. Obtain the firmware file from the Brocade website at http://www.brocade.com and store the
file on the FTP or SSH server or the USB memory device.
3. Unpack the compressed files preserving directory structures.
The firmware is in the form of RPM packages with names defined in a .plist file. The .plist file
contains specific firmware information and the names of packages of the firmware to be
downloaded.
4. Connect to the switch and log in using an account with admin permissions.
5. Issue the firmwareShow command to check the current firmware version on connected
switches. Upgrade the firmware on the connected switches, if necessary, before proceeding
with upgrading this switch.
Refer to “Connected switches” on page 293 for details.
6. Enter the firmwareDownload command and respond to the prompts.

NOTE

If DNS is enabled and a server name instead of a server IP address is specified in the
command line, firmwareDownload determines whether IPv4 or IPv6 should be used.
To be able to mention the FTP server by name, you must enter at least one DNS server using
the dnsConfig command.
7.

At the “Do you want to continue [y/n]” prompt, enter y.

8. After the HA reboot, connect to the switch and log in again using an account with admin
permissions.
9. Enter the firmwareDownloadStatus command to determine if the firmware download process
has completed.
10. After the firmware commit is completed, which takes several minutes, enter the firmwareShow
command to verify the firmware level of both partitions is the same.
Example of an interactive firmware download
switch:root> firmwaredownload
Server Name or IP Address: 10.31.2.25
User Name: releaseuser
File Name: /home/SAN/fos/v7.2.0/v7.2.0
Network Protocol(1-auto-select, 2-FTP, 3-SCP, 4-SFTP) [1]: 4
Verifying if the public key authentication is available.Please wait ...
The public key authentication is not available.
Password:
Server IP: 10.31.2.25, Protocol IPv4
Checking system settings for firmwaredownload...

Fabric OS Administrator’s Guide
53-1002920-02

295

10

Firmware download on a Backbone

Firmware download on a Backbone
ATTENTION
To successfully download firmware, you must have an active Ethernet connection on each CP.
You can download firmware to a Backbone without disrupting the overall fabric if the two CP blades
are installed and fully synchronized. Use the haShow command to verify that the CPs are
synchronized prior to beginning the firmware download process. If only one CP blade is inserted,
powered on, or plugged into the network, you can run firmwareDownload –s to upgrade the CP.
If the CPs are not in sync, you can run firmwareDownload –s on each of the CPs to upgrade them.
These operations are disruptive. If the CPs are not in sync, run the haSyncStart command. If the
CPs are still not in sync, refer to the Fabric OS Troubleshooting and Diagnostics Guide. If the
troubleshooting information fails to help resolve the issue, contact your switch service provider.
During the upgrade process, the Backbone fails over to its standby CP blade and the IP address for
the Backbone moves to that CP blade's Ethernet port. This may cause informational ARP address
reassignment messages to appear on other switches in the fabric. This is normal behavior,
because the association between the IP addresses and MAC addresses has changed.

Backbone firmware download process overview
The following summary describes the default behavior of the firmwareDownload command (without
options) on a Backbone. After you enter the firmwareDownload command on the active CP blade
the following actions occur.
1. The standby CP blade downloads firmware.
2. The standby CP blade reboots and comes up with the new Fabric OS.
3. The active CP blade synchronizes its state with the standby CP blade.
4. The active CP blade forces a failover and reboots to become the standby CP blade.
5. The new active CP blade synchronizes its state with the new standby CP blade.
6. The new standby CP blade (the active CP blade before the failover) downloads firmware.
7.

The new standby CP blade reboots and comes up with the new Fabric OS.

8. The new active CP blade synchronizes its state with the new standby CP blade.
9. The firmwareCommit command runs automatically on both CP blades.

CAUTION
After you start the process, do not enter any disruptive commands (such as reboot) that interrupt
the process. The entire firmware download and commit process takes approximately 17 minutes.
If there is a problem, wait for the timeout (30 minutes for network problems) before issuing the
firmwareDownload command again. Disrupting the process can render the switch inoperable and
require you to seek help from your switch service provider.
Do not disconnect the switch from power during the process. The switch could be inoperable
when rebooted.

296

Fabric OS Administrator’s Guide
53-1002920-02

Firmware download on a Backbone

10

Upgrading firmware on Backbones (including blades)
There is only one chassis management IP address for the Brocade Backbones.

NOTE

By default, the firmwareDownload command automatically upgrades both the active and the
standby CPs and all co-CPs on the CP blades in the Brocade Backbones. It automatically upgrades
all AP blades in the Brocade Backbones using autoleveling.
1. Verify that the Ethernet interfaces located on CP0 and CP1 are plugged into your network.
2. Verify that the FTP, SFTP, or SSH server is running on the host server and that you have a user
ID on that server.
3. Obtain the firmware file from the Brocade website at http://www.brocade.com and store the
file on the FTP or SSH server.
4. Unpack the compressed files preserving directory structures.
The firmware is in the form of RPM packages with names defined in a .plist file. The .plist file
contains specific firmware information and the names of packages of the firmware to be
downloaded.
5. Connect to the chassis IP management interface or active CP and log in using an account with
admin permissions.
6. Use the firmwareShow command to check the current firmware version on connected
switches. Upgrade the firmware, if necessary, before proceeding with upgrading this switch.
Refer to “Connected switches” on page 293.
7.

Enter the haShow command to confirm that the two CP blades are synchronized.
In the following example, the active CP blade is CP0 and the standby CP blade is CP1:
ecp:admin> hashow
Local CP (Slot 5, CP0): Active, Warm Recovered
Remote CP (Slot 6, CP1): Standby, Healthy
HA enabled, Heartbeat Up, HA State synchronized

CP blades must be synchronized and running Fabric OS v7.1.0 or later to provide a
nondisruptive download. If the two CP blades are not synchronized, enter the haSyncStart
command to synchronize them. If the CPs still are not synchronized, contact your switch
service provider.
For further troubleshooting, refer to the Fabric OS Troubleshooting and Diagnostics Guide.
8. Enter the firmwareDownload command and respond to the interactive prompts.
9. At the “Do you want to continue [y/n]” prompt, enter y.
The firmware is downloaded to one CP blade at a time, beginning with the standby CP blade.
During the process, the active CP blade fails over. After the firmware is downloaded, a firmware
commit starts on both CP blades. The entire firmware download and commit process takes
approximately 17 minutes.

Fabric OS Administrator’s Guide
53-1002920-02

297

10

Firmware download on a Backbone

If an AP blade is present: At the point of the failover, an autoleveling process is activated.
Autoleveling is triggered when the active CP detects a blade that contains a different version of
the firmware, regardless of which version is older. Autoleveling downloads firmware to the AP
blade, swaps partitions, reboots the blade, and copies the new firmware from the primary
partition to the secondary partition. If you have multiple AP blades, they are updated
simultaneously; however, the downloads can occur at different rates.
Autoleveling takes place in parallel with the firmware download being performed on the CPs,
but does not impact performance. Fibre Channel traffic is not disrupted during autoleveling,
but GbE traffic on AP blades may be affected. If there is an active FCIP tunnel on the FX8-24
blade, the FCIP tunnel traffic will be impacted for at least two minutes.
ecp:admin> firmwaredownload
Type of Firmware (FOS, SAS, or any application) [FOS]:
Server Name or IP Address: 10.1.2.3
User Name: userfoo
File Name: /home/userfoo/v7.2.0
Network Protocol (1-auto-select, 2-FTP, 3-SCP, 4-SFTP)) [1]:
Password: 
Checking version compatibility...
Version compatibility check passed.
The following AP blades are installed in the system.
Slot Name
Versions
Traffic Disrupted
----------------------------------------------------------------2 FS8-18
v7.2.0
Encrypted Traffic
8 FX8-24
v7.2.0
GigE
This command will upgrade the firmware on both CPs and all AP blade(s) above.
If you want to upgrade firmware on a single CP only, please use -s option.
You may run firmwaredownloadstatus to get the status of this
command.
This command will cause a warm/non-disruptive boot on the active CP,
but will require that existing telnet, secure telnet or SSH sessions
be restarted.
Do you want to continue [Y]: y
The firmware is being downloaded to the Standby CP. It may take up to 10
minutes.

10. Optionally, after the failover, connect to the switch, and log in again as admin. Using a separate
session to connect to the switch, enter the firmwareDownloadStatus command to monitor the
firmware download status.
sw0:FID128:admin> firmwaredownloadstatus
[1]: Mon Jul 22 04:27:21 2013
Slot 7 (CP1, active): Firmware is being downloaded to the switch. This step
may take up to 30 minutes.
[2]: Mon Jul 22 04:34:58 2013
Slot 7 (CP1, active): Relocating an internal firmware image on the CP blade.
[3]: Mon Jul 22 04:35:29 2013
Slot 7 (CP1, active): The internal firmware image is relocated successfully.
[4]: Mon Jul 22 04:35:30 2013

298

Fabric OS Administrator’s Guide
53-1002920-02

Firmware download from a USB device

10

Slot 7 (CP1, active): Firmware has been downloaded to the secondary partition
of the switch.
[5]: Mon Jul 22 04:37:24 2013
Slot 7 (CP1, standby): The firmware commit operation has started. This may
take up to 10 minutes.
[6]: Mon Jul 22 04:41:59 2013
Slot 7 (CP1, standby): The commit operation has completed successfully.
[7]: Mon Jul 22 04:41:59 2013
Slot 7 (CP1, standby): Firmwaredownload command has completed successfully.
Use firmwareshow to verify the firmware versions.

11. Enter the firmwareShow command to display the new firmware versions.

Firmware download from a USB device
The Brocade 300, 5100, 5300, 6505, 6510, 6520, 7800, and VA-40FC switches and the Brocade
DCX, DCX-4S, or DCX 8510 Backbones support a firmware download from a Brocade-branded USB
device attached to the switch or active CP. Before the USB device can be accessed by the
firmwareDownload command, it must be enabled and mounted as a file system. The firmware
images to be downloaded must be stored under the relative path from
/usb/usbstorage/brocade/firmware or use the absolute path in the USB file system. Multiple
images can be stored under this directory. There is a firmwarekey directory where the public key
signed firmware is stored.
When the firmwareDownload command line option, -U (uppercase), is specified, the
firmwareDownload command downloads the specified firmware image from the USB device. When
specifying a path to a firmware image in the USB device, you can only specify the relative path to
/firmware or the absolute path.

Enabling the USB device
1. Log in to the switch using an account assigned to the admin role.
2. Enter the usbStorage -e command.

Viewing the USB file system
1. Log in to the switch using an account assigned to the admin role.
2. Enter the usbStorage -l command.
BrcdDCXBB:admin> usbstorage –l
firmware\
381MB
2013
v7.2.0\
381MB
2013
config\
0B
2013
support\
0B
2013
firmwarekey\
0B
2013
Available space on usbstorage 79%

Fabric OS Administrator’s Guide
53-1002920-02

Jul
Jul
Jul
Jul
Jul

22
22
22
22
22

15:33
10:39
15:33
15:33
15:33

299

10

FIPS support

Downloading from the USB device using the relative path
1. Log in to the switch using an account assigned to the admin role.
2. Enter the firmwareDownload -U command.
ecp:admin>firmwaredownload –U v7.2.0

Downloading from the USB device using the absolute path
1. Log in to the switch using an account assigned to the admin role.
2. Enter the firmwareDownload command with the -U operand.
ecp:admin>firmwaredownload –U /usb/usbstorage/brocade/firmware/v7.2.0

FIPS support
Federal Information Processing Standards (FIPS) specify the security standards needed to satisfy a
cryptographic module utilized within a security system for protecting sensitive information in the
computer and telecommunication systems. For more information about FIPS, refer to Chapter 8,
“Configuring Security Policies”.
Fabric OS v7.2.0 firmware is digitally signed using the OpenSSL utility to provide FIPS support. To
use the digitally signed software, you must configure the switch to enable signed firmware
download. If it is not enabled, the firmware download process ignores the firmware signature and
performs as before.
If signed firmware download is enabled, and if the validation succeeds, the firmware download
process proceeds normally. If the firmware is not signed or if the signature validation fails, firmware
download fails.
To enable or disable FIPS mode, refer to Chapter 8, “Configuring Security Policies”.

Public and private key management
For signed firmware, Brocade uses RSA with 1024-bit length key pairs, a private key and a public
key. The private key is used to sign the firmware files when the firmware is generated. The public
key is packaged in an RPM package as part of the firmware, and is downloaded to the switch. After
it is downloaded, it can be used to validate the firmware to be downloaded next time when you run
the firmwareDownload command.
The public key file on the switch contains only one public key. It is only able to validate firmware
signed using one corresponding private key. If the private key changes in future releases, you need
to change the public key on the switch by one of the following methods:

• By using the firmwareDownload command. When a new firmware is downloaded, firmware
download always replaces the public key file on the switch with what is in the new firmware.
This allows you to have planned firmware key changes.

• By using the firmwareKeyUpdate command. This command retrieves a specified public key file
from a specific server location and replaces the one on the switch. The information about
firmware versions and their corresponding public key files is documented in the release notes
or stored in a known location on the Brocade website. This command allows the customer to
handle unplanned firmware key changes.

300

Fabric OS Administrator’s Guide
53-1002920-02

FIPS support

10

NOTE

If FIPS mode is enabled, all logins should be handled through SSH or direct serial method, and the
transfer protocol should be SCP.

Updating the firmware key
1. Log in to the switch as admin.
2. Enter the firmwareKeyUpdate command and respond to the prompts.

The firmwareDownload command
The public key file must be packaged, installed, and run on your switch before you download a
signed firmware.
When firmware download installs a firmware file, it must validate the signature of the file. Different
scenarios are handled as follows:

• If a firmware file does not have a signature, how it is handled depends on the
“signed_firmware” parameter on the switch. If it is enabled, firmware download fails.
Otherwise, firmware download displays a warning message and proceeds normally. When
downgrading to non-FIPS-compliant firmware, the “signed_firmware” flag must be disabled.

• If the firmware file has a signature but the validation fails, firmware download fails. This means
the firmware is not from Brocade, or the contents have been modified.

• If the firmware file has a signature and the validation succeeds, firmware download proceeds
normally.
SAS, DMM, and third-party application images are not signed.

Configuring a switch for signed firmware
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the configure command.
3. Respond to the prompts as follows:
System Service
ssl attributes
snmp
attributes
rpcd attributes
cfgload
attributes

Press Enter to select default setting; default is no.
Press Enter to select default setting; default is no.
Press Enter to select default setting; default is no.
Press Enter to select default setting; default is no.
Select Yes. The following questions are displayed:
Enforce secure config Upload/Download: Select yes.
Enforce signed firmware download: Select yes.

Webtools
attributes
System

Fabric OS Administrator’s Guide
53-1002920-02

Press Enter to select default setting; default is no.
Press Enter to select default setting; default is no.

301

10

Testing and restoring firmware on switches

Power-on firmware checksum test
FIPS requires the checksums of the executables and libraries on the filesystem to be validated
before Fabric OS modules are launched. This is to make sure these files have not been changed
after they are installed.
When firmware RPM packages are installed during firmware download, the MD5 checksums of the
firmware files are stored in the RPM database on the filesystem. The checksums go through all of
the files in the RPM database. Every file compares its current checksum with the checksum that is
in the RPM database. If they are different, the command displays an output message informing you
of the difference.
Because the validation may take up to a few minutes, it is not performed during a hot code load. It
is only performed after a cold reboot of the switch.
For more information on FIPS, refer to Chapter 8, “Configuring Security Policies”.

Testing and restoring firmware on switches
NOTE

This section does not apply to SAS or storage applications applied to the FA4-18 AP blade.
Typically, users downgrade firmware after briefly evaluating a newer (or older) version and then
restore the original version of the firmware. Testing a new version of firmware in this manner
ensures that you do not replace existing firmware because the evaluated version occupies only one
partition on the switch.

ATTENTION
When you evaluate new firmware, make sure you disable all features that are not supported by the
original firmware before restoring to the original version.

Testing a different firmware version on a switch
1. Verify that the FTP, SFTP, or SSH server is running on the host server and that you have a user
ID on that server.
2. Obtain the firmware file from the Brocade website at http://www.brocade.com or the switch
support provider and store the file on the FTP or SSH server.
3. Unpack the compressed files preserving directory structures.
The firmware is in the form of RPM packages with names defined in a .plist file that contains
specific firmware information and the names of packages of the firmware to be downloaded.
4. Connect to the switch and log in using an account with admin permissions.
5. Enter the firmwareShow command to view the current firmware.
6. Enter the firmwareDownload -s command to update the firmware, and respond to the prompts.
Example of a firmware download to a single partition
ecp:admin> firmwareDownload -s
Type of Firmware (FOS, SAS, or any application) [FOS]:
Server Name or IP Address: 10.1.2.3
Network Protocol (1-auto-select, 2-FTP, 3-SCP, 4-SFTP) [1]:

302

Fabric OS Administrator’s Guide
53-1002920-02

Testing and restoring firmware on switches

10

User Name: userfoo
File Name: /home/userfoo/v7.2.0
Password: 
Do Auto-Commit after Reboot [Y]: n
Reboot system after download [N]: y
Firmware is being downloaded to the switch. This step may take up to 30
minutes.
Checking system settings for firmwaredownload...

The switch performs a reboot and comes up with the new firmware to be tested. Your current
switch session automatically disconnects.

ATTENTION
Downloading firmware to a switch can be disruptive to switch traffic.
7.

Connect to the switch, log in as admin, and enter the firmwareShow command to confirm that
the primary partition of the switch contains the new firmware.
You are now ready to evaluate the new version of firmware.

ATTENTION
Stop! If you want to restore the firmware, stop here and skip ahead to step 9; otherwise,
continue to step 8 to commit the firmware on the switch, which completes the firmware
download operations.
8. Commit the firmware.
a.

Enter the firmwareCommit command to update the secondary partition with new firmware.
Note that it takes several minutes to complete the commit operation.

b.

Enter the firmwareShow command to confirm both partitions on the switch contain the
new firmware.

ATTENTION
Stop! If you have completed step 8, then you have committed the firmware on the switch and
you have completed the firmware download procedure.
9. Restore the firmware.
a.

Enter the firmwareRestore command. The switch reboots and comes up with the original
firmware again.
A firmware commit automatically begins to copy the original firmware from the primary
partition to the secondary partition. At the end of the firmware commit process, both
partitions have the original firmware. Note that it takes several minutes to complete the
commit operation.

b.

Wait five minutes to ensure that all processes have completed and the switch is fully up
and operational.

c.

Log in to the switch. Enter the firmwareShow command and verify that both partitions on
the switch have the original firmware.

Fabric OS Administrator’s Guide
53-1002920-02

303

10

Testing and restoring firmware on Backbones

Testing and restoring firmware on Backbones
This procedure enables you to perform a firmware download on each CP and verify that the
procedure was successful before committing to the new firmware. The old firmware is saved in the
secondary partition of each CP until you enter the firmwareCommit command. If you decide to back
out of the installation prior to the firmware commit, you can enter the firmwareRestore command to
restore the former active Fabric OS firmware image.
The firmwareRestore command can only run if autocommit was disabled during the firmware
download. This command cannot be used to restore SAS and SA images.

NOTE

Brocade recommends that, under normal operating conditions, you maintain the same firmware
version on both CPs, and on both partitions of each CP. This organization enables you to evaluate
firmware before you commit. As a standard practice, do not run mixed firmware levels on CPs.

Testing different firmware versions on Backbones
1. Connect to the Brocade Backbone IP address.
2. Enter the ipAddrShow command and note the address of CP0 and CP1.
3. Enter the haShow command and note which CP is active and which CP is standby. Verify that
both CPs are in sync.
4. Enter the firmwareShow command and confirm that the current firmware on both partitions on
both CPs is listed as expected.
5. Exit the session.
6. Update the firmware on the standby CP.
a.

Connect to the Backbone and log in as admin to the standby CP.

b.

Enter the firmwareDownload -s command and respond to the prompts.
At this point, the firmware downloads to the standby CP only. When it has completed the
download to that CP, reboot it. The current Backbone session is disconnected.

7.

Fail over to the standby CP.
a.

Connect to the Backbone on the active CP.

b.

Enter the haShow command to verify that HA synchronization is complete. It takes a
minute or two for the standby CP to reboot and synchronize with the active CP.

c.

Enter the firmwareShow command to confirm that the primary partition of the standby CP
contains the new firmware.

d.

Enter the haFailover command. The active CP reboots and the current Backbone session
is disconnected.
If an AP blade is present: At the point of the failover, an autoleveling process is activated.
Refer to “Backbone firmware download process overview” on page 296 for details about
autoleveling.

304

Fabric OS Administrator’s Guide
53-1002920-02

Testing and restoring firmware on Backbones

10

8. Verify the failover.
a.

Connect to the Backbone on the active CP, which is the former standby CP.

b.

Enter the haShow command to verify that the HA synchronization is complete. It takes a
minute or two for the standby CP, which is the old active CP, to reboot and synchronize with
the active CP.

NOTE

If the CPs fail to synchronize, you can still proceed because the version being tested is already
present on the active CP, and subsequent steps ensure that the standby CP is updated to the
same version as the active CP.
c.

Confirm the evaluation version of firmware is now running on the active CP by entering the
firmwareShow command.

9. Update firmware on the standby CP.
a.

Connect to the Backbone on the standby CP, which is the former active CP.

b.

Enter the firmwareDownload command with the -s -b -n operands. This ensures that the
following steps are successful.
At this point the firmware downloads to the standby CP only and reboots it. The current
Backbone session is disconnected.

c.

Wait one minute for the standby CP to reboot, and then connect to the Backbone and log
in as admin.

d.

Enter the firmwareShow command to confirm that both primary partitions have the test
drive firmware in place.
You are now ready to evaluate the new version of firmware.

ATTENTION
Stop! If you want to restore the firmware, stop here and skip ahead to step 12; otherwise,
continue to step 10 to commit the firmware on both CPs, which completes the firmware
download.
10. Perform a commit on the standby CP.
From the current Backbone session on the standby CP, enter the firmwareCommit command to
update the secondary partition with new firmware. It takes several minutes to complete the
commit operation. Do not do anything on the Backbone while this operation is in process.
11. Perform a commit on the active CP.
a.

From the current Backbone session on the active CP, enter the firmwareShow command
and confirm that only the active CP secondary partition contains the old firmware.

b.

Enter the firmwareCommit command to update the secondary partition with the new
firmware. It takes several minutes to complete the commit operation. Do not do anything
on the Backbone while this operation is in process.

c.

Upon completion of the firmwareCommit command, enter the firmwareShow command to
confirm both partitions on both CPs contain the new firmware.

d.

Enter the haShow command to confirm that the HA state is in sync.

Fabric OS Administrator’s Guide
53-1002920-02

305

10

Validating a firmware download

ATTENTION
Stop! If you have completed step 11, then you have committed the firmware on both CPs and
you have completed the firmware download procedure.
12. Restore the firmware on the standby CP.
In the current Backbone session for the standby CP, enter the firmwareRestore command. The
standby CP reboots and the current Backbone session ends. Both partitions have the same
Fabric OS after several minutes.
13. Perform haFailover on the active CP.
a.

In the current Backbone session for the active CP, enter the haShow command to verify
that HA synchronization is complete. It takes a minute or two for the standby CP to reboot
and synchronize with the active CP.

b.

Enter the haFailover command. The active CP reboots and the current Backbone session
ends. The Backbone is now running the original firmware.

14. Restore firmware on the “new” standby CP.
a.

Wait one minute and connect to the Backbone on the new standby CP, which is the former
active CP.

b.

Enter the firmwareRestore command. The standby CP reboots and the current Backbone
session ends. Both partitions have the same Fabric OS after several minutes.

c.

Wait five minutes and log in to the Backbone. Enter the firmwareShow command and
verify that all partitions have the original firmware.
If an AP blade is present: Blade partitions always contain the same version of the firmware
on both partitions. The firmware is stored on the blade’s compact flash card and is always
synchronized with the active CP’s firmware. Thus, if you restore the active CP firmware, the
blade firmware is automatically downloaded (autoleveled) to become consistent with the
new CP firmware (the blade firmware is restored).

Your system is now restored to the original partitions on both CPs. Make sure that servers using the
fabric can access their storage devices.
If you want to upgrade a Backbone with only one CP in it, follow the procedures in “Testing and
restoring firmware on switches” on page 302. Be aware that upgrading a Backbone with only one
CP is disruptive to switch traffic.

Validating a firmware download
Validate the firmware download by running the following commands: firmwareShow,
firmwareDownloadStatus, nsShow, nsAllShow, and fabricShow.
All of the connected servers, storage devices, and switches should be present in the output of
these commands. If there is a discrepancy, it is possible that a device or switch cannot connect to
the fabric and further troubleshooting is necessary.

306

Fabric OS Administrator’s Guide
53-1002920-02

Validating a firmware download

TABLE 58

10

Commands used for validating a firmware download

Command

Description

firmwareShow

Displays the current firmware level on the switch. For Brocade Backbones, this
command displays the firmware loaded on both partitions (primary and secondary) for
both CPs and AP blades. Brocade recommends that you maintain the same firmware
level on both partitions of each CP within the Brocade Backbone. The firmwareShow
command displays the firmware version on each CP.

firmwareDownloadStatus Displays an event log that records the progress and status of events during Fabric OS,
SAS, and SA firmware download. The event log is created by the current
firmwareDownload command and is kept until another firmwareDownload command is
issued. There is a time stamp associated with each event. When downloading SAS or
SA in systems with two control processor (CP) cards, you can only run this command on
the active CP. When downloading Fabric OS, the event logs in the two CPs are
synchronized. This command can be run from either CP.
nsShow

Displays all devices directly connected to the switch that have logged in to the name
server. Make sure the number of attached devices after the firmware download is
exactly the same as the number of attached devices prior to the firmware download.

nsAllShow

Displays all devices connected to a fabric. Make sure the number of attached devices
after the firmware download is exactly the same as the number of attached devices
prior to the firmware download.

fabricShow

Displays all switches in a fabric. Make sure the number of switches in the fabric after
the firmware download is exactly the same as the number of attached devices prior to
the firmware download.

Fabric OS Administrator’s Guide
53-1002920-02

307

10

308

Validating a firmware download

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

11

Managing Virtual Fabrics

In this chapter
• Virtual Fabrics overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Logical switch overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Logical fabric overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Management model for logical switches . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Account management and Virtual Fabrics. . . . . . . . . . . . . . . . . . . . . . . . . .
• Supported platforms for Virtual Fabrics. . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Limitations and restrictions of Virtual Fabrics. . . . . . . . . . . . . . . . . . . . . . .
• Enabling Virtual Fabrics mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Disabling Virtual Fabrics mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Configuring logical switches to use basic configuration values . . . . . . . . .
• Creating a logical switch or base switch . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Executing a command in a different logical switch context . . . . . . . . . . . .
• Deleting a logical switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Adding and moving ports on a logical switch . . . . . . . . . . . . . . . . . . . . . . .
• Displaying logical switch configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Changing the fabric ID of a logical switch . . . . . . . . . . . . . . . . . . . . . . . . . .
• Changing a logical switch to a base switch . . . . . . . . . . . . . . . . . . . . . . . . .
• Setting up IP addresses for a logical switch . . . . . . . . . . . . . . . . . . . . . . . .
• Removing an IP address for a logical switch . . . . . . . . . . . . . . . . . . . . . . . .
• Configuring a logical switch to use XISLs. . . . . . . . . . . . . . . . . . . . . . . . . . .
• Changing the context to a different logical fabric . . . . . . . . . . . . . . . . . . . .
• Creating a logical fabric using XISLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

309
310
315
314
319
320
322
324
325
326
326
328
329
329
330
331
331
333
333
333
334
334

Virtual Fabrics overview
Virtual Fabrics is an architecture to virtualize hardware boundaries. Traditionally, SAN design and
management is done at the granularity of a physical switch. Virtual Fabrics allows SAN design and
management to be done at the granularity of a port.
Virtual Fabrics is a suite of related features that can be customized based on your needs.
The Virtual Fabrics suite consists of the following specific features:

• Logical switch
• Logical fabric
• Device sharing
Fabric OS Administrator’s Guide
53-1002920-02

309

11

Logical switch overview

This chapter describes the logical switch and logical fabric features. For information about device
sharing with Virtual Fabrics, refer to “FC-FC routing and Virtual Fabrics” on page 636.
For information about supported switches and port types, refer to “Supported platforms for
Virtual Fabrics” on page 320.
Virtual Fabrics and Admin Domains are mutually exclusive and are not supported at the same time
on a switch.

NOTE

A note on terminology: Virtual Fabrics is the name of the suite of features. A logical fabric is a type
of fabric that you can create using the Virtual Fabrics suite of features.

Logical switch overview
Traditionally, each switch and all the ports in the switch act as a single Fibre Channel switch (FC
switch) that participates in a single fabric. The logical switch feature allows you to divide a physical
chassis into multiple fabric elements. Each of these fabric elements is referred to as a logical
switch. Each logical switch functions as an independent self-contained FC switch.

NOTE
Each chassis can have multiple logical switches.

Default logical switch
To use the Virtual Fabrics features, you must first enable Virtual Fabrics on the switch. Enabling
Virtual Fabrics creates a single logical switch in the physical chassis. This logical switch is called
the default logical switch, and it initially contains all of the ports in the physical chassis.
Figure 21 shows a switch before and after enabling Virtual Fabrics. In this example, the switch has
10 ports, labeled P0 through P9.
Before enabling Virtual Fabrics

After enabling Virtual Fabrics

Physical chassis

Physical chassis

P0

P3

P6

P1

P4

P7

P2

P5

P8

FIGURE 21

310

P9

Default logical switch

P0

P3

P6

P1

P4

P7

P2

P5

P8

P9

Switch before and after enabling Virtual Fabrics

Fabric OS Administrator’s Guide
53-1002920-02

Logical switch overview

11

After you enable Virtual Fabrics, you can create up to seven additional logical switches, depending
on the switch model.
Figure 22 shows a Virtual Fabrics-enabled switch before and after it is divided into logical switches.
Before you create logical switches, the chassis appears as a single switch (default logical switch).
After you create logical switches, the chassis appears as multiple independent logical switches. All
of the ports continue to belong to the default logical switch until you explicitly move them to other
logical switches.
The default logical switch always exists. You can add and delete other logical switches, but you
cannot delete the default logical switch unless you disable Virtual Fabrics.
Before logical switch creation

After logical switch creation

Physical chassis

Logical switch 1
(Default logical switch)
P0

P2

P4

P6

P8

P1

P3

P5

P7

P9

Default logical switch

P0

P3

P6

P1

P4

P7

P2

P5

P8

P9

Logical switch 2

Logical switch 3

Logical switch 4

FIGURE 22

Switch before and after creating logical switches

Logical switches and fabric IDs
When you create a logical switch, you must assign it a fabric ID (FID). The fabric ID uniquely
identifies each logical switch within a chassis and indicates to which fabric the logical switch
belongs. You cannot define multiple logical switches with the same fabric ID within the chassis.
In Figure 23 on page 312, logical switches 2, 3, 4, and 5 are assigned FIDs of 1, 15, 8, and 20,
respectively. These logical switches belong to different fabrics, even though they are in the same
physical chassis. For example, you could not assign logical switch 5 a fabric ID of 15, because
logical switch 3 is already assigned FID 15 in the chassis.
The default logical switch is initially assigned FID 128. You can change this value later.

NOTE

Each logical switch is assigned one and only one FID. The FID identifies the logical fabric to which
the logical switch belongs.

Fabric OS Administrator’s Guide
53-1002920-02

311

11

Logical switch overview

Physical chassis
Logical switch 1
(Default logical switch)
(FID = 128)

Logical switch 2
(FID = 1)

Logical switch 3
(FID = 15)

Logical switch 4
(FID = 8)

Logical switch 5
(FID = 20)

FIGURE 23

Fabric IDs assigned to logical switches

Port assignment in logical switches
Initially, all ports belong to the default logical switch. When you create additional logical switches,
they are empty and you must assign ports to those logical switches. As you assign ports to a logical
switch, the ports are moved from the default logical switch to the newly created logical switch.
A given port can be in only one logical switch.
In Figure 24, the default logical switch initially has 10 ports, labeled P0 through P9. After logical
switches are created, the ports are assigned to specific logical switches. Note that ports 0, 1, 7,
and 8 have not been assigned to a logical switch and so remain assigned to the default logical
switch.
Before port assignment

After port assignment

Logical switch 1
(Default logical switch)

Logical switch 1
(Default logical switch)

P0

P2

P4

P6

P8

P1

P3

P5

P7

P9

P0

P1

P7

P8

P2
Logical switch 2

Logical switch 2
P3

P4
Logical switch 3

P9
Logical switch 3

P5

P6
Logical switch 4

FIGURE 24

312

Logical switch 4

Assigning ports to logical switches

Fabric OS Administrator’s Guide
53-1002920-02

Logical switch overview

11

A given port is always in one (and only one) logical switch. The following scenarios refer to the
chassis after port assignment in Figure 24:

• If you assign P2 to logical switch 2, you cannot assign P2 to any other logical switch.
• If you want to remove a port from a logical switch, you cannot delete it from the logical switch,
but must move it to a different logical switch. For example, if you want to remove P4 from
logical switch 3, you must assign it to a different logical switch: logical switch 2, logical
switch 4, or logical switch 1 (the default logical switch).

• If you assign a port to a logical switch, it is removed automatically from the logical switch it is
currently in. If you assign P3 to Logical switch 3, P3 is automatically removed from logical switch 2.

• If you do not assign a port to any logical switch, it remains in the default logical switch, as is the
case with ports 0, 1, 7, and 8.
Refer to “Adding and moving ports on a logical switch” on page 329 for instructions for assigning
and moving ports on logical switches.
A logical switch can have as many ports as are available in the chassis. In Figure 24, the chassis
has 10 ports. You could assign all 10 ports to a single logical switch, such as logical switch 2; if you
did this, however, no ports would be available for logical switches 3 and 4.
You can move only F_Ports and E_Ports from one logical switch to another. If you want to configure
a different type of port, such as a VE_Port or EX_Port, you must configure them after you move
them. Some types of ports cannot be moved from the default logical switch. Refer to “Supported
platforms for Virtual Fabrics” on page 320 for detailed information about these ports.

Logical switches and connected devices
You can connect devices to logical switches, as shown in Figure 25 on page 314. In logical
switch 2, P2 is an F_Port that is connected to H1. In logical switch 3, P4 is an F_Port that is
connected to D1. H1 and D1 cannot communicate with each other because they are in different
fabrics, even though they are both connected to the same physical chassis.
You can also connect other switches to logical switches. In Figure 25, P6 is an E_Port that forms an
inter-switch link (ISL) between logical switch 4 and the non-Virtual Fabrics switch. Logical switch 4
is the only logical switch that can communicate with the non-Virtual Fabrics switch and D2,
because the other logical switches are in different fabrics.

Fabric OS Administrator’s Guide
53-1002920-02

313

11

Management model for logical switches

Physical chassis
Logical switch 1
P1
(Default logical switch)
Fabric ID 128

Logical switch 2
Fabric ID 1

H1

P2
P3

D1
P4
Logical switch 3
Fabric ID 15

Logical switch 4
Fabric ID 8

P5

P6

D2
ISL

Switch

FIGURE 25

Logical switches connected to devices and non-Virtual Fabrics switch

Figure 26 shows a logical representation of the physical chassis and devices in Figure 25. As
shown in Figure 26, the devices are isolated into separate fabrics.
H1
Switch 1

D1
D2

Fabric 128
Switch 4
Switch 2

Fabric 1

FIGURE 26

Switch 3

Fabric 15

Fabric 8

Logical switches in a single chassis belong to separate fabrics

For information on allowing device sharing across fabrics in a Virtual Fabrics environment, refer to
“FC-FC routing and Virtual Fabrics” on page 636.

Management model for logical switches
The operations you can perform on a logical switch depend on the context you are in. Some
operations affect only a single logical switch, and some operations affect the entire physical
chassis.

314

Fabric OS Administrator’s Guide
53-1002920-02

Logical fabric overview

11

All user operations are classified into one of the following:

• Chassis management operations
These are operations that span logical switch boundaries, such as:

-

Logical switch configuration (creating, deleting, or modifying logical switches)
Account management (determining which accounts can access which logical switches)
Field-replaceable unit (FRU) management (slot commands, such as slotShow)
Firmware management (firmware upgrade, HA failover)

• Logical switch operations
These are operations that are limited to the logical switch, such as displaying or changing port
states. Logical switch operations include all operations that are not covered in the chassis
management operations.
When you log in, you are assigned an active context, or active logical switch. This context filters the
view that you get, and determines which ports you can see. You can change the active context. For
example, if you are working with logical switch 1, you can change the context to logical switch 5.
When you change the context to logical switch 5, you only see the ports that are assigned to that
logical switch. You do not see any of the other ports in the chassis.
The scope of logical switch operations is defined by the active context. When you are in the context
of a logical switch, you can perform port, switch, and fabric-level operations, subject to Role-Based
Access Control (RBAC) rules.
If you have permission to execute chassis-level commands, you can do so, regardless of which
logical switch context you are in.

Logical fabric overview
A logical fabric is a fabric that contains at least one logical switch. The four fabrics shown in
Figure 25 and Figure 26 are logical fabrics because they each have at least one logical switch.
You can connect logical switches to non-Virtual Fabrics switches and to other logical switches.You
connect logical switches to non-Virtual Fabrics switches using an ISL, as shown in Figure 25.
You connect logical switches to other logical switches in two ways:

• Using ISLs
• Using base switches and extended ISLs (XISLs)

Logical fabric and ISLs
Figure 27 shows two physical chassis divided into logical switches. In Figure 27, ISLs are used to
connect the logical switches with FID 1 and the logical switches with FID 15. The logical switches
with FID 8 are each connected to a non-Virtual Fabrics switch. The two logical switches and the
non-Virtual Fabrics switch are all in the same fabric, with FID 8.

Fabric OS Administrator’s Guide
53-1002920-02

315

11

Logical fabric overview

Physical chassis 2

Physical chassis 1
P1
Logical switch 1
(Default logical switch)
Fabric ID 128

P1

P2

P2

P3

P3

P4

P5

Logical switch 2
Fabric ID 1

Logical switch 3
Fabric ID 15

Logical switch 4
Fabric ID 8

P5

P6

P6

P8

Logical switch 5
(Default logical switch)
Fabric ID 128

Logical switch 6
Fabric ID 1

P4

P7
Logical switch 7
Fabric ID 15

Logical switch 8
Fabric ID 8

P9

Switch

FIGURE 27

Logical switches connected to other logical switches through physical ISLs

Figure 28 shows a logical representation of the configuration in Figure 27.
Fabric 15

Fabric 128
SW3

SW1

SW7

SW5

Fabric 8
Fabric 1

SW4

SW2
SW8
SW6

FIGURE 28

Logical switches connected to form logical fabrics

The ISLs between the logical switches are dedicated ISLs because they carry traffic only for a single
logical fabric. In Figure 27, Fabric 128 has two switches (the default logical switches), but they
cannot communicate with each other because they have no ISLs between them and they cannot
use the ISLs between the other logical switches.

NOTE
Only logical switches with the same FID can form a fabric. If you connect two logical switches with
different FIDs, the link between the switches segments.

Base switch and extended ISLs
Another way to connect logical switches is to use extended ISLs and base switches.

316

Fabric OS Administrator’s Guide
53-1002920-02

Logical fabric overview

11

When you divide a chassis into logical switches, you can designate one of the switches to be a base
switch. A base switch is a special logical switch that is used for interconnecting the physical
chassis. A base switch has the following properties:

• ISLs connected through the base switch can be used for communication among the other
logical switches.

• Base switches do not support direct device connectivity. A base switch can have only E_Ports,
VE_Ports, EX_Ports, or VEX_Ports, but no F_Ports.

• The base switch provides a common address space for communication between different
logical fabrics.

• A base switch can be configured for the preferred domain ID just like a non-Virtual Fabrics
switch.

• You can have only one base switch in a physical chassis.
A base switch can be connected to other base switches through a special ISL, called a shared ISL
or extended ISL (XISL). An extended ISL connects base switches. The XISL is used to share traffic
among different logical fabrics.
Fabric formation across an XISL is based on the FIDs of the logical switches.
Figure 29 shows two physical chassis divided into logical switches. Each chassis has one base
switch. An ISL connects the two base switches. This ISL is an extended ISL (XISL) because it
connects base switches.
Physical chassis 1

Physical chassis 2

P1
Logical switch 1
(Default logical switch)
Fabric ID 128

P1

P2

P2

Logical switch 2
Fabric ID 1

Logical switch 5
(Default logical switch)
Fabric ID 128

Logical switch 6
Fabric ID 1

P4

P7
Logical switch 3
Fabric ID 15

P6

P5
XISL

Base switch
Fabric ID 8

FIGURE 29

P6

P8

Logical switch 7
Fabric ID 15

Base switch
Fabric ID 8

P9

Base switches connected by an XISL

Traffic between the logical switches can now flow across this XISL. The traffic can flow only
between logical switches with the same fabric ID. For example, traffic can flow between logical
switch 2 in chassis 1 and logical switch 6 in chassis 2, because they both have FID 1. Traffic cannot
flow between logical switch 2 and logical switch 7, because they have different fabric IDs (and are
thus in different fabrics).
Think of the logical switches as being connected with logical ISLs, as shown in Figure 30. In this
diagram, the logical ISLs are not connected to ports because they are not physical cables. They are
a logical representation of the switch connections that are allowed by the XISL.

Fabric OS Administrator’s Guide
53-1002920-02

317

11

Logical fabric overview

FIGURE 30

Logical ISLs connecting logical switches

To use the XISL, the logical switches must be configured to allow XISL use. By default, they are
configured to do so; you can change this setting, however, using the procedure described in
“Configuring a logical switch to use XISLs” on page 333.

NOTE

It is a good practice to configure at least two XISLs, for redundancy.
You can also connect logical switches using a combination of ISLs and XISLs, as shown in
Figure 31. In this diagram, traffic between the logical switches in FID 1 can travel over either the
ISL or the XISL. Traffic between the other logical switches travels only over the XISL.

FIGURE 31

Logical fabric using ISLs and XISLs

By default, the physical ISL path is favored over the logical path (over the XISL) because the
physical path has a lower cost. This behavior can be changed by configuring the cost of the
dedicated physical ISL to match the cost of the logical ISL.

318

Fabric OS Administrator’s Guide
53-1002920-02

Account management and Virtual Fabrics

11

ATTENTION
If you disable a base switch, all of the logical ISLs are broken and the logical switches cannot
communicate with each other unless they are connected by a physical ISL.

Base fabric
Base switch ports on different chassis can be connected together to form a fabric, called a base
fabric. Similar to other logical switches, the base switches must have the same FID to be
connected. If the base switches have different FIDs, the link between the switches is disabled.
The base fabric follows normal routing policies. As long as physical connectivity is available, the
base fabric maintains connectivity for the logical fabrics.

Logical ports
As shown in Figure 31, logical ISLs are formed to connect logical switches. A logical port represents
the ports at each end of a logical ISL. A logical port is a software construct only and does not
correspond to any physical port.
Most port commands are not supported on logical ports. For example, you cannot change the state
or configuration of a logical port.
The World Wide Name (WWN) for logical ports is in NAA=5 format, using the following syntax:
5n:nn:nn:nz:zz:zz:zx:xx
The NAA=5 syntax uses the following variables:

• nnnnnn is the Brocade Organizationally Unique Identifier (OUI).
• zzzzzz is the logical fabric serial number.
• xxx is the logical port number, in the range 0 through FFF.

Logical fabric formation
Fabric formation is not based on connectivity, but on the FIDs of the logical switches. The basic
order of fabric formation is as follows:
1. Base fabric forms.
2. Logical fabrics form when the base fabric is stable.
3. Traffic is initiated between the logical switches.
4. Devices begin recognizing one another.

Account management and Virtual Fabrics
When user accounts are created, they are assigned a list of logical fabrics to which they can log in and
a home logical fabric (home FID). When you connect to a physical chassis, the home FID defines the
logical switch to which you are logged in by default. You can change to a different logical switch
context, as described in “Changing the context to a different logical fabric” on page 334.

Fabric OS Administrator’s Guide
53-1002920-02

319

11

Supported platforms for Virtual Fabrics

When you are logged in to a logical switch, the system prompt changes to display the FID of that
switch. The following are example prompts for when you are logged in to the default logical switch
(FID = 128) and a user-defined logical switch (FID = 15):
switch:FID128:admin>
switch:FID15:admin>

Refer to Chapter 6, “Managing User Accounts,” for information about creating user accounts and
assigning FIDs to user accounts.

Supported platforms for Virtual Fabrics
The following platforms are Virtual Fabrics-capable:

•
•
•
•
•
•
•
•
•

Brocade 5100
Brocade 5300
Brocade 6510
Brocade 6520
Brocade 7800
Brocade VA-40FC, in Native mode only
Brocade DCX
Brocade DCX-4S
Brocade DCX 8510 family

Some restrictions apply to the ports, depending on the port type and blade type. The following
sections explain these restrictions.

Supported port configurations in the fixed-port switches
There are no restrictions on the ports in the Brocade 5100, 5300, 6510, 6520, and VA-40FC;
however, the following rules apply:

• Any port can belong to any logical switch (including the base switch and default logical switch),
with the exception that F_Ports cannot belong to the base switch.

• The default logical switch can use XISLs, except on Brocade DCX or DCX-4S devices.
• The default logical switch can also be a base switch.

Restrictions on fixed-port switches
Brocade 7800— Although it can be divided into four logical switches, you cannot use an XISL on
this switch because a base switch is not supported on this device.

320

Fabric OS Administrator’s Guide
53-1002920-02

Supported platforms for Virtual Fabrics

11

Supported port configurations in Brocade Backbones
Some of the ports in the Brocade DCX and DCX 8510 Backbone families are not supported on all
types of logical switches. Table 59 lists the blades and ports that are supported on each type of
logical switch.

TABLE 59

Blade and port types supported on logical switches

Blade type

Default logical switch

User-defined logical switch

Base switch

FC8-16
FC8-32
FC8-32E
FC8-48
FC8-48E
FC16-32
FC16-48

Yes (F, E)

Yes (F, E)

Yes (E, EX)

FC8-64

Yes (F, E)1

Yes (F, E)

Yes (E, EX)2

FS8-18

Yes (F, E)

No

No

FCOE10-24

Yes (F, E)

No

No

FX8-24: FC ports
GE ports

Yes (F, E)
Yes (VE)

Yes (F, E,)
Yes (VE)

Yes (E, EX)
Yes (VE, VEX)

ICL ports

Yes (E)

Yes (E)

Yes (E, EX)3

1.

In the Brocade DCX and DCX 8510-8, ports 56–63 of the FC8-64 blade are not supported as E_Ports on the
default logical switch. The Brocade DCX-4S and DCX 8510-4 do not have this limitation.

2.

In the Brocade DCX and DCX 8510-8, ports 48–63 of the FC8-64 blade are not supported in the base switch.
The Brocade DCX-4S and DCX 8510-4 do not have this limitation.

3.

EX_Ports on an ICL are supported only in the Brocade DCX 8510 Backbone family.

Restrictions on Brocade Backbones
The following restrictions apply to Brocade Backbones:

• EX_Ports and VEX_Ports can be in only the base switch.
• ICL ports cannot be in a logical switch that is using XISLs.
• All of the user ports in an ICL cable must be in the same logical switch. Distributing the user
ports within the same cable across multiple logical switches is not supported.

•
•
•
•

ICL ports that are configured as EX_Ports can be in only the base switch.
The default logical switch cannot use XISLs.
The default logical switch cannot be designated as the base switch.
In Fabric OS v7.0.0 and later, VE_Ports on the FX8-24 blade are supported on a logical switch
that is using an XISL, and on the base switch as an XISL.

NOTE

For the FX8-24 blade, if XISL use is enabled it is not recommended that you configure VE_Ports
on both the logical switch and the base switch, because FCIP tunnels support only two hops
maximum.

Fabric OS Administrator’s Guide
53-1002920-02

321

11

Limitations and restrictions of Virtual Fabrics

Virtual Fabrics interaction with other Fabric OS features
Table 60 lists some Fabric OS features and considerations that apply when using Virtual Fabrics.

TABLE 60

Virtual Fabrics interaction with Fabric OS features

Fabric OS feature

Virtual Fabrics interaction

Access Gateway

Virtual Fabrics is not supported on a switch if AG mode is enabled.

Admin Domains

Virtual Fabrics and Admin Domains are mutually exclusive and are not supported at the
same time on a switch. To use Admin Domains, you must first disable Virtual Fabrics; to
use Virtual Fabrics, you must first delete all Admin Domains.
Refer to “Deleting all user-defined Admin Domains non-disruptively” on page 502 for
information on deleting Admin Domains without disrupting device-to-device
communication.

Configuration upload
and download

Virtual Fabrics uses a configuration file that is different from the configuration file used
to download system configuration parameters. Refer to Chapter 9, “Maintaining the
Switch Configuration File,” for more information about how Virtual Fabrics affects the
configuration file.

Encryption

Encryption functionality using the FS8-18 blade is available only on the default logical
switch.

FC-FC Routing Service

All EX_Ports must reside in a base switch.
You cannot attach EX_Ports to a logical switch that has XISL use enabled. You must use
ISLs to connect the logical switches in an edge fabric.
Refer to Chapter 26, “Using FC-FC Routing to Connect Fabrics,” for more information
about Virtual Fabrics and FC-FC routing.

FICON

Up to two logical switches per chassis can run FICON Management Server (CUP), but the
FICON logical switch can use both ISLs and XISLs.

Licensing

Licenses are applicable for all logical switches in a chassis.

Performance
monitoring

Performance monitors are supported in a limited number of logical switches, depending
on the platform type. Refer to Chapter 23, “Monitoring Fabric Performance,” for more
information about performance monitoring when Virtual Fabrics is enabled.

QoS

QoS VCs are maintained across the base fabric. Refer to Chapter 14, “Optimizing Fabric
Behavior,” for more information about using the Adaptive Networking features with
Virtual Fabrics.

Traffic Isolation

Traffic Isolation zones with failover disabled are not supported in logical fabrics. Refer to
Chapter 14, “Optimizing Fabric Behavior,” for additional information about using TI Zones
with Virtual Fabrics.

Limitations and restrictions of Virtual Fabrics
Before you use the Virtual Fabrics feature, you should be aware of the restrictions and limitations
regarding QSFP ports and the maximum number of logical switches per chassis.
All four ports belonging to a QSFP must be moved to the same logical switch.

322

Fabric OS Administrator’s Guide
53-1002920-02

Limitations and restrictions of Virtual Fabrics

11

The maximum number of logical switches per chassis varies depending on the switch model.
Table 61 lists the supported platforms and the maximum number of logical switches (including the
default logical switch) supported on each.

TABLE 61

Maximum number of logical switches per chassis

Platform

Maximum number of logical switches

Brocade DCX

8

Brocade DCX-4S

8

Brocade DCX 8510 family

8

Brocade 5300

4

Brocade 5100

3

Brocade 6510

4

Brocade 6520

4

Brocade 7800

4

Brocade VA-40FC

3

Refer to “Supported port configurations in Brocade Backbones” on page 321 for restrictions on the
default logical switch.
If a blade slot is being decommissioned and has ports configured in logical switches, it is
recommended that the logical port assignments be removed from that blade before removing the
blade. This ensures a seamless transition for any new port or AP blade that might occupy that slot
in the future. This does not apply if you are simply replacing a blade of the same type.

Restrictions on XISLs
The Allow XISL Use option under the configure command, allows a logical switch to use XISLs in the
base switch as well as any standard ISLs that are connected to that logical switch. To allow or
disallow XISL use for a logical switch, see “Configuring a logical switch to use XISLs” on page 333.
The following restrictions apply to XISL use. XISL use is not permitted in any of the following
scenarios:

•
•
•
•

The logical switch has ICL ports.
The logical switch is the default logical switch in the Brocade DCX, DCX-4S, or DCX 8510 family.
The logical switch is a base switch.
The logical switch is an edge switch for an FC router.
In this case, if the logical switch is enabled, you cannot allow XISL use. If the logical switch is
disabled or has not yet joined the edge fabric, you can allow XISL use; however, fabric
segmentation occurs when the logical switch is enabled or is connected to an edge fabric.

NOTE

Using XISL and fmsmode at the same time is permitted, but this combination will only work in a
one-hop topology.

Fabric OS Administrator’s Guide
53-1002920-02

323

11

Enabling Virtual Fabrics mode

Restrictions on moving ports
The following are restrictions on moving ports among logical switches:

• FC ports cannot be moved if any one of the following features is enabled:
- Long distance
- QoS
- F_Port buffers
- F_Port trunking
• Before moving VE_Ports, you must remove the VE_Port tunnel configuration.
• VE_Ports on the FX8-24 blade can be moved to any logical switch independent of the location
of the physical GE port.

• If you move existing EX_Ports or VEX_Ports to any logical switch other than the base switch,
these ports are automatically disabled.

Enabling Virtual Fabrics mode
A fabric is said to be in Virtual Fabrics mode (VF mode) when the Virtual Fabrics feature is enabled.
Before you can use the Virtual Fabrics features, such as logical switch and logical fabric, you must
enable VF mode.
VF mode is enabled by default.

NOTE
When you enable VF mode, the control processors (CPs) are rebooted and all EX_Ports are disabled
after the reboot.
Use the following procedure to enable Virtual Fabrics mode:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Use the fosConfig command to check whether VF mode is enabled:
fosconfig --show

3. Delete all Admin Domains, as described in “Deleting all user-defined Admin Domains
non-disruptively” on page 502.
4. Use the fosConfig command to enable VF mode:
fosconfig --enable vf

5. Enter y at the prompt.
Example

The following example checks whether VF mode is enabled or disabled and then enables it.
switch:admin> fosconfig --show
FC Routing service:
iSCSI service:
iSNS client service:
Virtual Fabric:
Ethernet Switch Service:

324

disabled
Service not supported on this Platform
Service not supported on this Platform
disabled
Service not supported on this Platform

Fabric OS Administrator’s Guide
53-1002920-02

Disabling Virtual Fabrics mode

11

switch:admin> fosconfig --enable vf
WARNING: This is a disruptive operation that requires a reboot to take
effect.
All EX ports will be disabled upon reboot.
Would you like to continue [Y/N] y
VF has been enabled. Your system is being rebooted.

Disabling Virtual Fabrics mode
When you disable VF mode, the following occurs:

• The CPs are rebooted.
• If F_Port trunking is enabled on ports in the default switch, the F_Port trunking information is
deleted.

ATTENTION
If you want to use Admin Domains in a fabric, you must first disable VF mode.
Use the following procedure to disable Virtual Fabrics mode:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Use the fosConfig command to check whether VF mode is disabled:
fosconfig --show

3. Move all ports to the default logical switch.
lscfg --config 128 -slot slot -port port

4. Delete all of the non-default logical switches.
lscfg --delete fabricID

5. Use the fosConfig command to disable VF mode:
fosconfig --disable vf

6. Enter y at the prompt.
Example

The following example checks whether VF mode is enabled or disabled and then disables it.
switchA:FID128:admin> fosconfig
FC Routing service:
iSCSI service:
iSNS client service:
Virtual Fabric:
Ethernet Switch Service

--show
disabled
Service not supported on this Platform
Service not supported on this Platform
enabled
Service not supported on this Platform

switch:admin> fosconfig --disable vf
WARNING: This is a disruptive operation that requires a reboot to take
effect.
Would you like to continue [Y/N] y

Fabric OS Administrator’s Guide
53-1002920-02

325

11

Configuring logical switches to use basic configuration values

Configuring logical switches to use basic configuration values
All switches in the fabric are configured to use the same basic configuration values. When you
create logical switches, the logical switches might have different configuration values than the
default logical switch. Use the following procedure to ensure that newly created logical switches
have the same basic configuration values as the default logical switch.

NOTE

For most users, you do not need to run this procedure. Contact your switch service provider to
determine if you need to use this procedure.
You need to run this procedure only once on each chassis, after you enable Virtual Fabrics but
before you create logical switches. The configuration settings are then preserved across reboots
and firmware upgrades and downgrades.
Use the following procedure to configure logical switches to use basic configuration values:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Enter the configureChassis command to ensure that newly created logical switches have the
same basic configuration values as the default logical switch:
configurechassis

3. Enter n at the prompts to configure system and cfgload attributes. Enter y at the prompt to
configure custom attributes.
System (yes, y, no, n): [no] n
cfgload attributes (yes, y, no, n): [no] n
Custom attributes (yes, y, no, n): [no] y

4. Enter the appropriate value at the Config Index prompt. Contact your switch service provider to
determine the appropriate value.
Config Index (0 to ignore): (0..1000) [3]:

Creating a logical switch or base switch
When the logical switch is created, it is automatically enabled and is empty—that is, it does not
have any ports. After creating the logical switch, you must disable the switch to configure it and set
the domain ID. You then assign ports to the logical switch.
Optionally, you can define the logical switch to be a base switch. Each chassis can have only one
base switch.

NOTE
Domain ID conflicts are detected before fabric ID conflicts. If you have both a domain ID conflict and
a fabric ID conflict, only the domain ID conflict is reported.
Use the following procedure to create a logical switch or a base switch:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Enter the lsCfg command to create a logical switch:
lscfg --create fabricID [ -base ] [ -force ]

326

Fabric OS Administrator’s Guide
53-1002920-02

Creating a logical switch or base switch

11

In the command syntax, fabricID is the fabric ID that is to be associated with the logical switch.
Specify the -base option if the logical switch is to be a base switch.
Specify the -force option to execute the command without any user prompts or confirmation.
3. Set the context to the new logical switch.
setcontext fabricID (or switchname)

The fabricID parameter is the FID of the logical switch you just created. The switchname
parameter is the name assigned to the logical switch.
You can only use one parameter at a time.
4. Disable the logical switch.
switchdisable

5. Configure the switch attributes, including assigning a unique domain ID.
configure

6. Enable the logical switch.
switchenable

7.

Assign ports to the logical switch, as described in “Adding and moving ports on a logical
switch” on page 329.

Example

The following example creates a logical switch with FID 4, and then assigns domain ID 14 to it.
sw0:FID128:admin> lscfg --create 4
A Logical switch with FID 4 will be created with default configuration.
Would you like to continue [y/n]?:y
About to create switch with fid=4. Please wait...
Logical Switch with FID (4) has been successfully created.
Logical Switch has been created with default configurations.
Please configure the Logical Switch with appropriate switch
and protocol settings before activating the Logical Switch.
sw0:FID128:admin> setcontext 4
switch_4:FID4:admin> switchdisable
switch_4:FID4:admin> configure
Configure...
Fabric parameters (yes, y, no, n): [no] yes
Domain: (1..239) [1] 100
Select Addressing Mode:
(1 = Zero Based Area Assignment,
2 = Port Based Area Assignment): (1..2) [2] 2
WWN Based persistent PID (yes, y, no, n): [no]
(output truncated)
WARNING: The domain ID will be changed. The port level zoning may be affected
switch_4:FID4:admin> switchenable

Fabric OS Administrator’s Guide
53-1002920-02

327

11

Executing a command in a different logical switch context

Executing a command in a different logical switch context
This procedure describes how to execute a command for a logical switch while you are in the
context of a different logical switch. You can also execute a command for all the logical switches in
a chassis.
The command is not executed on those logical switches for which you do not have permission.
Use the following procedure to execute a command in a different logical switch context:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Enter one of the following commands:

• To execute a command in a different logical switch context:
fosexec --fid fabricID -cmd "command"

• To execute the command on all logical switches:
fosexec --fid all -cmd "command"

Example 1: Executing the switchShow command in a different logical switch context
sw0:FID128:admin> fosexec --fid 4 -cmd "switchshow"
--------------------------------------------------"switchshow" on FID 4:
switchName:
switchType:
switchState:
switchMode:
switchRole:
switchDomain:
switchId:
switchWwn:
zoning:
switchBeacon:
FC Router:
Fabric Name:
Allow XISL Use:
LS Attributes:

switch_4
66.1
Online
Native
Principal
14
fffc0e
10:00:00:05:1e:82:3c:2b
OFF
OFF
OFF
Fab4
ON
[FID: 4, Base Switch: No, Default Switch: No, Address Mode 0]

Index Port Address Media Speed State
Proto
==============================================
22 22
0e1600
-N8
No_Module
FC Disabled
23 23
0e1700
-N8
No_Module
FC Disabled

Example 2: Executing the fabricShow command on all logical switches
sw0:FID128:admin> fosexec --fid all -cmd "fabricshow"
--------------------------------------------------"fabricshow" on FID 128:
Switch ID
Worldwide Name
Enet IP Addr
FC IP Addr
Name
------------------------------------------------------------------------97: fffc61 10:00:00:05:1e:82:3c:2a 10.32.79.105
0.0.0.0
>"sw0"
--------------------------------------------------"fabricshow" on FID 4:

328

Fabric OS Administrator’s Guide
53-1002920-02

Deleting a logical switch

11

Switch ID
Worldwide Name
Enet IP Addr
FC IP Addr
Name
------------------------------------------------------------------------14: fffc0e 10:00:00:05:1e:82:3c:2b 10.32.79.105
0.0.0.0
>"switch_4"
(output truncated)

Deleting a logical switch
The following rules apply to deleting a logical switch:

• You must remove all ports from the logical switch before deleting it.
• You cannot delete the default logical switch.
NOTE

If you are in the context of the logical switch you want to delete, you are automatically logged out
when the fabric ID changes. To avoid being logged out, make sure you are in the context of a different
logical switch from the one you are deleting.
Use the following procedure to delete a logical switch:
1. Connect to the physical chassis and log in using an account with admin permissions.
2. Remove all ports from the logical switch as described in “Adding and moving ports on a logical
switch.”
3. Enter the lsCfg command to delete the logical switch:
lscfg --delete fabricID

The fabricID parameter is the fabric ID of the logical switch to be deleted.
Example of deleting the logical switch with FID 7
switch_4:FID4:admin> lscfg --delete 7
A Logical switch with FID 7 will be deleted.
Would you like to continue [y/n]?:y
All active login sessions for FID 7 have been terminated.
Switch successfully deleted.

Adding and moving ports on a logical switch
This procedure explains how to add and move ports on logical switches.
You add ports to a logical switch by moving the ports from one logical switch to another. See
“Supported platforms for Virtual Fabrics” on page 320 for port restrictions.
If you want to remove a port from a logical switch, you cannot remove it from the logical switch; you
must move the port to a different logical switch.
When you move a port from one logical switch to another, the port is automatically disabled. Any
performance monitors that were installed on the port are deleted. If monitors are required on the
port in its new location, you must manually reinstall them on the port after the move.
Notes
• If the logical switch to which the port is moved has fabric mode Top Talkers enabled, then if the
port is an E_Port, fabric mode Top Talker monitors are automatically installed on that port.

Fabric OS Administrator’s Guide
53-1002920-02

329

11

Displaying logical switch configuration

• If you are deploying ICLs in the base switch, all ports associated with those ICLs must be
assigned to the base switch. If you are deploying ICLs to connect to default switches (that is,
XISL use is not allowed), the ICL ports should be assigned (or left) in the default logical switch.
Use the following procedure to add or move ports on a logical switch:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Enter the lsCfg command to move ports from one logical switch to another:
lscfg --config fabricID -slot slot -port port

The ports are assigned to the logical switch specified by fabricID and are removed from the
logical switch on which they are currently configured.
If the -port option is omitted, all ports on the specified slot are assigned to the logical switch.

NOTE

On the Brocade DCX and DCX 8510-8, the lscfg command does not allow you to add ports 48–
63 of the FC8-64 blade to the base switch. These ports are not supported on the base switch.
The Brocade DCX-4S and DCX 8510-4 do not have this limitation.
3. Enter y at the prompt.
The ports are automatically disabled, then removed from their current logical switch, and
assigned to the logical switch specified by fabricID.
Example of assigning ports 18 through 20 to the logical switch with FID 5
sw0:FID128:admin> lscfg --config 5 -port 18-20
This operation requires that the affected ports be disabled.
Would you like to continue [y/n]?: y
Making this configuration change. Please wait...
Configuration change successful.
Please enable your ports/switch when you are ready to continue.

Displaying logical switch configuration
Use the following procedure to display the configuration for a logical switch:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Enter the lsCfg command to display a list of all logical switches and the ports assigned to them:
lscfg --show [ -provision ]

If the -provision option is specified, all ports on all slots are displayed, regardless of the slot
status.
Example displaying a list of all of the logical switches and the ports assigned to them
sw0:FID128:admin> lscfg --show
Created switches:

128(ds)

4

5

Port
0
1
2
3
4
5
6
7
8
9
------------------------------------------------------------------FID
128 | 128 | 128 | 128 | 128 | 128 | 128 | 128 |
5 |
5 |
(output truncated)

330

Fabric OS Administrator’s Guide
53-1002920-02

Changing the fabric ID of a logical switch

11

Changing the fabric ID of a logical switch
The following procedure describes how you can change the fabric ID of an existing logical switch.
The fabric ID indicates in which fabric the logical switch participates. By changing the fabric ID, you
are moving the logical switch from one fabric to another.
Changing the fabric ID requires permission for chassis management operations. You cannot
change the FID of your own logical switch context.

NOTE

If you are in the context of the logical switch with the fabric ID you want to change, you are
automatically logged out when the fabric ID changes. To avoid being logged out, make sure you are
in the context of a different logical switch from the one with the fabric ID you are changing.
Use the following procedure to change the fabric ID of a logical switch:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the lsCfg command to change the fabric ID of a logical switch:
lscfg --change fabricID -newfid newFID

3. Enter y at the prompt.
4. Enable the logical switch.
fosexec --fid newFID -cmd "switchenable"

Example of changing the fabric ID on the logical switch from 5 to 7
sw0:FID128:admin> lscfg --change 5 -newfid 7
Changing of a switch fid requires that the switch be disabled.
Would you like to continue [y/n]?: y
Disabling switch...
All active login sessions for FID 5 have been terminated.
Checking and logging message: fid = 5.
Please enable your switch.
sw0:FID128:admin> fosexec --fid 7 -cmd "switchenable"
--------------------------------------------------"switchenable" on FID 7:

Changing a logical switch to a base switch
Use the following procedure to change a logical switch to a base switch.
1. Connect to the switch and log in using an account with the chassis-role permission.
2. Set the context to the logical switch you want to change, if you are not already in that context.
setcontext fabricID (or switchname)

The fabricID parameter is the FID of the logical switch you want to change to a base switch.
The switchname parameter is the name assigned to the logical switch.
You can only use one parameter at a time.
3. Configure the switch to not allow XISL use, as described in “Configuring a logical switch to use
XISLs” on page 333.

Fabric OS Administrator’s Guide
53-1002920-02

331

11

Changing a logical switch to a base switch

4. Enter the lsCfg command to change the logical switch to a base switch:
lscfg --change fabricID -base

The fabricID parameter is the fabric ID of the logical switch with the attributes you want to
change.
5. Enable the switch.
switchenable

Example of changing the logical switch with FID 7 to a base switch
sw0:FID128:admin> setcontext 7
switch_25:FID7:admin> switchshow
switchName:
switch_25
switchType:
66.1
switchState:
Online
switchMode:
Native
switchRole:
Principal
switchDomain:
30
switchId:
fffc1e
switchWwn:
10:00:00:05:1e:82:3c:2c
zoning:
OFF
switchBeacon:
OFF
FC Router:
OFF
Fabric Name:
MktFab7
Allow XISL Use: ON
LS Attributes: [FID: 7, Base Switch: No, Default Switch: No, Address Mode 0]
(output truncated)
switch_25:FID7:admin> configure
Not all options will be available on an enabled switch.
To disable the switch, use the "switchDisable" command.
Configure...
Fabric parameters (yes, y, no, n): [no] y
WWN Based persistent PID (yes, y, no, n): [no]
Allow XISL Use (yes, y, no, n): [yes] n
WARNING!! Disabling this parameter will cause removal of LISLs to
other logical switches. Do you want to continue? (yes, y, no, n): [no] y
System services (yes, y, no, n): [no]
switch_25:FID7:admin> lscfg --change 7 -base
Creation of a base switch requires that the proposed new base switch on this
system be disabled.
Would you like to continue [y/n]?: y
Disabling the proposed new base switch...
Disabling switch fid 7
Please enable your switches when ready.
switch_25:FID7:admin> switchenable

332

Fabric OS Administrator’s Guide
53-1002920-02

Setting up IP addresses for a logical switch

11

Setting up IP addresses for a logical switch
Each physical chassis has one common IP address that is shared by all of the logical switches in
the chassis. You can also set up individual IPv4 addresses for each logical switch.
IPv4 addresses assigned to individual Virtual Fabrics are assigned to IP over Fibre Channel (IPFC)
network interfaces. In Virtual Fabrics environments, a single chassis can be assigned to multiple
fabrics, each of which is logically distinct and separate from one another. Each IPFC point of
connection to a given chassis needs a separate IPv4 address and prefix to be accessible to a
management host.
For a management host to access logical switches, the host bus adapter (HBA) must be able to
connect with the common, shared IP address and the individual IPv4 addresses configured for
each logical switch.

NOTE

IPv6 is not supported when setting the IPFC interface for Virtual Fabrics.
Use the following procedure to set up IP addresses for a logical switch:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the ipAddrSet -ls command. For the --add parameter, specify the network information in
dotted-decimal notation for the Ethernet IPv4 address with a Classless Inter-Domain Routing
(CIDR) prefix.
The following example sets an IP address for a logical switch in a Virtual Fabric with an FID of
123 in non-interactive mode with the CIDR prefix:
switch:admin> ipaddrset -ls 123 --add 11.1.2.4/24

Removing an IP address for a logical switch
Use the following procedure to delete an IP address for a logical switch:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the ipAddrSet -ls FID - -delete command.
switch:admin> ipaddrset -ls 123 –delete

Configuring a logical switch to use XISLs
When you create a logical switch, it is configured to use XISLs by default. Use the following
procedure to allow or disallow the logical switch to use XISLs in the base fabric.
XISL use is not supported in some cases. See “Limitations and restrictions of Virtual Fabrics” on
page 322 for restrictions on XISL use.
Use the following procedure to configure a logical switch to use XISLs:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Use the setContext command to set the context to the logical switch you want to manage, if you
are not already in that context.

Fabric OS Administrator’s Guide
53-1002920-02

333

11

Changing the context to a different logical fabric

setcontext fabricID (or switchname)

The fabricID parameter is the FID of the logical switch you want to switch to and manage.
The switchname parameter is the name assigned to the logical switch.
You can only use one parameter at a time.
3. Use the switchShow command and check the value of the Allow XISL Use parameter.
4. Enter the configure command:
configure

5. Enter y after the Fabric Parameters prompt:
Fabric parameters (yes, y, no, n): [no] y

6. Enter y at the Allow XISL Use prompt to allow XISL use; enter n at the prompt to disallow XISL
use:
Allow XISL Use (yes, y, no, n): y

7.

Respond to the remaining prompts or press Ctrl-d to accept the other settings and exit.

Changing the context to a different logical fabric
You can change the context to a different logical fabric. Your user account must have permission to
access the logical fabric.
Use the following procedure to change the context to a different logical fabric:
1. Connect to the physical chassis and log in using an account with the chassis-role permission.
2. Use the setContext command to switch to a different logical switch in the chassis:
setcontext fabricID (or switchname)

The fabricID parameter is the FID of the logical switch you want to switch to and manage.
The switchname parameter is the name assigned to the logical switch.
You can only use one parameter at a time.
The fabricID parameter is the fabric ID of the logical switch you want to switch to and manage.
Example of changing the context from FID 128 to FID 4

In this example, notice that the prompt changes when you change to a different logical fabric.
sw0:FID128:admin> setcontext 4
switch_4:FID4:admin>

Creating a logical fabric using XISLs
This procedure describes how to create a logical fabric using multiple chassis and XISLs and refers
to the configuration shown in Figure 32 as an example.

334

Fabric OS Administrator’s Guide
53-1002920-02

Creating a logical fabric using XISLs

FIGURE 32

11

Example of logical fabrics in multiple chassis and XISLs

Use the following procedure to create a logical fabric using XISLs:
1. Set up the base switches in each chassis:
a.

Connect to the physical chassis and log in using an account with the chassis-role
permission.

b.

Enable the Virtual Fabrics feature, if it is not already enabled. See “Enabling
Virtual Fabrics mode” on page 324 for instructions.
Enabling Virtual Fabrics automatically creates the default logical switch, with FID 128. All
ports in the chassis are assigned to the default logical switch.

c.

Create a base switch and assign it a fabric ID that will become the FID of the base fabric.
See “Creating a logical switch or base switch” on page 326 for instructions on creating a
base switch.
For the example shown in Figure 32, you would create a base switch with fabric ID 8.

d.

Assign ports to the base switch, as described in “Adding and moving ports on a logical
switch” on page 329.

e.

Repeat step a through step d in all chassis that are to participate in the logical fabric.

2. Physically connect ports in the base switches to form XISLs.
3. Enable all of the base switches. This forms the base fabric.
4. Configure the logical switches in each chassis:
a.

Connect to the physical chassis and log in using an account with the chassis-role
permission.

b.

Create a logical switch and assign it a fabric ID for the logical fabric. This FID must be
different from the FID in the base fabric. See “Creating a logical switch or base switch” on
page 326 for instructions.

Fabric OS Administrator’s Guide
53-1002920-02

335

11

Creating a logical fabric using XISLs

For the example shown in Figure 32, you would create a logical switch with FID 1 and a
logical switch with FID 15.
c.

Assign ports to the logical switch, as described in “Adding and moving ports on a logical
switch” on page 329.

d.

Physically connect devices and ISLs to these ports on the logical switch.

e.

(Optional) Configure the logical switch to use XISLs, if it is not already XISL-capable. See
“Configuring a logical switch to use XISLs” on page 333 for instructions.
By default, newly created logical switches are configured to allow XISL use.

f.

Repeat step a through step e in all chassis that are to participate in the logical fabric,
using the same fabric ID whenever two switches need to be part of a single logical fabric.

5. Enable all logical switches by entering the switchEnable command on each logical switch that
you created in step 4 (the base switches are already enabled).
The logical fabric is formed.
The fabricShow command displays all logical switches configured with the same fabric ID as the
local switch and all non-Virtual Fabrics switches connected through ISLs to these logical switches.
The switchShow command displays logical ports as E_Ports, with -1 for the slot and the user port
number for the slot port.

336

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

12

Administering Advanced Zoning

In this chapter
• Zone types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zoning overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Broadcast zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zone aliases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zone creation and maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Default zoning mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zone database size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zone configurations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zone object maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zone configuration management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Security and zoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zone merging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Concurrent zone transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

337
338
343
346
350
360
362
362
368
370
371
371
376

Zone types
Zones enable you to partition your fabric into logical groups of devices that can access each other.
These are “regular” or “standard” zones. Unless otherwise specified, all references to zones in this
chapter refer to these regular zones. Beyond this, Fabric OS has the following types of special zones:

• Broadcast zones
Controls which devices receive broadcast frames. A broadcast zone restricts broadcast
packets to only those devices that are members of the broadcast zone. Refer to “Broadcast
zones” on page 343 for more information.

• Frame redirection zones
Re-route frames between an initiator and a target through a Virtual Initiator and Virtual Target
for special processing or functionality, such as for storage virtualization or encryption. Refer to
“Frame Redirection” on page 132 for more information.

• LSAN zones
Provide device connectivity between fabrics without merging the fabrics. Refer to “LSAN zone
configuration” on page 620 for more information.

Fabric OS Administrator’s Guide
53-1002920-02

337

12

Zoning overview

• QoS zones
Assign high or low priority to designated traffic flows. QoS zones are regular zones with
additional QoS attributes specified by adding a QoS prefix to the zone name. Refer to “QoS” on
page 415 for more information.

• Traffic Isolation zones (TI zones)
Isolate traffic to a specific, dedicated path through the fabric. Refer to Chapter 13, “Traffic
Isolation Zoning,” for more information.

Zoning overview
Zoning is a fabric-based service that enables you to partition your storage area network (SAN) into
logical groups of devices that can access each other.
For example, you can partition your SAN into two zones, winzone and unixzone, so that your
Windows servers and storage do not interact with your UNIX servers and storage. You can use
zones to logically consolidate equipment for efficiency or to facilitate time-sensitive functions; for
example, you can create a temporary zone to back up nonmember devices.
A device in a zone can communicate only with other devices connected to the fabric within the
same zone. A device not included in the zone is not available to members of that zone. When
zoning is enabled, devices that are not included in any zone configuration are inaccessible to all
other devices in the fabric.
Zones can be configured dynamically. They can vary in size, depending on the number of
fabric-connected devices, and devices can belong to more than one zone.
Consider Figure 33 on page 339, which shows configured zones, Red, Green, and Blue.

•
•
•
•

Server 1 can communicate only with the Storage 1 device.
Server 2 can communicate only with the RAID and Storage 2 devices.
Server 3 can communicate with the RAID and Storage 1 devices.
The Storage 3 is not assigned to a zone; no other zoned fabric device can access it.

NOTE
When using a mixed fabric—that is, a fabric containing two or more switches running different
release levels of fabric operating systems—you should use the switch with the latest Fabric OS level
to perform zoning tasks.
To list the commands associated with zoning, use the zoneHelp command. For detailed information
on the zoning commands used in the procedures, refer to the Fabric OS Command Reference.

338

Fabric OS Administrator’s Guide
53-1002920-02

Zoning overview

12

JBOD
Loop 2
Server2
Blue zone

Fibre Channel Fabric

RAID

Hub
Server1

Loop 1

Red zone

FIGURE 33

Server3

Green zone

Zoning example

Approaches to zoning
Table 62 lists the various approaches you can take when implementing zoning in a fabric.

TABLE 62

Approaches to fabric-based zoning

Zoning approach

Description

Recommended approach
Single HBA

Zoning by single HBA most closely re-creates the original SCSI bus. Each zone created has only
one HBA (initiator) in the zone; each of the target devices is added to the zone. Typically, a zone
is created for the HBA and the disk storage ports are added. If the HBA also accesses tape
devices, a second zone is created with the HBA and associated tape devices in it. In the case of
clustered systems, it could be appropriate to have an HBA from each of the cluster members
included in the zone; this is equivalent to having a shared SCSI bus between the cluster
members and assumes that the clustering software can manage access to the shared devices.
In a large fabric, zoning by single HBA requires the creation of possibly hundreds of zones;
however, each zone contains only a few members. Zone changes affect the smallest possible
number of devices, minimizing the impact of an incorrect zone change. This zoning philosophy
is the preferred method.

Alternative approaches
Application

Fabric OS Administrator’s Guide
53-1002920-02

Zoning by application typically requires zoning multiple, perhaps incompatible, operating
systems into the same zones. This method of zoning creates the possibility that a minor server
in the application suite could disrupt a major server (such as a Web server disrupting a data
warehouse server). Zoning by application can also result in a zone with a large number of
members, meaning that more notifications, such as registered state change notifications
(RSCNs), or errors, go out to a larger group than necessary.

339

12

Zoning overview

TABLE 62

Approaches to fabric-based zoning (Continued)

Zoning approach

Description

Operating
system

Zoning by operating system has issues similar to zoning by application. In a large site, this type
of zone can become very large and complex. When zone changes are made, they typically
involve applications rather than a particular server type. If members of different operating
system clusters can see storage assigned to another cluster, they might attempt to own the
other cluster’s storage and compromise the stability of the clusters.

Port allocation

Avoid zoning by port allocation unless the administration team has very rigidly enforced
processes for port and device allocation in the fabric. Zoning by port allocation does, however,
provide some positive features. For instance, when a storage port, server HBA, or tape drive is
replaced, the change of WWN for the new device is of no consequence. As long as the new
device is connected to the original port, it continues to have the same access rights. The ports
on the edge switches can be pre-associated to storage ports, and control of the fan-in ratio (the
ratio of the input port to output port) can be established. With this pre-assigning technique, the
administrative team cannot overload any one storage port by associating too many servers with
it.

Not recommended
No fabric zoning

Using no fabric zoning is the least desirable zoning option because it allows devices to have
unrestricted access on the fabric. Additionally, any device attached to the fabric, intentionally or
maliciously, likewise has unrestricted access to the fabric. This form of zoning should be utilized
only in a small and tightly controlled environment, such as when host-based zoning or LUN
masking is deployed.

Zone objects
A zone object is any device in a zone, such as:

• Physical port number or port index on the switch
• Node World Wide Name (N-WWN)
• Port World Wide Name (P-WWN)
Zone objects identified by port number or index number are specified as a pair of decimal numbers
in the form D,I, where D is the domain ID of the switch and I is the index number on that switch in
relation to the port you want to specify.
For example, in Backbones, “4,30” specifies port 14 in slot number 2 (domain ID 4, port index 30).
On fixed-port models, “3,13” specifies port 13 in switch domain ID 3.
The following issues affect zone membership based on the type of zone object:

• When a zone object is the physical port number, then all devices connected to that port are in
the zone.

• World Wide Names are specified as 8-byte (16-digit) hexadecimal numbers, separated by
colons (:), for example, 10:00:00:90:69:00:00:8a.

• When a zone object is the node WWN name, only the specified device is in the zone.
• When a zone object is the port WWN name, only the single port is in the zone.
The types of zone objects used to define a zone can be mixed. For example, a zone defined with the
zone objects 2,12; 2,14; 10:00:00:80:33:3f:aa:11 contains the devices connected to domain 2,
ports 12 and 14, and a device with the WWN 10:00:00:80:33:3f:aa:11 (either node name or port
name) that is connected on the fabric.

340

Fabric OS Administrator’s Guide
53-1002920-02

Zoning overview

12

Zoning schemes
You can establish a zone by identifying zone objects using one or more of the following zoning
schemes:

• Domain,index (D,I)
All members are specified by domain ID, port number, or domain, index number pairs or
aliases.

• World Wide Name (WWN)
All members are specified only by World Wide Names (WWNs) or aliases of WWNs. They can be
node or port versions of the WWN.

• Mixed zoning
A zone containing members specified by a combination of domain,port or domain,index or
aliases, and WWNs or aliases of WWNs.
In any scheme, you can identify zone objects using aliases.

Zone configurations
A zone configuration is a group of one or more zones. A zone can be included in more than one
zone configuration. When a zone configuration is in effect, all zones that are members of that
configuration are in effect.
Several zone configurations can reside on a switch at once, and you can quickly alternate between
them. For example, you might want to have one configuration enabled during the business hours
and another enabled overnight. However, only one zone configuration can be enabled at a time.
The different types of zone configurations are:

• Defined configuration
The complete set of all zone objects defined in the fabric.

• Effective configuration
A single zone configuration that is currently in effect. The effective configuration is built when
you enable a specified zone configuration.

• Saved configuration
A copy of the defined configuration plus the name of the effective configuration, which is saved
in flash memory. (You can also provide a backup of the zone configuration and restore the zone
configuration.) There might be differences between the saved configuration and the defined
configuration if you have modified any of the zone definitions and have not saved the
configuration.

• Disabled configuration
The effective configuration is removed from flash memory.
If you disable the effective configuration, the Advanced Zoning feature is disabled on the fabric,
and all devices within the fabric can communicate with all other devices (unless you previously set
up a default zone, as described in “Default zoning mode” on page 360). This does not mean that
the zone database is deleted, however, only that there is no configuration active in the fabric.
On power-up, the switch automatically reloads the saved configuration. If a configuration was active
when it was saved, the same configuration is reinstated on the local switch.

Fabric OS Administrator’s Guide
53-1002920-02

341

12

Zoning overview

Zoning enforcement
Zoning enforcement describes a set of predefined rules that the switch uses to determine where to
send incoming data. Fabric OS uses hardware-enforced zoning. Hardware-enforced zoning means
that each frame is checked by hardware (the ASIC) before it is delivered to a zone member and is
discarded if there is a zone mismatch. When hardware-enforced zoning is active, the Fabric OS
switch monitors the communications and blocks any frames that do not comply with the effective
zone configuration. The switch performs this blocking at the transmit side of the port on which the
destination device is located.
There are two methods of hardware enforcement:

• Frame-based hardware enforcement: All frames are checked by the hardware.
• Session-based hardware enforcement: The only frames checked by hardware are the ELS
frames (such as PLOGI and RNID) used to establish a session.
The hardware-enforcement method used depends on how the zones are configured.
A zone can contain all WWN members, or all D,I members, or a combination of WWN and D,I
members.
Frame-based hardware enforcement is in effect if all members of a zone are identified the same
way, either using WWNs or D,I notation, but not both. If the zone includes aliases, then the aliases
must also be defined the same way as the zone.
Session-based hardware enforcement is in effect if the zone has a mix of WWN and D,I members.
If a port is in multiple zones, and is defined by WWN in one zone and by D,I in another, then
session-based hardware enforcement is in effect.

Identifying the enforced zone type
Use the following procedure to identify zones and zone types:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portZoneShow command.

Considerations for zoning architecture
Table 63 lists considerations for zoning architecture.

TABLE 63

342

Considerations for zoning architecture

Item

Description

Type of zoning
enforcement: frameor session-based

If security is a priority, frame-based hardware enforcement is recommended. The best way
to do this is to use WWN identification exclusively for all zoning configurations.

Use of aliases

The use of aliases is optional with zoning. Using aliases requires structure when defining
zones. Aliases aid administrators of zoned fabrics in understanding the structure and
context.

Fabric OS Administrator’s Guide
53-1002920-02

Broadcast zones

TABLE 63

12

Considerations for zoning architecture (Continued)

Item

Description

Effect of changes in a
production fabric

Zone changes in a production fabric can result in a disruption of I/O under conditions
when an RSCN is issued because of the zone change and the HBA is unable to process the
RSCN fast enough. Although RSCNs are a normal part of a functioning SAN, the pause in
I/O might not be acceptable. For these reasons, you should perform zone changes only
when the resulting behavior is predictable and acceptable. Ensuring that the HBA drivers
are current can shorten the response time in relation to the RSCN.

Testing

Before implementing a new zone, you should run the Zone Analyzer from Web Tools to
isolate any possible problems. This is especially useful as fabrics increase in size.

Confirming operation

After changing or enabling a zone configuration, you should confirm that the nodes and
storage can identify and access one another. Depending on the platform, you might need
to reboot one or more nodes in the fabric with the new changes.

Zoning can be implemented and administered from any switch in the fabric, although it is
recommended that you use a switch running the latest Fabric OS version.
The zone configuration is managed on a fabric basis. When a change in the configuration is saved,
enabled, or disabled according to the transactional model, it is automatically (by closing the
transaction) distributed to all switches in the fabric, preventing a single point of failure for zone
information.

NOTE

Zoning commands make changes that affect the entire fabric. When executing fabric-level
configuration tasks, allow time for the changes to propagate across the fabric before executing any
subsequent commands. For a large fabric, you should wait several minutes between commands.

Best practices for zoning
The following are recommendations for using zoning:

• Always zone using the switch with the latest Fabric OS release level.
Switches with earlier Fabric OS versions do not have the capability to view all the functionality
that a newer Fabric OS version provides, as functionality is backwards-compatible but not
forwards-compatible.

• Zone using the core switch versus an edge switch.
• Zone using a Backbone rather than a switch.
A Backbone has more resources to handle zoning changes and implementations.

Broadcast zones
Fibre Channel allows sending broadcast frames to all Nx_Ports if the frame is sent to a broadcast
well-known address (FFFFFF); however, many target devices and HBAs cannot handle broadcast
frames. To control which devices receive broadcast frames, you can create a special zone, called a
broadcast zone, that restricts broadcast packets to only those devices that are members of the
broadcast zone.

Fabric OS Administrator’s Guide
53-1002920-02

343

12

Broadcast zones

If there are no broadcast zones or if a broadcast zone is defined but not enabled, broadcast frames
are not forwarded to any F_Ports. If a broadcast zone is enabled, broadcast frames are delivered
only to those logged-in Nx_Ports that are members of the broadcast zone and are also in the same
zone (regular zone) as the sender of the broadcast packet.
Devices that are not members of the broadcast zone can send broadcast packets, even though
they cannot receive them.
A broadcast zone can have domain,port, WWN, and alias members.
Broadcast zones do not function in the same way as other zones. A broadcast zone does not allow
access within its members in any way. If you want to allow or restrict access between any devices,
you must create regular zones for that purpose. If two devices are not part of a regular zone, they
cannot exchange broadcast or unicast packets.
To restrict broadcast frames reaching broadcast-incapable devices, create a broadcast zone and
populate it with the devices that are capable of handling broadcast packets. Devices that cannot
handle broadcast frames must be kept out of the broadcast zone so that they do not receive any
broadcast frames.
You create a broadcast zone the same way you create any other zone except that a broadcast zone
must have the name “broadcast” (case-sensitive). You set up and manage broadcast zones using
the standard zoning commands, described in “Zone creation and maintenance” on page 350.

Broadcast zones and Admin Domains
Each Admin Domain can have only one broadcast zone. However, all of the broadcast zones from
all of the Admin Domains are considered as a single consolidated broadcast zone.
Broadcast packets are forwarded to all the ports that are part of the broadcast zone for any Admin
Domain, have membership in that Admin Domain, and are zoned together (in a regular zone) with
the sender of the broadcast frame.
Figure 34 illustrates how broadcast zones work with Admin Domains. Figure 34 shows a fabric with
five devices and two Admin Domains, AD1 and AD2. Each Admin Domain has two devices and a
broadcast zone.

344

Fabric OS Administrator’s Guide
53-1002920-02

Broadcast zones

12

"3,1"

"1,1"

"4,1"

"2,1"

AD1

AD2
broadcast
"2,1; 3,1; 4,1"

broadcast
"1,1; 3,1; 5,1"

"5,1"
"1,1"

"3,1; 4,1"
broadcast
"1,1; 3,1; 4,1"

FIGURE 34

Broadcast zones and Admin Domains

The dotted box represents the consolidated broadcast zone, which contains all of the devices that
can receive broadcast packets. The actual delivery of broadcast packets is also controlled by the
Admin Domain and zone enforcement logic. The consolidated broadcast zone is not an actual zone,
but is just an abstraction used for explaining the behavior.

• The broadcast zone for AD1 includes member devices “1,1”, “3,1” and “5,1”; however, “3,1”
and “5,1” are not members of AD1. Consequently, from the AD1 broadcast zone, only “1,1” is
added to the consolidated broadcast zone.

• The broadcast zone for AD2 includes member devices “2,1”, “3,1”, and “4,1”. Even though
“2,1” is a member of AD1, it is not a member of AD2 and so is not added to the consolidated
broadcast zone.

• Device “3,1” is added to the consolidated broadcast zone because of its membership in the
AD2 broadcast zone.
When a switch receives a broadcast packet, it forwards the packet only to those devices that are
zoned with the sender and are also part of the consolidated broadcast zone.
You can check whether a broadcast zone has any invalid members that cannot be enforced in the
current AD context. Refer to “Validating a zone” on page 358 for complete instructions.

Broadcast zones and FC-FC routing
If you create broadcast zones in a metaSAN consisting of multiple fabrics connected through an FC
router, the broadcast zone must include the IP device that exists in the edge or backbone fabric as
well as the proxy device in the remote fabric. Refer to Chapter 26, “Using FC-FC Routing to Connect
Fabrics,” for information about proxy devices and the FC router.

Fabric OS Administrator’s Guide
53-1002920-02

345

12

Zone aliases

High availability considerations with broadcast zones
If a switch has broadcast zone-capable firmware on the active CP (Fabric OS v5.3.x or later) and
broadcast zone-incapable firmware on the standby CP (Fabric OS version earlier than v5.3.0), then
you cannot create a broadcast zone because the zoning behavior would not be the same across an
HA failover. If the switch failed over, then the broadcast zone would lose its special significance and
would be treated as a regular zone.

Loop devices and broadcast zones
Delivery of broadcast packets to individual devices in a loop is not controlled by the switch.
Consequently, adding loop devices to a broadcast zone does not have any effect. If a loop device is
part of a broadcast zone, then all devices in that loop receive broadcast packets.
Best practice: All devices in a single loop should have uniform broadcast capability. If all the
devices in the loop can handle broadcast frames, then add the FL_Port to the broadcast zone.

Broadcast zones and default zoning mode
The default zoning mode defines the device accessibility behavior if zoning is not implemented or if
there is no effective zone configuration. The default zoning mode has two options:

• All Access—All devices within the fabric can communicate with all other devices.
• No Access—Devices in the fabric cannot access any other device in the fabric.
If a broadcast zone is active, even if it is the only zone in the effective configuration, the default
zone setting is not in effect.
If the effective configuration has only a broadcast zone, then the configuration appears as a No
Access configuration. To change this configuration to All Access, you must put all the available
devices in a regular zone.
Refer to “Default zoning mode” on page 360 for additional information about default zoning.

Zone aliases
A zone alias is a name assigned to a logical group of ports or WWNs. By creating an alias, you can
assign a familiar name to a device or group multiple devices into a single name. This simplifies
cumbersome data entry and allows an intuitive naming structure (such as using “NT_Hosts” to
define all NT hosts in the fabric). Using zone aliases eliminates the need for long lists of individual
zone member names.
Zone aliases also simplify repetitive entry of zone objects such as port numbers or a WWN. For
example, you can use the name “Eng” as an alias for “10:00:00:80:33:3f:aa:11”.
Naming zones for the initiator they contain can also be useful. For example, if you use the alias
SRV_MAILSERVER_SLT5 to designate a mail server in PCI slot 5, then the alias for the associated
zone is ZNE_MAILSERVER_SLT5. This clearly identifies the server host bus adapter (HBA)
associated with the zone.

346

Fabric OS Administrator’s Guide
53-1002920-02

Zone aliases

12

Zone configuration naming is flexible. One configuration should be named PROD_fabricname,
where fabricname is the name that the fabric has been assigned. The purpose of the PROD
configuration is to easily identify the configuration that can be implemented and provide the most
generic services. If other configurations are used for specialized purposes, names such as
“BACKUP_A,” “RECOVERY_2,” and “TEST_18jun02” can be used. If you are creating a new alias
using aliCreate w, “1,1”, and a user in another Telnet session executes cfgEnable (or cfgDisable, or
cfgSave), the other user’s transaction will abort your transaction and you will receive an error
message. Creating a new alias while there is a zone merge taking place may also abort your
transaction. For more details about zone merging and zone merge conflicts, refer to “Zone
merging” on page 371.
Virtual Fabrics considerations: Alias definitions should not include logical port numbers. Zoning is
not enforced on logical ports.

Creating an alias
Use the following procedure to create an alias.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aliCreate command, using the following syntax:
alicreate "aliasname", "member[; member...]"

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example
switch:admin> alicreate "array1", "2,32; 2,33; 2,34; 4,4"
switch:admin> alicreate "array2", "21:00:00:20:37:0c:66:23; 4,3"
switch:admin> alicreate "loop1", "4,6"
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the
Effective configuration and the Defined configuration
inconsistent. The inconsistency will result in different
Effective Zoning configurations for switches in the fabric if
a zone merge or HA failover happens. To avoid inconsistency
it is recommended to commit the configurations using the
'cfgenable' command.
Do you still want to proceed with saving the Defined
zoning configuration only? (yes, y, no, n): [no] y

Adding members to an alias
Use the following procedure to add a member to an alias.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aliAdd command, using the following syntax:
aliadd "aliasname", "member[; member...]"

Fabric OS Administrator’s Guide
53-1002920-02

347

12

Zone aliases

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example
switch:admin> aliadd "array1", "1,2"
switch:admin> aliadd "array2", "21:00:00:20:37:0c:72:51"
switch:admin> aliadd "loop1", "5,6"
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the
Effective configuration and the Defined configuration
inconsistent. The inconsistency will result in different
Effective Zoning configurations for switches in the fabric if
a zone merge or HA failover happens. To avoid inconsistency
it is recommended to commit the configurations using the
'cfgenable' command.
Do you still want to proceed with saving the Defined
zoning configuration only? (yes, y, no, n): [no] y

Removing members from an alias
Use the following procedure to removing a member from an alias.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aliRemove command, using the following syntax:
aliremove "aliasname", "member[; member...]"

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example
switch:admin> aliremove "array1", "1,2"
switch:admin> aliremove "array2", "21:00:00:20:37:0c:72:51"
switch:admin> aliremove "loop1", "4,6"
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the
Effective configuration and the Defined configuration
inconsistent. The inconsistency will result in different
Effective Zoning configurations for switches in the fabric if
a zone merge or HA failover happens. To avoid inconsistency
it is recommended to commit the configurations using the
'cfgenable' command.
Do you still want to proceed with saving the Defined
zoning configuration only? (yes, y, no, n): [no] y

348

Fabric OS Administrator’s Guide
53-1002920-02

Zone aliases

12

Deleting an alias
Use the following procedure to delete an alias.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aliDelete command, using the following syntax.
alidelete "aliasname"

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example
switch:admin> alidelete "array1"
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the
Effective configuration and the Defined configuration
inconsistent. The inconsistency will result in different
Effective Zoning configurations for switches in the fabric if
a zone merge or HA failover happens. To avoid inconsistency
it is recommended to commit the configurations using the
'cfgenable' command.
Do you still want to proceed with saving the Defined
zoning configuration only? (yes, y, no, n): [no] y

Viewing an alias in the defined configuration
Use the following procedure to view an alias in the configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the aliShow command to display zone configuration information.
alishow [--ic]["pattern"][, mode]

If no parameters are specified, the entire zone database (both the defined and effective
configuration) is displayed.
Example

The following example shows all zone aliases beginning with “arr”:
switch:admin> alishow "arr*"
alias: array1 21:00:00:20:37:0c:76:8c
alias: array2 21:00:00:20:37:0c:66:23

The following example shows all zone aliases beginning with “arr”, regardless of the case:
switch:admin> alishow --ic "arr*"
alias: array1 20:e0:00:05:33:11:1f:00
alias: array2 2f:11:00:05:33:c1:37:a2

Fabric OS Administrator’s Guide
53-1002920-02

349

12

Zone creation and maintenance

Zone creation and maintenance
Fabric OS allows you to create zones to better manage devices.

NOTE

Broadcast Zone: To create a broadcast zone, use the reserved name “broadcast”. Do not give a
regular zone the name of “broadcast”. Refer to “Broadcast zones” on page 343 for additional
information about this special type of zone.

NOTE

Virtual Fabrics considerations: Zone definitions should not include logical port numbers. Zoning is
not enforced on logical ports.

Displaying existing zones
Use the following procedure to display a list of existing zones.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgShow command.
Example displaying existing zones
switch:admin> cfgshow
Defined configuration:
zone: matt 30:06:00:07:1e:a2:10:20; 3,2
alias:
bawn
3,5; 4,8
alias:
bolt
10:00:00:02:1f:02:00:01
alias:
bond
10:00:05:1e:a9:20:00:01; 3,5
alias:
brain 11,4; 22,1; 33,6
alias:
jake
4,7; 8,9; 14,11
alias:
jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias:
jones 7,3; 4,5
alias:
zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)

Creating a zone
ATTENTION
The zoneCreate command will add all zone member aliases that match the “aliasname_pattern” in
the zone database to the new zone.
Use the following procedure to create a zone.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zoneCreate command, using either of the following syntaxes:
zonecreate "zonename", "member[; member...]"
zonecreate "zonename", "aliasname_pattern*[;members]"

350

Fabric OS Administrator’s Guide
53-1002920-02

Zone creation and maintenance

12

NOTE

The zoneCreate command supports partial pattern matching (“wildcards”) of zone
member aliases. This allows you to add multiple aliases that match the
“aliasname_pattern” in the command line.

To create a broadcast zone, use the reserved name “broadcast”.
3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
4. Enter the cfgShow command to view the changes.
Example creating a new zone
switch:admin> zonecreate sloth, "b*; 10:00:00:00:01:1e:20:20"
switch:admin> cfgsave
switch:admin> cfgshow
Defined configuration:
zone: matt
30:06:00:07:1e:a2:10:20; 3,2
zone: sloth
bawn; bolt; bond; brain; 10:00:00:00:01:1e:20:20
alias: bawn
3,5; 4,8
alias: bolt
10:00:00:02:1f:02:00:01
alias: bond
10:00:05:1e:a9:20:00:01; 3,5
alias: brain 11,4; 22,1; 33,6
alias: jake
4,7; 8,9; 14,11
alias: jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias: jones 7,3; 4,5
alias: zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)

Adding devices (members) to a zone
ATTENTION
The zoneAdd command will add all zone member aliases that match the “aliasname_pattern” in the
zone database to the specified zone.
Use the following procedure to add members to a zone:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zoneAdd command, using either of the following syntaxes:
zoneadd "zonename", "member[; member...]"
zoneadd "zonename", "aliasname_pattern*[;members]"

NOTE

The zoneAdd command supports partial pattern matching (“wildcards”) of zone member
aliases. This allows you to add multiple aliases that match the “aliasname_pattern” in the
command line.

3. Enter the cfgSave command to save the change to the defined configuration.

Fabric OS Administrator’s Guide
53-1002920-02

351

12

Zone creation and maintenance

The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
4. Enter the cfgShow command to view the changes.
Example adding members to a zone
switch:admin> zoneadd matt, "ze*; bond*; j*"
switch:admin> cfgsave
switch:admin> cfgshow
Defined configuration:
zone: matt
30:06:00:07:1e:a2:10:20; 3,2; zeus; bond; jake; jeff; jones
zone: sloth bawn; bolt; bond; brain; 10:00:00:00:01:1e:20:20
alias:
bawn
3,5; 4,8
alias:
bolt
10:00:00:02:1f:02:00:01
alias:
bond
10:00:05:1e:a9:20:00:01; 3,5
alias:
brain 11,4; 22,1; 33,6
alias:
jake
4,7; 8,9; 14,11
alias:
jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias:
jones 7,3; 4,5
alias:
zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)

Removing devices (members) from a zone
ATTENTION
The zoneRemove command will remove all zone member aliases that match the
“aliasname_pattern” in the zone database from the specified zone.
Use the following procedure to remove members from a zone:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zoneRemove command, using either of the following syntaxes:
zoneremove "zonename", "member[; member...]"
zoneremove "zonename", "aliasname_pattern*[;members]"

NOTE

This command supports partial pattern matching (“wildcards”) of zone member aliases.
This allows you to remove multiple aliases that match the “aliasname_pattern” in the
command line.

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
4. Enter the cfgShow command to view the changes.
Example removing members from a zone
switch:admin> cfgshow
Defined configuration:

352

Fabric OS Administrator’s Guide
53-1002920-02

Zone creation and maintenance

12

zone: matt
zeus; bond; jake; jeff; jones; 3,2; 30:06:00:07:1e:a2:10:20
zone: sloth bawn; bolt; bond; brain; 10:00:00:00:01:1e:20:20
alias:
bawn
3,5; 4,8
alias:
bolt
10:00:00:02:1f:02:00:01
alias:
bond
10:00:05:1e:a9:20:00:01; 3,5
alias:
brain 11,4; 22,1; 33,6
alias:
jake
4,7; 8,9; 14,11
alias:
jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias:
jones 7,3; 4,5
alias:
zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)
switch:admin>
switch:admin> zoneremove matt,"30:06:00:07:1e:a2:10:20; ja*; 3,2"
switch:admin> cfgsave
switch:admin> cfgshow
Defined configuration:
zone: matt
zeus; bond; jeff; jones zone: sloth bawn; bolt; bond; brain;
10:00:00:00:01:1e:20:20
alias:
bawn
3,5; 4,8
alias:
bolt
10:00:00:02:1f:02:00:01
alias:
bond
10:00:05:1e:a9:20:00:01; 3,5
alias:
brain 11,4; 22,1; 33,6
alias:
jake
4,7; 8,9; 14,11
alias:
jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias:
jones 7,3; 4,5
alias:
zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)

Replacing zone members
Fabric OS allows you to replace one zone member with another zone member using a CLI
command. This command takes two inputs. The first is the member to be replaced and the second
is the new member. These inputs can be formatted only with WWN or D,I zoning schemes.
Notes and restrictions
• To make a configuration change effective, a cfgEnable command should be issued after the
zoneObjectReplace command. Otherwise, the changes will be in the transaction buffer but not
committed.

• Only members of regular zones and aliases (those identified using either D,I or WWN) can be
replaced using zoneObjectReplace.

• The zoneObjectReplace command is not applicable for Frame Redirect (FR) and Traffic
Isolation (TI) zones. Only members of regular zones can be replaced using this command.

• The zoneObjectReplace command does not work with aliases. Alias members (that is,
members inside an alias) can be replaced using zoneObjectReplace, but an alias itself cannot
be directly replaced. To achieve the effect of replacement, create a new alias (with the desired
new name) containing the same members, and then delete the old alias.
Use the following procedure to replace members in a zone.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zoneObjectReplace command, using the following syntax:
zoneobjectreplace old wwn/D,I new wwn/D,I

Fabric OS Administrator’s Guide
53-1002920-02

353

12

Zone creation and maintenance

NOTE

The zoneObjectReplace command does not support partial pattern matching (“wildcards”)
of zone member aliases.

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
4. Enter the cfgShow command to view the changes.
Example replacing zone members
switch:admin> cfgshow
Defined configuration:
zone: matt
zeus; bond; jeff;jones; 11, 2
zone: sloth bawn; bolt; bond; brain; brit; bru; 10:00:00:00:01:1e:20:20
alias:
bawn
3,5
alias:
bolt
10:00:00:02:1f:02:00:01
alias:
bond
10:00:05:1e:a9:20:00:01; 3,5
alias:
brain 11,4; 22,1; 33,6
alias:
jake
4,7; 8,9; 14,11
alias:
jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias:
jones 7,3; 4,5
alias:
zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)
switch:admin>
switch:admin> zoneobjectreplace 11,2 4,8
switch:admin> cfgsave
switch:admin> cfgshow
Defined configuration:
zone: matt
zeus; bond; jeff; 4,8
zone: sloth bawn; bolt; bond; brain; 10:00:00:00:01:1e:20:20
alias:
bawn
3,5
alias:
bolt
10:00:00:02:1f:02:00:01
alias:
bond
10:00:05:1e:a9:20:00:01; 3,5
alias:
brain 11,4; 22,1; 33,6
alias:
jake
4,7; 8,9; 14,11
alias:
jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias:
jones 7,3; 4,5
alias:
zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)

CAUTION
Executing this command replaces all instances of the older member with the new member in the
entire zone database.

354

Fabric OS Administrator’s Guide
53-1002920-02

Zone creation and maintenance

12

Deleting a zone
Use the following procedure to delete a zone.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zoneDelete command, using the following syntax:
zonedelete "zonename"

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example deleting zone members
switch:admin> cfgshow
Defined configuration:
zone: matt
zeus; bond; jeff; 4,8
zone: sloth bawn; bolt; bond; brain; brit; bru; 10:00:00:00:01:1e:20:20
alias:
bawn
3,5
alias:
bolt
10:00:00:02:1f:02:00:01
alias:
bond
10:00:05:1e:a9:20:00:01; 3,5
alias:
brain 11,4; 22,1; 33,6
alias:
jake
4,7; 8,9; 14,11
alias:
jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias:
jones 7,3; 4,5
alias:
zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)
switch:admin>
switch:admin> zonedelete sloth
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the Effective configuration
and the Defined configuration inconsistent. The inconsistency will result in
different Effective Zoning configurations for switches in the fabric if a zone
merge or HA failover happens. To avoid inconsistency it is recommended to
commit the configurations using the 'cfgenable' command.
Do you still want to proceed with saving the Defined zoning configuration
only? (yes, y, no, n): [no] y
switch:admin>
switch:admin> cfgshow
Defined configuration:
zone: matt
zeus; bond; jeff; 4,8
alias:
bawn
3,5
alias:
bolt
10:00:00:02:1f:02:00:01
alias:
bond
10:00:05:1e:a9:20:00:01; 3,5
alias:
brain 11,4; 22,1; 33,6
alias:
jake
4,7; 8,9; 14,11
alias:
jeff
30:00:00:05:1e:a1:cd:02; 40:00:00:05:1e:a1:cd:04
alias:
jones 7,3; 4,5
alias:
zeus
4,7; 6,8; 9,2
Effective configuration:
No Effective configuration: (No Access)

Fabric OS Administrator’s Guide
53-1002920-02

355

12

Zone creation and maintenance

Viewing a zone in the defined configuration
Use the following procedure to view a zone in the configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zoneShow command, using the following syntax:
zoneshow[--sort] ["pattern"] [, mode]

If no parameters are specified, the entire zone database (both the defined and effective
configuration) is displayed.
Example

The following example shows all zones beginning with A, B, or C, in ascending order:
switch:admin> zoneshow --sort "[A-C]*"
zone: Blue_zone 1,1; array1; 1,2; array2
zone: Bobs_zone 4,5; 4,6; 4,7; 4,8; 4,9

Viewing zone configuration names without case distinction
Use the following procedure to view selected zone configuration names for a given pattern without
case distinction.
1. Connect to the switch and log in using an account with admin permissions.
2. Use the zoneShow command to view configuration names.
zoneshow[--ic] ["pattern"] [, mode]

Example

The following example shows all green zones using pattern search, regardless of the case:
switch:admin> zoneshow --ic GREEN*
zone: GREEN 44,4; 21:00:00:20:37:0c:71:02; 8,9
zone: green 2,2; 2,3; 21:00:00:20:37:0c:76:8c

Examining changes in the zone database
Fabric OS allows you to check for and display any differences between the transaction buffer and
the committed database by appending the options --transdiffs and --transdiffsonly to the
zoneShow and cfgShow commands.
The options use the format in the following commands:
zoneShow –-transdiffs
zoneShow –-transdiffsonly
cfgShow --transdiffs
cfgShow –-transdiffsonly

To reflect the changes made to the zone database (a new zone is added or an existing zone is
deleted, or a zone member is added or deleted or any other valid zone database entity is modified),
the following notation is used.

• An asterisk (*) at the start indicates a change in that zone, zone configuration, alias or any
other entity in the zone database.

• A plus sign (+) before any entity (an alias or a zone name or a configuration) indicates that it is
a newly added entity.

356

Fabric OS Administrator’s Guide
53-1002920-02

Zone creation and maintenance

12

• A minus sign (–) before any entity indicates that this entity has been deleted. If zone members
are added as well as deleted in a zone configuration, then a plus sign and a minus sign (+-) will
be displayed before the member and a * sign will be displayed before the zone name.

• A plus sign (+) before any member of an alias or zone name or any other entity indicates this
member has been added, and a minus sign (–) indicates the particular member has been deleted.
In the case of TI zones, for inter-fabric links for example, “5,-1” is a valid zone member.
Notice that the minus sign (-) comes before the port index. If this was a deleted zone member, it
would have been shown as "5,-1". A minus sign (-) before a domain ID indicates that this TI zone
member has been deleted.
Example displaying existing zone database
switch:admin> cfgshow
Defined configuration:
cfg: fabric_cfg Blue_zone
zone: Blue_zone
1,1; array1; 1,2; array2
zone: green_zone
1,1; 1,2
alias: array1 21:00:00:20:37:0c:76:8c; 21:00:00:20:37:0c:71:02
alias: array2 21:00:00:20:37:0c:76:22; 21:00:00:20:37:0c:76:28
Effective configuration:
cfg: fabric_cfg
zone: Blue_zone
1,1
21:00:00:20:37:0c:76:8c
21:00:00:20:37:0c:71:02
1,2

Example adding a new zone ‘red_zone’, deleting “1,1” and adding “6,15” to green_zone
switch:admin> cfgshow --transdiffs
Defined configuration:
cfg: fabric_cfg Blue_zone
zone: Blue_zone
1,1; array1; 1,2; array2
*zone: green_zone
-1,1; 1,2; +6, 15
*zone: +red_zone
5,1; 4,2
alias: array1 21:00:00:20:37:0c:76:8c; 21:00:00:20:37:0c:71:02
alias: array2 21:00:00:20:37:0c:76:22; 21:00:00:20:37:0c:76:28
Effective configuration:
cfg: fabric_cfg
zone: Blue_zone
1,1
21:00:00:20:37:0c:76:8c
21:00:00:20:37:0c:71:02
1,2

Example cfgShow --transdiffsonly output for the previous example
switch:admin> cfgshow --transdiffsonly
*zone: green_zone -1,1; 1,2; +6,15
*zone: +red_zone 5,1; 4,2
switch:admin>

Fabric OS Administrator’s Guide
53-1002920-02

357

12

Zone creation and maintenance

Validating a zone
Use the following procedure to validate a zone.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgShow command to view the zone configuration objects you want to validate.
switch:admin> cfgShow
Defined configuration:
cfg: USA_cfg Purple_zone; White_zone; Blue_zone
zone: Blue_zone
1,1; array1; 1,2; array2
zone: Purple_zone
1,0; loop1
zone: White_zone
1,3; 1,4
alias: array1 21:00:00:20:37:0c:76:8c; 21:00:00:20:37:0c:71:02
alias: array2 21:00:00:20:37:0c:76:22; 21:00:00:20:37:0c:76:28
alias: loop1
21:00:00:20:37:0c:76:85; 21:00:00:20:37:0c:71:df

3. Enter the zone --validate command to list all zone members that are not part of the current
zone enforcement table. Note that zone configuration names are case-sensitive; blank spaces
are ignored.
switch:admin> zone --validate "White_zone"

4. Enter the following command to validate all zones in the zone database in the defined
configuration.
switch:admin> zone --validate -m 1
Defined configuration:
cfg:
cfg1
zone1
cfg:
cfg2
zone1; zone2
zone: zone1
1,1; ali1
zone: zone2
1,1; ali2
alias: ali1
10:00:00:05:1e:35:81:7f*; 10:00:00:05:1e:35:81:7d*
alias: ali2
10:00:00:05:1e:35:81:09*; 10:00:00:05:1e:35:81:88*
-----------------------------------~ - Invalid configuration
* - Member does not exist

The mode flag -m can be used to specify the zone database location. Supported mode flag
values are:

• 0 - Zone database from the current transaction buffer
• 1 - Zone database stored from the persistent storage
• 2 - Currently effective zone database.
If no mode options are given, the validated output of all three buffers is shown.
If the -f option is specified, all the zone members that are not enforceable would be expunged
in the transaction buffer. This pruning operation always happens on the transaction and
defined buffers. You cannot specify a mode option or specify a zone object as an argument
with the -f option. This mode flag should be used after the zone has been validated.
If the -i option is specified, all zone members for a given pattern are listed without case
distinction

358

Fabric OS Administrator’s Guide
53-1002920-02

Zone creation and maintenance

12

Example validating the zone members beginning with gre, regardless of the case
switch:admin> zone --validate -i gre*
Defined configuration:
zone:
GREEN
44, 4; 21:00:00:20:37:0c:71:02; 8,9
zone: green 2,2*; 2,3*; 21:00:00:20:37:0c:76:8c*
Effective configuration:
zone: green 2,2*
2,3*
21:00:00:20:37:0c:76:8c*
-----------------------------------~ - Invalid configuration
* - Member does not exist
# - Invalid usage of broadcast zone

Inconsistencies between the defined and effective configurations
If you edit zone objects in the defined configuration that also exist in the effective configuration and
then issue the cfgSave command, a warning message stating that a mismatch is observed
between the defined and effective configurations is posted, and you are asked to confirm that you
want cfgSave to continue. If you enter “y”, then the updated configuration will be saved; if you enter
“n”, then the updated configuration will be discarded.
Example warning message
switch: admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the
Effective configuration and the Defined configuration inconsistent.
The inconsistency will result in different Effective Zoning
configurations for switches in the fabric if a zone merge or
HA failover happens. To avoid inconsistency it is recommended to
commit the configurations using the 'cfgenable' command.
Do you want to proceed with saving the Defined
zoning configuration only? (yes, y, no, n): [no] yes

If you enter yes, and the cfgSave operation completes successfully, then the following RASlog
message [ZONE-1062] will be posted.
[ZONE-1062], 620/181, FID 128, WARNING, sw0, Defined and Effective zone
configurations are inconsistent, ltime:2012/09/03-23:18:30:983609

You can then either re-enable the updated configuration or revert to the older configuration. If there
is no impact to the effective configuration with the latest update to the defined configuration, then
the following message will be displayed.
“You are about to save the Defined zoning configuration. This action will only
save the changes on Defined configuration.
Do you want to proceed?” (yes, y, no, n): [no] y

Example of inconsistent defined and effective configuration warning to use
switch: admin> zoneShow
Defined configuration:
cfg:
cfg1
zone1; zone2
zone:
zone1
10:00:00:00:00:00:00:01; 10:00:00:00:00:00:00:02
zone:
zone2
1,1; 1,2
Effective configuration:
cfg:
cfg1
zone:
zone1
10:00:00:00:00:00:00:01
10:00:00:00:00:00:00:02

Fabric OS Administrator’s Guide
53-1002920-02

359

12

Default zoning mode

zone:

zone2

1,1; 1,2

switch: admin> zoneadd zone1, 10:00:00:00:00:00:00:03
switch: admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the
Effective configuration and the Defined configuration inconsistent.
The inconsistency will result in different Effective Zoning
configurations for switches in the fabric if a zone merge or
HA failover happens. To avoid inconsistency it is recommended to
commit the configurations using the 'cfgenable' command.
Do you want to proceed with saving the Defined zoning
configuration only? (yes, y, no, n): [no] y
Updating flash ...
switch:admin> zoneShow
Defined configuration:
cfg:
cfg1
zone1; zone2
zone:
zone1
10:00:00:00:00:00:00:01; 10:00:00:00:00:00:00:02;
10:00:00:00:00:00:00:03
zone:
zone2
1,1; 1,2
Effective configuration:
cfg:
cfg1
zone:
zone1
10:00:00:00:00:00:00:01; 10:00:00:00:00:00:00:02
zone:
zone2
1,1; 1,2

Default zoning mode
The default zoning mode controls device access if zoning is not implemented or if there is no
effective zone configuration. The default zoning mode has two options:

• All Access—All devices within the fabric can communicate with all other devices.
• No Access—Devices in the fabric cannot access any other device in the fabric.
The default zone mode applies to the entire fabric, regardless of switch model.
The default setting is “All Access”.
Typically, when you disable the zoning configuration in a large fabric with thousands of devices, the
name server indicates to all hosts that they can communicate with each other. Each host can
receive an enormous list of PIDs, and ultimately cause other hosts to run out of memory or crash.
To ensure that all devices in a fabric do not see each other during a configuration disable
operation, set the default zoning mode to No Access.

NOTE

For switches in large fabrics, the default zone mode should be set to No Access. You cannot disable
the effective configuration if the default zone mode is All Access and you have more than 120
devices in the fabric.
Admin Domain considerations: If you want to use Admin Domains, you must set the default zoning
mode to No Access prior to setting up the Admin Domains. You cannot change the default zoning
mode to All Access if user-specified Admin Domains are present in the fabric.

360

Fabric OS Administrator’s Guide
53-1002920-02

Default zoning mode

12

Setting the default zoning mode
NOTE
You should not change the default zone mode from “No Access” to “All Access” if there is no effective
zone configuration and more than 120 devices are connected to the fabric.
Use the following procedure to set the default zoning mode.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgActvShow command to view the current zone configuration.
3. Enter the defZone command with one of the following options:
defzone --noaccess

or
defzone --allaccess

This command initiates a transaction (if one is not already in progress).
4. Enter the cfgSave, cfgEnable, or cfgDisable command to commit the change and distribute it to
the fabric. The change will not be committed and distributed across the fabric if you do not
enter one of these commands.
Example
switch:admin> defzone --noaccess
You are about to set the Default Zone access mode to No Access
Do you want to set the Default Zone access mode to No Access ? (yes, y, no, n):
[no] y
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the Effective configuration
and the Defined configuration inconsistent. The inconsistency will result in
different Effective Zoning configurations for switches in the fabric if a zone
merge or HA failover happens. To avoid inconsistency it is recommended to
commit the configurations using the 'cfgenable' command.
Do you still want to proceed with saving the Defined
zoning configuration only? (yes, y, no, n): [no] y
Updating flash ...

Viewing the current default zone access mode
Use the following procedure to view the current default zone access mode.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the defZone --show command.

NOTE

If you perform a firmware download of an older release, then the current default zone access state
will appear as it did prior to the download. For example, if the default zoning mode was No Access
before the download, it will remain as No Access afterward.

Fabric OS Administrator’s Guide
53-1002920-02

361

12

Zone database size

Zone database size
The maximum size of a zone database is the upper limit for the defined configuration, and it is
determined by the amount of flash memory available for storing the defined configuration.
Use the cfgSize command to display the zone database size.
The supported maximum zone database size is 2 MB for systems running only Brocade DCX,
DCX-4S, and DCX 8510 platforms. The presence of any other platform reduces the maximum zone
database size to 1 MB.
Virtual Fabrics considerations: If Virtual Fabrics is enabled, the sum of the zone database sizes on
all of the logical fabrics must not exceed the maximum size allowed for the chassis (1 MB). The
maximum size limit is enforced per-partition, but is not enforced chassis-wide. If the chassis size
limit is exceeded, you are not informed of this and unpredictable behavior may occur. It is your
responsibility to keep track of the chassis-wide zone database size.

ATTENTION
In a fabric with some switches running Fabric OS 7.0.0 or later and some switches running Fabric OS
versions earlier than 7.0.0, if you execute the cfgSave or cfgEnable command from a pre-7.0.0
switch, a zone database size of 128 KB is enforced.
To avoid this problem, use the switch with the latest Fabric OS version to perform zoning tasks, as
described in “Best practices for zoning” on page 343. Alternatively, make sure that your pre-7.0.0
switches are upgraded with the latest patch release.

Zone configurations
You can store a number of zones in a zone configuration database. The maximum number of items
that can be stored in the zone configuration database depends on the following criteria:

• Number of switches in the fabric.
• Number of bytes for each item name. The number of bytes required for an item name depends
on the specifics of the fabric, but cannot exceed 64 bytes for each item.
When enabling a new zone configuration, ensure that the size of the defined configuration does not
exceed the maximum configuration size supported by all switches in the fabric. This is particularly
important if you downgrade to a Fabric OS version that supports a smaller zone database than the
current Fabric OS. In this scenario, the zone database in the current Fabric OS would have to be
changed to the smaller zone database before the downgrade.
You can use the cfgSize command to check both the maximum available size and the currently
saved size on all switches. If you think you are approaching the maximum, you can save a partially
completed zone configuration and use the cfgSize command to determine the remaining space.
The cfgSize command reports the maximum available size on the current switch only. It cannot
determine the maximum available size on other switches in the fabric.

NOTE

The minimum zone database size is 4 bytes, even if the zone database is empty.
For important considerations for managing zoning in a fabric, and more details about the maximum
zone database size for each version of the Fabric OS, refer to “Zone database size” on page 362.

362

Fabric OS Administrator’s Guide
53-1002920-02

Zone configurations

12

If you create or make changes to a zone configuration, you must enable the configuration for the
changes to take effect.

Creating a zone configuration
Use the following procedure to create a zone configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgCreate command, using the following syntax:
cfgcreate "cfgname", "member[; member...]"

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example
switch:admin> cfgcreate "NEW_cfg", "purplezone; bluezone; greenzone"
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the Effective configuration
and the Defined configuration inconsistent. The inconsistency will result in
different Effective Zoning configurations for switches in the fabric if a zone
merge or HA failover happens. To avoid inconsistency it is recommended to
commit the configurations using the 'cfgenable' command.
Do you still want to proceed with saving the Defined zoning configuration
only? (yes, y, no, n): [no] y

Adding zones to a zone configuration
Use the following procedure to add members to a zone configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgAdd command, using the following syntax:
cfgadd "cfgname", "member[; member...]"

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example
switch:admin> cfgadd "newcfg", "bluezone"
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the Effective configuration
and the Defined configuration inconsistent. The inconsistency will result in
different Effective Zoning configurations for switches in the fabric if a zone
merge or HA failover happens. To avoid inconsistency it is recommended to
commit the configurations using the 'cfgenable' command.

Fabric OS Administrator’s Guide
53-1002920-02

363

12

Zone configurations

Do you still want to proceed with saving the Defined zoning configuration
only? (yes, y, no, n): [no] y

Removing members from a zone configuration
Use the following procedure to remove members from a zone configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgRemove command, using the following syntax:
cfgremove "cfgname", "member[; member...]"

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example
switch:admin> cfgremove "NEW_cfg", "purplezone"
switch:admin> cfgsave
WARNING!!!
The changes you are attempting to save will render the Effective configuration
and the Defined configuration inconsistent. The inconsistency will result in
different Effective Zoning configurations for switches in the fabric if a zone
merge or HA failover happens. To avoid inconsistency it is recommended to
commit the configurations using the 'cfgenable' command.
Do you still want to proceed with saving the Defined zoning configuration
only? (yes, y, no, n): [no] y

Enabling a zone configuration
The following procedure ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this procedure is run, the
transaction on the other switch is automatically aborted. A message displays on the other switches
to indicate that the transaction was aborted.
Use the following procedure to enable a zone configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgenable command, using the following syntax:
cfgenable "cfgname"

3. Enter y at the prompt.
Example
switch:admin> cfgenable "USA_cfg"
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the current
configuration selected. If the update includes changes to one or more traffic
isolation zones, the update may result in localized disruption to traffic on
ports associated with the traffic isolation zone changes.
Do you want to enable 'USA_cfg' configuration (yes, y, no, n): [no] y
zone config "USA_cfg" is in effect
Updating flash ...

364

Fabric OS Administrator’s Guide
53-1002920-02

Zone configurations

12

Disabling a zone configuration
When you disable the current zone configuration, the fabric returns to non-zoning mode. All devices
can then access each other or not, depending on the default zone access mode setting.

NOTE
If the default zoning mode is set to All Access and more than 120 devices are connected to the
fabric, you cannot disable the zone configuration because this would enable All Access mode and
cause a large number of requests to the switch. In this situation, set the default zoning mode to
No Access prior to disabling the zone configuration. Refer to “Default zoning mode” on page 360 for
information about setting this mode to No Access.
The following procedure ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this procedure is run, the
transaction on the other switch is automatically aborted. A message displays on the other switches
to indicate that the transaction was aborted.
Use the following procedure to disable a zone configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgDisable command.
3. Enter y at the prompt.
Example
switch:admin> cfgdisable
You are about to disable zoning configuration. This action will disable any
previous zoning configuration enabled.
Do you want to disable zoning configuration? (yes, y, no, n): [no] y

Deleting a zone configuration
Use the following procedure to delete a zone configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgDelete command, using the following syntax:
cfgdelete "cfgname"

3. Enter the cfgSave command to save the change to the defined configuration.
The cfgSave command ends and commits the current zoning transaction buffer to nonvolatile
memory. If a transaction is open on a different switch in the fabric when this command is run,
the transaction on the other switch is automatically aborted. A message displays on the other
switches to indicate that the transaction was aborted.
Example
switch:admin> cfgdelete "testcfg"
switch:admin> cfgsave
You are about to save the Defined zoning configuration. This action will only
save the changes on Defined configuration.
Do you want to save Defined zoning configuration only? (yes, y, no, n): [no] y

Fabric OS Administrator’s Guide
53-1002920-02

365

12

Zone configurations

Abandoning zone configuration changes
To abandon zone configuration changes, enter the cfgTransAbort command.
When this command is executed, all changes since the last save operation (performed with the
cfgSave, cfgEnable, or cfgDisable command) are cleared.
Example assuming that the removal of a member from zone1 was done in error
switch:admin> zoneremove "zone1","3,5"
switch:admin> cfgtransabort

Viewing all zone configuration information
If you do not specify an operand when executing the cfgShow command to view zone configurations,
then all zone configuration information (both defined and effective) displays. If there is an
outstanding transaction, then the newly edited zone configuration that has not yet been saved is
displayed. If there are no outstanding transactions, then the committed zone configuration displays.
Use the following procedure to view all zone configuration information.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgShow command with no operands.
Example
switch:admin> cfgshow
Defined configuration:
cfg:
USA1
Blue_zone
cfg:
USA_cfg Purple_zone; Blue_zone
zone: Blue_zone
1,1; array1; 1,2; array2
zone: Purple_zone
1,0; loop1
alias: array1 21:00:00:20:37:0c:76:8c; 21:00:00:20:37:0c:71:02
alias: array2 21:00:00:20:37:0c:76:22; 21:00:00:20:37:0c:76:28
alias: loop1
21:00:00:20:37:0c:76:85; 21:00:00:20:37:0c:71:df
Effective configuration:
cfg:
USA_cfg
zone: Blue_zone
1,1
21:00:00:20:37:0c:76:8c
21:00:00:20:37:0c:71:02
1,2
21:00:00:20:37:0c:76:22
21:00:00:20:37:0c:76:28
zone: Purple_zone
1,0
21:00:00:20:37:0c:76:85
21:00:00:20:37:0c:71:df

366

Fabric OS Administrator’s Guide
53-1002920-02

Zone configurations

12

Viewing selected zone configuration information
Use the following procedure to view the selected zone configuration information.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgShow command and specify a pattern.
cfgshow [--ic][,"pattern"] [, mode]

Example displaying all zone configurations that start with “Test”
switch:admin> cfgshow "Test*"
cfg:
Test1
Blue_zone
cfg:
Test_cfg Purple_zone; Blue_zone

Example displaying all zone configurations that start with “Test”, regardless of the case
switch:admin> cfgshow --ic "Test*"
cfg:
Test1
Blue_zone
cfg:
Test_2 Red zone; Blue_zone

Viewing the configuration in the effective zone database
Use the following procedure to view the configuration in the effective zone database.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgActvShow command.
Example
switch:admin> cfgactvshow
Effective configuration:
cfg:
NEW_cfg
zone: Blue_zone
1,1
21:00:00:20:37:0c:76:8c
21:00:00:20:37:0c:71:02
1,2
21:00:00:20:37:0c:76:22
21:00:00:20:37:0c:76:28
zone: Purple_zone
1,0
21:00:00:20:37:0c:76:85
21:00:00:20:37:0c:71:df

Clearing all zone configurations
Use the following procedure to clear all zone configurations.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgClear command to clear all zone information in the transaction buffer.

ATTENTION
Be careful using the cfgClear command because it deletes the defined configuration.
switch:admin> cfgclear
The Clear All action will clear all Aliases, Zones, FA Zones and configurations
in the Defined configuration.

Fabric OS Administrator’s Guide
53-1002920-02

367

12

Zone object maintenance

Run cfgSave to commit the transaction or cfgTransAbort to cancel the transaction.
Do you really want to clear all configurations? (yes, y, no, n): [no]

3. Enter one of the following commands, depending on whether an effective zone configuration
exists:

• If no effective zone configuration exists, use the cfgSave command.
• If an effective zone configuration exists, use the cfgDisable command to disable and clear
the zone configuration in nonvolatile memory for all switches in the fabric.

Zone object maintenance
The following procedures describe how to copy, delete, and rename zone objects. Depending on
the operation, a zone object can be a zone member, a zone alias, a zone, or a zone configuration.

Copying a zone object
When you copy a zone object, the resulting object has the same name as the original. The zone
object can be a zone configuration, a zone alias, or a zone.
Use the following procedure to copy a zone object.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgShow command to view the zone configuration objects you want to copy.
cfgshow "pattern"[, mode]

For example, to display all zone configuration objects that start with “Test”:
switch:admin> cfgshow "Test*"
cfg:
Test1
Blue_zone
cfg:
Test_cfg Purple_zone; Blue_zone

3. Enter zone --copy specifying the zone objects you want to copy along with the new object
name.

NOTE

Zone configuration names are case-sensitive, blank spaces are ignored, and the zone --copy
command works in any Admin Domain except AD255.
switch:admin> zone --copy Test1 US_Test1

4. Enter the cfgShow command to verify the new zone object is present.
switch:admin> cfgshow "Test*"
cfg:
Test1
Blue_zone
cfg:
Test_cfg Purple_zone; Blue_zone
switch:admin> cfgShow "US_Test1"
cfg:
US_Test1 Blue_zone

5. If you want the change preserved when the switch reboots, use cfgSave to save it to nonvolatile
(flash) memory.
6. Enter cfgEnable for the appropriate zone configuration to make the change effective.

368

Fabric OS Administrator’s Guide
53-1002920-02

Zone object maintenance

12

Deleting a zone object
The following procedure removes all references to a zone object and then deletes the zone object.
The zone object can be a zone member, a zone alias, or a zone.
Use the following procedure to delete a zone object.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgShow command to view the zone configuration objects you want to delete.
switch:admin> cfgShow
Defined configuration:
cfg: USA_cfg Purple_zone; White_zone;
zone: Blue_zone
1,1; array1; 1,2; array2
zone: Purple_zone
1,0; loop1
zone: White_zone
1,3; 1,4
alias: array1 21:00:00:20:37:0c:76:8c;
alias: array2 21:00:00:20:37:0c:76:22;
alias: loop1
21:00:00:20:37:0c:76:85;

Blue_zone

21:00:00:20:37:0c:71:02
21:00:00:20:37:0c:76:28
21:00:00:20:37:0c:71:df

Effective configuration:
cfg: USA_cfg
zone: Blue_zone
1,1
21:00:00:20:37:0c:76:8c
21:00:00:20:37:0c:71:02
1,2
21:00:00:20:37:0c:76:22
21:00:00:20:37:0c:76:28
zone: Purple_zone
1,0
21:00:00:20:37:0c:76:85
21:00:00:20:37:0c:71:df

3. Enter zone --expunge to delete the zone object.

NOTE

Zone configuration names are case-sensitive, blank spaces are ignored, and the
zone --expunge command works in any Admin Domain except AD255.
switch:admin> zone --expunge "White_zone"
You are about to expunge one configuration or member. This action could result
in removing many zoning configurations recursively. [Removing the last member
of a configuration removes the configuration.]
Do you want to expunge the member? (yes, y, no, n): [no] yes

4. Enter yes at the prompt.
5. Enter cfgShow to verify the deleted zone object is no longer present.
6. If you want the change preserved when the switch reboots, use cfgSave to save it to nonvolatile
(flash) memory.
7.

Enter cfgEnable for the appropriate zone configuration to make the change effective.

Fabric OS Administrator’s Guide
53-1002920-02

369

12

Zone configuration management

Renaming a zone object
Use the following procedure to rename a zone object.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter cfgShow to view the zone configuration objects you want to rename.
switch:admin> cfgShow
Defined configuration:
cfg: USA_cfg Purple_zone; White_zone; Blue_zone
zone: Blue_zone
1,1; array1; 1,2; array2
zone: Purple_zone
1,0; loop1
zone: White_zone
1,3; 1,4
alias: array1 21:00:00:20:37:0c:76:8c; 21:00:00:20:37:0c:71:02
alias: array2 21:00:00:20:37:0c:76:22; 21:00:00:20:37:0c:76:28
alias: loop1
21:00:00:20:37:0c:76:85; 21:00:00:20:37:0c:71:df

3. Enter zoneObjectRename to rename zone configuration objects.

NOTE

Zone configuration names are case-sensitive, blank spaces are ignored, and the
zoneObjectRename command works in any Admin Domain except AD255.
switch:admin> zoneObjectRename "White_zone", "Purple_zone"

4. Enter the cfgShow command to verify the renamed zone object is present.
5. If you want the change preserved when the switch reboots, enter the cfgSave command to
save it to nonvolatile (flash) memory.
6. Enter the cfgEnable command for the appropriate zone configuration to make the change
effective.

Zone configuration management
You can add, delete, or remove individual elements in an existing zone configuration to create an
appropriate configuration for your SAN environment. After the changes have been made, save the
configuration to ensure the configuration is permanently saved in the switch and that the
configuration is replicated throughout the fabric.
The switch configuration file can also be uploaded to the host for archiving and it can be
downloaded from the host to a switch in the fabric. Refer to “Configuration file backup” on
page 279, “Configuration file restoration” on page 280, or the configUpload and configDownload
commands in the Fabric OS Command Reference for additional information on uploading and
downloading the configuration file.

370

Fabric OS Administrator’s Guide
53-1002920-02

Security and zoning

12

Security and zoning
Zones provide controlled access to fabric segments and establish barriers between operating
environments. They isolate systems with different uses, protecting individual systems in a
heterogeneous environment; for example, when zoning is in secure mode, no merge operations
occur.
Brocade Advanced Zoning is configured on the primary fabric configuration server (FCS). The
primary FCS switch makes zoning changes and other security-related changes. The primary FCS
switch also distributes zoning to all other switches in the secure fabric. All existing interfaces can
be used to administer zoning.
You must perform zone management operations from the primary FCS switch using a zone
management interface, such as Telnet or Web Tools. You can alter a zone database, provided you
are connected to the primary FCS switch.
When two secure fabrics join, the traditional zone merge does not occur. Instead, a zone database
is downloaded from the primary FCS switch of the merged secure fabric. When E_Ports are active
between two switches, the name of the FCS server and a zoning policy set version identifier are
exchanged between the switches. If the views of the two secure fabrics are the same, the fabric’s
primary FCS server downloads the zone database and security policy sets to each switch in the
fabric. If there is a view conflict, the E_Ports are segmented due to incompatible security data.
All zones should use frame-based hardware enforcement; the best way to do this is to use WWN
identification exclusively for all zoning configurations.

Zone merging
When a new switch is added to the fabric, it automatically takes on the zone configuration
information from the fabric. You can verify the zone configuration on the switch using the procedure
described in “Viewing the configuration in the effective zone database” on page 367.
If you are adding a switch that is already configured for zoning, clear the zone configuration on that
switch before connecting it to the zoned fabric. Refer to “Clearing all zone configurations” on
page 367 for instructions.
Adding a new fabric that has no zone configuration information to an existing fabric is very similar
to adding a new switch. All switches in the new fabric inherit the zone configuration data. If the
existing fabric has an effective zone configuration, then the same configuration becomes the
effective configuration for the new switches.
Before the new fabric can merge successfully, it must pass the following criteria:

• Before merging
To facilitate merging, check the following before merging switches or fabrics:

-

Defaultzone: The switches must adhere to the default zone merge rules, as described in
“Zone merging scenarios” on page 373.

-

Effective and defined zone configuration match: Ensure that the effective and defined
zone configurations match. If they do not match, and you merge with another switch, the
merge may be successful, but unpredictable zoning and routing behavior can occur.

Fabric OS Administrator’s Guide
53-1002920-02

371

12

Zone merging

• Merging and segmentation
The fabric is checked for segmentation during power-up, when a switch is disabled or enabled,
or when a new switch is added.
The zone configuration database is stored in nonvolatile memory by the cfgSave command. All
switches in the fabric have a copy of this database. When a change is made to the defined
configuration, the switch where the changes were made must close its transaction for the
changes to be propagated throughout the fabric.
If you have implemented default zoning, you must set the switch you are adding into the fabric
to the same default zone mode setting as the rest of the fabric to avoid segmentation.

• Merging rules
Observe these rules when merging zones:

-

Local and adjacent configurations: If the local and adjacent zone database configurations
are the same, they will remain unchanged after the merge.

-

Effective configurations: If there is an effective configuration between two switches, the
effective zone configurations must match.

-

Zone object naming: If a zoning object has the same name in both the local and adjacent
defined configurations, the object types and member lists must match. When comparing
member lists, the content and order of the members are important.

-

Objects in adjacent configurations: If a zoning object appears in an adjacent defined
configuration, but not in the local defined configuration, the zoning object is added to the
local defined configuration. The modified zone database must fit in the nonvolatile
memory area allotted for the zone database.

-

Local configuration modification: If a local defined configuration is modified because of a
merge, the new zone database is propagated to other the switches within the merge
request.

-

TI zones: If there is an effective configuration between two switches and TI zones are
present on either switch, the TI zones are not automatically activated after the merge.
Check the TI zone enabled status using the zone --show command, and if the status does
not match across switches, issue the cfgEnable command.

• Merging two fabrics
Both fabrics have identical zones and configurations enabled, including the default zone
mode. The two fabrics will join to make one larger fabric with the same zone configuration
across the newly created fabric.
If the two fabrics have different zone configurations, they will not be merged. If the two fabrics
cannot join, the ISL between the switches will segment.

• Merge conflicts
When a merge conflict is present, a merge will not take place and the ISL will segment. Use the
switchShow or errDump commands to obtain additional information about possible merge
conflicts, because many non-zone-related configuration parameters can cause conflicts. Refer
to the Fabric OS Command Reference for detailed information about these commands.
If the fabrics have different zone configuration data, the system attempts to merge the two
sets of zone configuration data. If the zones cannot merge, the ISL will be segmented.

372

Fabric OS Administrator’s Guide
53-1002920-02

Zone merging

12

A merge is not possible if any of the following conditions exist:

-

Configuration mismatch: Zoning is enabled in both fabrics and the zone configurations
that are enabled are different in each fabric.

-

Type mismatch: The name of a zone object in one fabric is used for a different type of zone
object in the other fabric.

-

Content mismatch: The definition of a zone object in one fabric is different from the
definition of the zone object with the same name in the other fabric.

-

Zone database size: The zone database size exceeds the maximum limit of another switch.

NOTE

If the zone set members on two switches are not listed in the same order, the configuration is
considered a mismatch, resulting in the switches being segmented from the fabric. For
example, cfg1 = z1; z2 is different from cfg1 = z2; z1, even though members of the
configuration are the same. If zone set members on two switches have the same names
defined in the configuration, make sure zone set members are listed in the same order.

Fabric segmentation and zoning
If the connections between two fabrics are no longer available, the fabric segments into two
separate fabrics. Each new fabric retains the same zone configuration.
If the connections between two fabrics are replaced and no changes have been made to the zone
configuration in either of the two fabrics, then the two fabrics merge back into one single fabric. If
any changes that cause a conflict have been made to either zone configuration, then the fabrics
may segment.

Zone merging scenarios
The following tables provide information on merging zones and the expected results:

•
•
•
•
•
•
TABLE 64

Table 64 on page 373: Defined and effective configurations
Table 65 on page 374: Different content
Table 66 on page 375: Different names
Table 67 on page 375: TI zones
Table 68 on page 376: Default access mode
Table 69 on page 376: Mixed Fabric OS versions

Zone merging scenarios: Defined and effective configurations

Description

Switch A

Switch B

Expected results

Switch A has a defined configuration.
Switch B does not have a defined
configuration.

defined:
cfg1:
zone1: ali1; ali2
effective: none

defined: none
effective: none

Configuration from Switch A to propagate
throughout the fabric in an inactive state,
because the configuration is not enabled.

Switch A has a defined and effective
configuration.
Switch B has a defined configuration
but no effective configuration.

defined: cfg1
zone1: ali1; ali2
effective: cfg1:

defined: cfg1
zone1: ali1; ali2
effective: none

Configuration from Switch A to propagate
throughout the fabric. The configuration is
enabled after the merge in the fabric.

Fabric OS Administrator’s Guide
53-1002920-02

373

12

Zone merging

TABLE 64

Zone merging scenarios: Defined and effective configurations (Continued)

Description

Switch A

Switch B

Expected results

Switch A and Switch B have the same
defined configuration. Neither have an
effective configuration.

defined: cfg1
zone1: ali1; ali2
effective: none

defined: cfg1
zone1: ali1; ali2
effective: none

No change (clean merge).

Switch A and Switch B have the same
defined and effective configuration.

defined: cfg1
zone1: ali1; ali2
effective: cfg1:

defined: cfg1
zone1: ali1; ali2
effective: cfg1:

No change (clean merge).

Switch A does not have a defined
configuration.
Switch B has a defined configuration.

defined: none
effective: none

defined:cfg1
zone1: ali1; ali2
effective: none

Switch A will absorb the configuration from the
fabric.

Switch A does not have a defined
configuration.
Switch B has a defined configuration.

defined: none
effective: none

defined:cfg1
zone1: ali1; ali2
effective: cfg1

Switch A will absorb the configuration from the
fabric, with cfg1 as the effective configuration.

Switch A and Switch B have the same
defined configuration. Only Switch B
has an effective configuration.

defined: cfg1
zone1: ali1; ali2
effective: none

defined: cfg1
zone1: ali1; ali2
effective: cfg1

Clean merge, with cfg1 as the effective
configuration.

Switch A and Switch B have different
defined configurations. Neither have an
enabled zone configuration.

defined: cfg2
zone2: ali3; ali4
effective: none

defined: cfg1
zone1: ali1; ali2
effective: none

Clean merge. The new configuration will be a
composite of the two.
defined: cfg1
zone1: ali1; ali2
cfg2:
zone2: ali3; ali4
effective: none

Switch A and Switch B have different
defined configurations. Switch B has an
effective configuration.

defined: cfg2
zone2: ali3; ali4
effective: none

defined: cfg1
zone1: ali1; ali2
effective: cfg1

Clean merge. The new configuration will be a
composite of the two, with cfg1 as the
effective configuration.

Switch A does not have a defined
configuration.
Switch B has a defined configuration
and an effective configuration, but the
effective configuration is different from
the defined configuration.

defined: none
effective: none

defined: cfg1
zone1: ali1; ali2
effective: cfg1
zone1: ali1; ali2
zone2: ali3, ali4

Clean merge. Switch A absorbs the defined
configuration from the fabric, with cfg1 as the
effective configuration.
In this case, however, the effective
configurations for Switch A and Switch B are
different. You should issue a cfgenable from
the switch with the proper effective
configuration.

TABLE 65

Zone merging scenarios: Different content

Description

Switch A

Switch B

Expected results

Effective configuration mismatch.

defined: cfg1
zone1: ali1; ali2
effective: cfg1
zone1: ali1; ali2

defined: cfg2
zone2: ali3; ali4
effective: cfg2
zone2: ali3; ali4

Fabric segments due to: Zone Conflict cfg
mismatch

Configuration content mismatch.

defined: cfg1
zone1: ali1; ali2
effective: irrelevant

defined: cfg1
zone1: ali3; ali4
effective: irrelevant

Fabric segments due to: Zone Conflict content
mismatch

374

Fabric OS Administrator’s Guide
53-1002920-02

Zone merging

TABLE 66

12

Zone merging scenarios: Different names

Description

Switch A

Switch B

Expected results

Same content, different effective cfg
name.

defined: cfg1
zone1: ali1; ali2
effective: cfg1
zone1: ali1; ali2

defined:cfg2
zone1: ali1; ali2
effective: cfg2
zone1: ali1; ali2

Fabric segments due to: Zone Conflict cfg
mismatch

Same content, different zone name.

defined: cfg1
zone1: ali1; ali2
effective: irrelevant

defined: cfg1
zone2: ali1; ali2
effective: irrelevant

Fabric segments due to: Zone Conflict content
mismatch

Same content, different alias name.

defined: cfg1
ali1: A; B
effective: irrelevant

defined:cfg1
ali2: A; B
effective: irrelevant

Fabric segments due to: Zone Conflict content
mismatch

Same alias name, same content,
different order.

defined: cfg1
ali1: A; B; C
effective: irrelevant

defined: cfg1
ali1: B; C; A
effective: irrelevant

Fabric segments due to: Zone Conflict content
mismatch

Same name, different types.

effective: zone1:
MARKETING

effective: cfg1:
MARKETING

Fabric segments due to: Zone Conflict type
mismatch

Same name, different types.

effective: zone1:
MARKETING

effective: alias1:
MARKETING

Fabric segments due to: Zone Conflict type
mismatch

Same name, different types.

effective: cfg1:
MARKETING

effective: alias1:
MARKETING

Fabric segments due to: Zone Conflict type
mismatch

TABLE 67

Zone merging scenarios: TI zones

Description

Switch A

Switch B

Expected results

Switch A does not have Traffic Isolation
(TI) zones.
Switch B has TI zones.

defined: cfg1
effective: cfg1

defined: cfg1
TI_zone1
effective: cfg1

Clean merge. TI zones are not automatically
activated after the merge.

Switch A has TI zones.
Switch B has identical TI zones.

defined: cfg1
TI_zone1
effective: cfg1

defined: cfg1
TI_zone1
effective: cfg1

Clean merge. TI zones are not automatically
activated after the merge.

Switch A has a TI zone.
Switch B has a different TI zone.

defined: cfg1
TI_zone1

defined: cfg1
TI_zone2

Fabric segments due to: Zone Conflict cfg
mismatch. Cannot merge switches with
different TI zone configurations.

Switch A has Enhanced TI zones.
Switch B is running Fabric OS v6.4.0 or
later.

defined: cfg1
TI_zone1
TI_zone2

defined: none

Clean merge. TI zones are not automatically
activated after the merge.

Switch A has Enhanced TI zones.
Switch B is running a Fabric OS version
earlier than v6.4.0.

defined: cfg1
TI_zone1
TI_zone2

defined: none

Fabric segments because all switches in the
fabric must be running Fabric OS v6.4.0 or
later to support Enhanced TI zones.

Fabric OS Administrator’s Guide
53-1002920-02

375

12

Concurrent zone transactions

TABLE 68

Zone merging scenarios: Default access mode

Description

Switch A

Switch B

Expected results

Different default zone access mode
settings.

defzone: allaccess

defzone: noaccess

Clean merge — noaccess takes precedence
and defzone configuration from Switch B
propagates to fabric.
defzone: noaccess

Same default zone access mode
settings.

defzone: allaccess

defzone: allaccess

Clean merge — defzone configuration is
allaccess in the fabric.

Same default zone access mode
settings.

defzone: noaccess

defzone: noaccess

Clean merge — defzone configuration is
noaccess in the fabric.

Effective zone configuration.

No effective
configuration.
defzone = allaccess

effective: cfg2
defzone: allaccess or
noaccess

Clean merge — effective zone configuration
and defzone mode from Switch B propagates
to fabric.

Effective zone configuration.

No effective
configuration.
defzone = noaccess

effective: cfg2
defzone: allaccess

Fabric segments because Switch A has a
hidden zone configuration (no access)
activated and Switch B has an explicit zone
configuration activated.

Effective zone configuration

effective: cfg1
defzone: noaccess

No effective
configuration.
defzone: noaccess

Clean merge — effective zone configuration
from Switch A propagates to fabric.

Effective zone configuration

effective: cfg1
defzone: allaccess

No effective
configuration.
defzone: noaccess

Fabric segments. You can resolve the zone
conflict by changing defzone to noaccess on
Switch 1.

TABLE 69

Zone merging scenarios: Mixed Fabric OS versions

Description

Switch A

Switch B

Expected results

Switch A is running Fabric OS 7.0.0 or
later.
Switch B is running a Fabric OS version
earlier than 7.0.0.

effective: cfg1
defzone = allaccess

No effective
configuration.
defzone - noaccess

Fabric segments due to zone conflict.

Switch A is running Fabric OS 7.0.0 or
later.
Switch B is running a Fabric OS version
earlier than 7.0.0.

No effective
configuration.
defzone = noaccess

effective: cfg2
defzone - allaccess

Fabric segments due to zone conflict.

NOTE

When merging mixed versions of Fabric OS where both sides have default zone mode No Access set,
the merge results vary depending on which switch initiates the merge.

Concurrent zone transactions
While working on zone sets, a special workspace is provided to allow you to manipulate the zone
sets of your choice. These changes are not put into effect until they are committed to the database.
Once they are committed, they will replace the existing active zone sets with the new zone sets or
create more zone sets in the defined database. When updates to the zoning database are being
made by multiple users, Fabric OS warns you about the situation and allows you to choose which
operation will prevail.

376

Fabric OS Administrator’s Guide
53-1002920-02

Concurrent zone transactions

12

Example of how users are warned if there is already a pending zoning transaction in the fabric
u30:FID128:admin> zonecreate z2, "2,3"
WARNING!!
Multiple open transactions are pending in this fabric. Only one
transaction can be saved. Please abort all unwanted transactions
using the cfgtransabort command. Use the cfgtransshow --opentrans
command to display a list of domains with open transactions

If no other transaction is open in this fabric, no message is shown.
Example of what is shown if there is not a pending zoning transaction in the fabric
sw0:FID128:admin> zonecreate z7, "4,5;10,3"
sw0:FID128:admin>

Similar messages are shown for cfgSave and cfgEnable:
u30:FID128:admin> cfgenable cfg
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected. If the update includes changes
to one or more traffic isolation zones, the update may result in
localized disruption to traffic on ports associated with
the traffic isolation zone changes
Multiple open transactions are pending in this fabric. Only one
transaction can be saved. Please abort all unwanted transactions
using the cfgtransabort command. Use the cfgtransshow --opentrans
command to display a list of domains with open transactions
Do you want to enable 'cfg' configuration (yes, y, no, n): [no]

u30:FID128:admin> cfgsave
You are about to save the Defined zoning configuration.
This action will only save the changes on Defined configuration.
Multiple open transactions are pending in this fabric. Only one
transaction can be saved. Please abort all unwanted transactions
using the cfgtransabort command. Use the cfgtransshow --opentrans
command to display a list of domains with open transactions
Do you want to save the Defined zoning configuration only?
(yes, y, no, n): [no] n

Viewing zone database transactions
You can use the cfgTransShow command to list all the domains in the fabric with open
transactions, as shown in the following syntax:
cfgTransShow [ |–-opentrans | --help]

Example:
switch:admin> cfgtransshow
Current transaction token is 0x571010459
It is abortable
switch:admin> cfgtransshow --help
Usage:
cfgTransShow : Displays local open transaction token details
cfgTransShow --openTrans : Displays list of Domains with Open Transactions
cfgTransShow --help : Help
switch:admin> cfgtransshow --opentrans

Fabric OS Administrator’s Guide
53-1002920-02

377

12

Concurrent zone transactions

Current transaction token is 0x3109
It is abortable
Transactions Detect: Capable
Current Open Transactions
Domain List:
-----------1 2 3 4

378

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

13

Traffic Isolation Zoning

In this chapter
• Traffic Isolation Zoning overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• TI zone failover. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Enhanced TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Traffic Isolation Zoning over FC routers . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Fabric-Level Traffic Isolation in a backbone fabric . . . . . . . . . . . . . . . . . . .
• General rules for TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Supported configurations for Traffic Isolation Zoning. . . . . . . . . . . . . . . . .
• Limitations and restrictions of Traffic Isolation Zoning. . . . . . . . . . . . . . . .
• Admin Domain considerations for Traffic Isolation Zoning. . . . . . . . . . . . .
• Virtual Fabrics considerations for Traffic Isolation Zoning . . . . . . . . . . . . .
• Traffic Isolation Zoning over FC routers with Virtual Fabrics . . . . . . . . . . .
• Creating a TI zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Modifying TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Changing the state of a TI zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Deleting a TI zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Displaying TI zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Troubleshooting TI zone routing problems. . . . . . . . . . . . . . . . . . . . . . . . . .
• Setting up TI zones over FCR (sample procedure) . . . . . . . . . . . . . . . . . . .

379
380
384
386
390
394
396
398
398
399
401
402
405
406
407
407
408
409

Traffic Isolation Zoning overview
The Traffic Isolation Zoning feature allows you to control the flow of interswitch traffic by creating a
dedicated path for traffic flowing from a specific set of source ports (N_Ports). For example, you
might use Traffic Isolation Zoning for the following scenarios:

• To dedicate an ISL to high priority, host-to-target traffic.
• To force high volume, low priority traffic onto a given ISL to limit the effect on the fabric of this
high traffic pattern.

• To ensure that requests and responses of FCIP-based applications such as tape pipelining use
the same VE_Port tunnel across a metaSAN.
Traffic Isolation Zoning does not require a license.

Fabric OS Administrator’s Guide
53-1002920-02

379

13

TI zone failover

Traffic isolation is implemented using a special zone, called a Traffic Isolation zone (TI zone).
A TI zone indicates the set of N_Ports and E_Ports to be used for a specific traffic flow. When a
TI zone is activated, the fabric attempts to isolate all inter-switch traffic entering from a member
of the zone to only those E_Ports that have been included in the zone. The fabric also attempts to
exclude traffic not in the TI zone from using E_Ports within that TI zone.
Figure 35 shows a fabric with a TI zone consisting of the following:

• N_Ports:
• E_Ports:

“1,7”, “1,8”, “4,5”, and “4,6”
“1,1”, “3,9”, “3,12”, and “4,7”

The dotted line indicates the dedicated path between the initiator in Domain 1 to the target in
Domain 4.
Domain 1

Domain 3

7
8
9

1

9

2

10
12

7
6
5
= Dedicated Path
4

= Ports in the TI zone
Domain 4

FIGURE 35

Traffic Isolation zone creating a dedicated path through the fabric

In Figure 35, all traffic entering Domain 1 from N_Ports 7 and 8 is routed through E_Port 1.
Similarly, traffic entering Domain 3 from E_Port 9 is routed to E_Port 12, and traffic entering
Domain 4 from E_Port 7 is routed to the devices through N_Ports 5 and 6. Traffic coming from
other ports in Domain 1 would not use E_Port 1, but would use E_Port 2 instead.
Use the zone command to create and manage TI zones.

TI zone failover
A TI zone can have failover enabled or disabled.
Disable failover if you want to guarantee that TI zone traffic uses only the dedicated path, and that
no other traffic can use the dedicated path.
Enable failover if you want traffic to have alternate routes if the dedicated path cannot be used,
and if you want other traffic to be able to use the dedicated path if the non-dedicated paths cannot
be used.

380

Fabric OS Administrator’s Guide
53-1002920-02

TI zone failover

13

ATTENTION
If failover is disabled, use care when planning your TI zones so that non-TI zone devices are not
isolated. If this feature is not used correctly, it can cause major fabric disruptions that are difficult
to resolve. See “Additional considerations when disabling failover” on page 381 for additional
information about using this feature.
Table 70 compares the behavior of traffic when failover is enabled and disabled.

TABLE 70

Traffic behavior when failover is enabled or disabled in TI zones

Failover enabled

Failover disabled

If the dedicated path is not the shortest path or if the
dedicated path is broken, the TI zone traffic will use a
non-dedicated path instead.

If the dedicated path is not the shortest path or if the
dedicated path is broken, traffic for that TI zone is
halted until the dedicated path is fixed.

Non-TI zone traffic will use the dedicated path if no
other paths through the fabric exist, or if the
non-dedicated paths are not the shortest paths.

Non-TI zone traffic will never use the dedicated path,
even if the dedicated path is the shortest path or if
there are no other paths through the fabric.

For example, in Figure 35 on page 380, if the dedicated ISL between Domain 1 and Domain 3 goes
offline, then the following occurs, depending on the failover option:

• If failover is disabled for the TI zone, the TI zone traffic is halted until the ISL between
Domain 1 and Domain 3 is back online.

• If failover is enabled for the TI zone, the TI zone traffic is routed from Domain 1 to Domain 3
through E_Ports “1,2” and “3,10”.

NOTE

When TI zone traffic enters the non-TI path, the TI zone traffic continues to flow through that
path. In this example, when the TI zone traffic is routed through E_Ports “1,2” and “3,10”, that
traffic continues through the non-TI path between domains 3 and 4, even though the TI path
between domains 3 and 4 is not broken.
If the non-dedicated ISL between Domain 1 and Domain 3 goes offline, then the following occurs,
depending on the failover option:

• If failover is disabled for the TI zone, non-TI zone traffic is halted until the non-dedicated ISL
between Domain 1 and Domain 3 is back online.

• If failover is enabled for the TI zone, non-TI zone traffic is routed from Domain 1 to Domain 3
through the dedicated ISL.

NOTE

When non-TI zone traffic enters the TI path, the non-TI zone traffic continues to flow through
that path. In this example, when the non-TI zone traffic is routed through E_Ports “1,1” and
“3,9”, that traffic continues through E_Ports “3,12” and “4,7”, even though the non-dedicated
ISL between domains 3 and 4 is not broken.

Additional considerations when disabling failover
If failover is disabled, be aware of the following considerations:

• This feature is intended for use in simple linear fabric configurations, such as that shown in
Figure 35 on page 380.

Fabric OS Administrator’s Guide
53-1002920-02

381

13

TI zone failover

• Ensure that there are non-dedicated paths through the fabric for all devices that are not in a TI
zone.

• If you create a TI zone with just E_Ports, failover must be enabled. If failover is disabled, the
specified ISLs will not be able to route any traffic.

• If the path between devices in a TI zone is broken, no inter-switch RSCNs are generated. Each
switch that is part of the TI zone generates RSCNs to locally attached devices that are part of
the TI zone and are registered to receive RSCNs.

• Ensure that there are multiple paths between switches.
Disabling failover locks the specified route so that only TI zone traffic can use it. Non-TI zone
traffic is excluded from using the dedicated path.

• You should enable failover-enabled TI zones before enabling failover-disabled TI zones, to avoid
dropped frames.
When you issue the cfgEnable command to enable the zone configuration, if you have failover
disabled zones, do the following:
1. Temporarily change failover-disabled TI zones to failover-enabled.
2. Enable the zones (cfgEnable).
3. Reset all the zones you changed in step 1 to failover-disabled.
4. Enable the zones again (cfgEnable).
These steps are listed in the procedures in this section.

• It is recommended that TI zone definitions and regular zone definitions match.
• Domain controller frames can use any path between switches. Disabling failover does not
affect Domain Controller connectivity.
For example, in Figure 36, if failover is disabled, Domain 2 can continue to send domain
controller frames to Domain 3 and 4, even though the path between Domain 1 and Domain 3
is a dedicated path. Domain controller frames include zone updates and name server queries.
Domain 1
8

Domain 3
1

9

9
12
3

15

7
6

= Dedicated Path
= Ports in the TI zone

5
Domain 2

FIGURE 36

Domain 4

Fabric incorrectly configured for TI zone with failover disabled

• It is recommended that the insistent Domain ID feature be enabled; if a switch changes its
active domain ID, the route is broken. See the configure command in the Fabric OS Command
Reference for information about setting insistent Domain ID.

382

Fabric OS Administrator’s Guide
53-1002920-02

TI zone failover

13

FSPF routing rules and traffic isolation
All traffic must use the lowest cost path. FSPF routing rules take precedence over the TI zones, as
described in the following situations.
If the dedicated ISL is not the lowest cost path ISL, then the following rules apply:

• If failover is enabled, the traffic path for the TI zone is broken, and TI zone traffic uses the
lowest cost path instead.

• If failover is disabled, the TI zone traffic is blocked.
If the dedicated ISL is the only lowest cost path ISL, then the following rules apply:

• If failover is enabled, non-TI zone traffic as well as TI zone traffic uses the dedicated ISL.
• If failover is disabled, non-TI zone traffic is blocked because it cannot use the dedicated ISL,
which is the lowest cost path.
For example, in Figure 37, there is a dedicated path between Domain 1 and Domain 3, and
another, non-dedicated, path that passes through Domain 2. If failover is enabled, all traffic will
use the dedicated path, because the non-dedicated path is not the shortest path. If failover is
disabled, non-TI zone traffic is blocked because the non-dedicated path is not the shortest path.
Domain 1
8

Domain 3
1

9

9
14

12

3

15

7
16

= Dedicated Path
= Ports in the TI zone

6
5

Domain 2

FIGURE 37

Domain 4

Dedicated path is the only shortest path

In Figure 38 on page 384, a dedicated path between Domain 1 and Domain 4 exists, but is not the
shortest path. In this situation, if failover is enabled, the TI zone traffic uses the shortest path, even
though the E_Ports are not in the TI zone. If failover is disabled, the TI zone traffic stops until the
dedicated path is configured to be the shortest path.

Fabric OS Administrator’s Guide
53-1002920-02

383

13

Enhanced TI zones

Domain 1
8

Domain 3
1

9

9
14

12

3

15

7
16

6

= Dedicated Path
= Ports in the TI zone

5
Domain 4

Domain 2

FIGURE 38

Dedicated path is not the shortest path

NOTE
For information about setting or displaying the FSPF cost of a path, see the linkCost and
topologyShow commands in the Fabric OS Command Reference.

Enhanced TI zones
In Fabric OS v6.4.0 and later, ports can be in multiple TI zones at the same time. Zones with
overlapping port members are called enhanced TI zones (ETIZ).
Figure 39 shows an example of two TI zones. Because these TI zones have an overlapping port
(3,8), they are enhanced TI zones.
Domain 1

Host 1

Domain 3
2

1

6

Target
8

7

Host 2
2
1
= ETIZ 1
= ETIZ 2
Domain 2

FIGURE 39

Enhanced TI zones

Enhanced TI zones are especially useful in FICON fabrics. See the FICON Administrator’s Guide for
example topologies using enhanced TI zones.
See “Additional configuration rules for enhanced TI zones” on page 396 for more information about
enhanced TI zones.

384

Fabric OS Administrator’s Guide
53-1002920-02

Enhanced TI zones

13

Illegal configurations with enhanced TI zones
When you create TI zones, ensure that all traffic from a port to all destinations on a remote domain
have the same path. Do not create separate paths from a local port to two or more ports on the
same remote domain.
If the TI zones are configured with failover disabled, some traffic will be dropped. If the TI zones are
configured with failover enabled, all traffic will go through, but half of the traffic will be routed
incorrectly according to the TI zone definitions.
A message is sent to the RASlog if a potential error condition is detected in the TI zone
configuration. You can also display a report of existing and potential problems with TI zone
configurations, as described in “Troubleshooting TI zone routing problems” on page 408.

Illegal ETIZ configuration: separate paths from a port to devices on same domain
Figure 40 shows two enhanced TI zones that are configured incorrectly because there are two
paths from a local port (port 8 on Domain 3) to two or more devices on the same remote domain
(ports 1 and 4 on Domain 1).
The TI zones are enhanced TI zones because they have an overlapping member (3,8). Each zone
describes a different path from the Target to Domain 1. Traffic is routed correctly from Host 1 and
Host 2 to the Target; however, traffic from the Target to the Hosts might not be.
Traffic from (3,8) destined for Domain 1 cannot go through both port 6 and port 7, so only one port
is chosen. If port 6 is chosen, frames destined for (1,4) will be dropped at Domain 1. If port 7 is
chosen, frames destined for (1,1) will be dropped.
Host 1

Domain 1
1
4

Domain 3
2

6

3

7

8

= ETIZ 1
= ETIZ 2

Host 2

FIGURE 40

Target

Illegal ETIZ configuration: two paths from one port to two devices on the same remote domain

Illegal ETIZ configuration: separate paths from a single port to the same domain
Figure 41 shows another example of an illegal ETIZ configuration. In this example, the two hosts
are on separate remote domains, but the path to each host goes through the same domain
(Domain 1).
This example contains two enhanced TI zones, with port (3,8) as the overlapping member:

• ETIZ 1 contains (1,1), (1,2), (3,6), (3,8)
• ETIZ 2 contains (2,1), (2,2), (1,4), (1,3), (3,7), (3,8)

Fabric OS Administrator’s Guide
53-1002920-02

385

13

Traffic Isolation Zoning over FC routers

In this example traffic from the Target to Domain 2 is routed correctly. Only one TI zone describes a
path to Domain 2. However, both TI zones describe different, valid paths from the Target to
Domain 1. Only one path will be able to get to (1,1). Traffic from port (3,8) cannot be routed to
Domain 1 over both (3,6) and (3,7), so one port will be chosen. If (3,7) is chosen, frames destined
for (1,1) will be dropped at Domain 1.
Domain 1

Host 1
1

Domain 3
2

6

3

7

Target
8

4

Host 2

2
1
= ETIZ 1
= ETIZ 2
Domain 2

FIGURE 41

Illegal ETIZ configuration: two paths from one port

Traffic Isolation Zoning over FC routers
This section describes how TI zones work with Fibre Channel routing (TI over FCR). See Chapter 26,
“Using FC-FC Routing to Connect Fabrics,” for information about FC routers, phantom switches, and
the FC-FC Routing Service.
Some VE_Port-based features, such as tape pipelining, require the request and corresponding
response traffic to traverse the same VE_Port tunnel across the metaSAN. To ensure that the
request and response traverse the same VE_Port tunnel, you must set up Traffic Isolation zones in
the edge and backbone fabrics.

• Set up a TI zone in an edge fabric to guarantee that traffic from a specific device in that edge
fabric is routed through a particular EX_Port or VEX_Port.

• Set up a TI zone in the backbone fabric to guarantee that traffic between two devices in
different fabrics is routed through a particular ISL (VE_Ports or E_Ports) in the backbone.
This combination of TI zones in the backbone and edge fabrics ensures that the traffic between
devices in different fabrics traverses the same VE_Port tunnel in a backbone fabric. Figure 42
shows how three TI zones form a dedicated path between devices in different edge fabrics.
The backbone fabric can contain one or more FC routers.

386

Fabric OS Administrator’s Guide
53-1002920-02

Traffic Isolation Zoning over FC routers

Edge fabric 1

Backbone
fabric

13

Edge fabric 2

= Dedicated path set up by TI zone in edge fabric 1
= Dedicated path set up by TI zone in edge fabric 2
= Dedicated path set up by TI zone in backbone fabric

FIGURE 42

Traffic Isolation Zoning over FCR

In addition to setting up TI zones, you must also ensure that the devices are in an LSAN zone so
that they can communicate with each other.
If failover is enabled and the TI path is not available, an alternate path is used. If failover is disabled
and the TI path is not available, then devices are not imported.

NOTE

For TI over FCR, all switches in the backbone fabric and in the edge fabrics must be running
Fabric OS v6.1.0 or later.

Fabric OS Administrator’s Guide
53-1002920-02

387

13

Traffic Isolation Zoning over FC routers

TI zones within an edge fabric
A TI zone within an edge fabric is used to route traffic between a real device and a proxy device
through a particular EX_Port. For example, in Figure 43, you can set up a TI zone to ensure that
traffic between Host 1 and the proxy target is routed through EX_Port 9.
Host 1

Domain 1
8

Front Domain 3
1

9

2

10

9
-1

Host 2

E_Ports

EX_Ports
-1

= Dedicated Path
= Ports in the TI zone
Xlate Domain 4

FIGURE 43

Proxy Target

TI zone in an edge fabric

In the TI zone, when you designate E_Ports between the front and xlate phantom switches,
you must use -1 in place of the “I” in the D,I notation. Both the front and xlate domains must be
included in the TI zone.
Using D,I notation, the members of the TI zone in Figure 43 are:

•
•
•
•

1,8
1,1
3,-1 (E_Port for the front phantom domain)
4,-1 (E_Port for the xlate phantom domain)

NOTE
In this configuration the traffic between the front and xlate domains can go through any path
between these two domains. The -1 does not identify any specific ISL. To guarantee a specific ISL,
you need to set up a TI zone within the backbone fabric.

388

Fabric OS Administrator’s Guide
53-1002920-02

Traffic Isolation Zoning over FC routers

13

TI zones within a backbone fabric
A TI zone within a backbone fabric is used to route traffic within the backbone fabric through a
particular ISL. For example, in Figure 44, a TI zone is set up in the backbone fabric to ensure that
traffic between EX_Ports “1,1” and “2,1” is routed through VE_Ports “1,4” and “2,7”.
Target 1

Target 2

WWN

WWN

Host

WWN

Target 3

Edge fabric 2

Edge fabric 1
Backbone fabric

1

3
1

2

4

VE_Ports

7

5

8

6

9

FC router 1

2

Edge fabric 3

3

FC router 2

= Dedicated Path
= Ports in the TI zone

FIGURE 44

TI zone in a backbone fabric

TI zones within the backbone fabric use the port WWN instead of D,I notation for devices that are to
communicate across fabrics. (You can use the portShow command to obtain the port WWN.)
Port WWNs should be used only in TI zones within a backbone fabric and should not be used in
other TI zones.
Using D,I and port WWN notation, the members of the TI zone in Figure 44 are:

•
•
•
•
•
•
•

1,1 (EX_Port for FC router 1)
1,4 (VE_Port for FC router 1)
2,7 (VE_Port for FC router 2)
2,1 (EX_Port for FC router 2)
10:00:00:00:00:01:00:00 (Port WWN for the host)
10:00:00:00:00:02:00:00 (Port WWN for target 1)
10:00:00:00:00:03:00:00 (Port WWN for target 2)

Fabric OS Administrator’s Guide
53-1002920-02

389

13

Fabric-Level Traffic Isolation in a backbone fabric

Limitations of TI zones over FC routers
Be aware of the following when configuring TI zones over FC routers:

• A TI zone defined within the backbone fabric does not guarantee that edge fabric traffic will
arrive at a particular EX_Port. You must set up a TI zone in the edge fabric to guarantee this.

• TI zones within the backbone fabric cannot contain more than one destination router port
(DRP) per each fabric. This means you cannot define more than one EX_Port to any one edge
fabric unless they are part of a trunk.

• Only one egress E_Port or VE_Port connected to the next hop can be defined within TI zones.
Only one ISL or trunk can be defined between two backbone switches.

• TI over FCR is supported only from edge fabric to edge fabric. Traffic isolation from backbone to
edge is not supported.

• Non-TI data traffic is not restricted from going through the TI path in the backbone fabric.
• For TI over FCR, failover must be enabled in the TI zones in the edge fabrics and in the
backbone fabric.

• TI over FCR is not supported with FC Fast Write.
• ETIZ over FCR is not supported.
• For the FC8-16, FC8-32, FC8-48, FC8-64, and FX8-24 blades only: If Virtual Fabrics is disabled,
two or more shared area EX_Ports connected to the same edge fabric should not be configured
in different TI zones. This configuration is not supported.

Fabric-Level Traffic Isolation in a backbone fabric
For Fibre Channel Routed (FCR) environments, you can use TI zoning if you want traffic isolation
only at the fabric level and not at the device level.
For example, two fabrics within a MetaSAN need to communicate only with each other. There is no
other traffic across the backbone that goes from either of these edge fabrics to any other edge
fabric in the MetaSAN. In this case, all of the traffic entering the FCR backbone from one of these
edge fabrics will go to the other edge fabric. If these two edge fabrics are connected to two different
backbone switches (FC routers), then traffic between these fabrics can be isolated to a specified
set of links within the backbone fabric using one of two methods:

• TI over FCR, which includes the PWWN of devices and maintains device level isolation
• TI zoning in the backbone, which provides fabric level isolation
If device-level isolation is needed from one edge fabric to another, then use TI over FCR using Port
World Wide Names (PWWNs). However, if there is no need for device-level isolation, but a need for
fabric-level isolation, then use Fabric-Level Traffic Isolation, described in this section.
TI over FCR is described in “Traffic Isolation Zoning over FC routers” on page 386.
If two edge fabrics are connected to two different backbone switches, then traffic between these
fabrics can be isolated to a specified set of links within the backbone fabric using TI zoning in the
backbone without including device PWWNs. This is called Fabric-Level Traffic Isolation, as shown in
Figure 45.

390

Fabric OS Administrator’s Guide
53-1002920-02

Fabric-Level Traffic Isolation in a backbone fabric

FIGURE 45

13

Fabric-level traffic isolation

In the figure, there are two links between each edge fabric and the backbone fabric, and there are
five links between the two FC routers in the backbone. Fabric ID 1 and Fabric ID 4 communicate
only with each other. Two backbone ISLs are dedicated to traffic between FID1 and FID4. These
dedicated ISL are indicted in red and blue.

Fabric-Level TI zones
Fabric-Level Traffic Isolation is accomplished through the use of TI zones. These zones define the
dedicated set of paths between the two fabrics. These paths are to be restricted to just the traffic
between the two fabrics.
Fabric-Level TI zones are defined in the backbone fabric, and include only EX_Ports and E_Ports in
the backbone fabric. The TI zones do not include device PWWNs or ports in the edge fabrics.
The TI zone must include every port in the path from ingress EX_Port to egress EX_Port. The TI zone
definitions must include all EX_Ports connected to the two edge fabrics. Unless all possible ingress
ports are included, some traffic will not be isolated to the desired paths.
Fabric-Level Traffic Isolation is not enforced on the egress EX_Ports. Any available egress IFL can
be used, regardless of whether it is in the Fabric-Level TI zone.
Note the following rules for creating Fabric-Level TI zones:

•
•
•
•
•

Include all EX_Ports connected to the two edge fabrics.
Include E_Ports for the path between the backbone switches.
Do not include E_Ports from the edge fabrics.
Do not include device PWWNs.
Ensure that failover is enabled.

Fabric OS Administrator’s Guide
53-1002920-02

391

13

Fabric-Level Traffic Isolation in a backbone fabric

There are two options for defining the Fabric-Level Traffic Isolation paths within TI zones.

• Create a separate TI zone for each path
• Combine all of the paths in a single TI zone
The option you select affects the failover behavior of the TI zones.

Failover behavior for Fabric-Level TI zones
Fabric-Level Traffic Isolation requires the TI zones in the backbone to have failover enabled. The
failover behavior differs depending on how you create the TI zones.
If you create a separate TI zone for each path:

• If one of the TI zone paths fails, then traffic on that path is re-routed to a non-dedicated path.
For example, in Figure 45 on page 391, if the BLUE path fails, then traffic associated with the
BLUE path is re-routed to a black path.
If you combine all of the paths in a single TI zone:

• If one of the paths in the TI zone fails, then traffic on that path is re-routed to another path in
the TI zone.
For example, in Figure 45 on page 391, if the BLUE path fails, then traffic associated with the
BLUE path is re-routed to the RED path.
Failover behavior for Fabric-Level TI zones is the same as for other TI zones, as described in “TI
zone failover” on page 380.

Creating a separate TI zone for each path
Create a separate TI zone for each path if you want TI traffic to failover to non-TI zone paths.
This example procedure creates two TI zones in the backbone fabric shown in Figure 45 on
page 391.
1. Create TI zones with failover enabled.
Each TI zone must include the ingress and egress EX_Ports, as well as the E_Ports between
the two backbone switches. Do not include the edge fabric E_Ports or device PWWNs.
switch:admin> zone --create -t ti TI_Zone_Red -p "20,5; 20,3; 30,7; 30,9"
switch:admin> zone --create -t ti TI_Zone_Blue -p "20,6; 20,4; 30,8; 30,10"

By default, a new TI zone is configured as “Activated” with failover enabled.
2. Display defined TI zones.
switch:admin> zone --show
Defined TI zone configuration:
TI Zone Name:

TI_Zone_Blue

Port List:

20,6; 20,4; 30,8; 30,10

Configured Status: Activated / Failover-Enabled
Enabled Status: Deactivated
TI Zone Name:

392

TI_Zone_Red

Fabric OS Administrator’s Guide
53-1002920-02

Fabric-Level Traffic Isolation in a backbone fabric

Port List:

13

20,5; 20,3; 30,7; 30,9

Configured Status: Activated / Failover-Enabled
Enabled Status: Deactivated

Note that although the configured status is “Activated”, the enabled status is “Deactivated”.
3. Activate TI zones.
switch:admin> cfgactvshow
Effective configuration:
cfg: 
…
switch:admin> cfgenable 
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected. If the update includes changes
to one or more traffic isolation zones, the update may result in
localized disruption to traffic on ports associated with
the traffic isolatoin zone changes
Do you want to enable ’TI_Config’ configuration (yes, y, no, n): [no] y
zone config "name" is in effect
Updating flash ...
switch:admin>
Switch:admin> zone --show
Defined TI zone configuration:
TI Zone Name:

TI_Zone_Blue

Port List:

20,6; 20,4; 30,8; 30,10

Configured Status: Activated / Failover-Enabled
Enabled Status: Activated / Failover-Enabled
TI Zone Name:

TI_Zone_Red

Port List:

20,5; 20,3; 30,7; 30,9

Configured Status: Activated / Failover-Enabled
Enabled Status: Activated / Failover-Enabled

Then enabled status now displays as “Activated”.

Creating a single TI zone for all paths
Create a single TI zone for all paths if you want TI traffic to failover other paths in the TI zone.
1. Create a single TI zone with failover enabled.
The TI zone must include the ingress and egress EX_Ports, as well as the E_Ports between the
two backbone switches. Do not include the edge fabric E_Ports or device PWWNs.
switch:admin> zone --create -t ti TI_Zone_ALL -p "20,3; 20,4; 20,5; 20,6;
30,7; 30,8; 30,9 30,10"

By default, a new TI zone is configured as “Activated” with failover enabled.

Fabric OS Administrator’s Guide
53-1002920-02

393

13

General rules for TI zones

2. Display defined TI zone.
switch:admin> zone --show
Defined TI zone configuration:
TI Zone Name:

TI_Zone_ALL

Port List:

20,3; 20,4 20,5; 20,6; 30,7; 30,8; 30,9; 30,10

Configured Status: Activated / Failover-Enabled
Enabled Status: Deactivated

Note that although the configured status is “Activated”, the enabled status is “Deactivated”.
3. Activate the TI zone.
switch:admin> cfgactvshow
Effective configuration:
cfg: 
…
switch:admin> cfgenable 
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected. If the update includes changes
to one or more traffic isolation zones, the update may result in
localized disruption to traffic on ports associated with
the traffic isolatoin zone changes
Do you want to enable ’TI_Config’ configuration (yes, y, no, n): [no] y
zone config "name" is in effect
Updating flash ...
switch:admin>
Switch:admin> zone --show
Defined TI zone configuration:
TI Zone Name:

TI_Zone_ALL

Port List:

20,3; 20,4; 20,5; 20,6; 30,7; 30,8; 30,9; 30,10

Configured Status: Activated / Failover-Enabled
Enabled Status: Activated / Failover-Enabled

Then enabled status now displays as “Activated”.

General rules for TI zones
The following general rules apply to TI zones:

• A TI zone must include E_Ports and N_Ports that form a complete, end-to-end route from
initiator to target.

• When an E_Port is a member of a TI zone that E_Port cannot have its indexed swapped with
another port.

• A given E_Port used in a TI zone should not be a member of more than one TI zone.
If multiple E_Ports are configured that are on the lowest cost route to a domain, the various
source ports for that zone are load-balanced across the specified E_Ports.

394

Fabric OS Administrator’s Guide
53-1002920-02

General rules for TI zones

13

• TI zones reside only in the defined configuration and not in the effective configuration. When
you make any changes to TI zones, including creating or modifying them, you must enable the
effective configuration for the changes to take effect, even if the effective configuration is
unchanged.

• A TI zone only provides traffic isolation and is not a “regular” zone.
• Routing rules imposed by TI zones with failover disabled override regular zone definitions.
Regular zone definitions should match TI zone definitions.

• FSPF supports a maximum of 16 paths to a given domain. This includes paths in a TI zone.
• Each TI zone is interpreted by each switch and each switch considers only the routing required
for its local ports. No consideration is given to the overall topology and to whether the TI zones
accurately provide dedicated paths through the whole fabric.
For example, in Figure 46, the TI zone was configured incorrectly and E_Port “3,9” was
erroneously omitted from the zone. The domain 3 switch assumes that traffic coming from
E_Port 9 is not part of the TI zone and so that traffic is routed to E_Port 11 instead of E_Port
12, if failover is enabled. If failover is disabled, the route is broken and traffic stops.
Domain 1
8

Domain 3
1

9

2

10

9
12
11

8

7
6

= Dedicated path
5
= Ports in the TI zone
Domain 4

FIGURE 46

TI zone misconfiguration

Traffic Isolation Zone violation handling for trunk ports
For any trunk group, all the members of the group need to belong to the TI zone to prevent routing
issues resulting from changes in the members of the trunk group. This applies to any E_Port or
F_Port trunk groups that are included in TI zones using failover disabled mode.
Fabric OS posts a RASlog message (ZONE-1061) if any of the ports part of a trunk group is not
added to the TI zone with failover disabled. Also, a CLI (zone --showTItrunkerrors) is provided to
check if all ports per switch in a TI zone are proper. This will help you identify missing trunk
members and take corrective actions.
Example RASlog message when any port in a trunk group is not in the TI zone
SW82:FID128:admin> zone
[ZONE-1061], 620/181, FID 128, WARNING, sw0, Some trunk members are missing
from failover disabled active TI zones.

The CLI essentially displays the details of the trunk members present in the TI zone and those
not present in the TI zone. These details are displayed per TI Zone basis.

Fabric OS Administrator’s Guide
53-1002920-02

395

13

Supported configurations for Traffic Isolation Zoning

Example RASlog message when --showTItrunkerrors is added to zone command
switch:admin> zone --showTItrunkerrors
TI Zone Name: brackets
E-Port Trunks
Trunk members in TI zone: 16 18
Trunk members not in TI zone: 17
F-Port Trunks
Trunk members in TI zone: 4 5
Trunk members not in TI zone: 6
TI Zone Name: loop
E-Port Trunks
Trunk members in TI zone: 0
Trunk members not in TI zone: 1
TI Zone Name: operand
E-Port Trunks
Trunk members in TI zone: 8
Trunk members not in TI zone: 9 10
E-Port Trunks
Trunk members in TI zone: 16
Trunk members not in TI zone: 17 18

Supported configurations for Traffic Isolation Zoning
The following configuration rules apply to TI zones:

• Ports in a TI zone must belong to switches that run Fabric OS v6.0.0 or later. For TI over FCR
zones, all switches and FC routers in both edge and backbone fabrics must be running
Fabric OS v6.1.0 or later.

• For the FC8-64 blade in the Brocade DCX and DCX 8510-8, ports 48–63 can be in a TI zone
only if all switches in that TI zone are running Fabric OS v6.4.0 or later. Ports 48–63 can still be
in a failover path for TI traffic.
The Brocade DCX-4S and DCX 8510-4 do not have this limitation.

• VE_Ports are supported in TI zones.
• TI Zoning is not supported in fabrics with switches running firmware versions earlier than
Fabric OS v6.0.0. However, the existence of a TI zone in such a fabric is backward-compatible
and does not disrupt fabric operation in switches running earlier firmware versions.
TI over FCR is not backward compatible with Fabric OS v6.0.x or earlier. The -1 in the
domain,index entries causes issues to legacy switches in a zone merge. Firmware downgrade
is prevented if TI over FCR zones exist.

Additional configuration rules for enhanced TI zones
Enhanced TI zones (ETIZ) have the following additional configuration rules:

• Enhanced TI zones are supported only if every switch in the fabric is ETIZ capable. A switch is
ETIZ capable if it meets the following qualifications:

396

-

The switch must be one of the supported platforms, as listed in “Supported hardware and
software” on page 35.

-

The switch must be running Fabric OS v6.4.0 or later.

Fabric OS Administrator’s Guide
53-1002920-02

Supported configurations for Traffic Isolation Zoning

13

• If the fabric contains a switch running an earlier version of Fabric OS, you cannot create an
enhanced TI zone. You cannot merge a downlevel switch into a fabric containing enhanced TI
zones, and you cannot merge a switch with enhanced TI zones defined into a fabric containing
switches that do not support ETIZ.

• Overlapping TI zones must have the same failover type. That is, both must be either failover
enabled or failover disabled.

NOTE
FC router domains are excluded from the ETIZ platform restrictions. You can create enhanced TI
zones with these switches in the fabric.

Trunking with TI zones
If you implement trunking and TI zones, you should keep the following points in mind:

• To include a trunk group in a TI zone, you must include all ports of the trunk in the TI zone.
• Trunked ISL ports cannot be members of more than one TI zone.
• The zone command includes an option to show TI trunk errors: zone --showTItrunkerrors
Description
This command parameter displays the details of the trunk members in the TI zone, separated
into present and not present, and displayed per TI Zone basis.
Sample output
switch:admin> zone --showTItrunkerrors
TI Zone Name: brackets
E-Port Trunks
Trunk members in TI zone: 16 18
Trunk members not in TI zone: 17
F-Port Trunks
Trunk members in TI zone: 4 5
Trunk members not in TI zone: 6
TI Zone Name: loop
E-Port Trunks
Trunk members in TI zone: 0
Trunk members not in TI zone: 1
TI Zone Name: operand
E-Port Trunks
Trunk members in TI zone: 8
Trunk members not in TI zone: 9 10
E-Port Trunks
Trunk members in TI zone: 16
Trunk members not in TI zone: 17 18

Fabric OS Administrator’s Guide
53-1002920-02

397

13

Limitations and restrictions of Traffic Isolation Zoning

Limitations and restrictions of Traffic Isolation Zoning
The following limitations and restrictions apply to Traffic Isolation Zoning:

• For switches running Fabric OS 6.1.0 or later, a maximum of 255 TI zones can be created in
one fabric. For switches running Fabric OS 6.0.x, no more than 239 TI zones should be
created.
A fabric merge resulting in greater than the maximum allowed TI zones results in merge failure
and the fabrics are segmented.

• A TI zone can be created using D,I (Domain, Index) notation only, except for TI zones in a
backbone fabric, which use port WWNs. See “Traffic Isolation Zoning over FC routers” on
page 386 for information about TI zones in a backbone fabric.

• To include a trunk group in a TI zone, you must include all ports of the trunk in the TI zone.
• If two N_Ports are online and have the same shared area, and one of them is configured in a
TI zone, then they both must be configured in that same TI zone. One of the online shared area
N_Ports should not remain outside the TI zone unless it is offline, then it may remain outside
the TI zone. This limitation does not apply to E_Ports that use the same shared area on FC4-48
and FC8-48 port blades.

• Ports that are in different TI zones cannot communicate with each other if failover is disabled.
• TI zone members that overlap must have the same TI failover policy across all TI zones to which
they belong. That is, if an overlapping member is part of a failover-disabled zone, then it can
belong only to other TI zones where the policy is also failover-disabled; the member cannot
overlap with failover-enabled TI zones.

• TI zones that have members with port index greater than 511 are not supported with Fabric OS
versions earlier than v6.4.0. If such a TI zone and Fabric OS version combination is detected,
a warning is issued. These configurations are not prevented, but their behavior is
unpredictable.

• When you merge two switches, if there is an effective configuration on the switches and
TI zones are present on either switch, the TI zones are not automatically activated after the
merge. Check the TI zone enabled status using the zone --show command, and if the TI Zone
Enabled status does not match across switches, issue the cfgEnable command.

• Use care when creating TI zones on ICL ports in topologies that span more than two switches
connected with ICLs. If a user-defined TI zone breaks the ICL connectivity requirements, a a
FSPF-1009 RASLOG entry and message is generated to notify you of this error condition.

ATTENTION
Removing a core blade when both ICL connections and lossless dynamic load sharing are enabled
may cause frame loss on a number of F_Ports.

Admin Domain considerations for Traffic Isolation Zoning
If you implement Admin Domains and TI zones, you should keep the following points in mind:

• TI zones are applicable only in AD0, and the E_Ports that are members of a TI zone must be in
the AD0 device list. Because TI zones must use D,I notation, the AD0 device list must be
declared using D,I notation for ports that are to be used in TI zones.

• A port used in a TI zone should not be a member of multiple Admin Domains.

398

Fabric OS Administrator’s Guide
53-1002920-02

Virtual Fabrics considerations for Traffic Isolation Zoning

13

• Use care if defining TI zones with ports that are shared across Admin Domains because of the
limitation that a given port can appear in only one TI zone.
Best practice: Do not use ports that are shared across Admin Domains in a TI zone.

Virtual Fabrics considerations for Traffic Isolation Zoning
This section describes how TI zones work with Virtual Fabrics. See Chapter 11, “Managing
Virtual Fabrics,” for information about the Virtual Fabrics feature, including logical switches and
logical fabrics.
TI zones can be created in a logical fabric like in regular fabrics, with the following exceptions:

• The disable failover option is not supported in logical fabrics that use XISLs.
Although logical switches that use XISLs allow the creation of a TI zone with failover disabled,
this is not a supported configuration. Base switches do not allow the creation of a TI zone with
failover disabled.

• To create a TI zone for a logical fabric that uses XISLs, you must create two TI zones: one in the
logical fabric and one in the base fabric. The combination of TI zones in the base fabric and
logical fabric sets the path through the base fabric for logical switches.
The TI zone in the logical fabric includes the extended XISL (XISL) port numbers, as well as the
F_Ports and ISLs in the logical fabric.
The TI zone in the base fabric reserves XISLs for a particular logical fabric. The base fabric TI zone
should also include ISLs that belong to logical switches participating in the logical fabric.
Figure 47 shows an initiator and target in a logical fabric (FID1). The dotted line indicates a
dedicated path between initiator and target. The dedicated path passes through the base fabric
over an XISL. (Figure 47 shows only physical ISLs, not logical ISLs.) To create the TI zones for this
dedicated path, you must create a TI zone in the logical fabric (FID 1) and one in the base fabric.
Host

Domain 8

8

9
1

2

3

4

LS3, FID1
Domain 3
Chassis 1

Target

Domain 9

LS4, FID3
Domain 4

10

XISL

12

11

XISL

13

6

8

7

LS1, FID1
Domain 5

Domain 7

Base switch
Domain 1

5

14
15

XISL

16

XISL

17

LS2, FID3
Domain 6

Chassis 2

Base switch
Domain 2

= Dedicated Path

FIGURE 47

Dedicated path with Virtual Fabrics

Figure 48 shows a logical representation of FID1 in Figure 47. To create the dedicated path, you
must create and activate a TI zone in FID1 that includes the circled ports shown in Figure 48.

Fabric OS Administrator’s Guide
53-1002920-02

399

13

Virtual Fabrics considerations for Traffic Isolation Zoning

Domain 8

Host

Domain 3
2

4

Domain 5

Domain 9

11

17

7

6

10

16

8

5

8

Target

9
1

3

= Dedicated Path
= Ports in the TI zones

FIGURE 48

Creating a TI zone in a logical fabric

You must also create and activate a TI zone in the base fabric to reserve the XISLs for the dedicated
path. In Figure 49, the XISLs highlighted (by a dotted line) in the base fabric can be reserved for
FID1 by defining and activating a base fabric TI zone that consists of ports 10, 12, 14, and 16. You
must also include ports 3 and 8, because they belong to logical switches participating in the logical
fabric. For the TI zone, it is as though ports 3 and 8 belong to Domains 1 and 2 respectively.
Domain 1

Domain 7
11

13

4
3

10

12

Domain 2
15

14

17

16

7
8

= Dedicated Path
= Ports in the TI zones

FIGURE 49

Creating a TI zone in a base fabric

Using D,I notation, the port numbers for the TI zones in the logical fabric and base fabric are as
follows:
Port members for the TI zone in logical fabric Port members for the TI zone in base fabric
8,8 F_Port
8,1 E_Port
3,3 E_Port
3,10 E_Port
5,16 E_Port
5,8 E_Port
9,5 E_Port
9,9 F_Port

1,3 E_Port for ISL in logical switch
1,10 E_Port for XISL
7,12 E_Port for XISL
7,14 E_Port for XISL
2,16 E_Port for XISL
2,8 E_Port for ISL in logical switch

Notice that the base fabric zone contains a reference to port 1,3 even though the base switch with
domain 1 does not have a port 3 in the switch. This number refers to the port in the chassis with
port index 3, which actually belongs to LS3 in FID 1.

400

Fabric OS Administrator’s Guide
53-1002920-02

Traffic Isolation Zoning over FC routers with Virtual Fabrics

13

Traffic Isolation Zoning over FC routers with Virtual Fabrics
This section describes how you can set up TI zones over FC routers in logical fabrics. Figure 50
shows two physical chassis configured into logical switches. The initiator in FID 1 communicates
with the target in FID 3 over the EX_Ports in the base switches.

1

10
F

2

F

E
3
E
4
5

EX

11

LS2, FID3
Domain 6

LS3, FID1
Domain 3

E
E

E

Base switch
Domain 1

EX

E

6

15

7

16

E

EX
Base switch
Domain 2

E

12

13
14

EX

= Dedicated Path
= Ports in the TI zones

FIGURE 50

Example configuration for TI zones over FC routers in logical fabrics

Figure 51 shows a logical representation of the configuration in Figure 50. This SAN is similar to
that shown in Figure 42 on page 387 and you would set up the TI zones in the same way as
described in “Traffic Isolation Zoning over FC routers” on page 386.

Edge fabric
Fabric 1
1
SW3

3
10

2
12

4
5

SW1

FIGURE 51

Fabric OS Administrator’s Guide
53-1002920-02

SW6
11

6
15 13

7
Backbone fabric

Edge fabric
Fabric 3

16

SW2

14

Logical representation of TI zones over FC routers in logical fabrics

401

13

Creating a TI zone

Creating a TI zone
You create and modify TI zones using the zone command. Other zoning commands, such as
zoneCreate, aliCreate, and cfgCreate, cannot be used to manage TI zones.
When you create a TI zone, you can set the state of the zone to activated or deactivated. By default
the zone state is set to activated; however, this does not mean that the zone is activated. After you
create the TI zone, you must enable the current effective configuration to enforce the new TI zone,
which is either activated or deactivated.
Virtual Fabric considerations: Because base fabrics do not contain end devices, they normally do
not have an effective zone configuration. To activate a TI zone in a base fabric, you should create a
“dummy” configuration, as described in “Creating a TI zone in a base fabric” on page 404.
When you create a TI zone, you can enable or disable failover mode. By default, failover mode is
enabled. If you want to change the failover mode after you create the zone, see “Modifying TI
zones” on page 405.
If you are creating a TI zone with failover disabled, note the following:

• Ensure that the E_Ports of the TI zone correspond to valid paths; otherwise, the route might be
missing for ports in that TI zone. You can use the topologyShow command to verify the paths.

• Ensure that sufficient non-dedicated paths through the fabric exist for all devices that are not
in a TI zone; otherwise, these devices might become isolated.
See “TI zone failover” on page 380 for information about disabling failover mode.
Use the following procedure to create a TI zone. If you are creating a TI zone in a base fabric, use
the procedure described in “Creating a TI zone in a base fabric” on page 404.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zone --create command:
zone --create -t objtype [-o optlist] name -p "portlist"

Be aware of the ramifications if you create a TI zone with failover mode disabled. See “TI zone
failover” on page 380 for information about disabling failover mode.
3. Perform the following steps if you have any TI zones with failover disabled. If all of your TI zones
are failover-enabled, skip to step 4.
a.

Change the failover option to failover enabled. This is a temporary change to avoid frame
loss during the transition.
zone --add -o f name

b.

Enable the zones.
cfgenable "current_effective_configuration"

c.

Reset the failover option to failover disabled. Then continue with step 4.
zone --add -o n name

4. Enter the cfgEnable command to reactivate your current effective configuration and enforce
the TI zones.
cfgenable "current_effective_configuration"

402

Fabric OS Administrator’s Guide
53-1002920-02

Creating a TI zone

13

Example TI zone creation

The following examples create a TI zone named “bluezone”, which contains E_Ports 1,1 and 2,4
and N_Ports 1,8 and 2,6.
To create a TI zone with failover enabled and in the activated state (default settings):
switch:admin> zone --create -t ti bluezone -p "1,1; 2,4; 1,8; 2,6"

To create a TI zone with failover enabled (the zone is set to the activated state by default):
switch:admin> zone --create -t ti -o f bluezone -p "1,1; 2,4; 1,8; 2,6"

To create a TI zone with failover disabled and the state set to activated:
switch:admin> zone --create -t ti -o an bluezone -p "1,1; 2,4; 1,8; 2,6"

To create a TI zone and set the state to deactivated (failover is enabled by default):
switch:admin> zone --create -t ti -o d bluezone -p "1,1; 2,4; 1,8; 2,6"

To create a TI zone with failover disabled and the state set to deactivated:
switch:admin> zone --create -t ti -o dn bluezone -p "1,1; 2,4; 1,8; 2,6"

To create a TI zone in the edge fabric with failover enabled and the state set to activated (default
settings):
switch:admin> zone --create -t ti bluezone -p "1,1; 1,8; 2,-1; 3,-1"

To create a TI zone in the backbone fabric with failover enabled and the state set to activated
(default settings):
switch:admin> zone --create -t ti backbonezone -p "10:00:00:04:1f:03:16:f2;
1,1; 1,4; 2,7; 2,1; 10:00:00:04:1f:03:18:f1, 10:00:00:04:1f:04:06:e2"

To create TI zones in a logical fabric, such as the one shown in Figure 48 on page 400:
Log in to the logical switch FID1, Domain 7 and create a TI zone in the logical fabric with FID=1:
LS1> zone --create -t ti -o f "ti_zone1" -p "8,8; 8,1; 3,3; 3,10; 5,16; 5,8;
9,5; 9,9"

Then create a TI zone in the base fabric, as described in “Creating a TI zone in a base fabric”.
Remember that your changes are not enforced until you enter the cfgEnable command, as shown
here:
switch:admin> cfgenable "USA_cfg"
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected.
If the update includes changes to one or more traffic isolation zones, the
update may result in localized disruption to traffic on ports associated with
the traffic isolation zone changes
Do you want to enable 'USA_cfg' configuration (yes, y, no, n): [no] y
zone config "USA_cfg" is in effect
Updating flash ...

Fabric OS Administrator’s Guide
53-1002920-02

403

13

Creating a TI zone

Creating a TI zone in a base fabric
1. Connect to the switch and log in using an account with admin permissions.
2. Create a “dummy” zone configuration in the base fabric. For example:
zone --create "z1", "1,1"
cfgcreate "base_config", z1

3. Enter the zone --create command to create the TI zone in the base fabric:
zone --create -t objtype -o f name -p "portlist"

The disable failover option is not supported in base fabrics.
4. Perform the following steps if you have any TI zones with failover disabled. If all of your TI zones
are failover-enabled, skip to step 5.
a.

Change the failover option to failover enabled. This is a temporary change to avoid frame
loss during the transition.
zone --add -o f name

b.

Enable the zones.
cfgenable "current_effective_configuration"

c.

Reset the failover option to failover disabled. Then continue with step 4.
zone --add -o n name

5. Enter the cfgEnable command to reactivate your current effective configuration and enforce
the TI zones.
cfgenable "base_config"

Example

The following example creates TI zones in the base fabric shown in Figure 49 on page 400:
BS_D1>
BS_D1>
BS_D1>
2,8"
BS_D1>

404

zonecreate "z1", "1,1"
cfgcreate "base_cfg", z1
zone --create -t ti -o f "ti_zone2" -p "1,3; 1,10; 7,12; 7,14; 2,16;
cfgenable "base_config"

Fabric OS Administrator’s Guide
53-1002920-02

Modifying TI zones

13

Modifying TI zones
Using the zone --add command, you can add ports to an existing TI zone, change the failover
option, or both.You can also activate or deactivate the TI zone.
Using the zone --remove command, you can remove ports from existing TI zones. If you remove the
last member of a TI zone, the TI zone is deleted.
After you modify the TI zone, you must enable the current effective configuration to enforce the
changes.

ATTENTION
If failover is disabled, do not allocate all ISLs in TI zones. Make sure sufficient non-dedicated paths
exist through the fabric for all devices that are not in a TI zone. See “TI zone failover” on page 380
for additional information about disabling failover mode.

NOTE

If you have overlapping TI zones and you want to change the failover option on these zones, you must
first remove the overlapping ports from the zones, then change the failover type, and finally re-add
the overlapping members.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter one of the following commands, depending on how you want to modify the TI zone.

• Enter the zone --add command to add ports or change the failover option for an existing TI
zone. You can also activate or deactivate the zone.
zone --add [-o optlist] name -p "portlist"
zone --add -o optlist name [-p "portlist"]

• Enter the zone --remove command to remove ports from an existing TI zone.
zone --remove name -p "portlist"

Be aware of the ramifications if you disable failover mode. See “TI zone failover” on page 380
for information about disabling failover mode.
3. Perform the following steps if you have any TI zones with failover disabled. If all of your TI zones
are failover-enabled, skip to step 4.
a.

Change the failover option to failover enabled. This is a temporary change to avoid frame
loss during the transition.
zone --add -o f name

b.

Enable the zones.
cfgenable "current_effective_configuration"

c.

Reset the failover option to failover disabled. Then continue with step 4.
zone --add -o n name

4. Enter the cfgEnable command to reactivate your current effective configuration and enforce
the TI zones.
cfgenable "current_effective_configuration"

Fabric OS Administrator’s Guide
53-1002920-02

405

13

Changing the state of a TI zone

Example of modifying a TI zone

To add port members to the existing TI zone bluezone:
switch:admin> zone --add bluezone -p "3,4; 3,6"

To add port members to the existing TI zone in a backbone fabric:
switch:admin> zone --add backbonezone -p "3,4; 3,6; 10:00:00:04:1f:03:16:f2;"

To disable failover on the existing TI zone bluezone:
switch:admin> zone --add -o n bluezone

To enable failover and add ports to TI zone greenzone:
switch:admin> zone --add -o f greenzone -p "3,4"

To remove ports from the TI zone bluezone:
switch:admin> zone --remove bluezone -p "3,4; 3,6"

Remember that your changes are not enforced until you enter the cfgEnable command.

Changing the state of a TI zone
You can change the state of a TI zone to activated or deactivated. Changing the state does not
activate or deactivate the zone. After you change the state of the TI zone, you must enable the
current effective configuration to enforce the change.
The TI zone must exist before you can change its state.
1. Connect to the switch and log in using an account with admin permissions.
2. Perform one of the following actions:

• To activate a TI zone, enter the zone --activate command.
zone --activate name

• To deactivate a TI zone, enter the zone --deactivate command.
zone --deactivate name

3. Enter the cfgEnable command to reactivate your current effective configuration and enforce
the TI zones.
cfgenable "current_effective_configuration"

Example of setting the state of a TI zone

To change the state of the existing TI zone bluezone to activated, type:
switch:admin> zone --activate bluezone

To change the state of the existing TI zone greenzone to deactivated, type:
switch:admin> zone --deactivate greenzone

Remember that your changes are not enforced until you enter the cfgEnable command.

406

Fabric OS Administrator’s Guide
53-1002920-02

Deleting a TI zone

13

Deleting a TI zone
Use the zone --delete command to delete a TI zone from the defined configuration. This command
deletes the entire zone; to only remove port members from a TI zone, use the zone --remove
command, as described in “Modifying TI zones” on page 405.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zone --delete command.
zone --delete name

You can delete multiple zones by separating the zone names with a semicolon and enclosing
them in quotation marks.
3. Enter the cfgEnable command to reactivate your current effective configuration and enforce
the TI zones.
cfgenable "current_effective_configuration"

Example of deleting a TI zone

To delete the TI zone bluezone, type:
switch:admin> zone --delete bluezone

Remember that your changes are not enforced until you enter the cfgEnable command.

Displaying TI zones
Use the zone --show command to display information about TI zones. This command displays the
following information for each zone:

•
•
•
•
•

Zone name
E_Port members
N_Port members
Configured status (the latest status, which may or may not have been activated by cfgEnable)
Enabled status (the status that has been activated by cfgEnable)

If you enter the cfgShow command to display information about all zones, the TI zones appear in
the defined zone configuration only and do not appear in the effective zone configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zone --show command.
zone --show [ name ] [-ascending]

Example displaying information about the TI zone purplezone
switch:admin> zone --show purplezone
Defined TI zone configuration:
TI Zone Name:
Port List:

redzone:
1,2; 1,3; 3,3; 4,5

Configured Status: Activated / Failover-Enabled
Enabled Status: Activated / Failover-Enabled

Fabric OS Administrator’s Guide
53-1002920-02

407

13

Troubleshooting TI zone routing problems

Example displaying information about all TI zones in the defined configuration in ascending order
switch:admin> zone --show -ascending
Defined TI zone configuration:
TI Zone Name:
Port List:

bluezone:
8,3; 8,5; 9,2; 9,3;

Configured Status: Deactivated / Failover-Disabled
Enabled Status: Activated / Failover-Enabled
TI Zone Name:
Port List:

greenzone:
2,2; 3,3; 4,11; 5,3;

Configured Status: Activated / Failover-Enabled
Enabled Status: Activated / Failover-Enabled
TI Zone Name:
Port List:

purplezone:
1,2; 1,3; 3,3; 4,5;

Configured Status: Activated / Failover-Enabled
Enabled Status: Deactivated / Failover-Enabled

Example displaying members for the zone "ti_red" in ascending order
switch:admin> zone --show -ascending ti_red
Defined TI zone configuration:
TI Zone Name:
Port List:

ti_red
3,3; 4,4; 5,5

Configured Status: Activated / Failover-Enabled
Enabled Status: Deactivated

Example displaying members for the zone "TI_zone", regardless of the case
switch:admin> zone --show -ic TI_zone*
Defined TI zone configuration:
TI Zone Name:
Port List:

TI_zone
7,8

Configured Status: Activated / Failover-Enabled
Enabled Status: Deactivated
TI Zone Name:
Port List:

ti_zone
3,3

Configured Status: Activated / Failover-Enabled
Enabled Status: Deactivated

Troubleshooting TI zone routing problems
Use the following procedure to generate a report of existing and potential problems with TI zones.
The report displays an error type.

• “ERROR” indicates a problem currently exists in the fabric.

408

Fabric OS Administrator’s Guide
53-1002920-02

Setting up TI zones over FCR (sample procedure)

13

• “WARNING” indicates that there is not currently a problem, given the current set of online
devices and reachable domains, but given the activated TI zone configuration, parallel
exclusive paths between a shared device and a remote domain have been detected, which
might cause a problem for devices that join the fabric later.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zone --showTIerrors command.
zone --showTIerrors

Here is an example report that would be generated for the illegal configuration shown in Figure 40
on page 385.
switch:admin> zone --showTIerrors
My Domain: 3
Error type:
ERROR
Affected Remote Domain: 1
Affected Local Port:
8
Affected TI Zones:
etiz1, etiz2
Affected Remote Ports: 1, 2, 3, 4

Setting up TI zones over FCR (sample procedure)
The following example shows how to set up TI zones over FCR to provide a dedicated path shown in
Figure 52. In this example, three TI zones are created: one in each of the edge fabrics and one in
the backbone fabric. The combination of these three TI zones creates a dedicated path for traffic
between Host 1 in edge fabric 1 and Targets 1 and 2 in edge fabric 2.
Host 1 has port WWN 10:00:00:00:00:08:00:00
Target 1 has port WWN 10:00:00:00:00:02:00:00
Target 2 has port WWN 10:00:00:00:00:03:00:00
Host 1

Target 1

Target 2

Domain ID = 1
Domain ID = 2
2

9

8
5

3

6
1

7
4

Edge fabric 1
Domain ID = 4

Backbone
fabric

Edge fabric 2
Domain ID = 9

= Dedicated path set up by TI zone in edge fabric 1
= Dedicated path set up by TI zone in edge fabric 2
= Dedicated path set up by TI zone in backbone fabric

FIGURE 52

Fabric OS Administrator’s Guide
53-1002920-02

TI over FCR example

409

13

Setting up TI zones over FCR (sample procedure)

NOTE

In the following procedure the three TI zones in the edge and backbone fabrics are all given the same
name, TI_Zone1. It is not required that the TI zones have the same name, but this is done to avoid
confusion. If several dedicated paths are set up across the FC router, the TI zones for each path can
have the same name.
1. In each edge fabric, set up an LSAN zone that includes Host 1, Target 1, and Target 2, so these
devices can communicate with each other. See Chapter 26, “Using FC-FC Routing to Connect
Fabrics,” for information about creating LSAN zones.
2. Log in to the edge fabric 1 and set up the TI zone.
a.

Enter the fabricShow command to display the switches in the fabric. From the output, you
can determine the front and translate domains.

E1switch:admin> fabricshow
Switch ID
Worldwide Name
Enet IP Addr
FC IP Addr
Name
------------------------------------------------------------------------1: fffc01 50:00:51:e3:95:36:7e:04 0.0.0.0
0.0.0.0
"fcr_fd_1"
4: fffc04 10:00:00:60:69:80:1d:bc 10.32.72.4
0.0.0.0
>"E1switch"
6: fffc06 50:00:51:e3:95:48:9f:a0 0.0.0.0
0.0.0.0
"fcr_xd_6_9"

The Fabric has 3 switches

b.

Enter the following commands to create and display a TI zone:
E1switch:admin> zone --create -t ti TI_Zone1 -p "4,8; 4,5, 1,-1; 6,-1"
E1switch:admin> zone --show
Defined TI zone configuration:
TI Zone Name:
Port List:

TI_Zone1
4,8; 4,5; 1,-1; 6,-1

Status: Activated

c.

Failover: Enabled

Enter the following commands to reactivate your current effective configuration and
enforce the TI zones.
E1switch:admin> cfgactvshow
Effective configuration:
cfg:
cfg_TI
zone: lsan_t_i_TI_Zone1
10:00:00:00:00:00:02:00:00
10:00:00:00:00:00:03:00:00
10:00:00:00:00:00:08:00:00
E1switch:admin> cfgenable cfg_TI
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected.
If the update includes changes to one or more traffic isolation zones, the
update may result in localized disruption to traffic on ports associated
with the traffic isolation zone changes
Do you want to enable 'cfg_TI' configuration (yes, y, no, n): [no] y
zone config "cfg_TI" is in effect
Updating flash ...

410

Fabric OS Administrator’s Guide
53-1002920-02

Setting up TI zones over FCR (sample procedure)

13

3. Log in to the edge fabric 2 and set up the TI zone.
a.

Enter the fabricShow command to display the switches in the fabric. From the output, you
can determine the front and translate domains.

E2switch:admin> fabricshow
Switch ID
Worldwide Name
Enet IP Addr
FC IP Addr
Name
------------------------------------------------------------------------1: fffc01 50:00:51:e3:95:36:7e:09 0.0.0.0
0.0.0.0
"fcr_fd_1"
4: fffc04 50:00:51:e3:95:48:9f:a1 0.0.0.0
0.0.0.0
"fcr_xd_6_9"
9: fffc09 10:00:00:05:1e:40:f0:7d 10.32.72.9
0.0.0.0
>"E2switch"
The Fabric has 3 switches

b.

Enter the following commands to create and display a TI zone:
E2switch:admin> zone --create -t ti TI_Zone1 -p "9,2; 9,3; 9,6; 1,-1; 4,-1"
E2switch:admin> zone --show
Defined TI zone configuration:
TI Zone Name:
Port List:

TI_Zone1
9,2; 9,3; 9,6; 1,-1; 4,-1

Status: Activated

c.

Failover: Enabled

Enter the following commands to reactivate your current effective configuration and
enforce the TI zones.
E2switch:admin> cfgactvshow
Effective configuration:
cfg:
cfg_TI
zone: lsan_t_i_TI_Zone1
10:00:00:00:00:00:02:00:00
10:00:00:00:00:00:03:00:00
10:00:00:00:00:00:08:00:00
E2switch:admin> cfgenable cfg_TI
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected.
If the update includes changes to one or more traffic isolation zones, the
update may result in localized disruption to traffic on ports associated
with the traffic isolation zone changes
Do you want to enable 'cfg_TI' configuration (yes, y, no, n): [no] y
zone config "cfg_TI" is in effect
Updating flash ...

4. Log in to the backbone fabric and set up the TI zone.
a.

Enter the following commands to create and display a TI zone:
BB_DCX_1:admin> zone --create -t ti TI_Zone1 -p "1,9; 1,1; 2,4; 2,7;
10:00:00:00:00:08:00:00; 10:00:00:00:00:02:00:00; 10:00:00:00:00:03:00:00"
BB_DCX_1:admin> zone --show
Defined TI zone configuration:
TI Zone Name:
TI_Zone1
Port List:
1,9; 1,1; 2,4; 2,7; 10:00:00:00:00:08:00:00;
10:00:00:00:00:02:00:00; 10:00:00:00:00:03:00:00
Status: Activated

Fabric OS Administrator’s Guide
53-1002920-02

Failover: Enabled

411

13

Setting up TI zones over FCR (sample procedure)

b.

Enter the following commands to reactivate your current effective configuration and
enforce the TI zones.
BB_DCX_1:admin> cfgactvshow
Effective configuration:
cfg:
cfg_TI
zone: lsan_t_i_TI_Zone1
10:00:00:00:00:00:02:00:00
10:00:00:00:00:00:03:00:00
10:00:00:00:00:00:08:00:00
BB_DCX_1:admin> cfgenable cfg_TI
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected.
If the update includes changes to one or more traffic isolation zones, the
update may result in localized disruption to traffic on ports associated
with the traffic isolation zone changes
Do you want to enable 'cfg_TI' configuration (yes, y, no, n): [no] y
zone config "cfg_TI" is in effect
Updating flash ...

412

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

14

Optimizing Fabric Behavior

In this chapter
• Adaptive Networking overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Ingress Rate Limiting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• QoS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• CS_CTL-based frame prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• QoS zone-based traffic prioritization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• QoS zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Setting QoS zone-based traffic prioritization . . . . . . . . . . . . . . . . . . . . . . . .
• Setting QoS zone-based traffic prioritization over FC routers . . . . . . . . . .
• Disabling QoS zone-based traffic prioritization . . . . . . . . . . . . . . . . . . . . . .

413
414
415
416
419
419
424
426
426

Adaptive Networking overview
Adaptive Networking is a suite of tools and capabilities that enable you to ensure optimized
behavior in the SAN. Under the worst congestion conditions, Adaptive Networking can maximize the
fabric behavior and provide necessary bandwidth for high-priority, mission-critical applications and
connections.
The Adaptive Networking suite includes the following features:

• Bottleneck detection
The bottleneck detection feature identifies devices attached to the fabric that are slowing
down traffic. Bottleneck detection does not require a license. Refer to Chapter 15, “Bottleneck
Detection,” for information about this feature.

• Top Talkers
The Top Talkers feature provides real-time information about the top n bandwidth-consuming
flows passing through a specific port in the network. Top Talkers requires a Fabric Vision
license or an Advanced Performance Monitoring license. Refer to “Top Talker monitors” on
page 562 for more information about this feature.

• Traffic Isolation Zoning
Traffic Isolation Zoning (TI zoning) allows you to control the flow of interswitch traffic by creating
a dedicated path for traffic flowing from a specific set of source ports (F_Ports). Traffic
Isolation Zoning does not require a license. Refer to Chapter 13, “Traffic Isolation Zoning,” for
more information about this feature.

Fabric OS Administrator’s Guide
53-1002920-02

413

14

Ingress Rate Limiting

• Ingress Rate Limiting
Ingress Rate Limiting restricts the speed of traffic from a particular device to the switch port.
Ingress Rate Limiting does not require a license. Refer to “Ingress Rate Limiting” on page 414
for more information about this feature.

• Quality of Service (QoS)
QoS allows you to categorize the traffic flow between a host and target as having a high,
medium, or low priority. QoS does not require a license. Refer to “QoS” on page 415 for more
information about this feature.
You can use the Adaptive Networking features together to optimize the performance of your fabric.
For example, you can use the features in the following ways:

• You can use Top Talkers to identify the SID/DID pairs that consume the most bandwidth and
can then configure them with certain QoS attributes so they get proper priority.

• If the bottleneck detection feature detects a latency bottleneck, you can use TI zones or QoS to
isolate latency device traffic from high-priority application traffic.

• If the bottleneck detection feature detects ISL congestion, you can use Ingress Rate Limiting to
slow down low-priority application traffic if it is contributing to the congestion.

Ingress Rate Limiting
Ingress Rate Limiting restricts the speed of traffic from a particular device to the switch port. Use
Ingress Rate Limiting for the following situations:

• To reduce existing congestion in the network or proactively avoid congestion.
• To enable you to offer flexible bandwidth-limit services based on requirements.
• To enable more important devices to use the network bandwidth during specific services, such
as network backup.
To limit the traffic, you set the maximum speed at which the traffic can flow through a particular
F_Port or FL_Port. For example, if you set the rate limit at 4 Gbps, then traffic from a particular
device is limited to a maximum of 4 Gbps.
Ingress Rate Limiting enforcement is needed only if the port can run at a speed higher than the
rate limit. For example, if the rate limit is 4 Gbps and the port is only a 2-Gbps port, then Ingress
Rate Limiting is not enforced.
The Ingress Rate Limiting configuration is persistent across reboots.
You should keep in mind the following considerations about Ingress Rate Limiting:

• Ingress Rate Limiting is applicable only to F_Ports and FL_Ports.
• QoS takes precedence over Ingress Rate Limiting.
• Ingress Rate Limiting is not enforced on trunked ports.

Virtual Fabrics considerations
If Virtual Fabrics is enabled and if a port is configured to have a certain rate limit value, you must
first disable the rate limit on the port before moving it to a different logical switch. Ports cannot be
moved when they have rate limit configured on them.

414

Fabric OS Administrator’s Guide
53-1002920-02

QoS

14

Limiting traffic from a particular device
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portCfgQos --setratelimit command.
portcfgqos --setratelimit [slot/]port ratelimit

Example of setting the rate limit on slot 3, port 9 to 4000 Mbps
portcfgqos --setratelimit 3/9 4000

Disabling Ingress Rate Limiting
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portCfgQos --resetratelimit command.
portcfgqos --resetratelimit [slot/]port

Example of disabling Ingress Rate Limiting on slot 3, port 9
portcfgqos --resetratelimit 3/9

QoS
Quality of Service (QoS) allows you to categorize the traffic flow between a host and a target as
having a high, medium, or low priority.
Fabric OS supports two types of prioritization:

• Class-Specific Control (CS_CTL)-based frame prioritization
Each frame between a host and a target is assigned a specific priority, depending on the value
of the CS_CTL field in the frame header.

• QoS zone-based traffic prioritization
All traffic between a host and a target is assigned a specific priority, depending on the name
you define for the QoS zone.
CS_CTL-based frame prioritization and QoS zone-based traffic prioritization are mutually exclusive.
If you enable CS_CTL-based frame prioritization on F_Ports or FL_Ports, then QoS zone-based
traffic prioritization cannot be used between any devices connected to the F_Ports or FL_Ports.
CS_CTL-based frame prioritization takes precedence over QoS zone-based traffic prioritization. If
you enable CS_CTL-based frame prioritization on F_Ports or FL_Ports that are defined in a QoS
zone, CS_CTL-based frame prioritization takes precedence over the QoS zones.
Table 71 shows a basic comparison between CS-CTL-based frame prioritization and QoS
zone-based traffic prioritization. Refer to “CS_CTL-based frame prioritization” on page 416 and
“QoS zone-based traffic prioritization” on page 419 for detailed information about each type of
prioritization scheme.

Fabric OS Administrator’s Guide
53-1002920-02

415

14

CS_CTL-based frame prioritization

TABLE 71

Comparison between CS_CTL-based and QoS zone-based prioritization

CS_CTL-based frame prioritization

QoS zone-based traffic prioritization

Must be manually enabled.

Automatically enabled.

No zones are required.

Requires you to create QoS zones.

Enabled on F_Ports or FL_Ports.

Enabled on E_Ports.

Takes precedence over QoS zone-based traffic
prioritization.

Is overridden by CS_CTL-based frame prioritization.

Priority is defined by CS-CTL field in frame header.

Priority is defined by name of QoS zone.

Prioritization is on a frame-basis.

Prioritization is on a flow-basis.

Setup steps:
• Enable QoS on F_Ports or FL_Ports.
• Enable QoS on E_Ports.

Setup steps:
Create QoS zones with host/target members.
Add the QoS zones to the zone configuration.
Save and then enable the zone configuration.
Enable QoS on E_Ports.

•
•
•
•

License requirements for QoS
Starting in Fabric OS 7.2.0, QoS does not require the Adaptive Networking license to be explicitly
installed. This license is automatically enabled for new switches and for existing switches that are
upgraded to Fabric OS 7.2.0 or later.
If you upgrade to Fabric OS 7.2.0 and you did not previously have an Adaptive Networking license,
then all ports that had QoS mode set to AE (automatically enabled) and were not using the QoS
feature are automatically set to OFF after the upgrade.

CS_CTL-based frame prioritization
CS_CTL-based frame prioritization allows you to prioritize the frames between a host and a target
as having high, medium, or low priority, depending on the value of the CS_CTL field in the FC frame
header.
The CS_CTL field in the FC header can be used to assign a priority to a frame. This field can be
populated by selected end devices (storage or host) and then honored by the switch, which assigns
the frame, based on the value in the CS_CTL field, to allocate appropriate resources throughout the
fabric. This method of establishing QoS is an alternative to the switch-controlled assignment that
uses zone-based QoS.

ATTENTION
Check with your host and storage manufacturers to determine whether they support Fibre Channel
CS_CTL prioritization on their devices.
High-, medium-, and low-priority frames are allocated to different sets of fabric resources.
High-priority frames are assigned more fabric resources than medium-priority frames, which in turn
are assigned more fabric resources than low-priority frames. The resources are allocated according
to the CS_CTL value, as shown in Table 72. The values are enabled by default to ensure backward
compatibility.

416

Fabric OS Administrator’s Guide
53-1002920-02

CS_CTL-based frame prioritization

TABLE 72

14

Mapping of CS_CTL values to QoS priority for frame prioritization in CS_CTL default mode

CS_CTL value

Priority

1–8

Low

9–16

Medium

17–24

High

Alternatively, the user can apply CS_CTL auto mode. The CS_CTL auto mode uses only three
CS_CTL values, as illustrated in Table 73.

TABLE 73

Mapping of CS_CTL values to QoS priority for frame prioritization in CS_CTL auto mode

CS_CTL value

Priority

1

Low

2

Medium

3

High

NOTE
The values in the tables represent chassis-level configurations. For configuration details, refer to
“Using CS_CTL auto mode at the chassis level” on page 418, and “Considerations for using
CS_CTL-based frame prioritization” on page 418.

Supported configurations for CS_CTL-based frame prioritization
CS_CTL-based frame prioritization is supported on all 8-Gbps and 16-Gbps platforms.
All switches in the fabric should be running Fabric OS v6.0.0 or later.

NOTE

If a switch is running a firmware version earlier than Fabric OS v6.0.0, the outgoing frames from that
switch lose their priority.

High availability considerations for CS_CTL-based frame prioritization
If the standby CP is running a Fabric OS version earlier than 6.3.0 and is synchronized with the active
CP, then you cannot enable CS_CTL-based frame prioritization on the active CP. If the standby CP is not
synchronized or if no standby CP exists, then enabling CS_CTL-based frame prioritization succeeds.

Enabling CS_CTL-based frame prioritization on ports
When you enable CS_CTL-based frame prioritization, you must enable it on both the source port and
the destination port, so that the frames returned from the destination port for a given exchange
always have the same CS_CTL prioritization as the frames originating from the source port.
1. Connect to the switch and log in to an account that has admin permissions.
2. Enable CS_CTL mode:
portcfgqos --enable [slot/]port csctl_mode

3. Enter y at the prompt to override QoS zone-based traffic prioritization.

Fabric OS Administrator’s Guide
53-1002920-02

417

14

CS_CTL-based frame prioritization

Disabling CS_CTL-based frame prioritization on ports
When you disable CS_CTL-based frame prioritization, QoS zone-based traffic prioritization is
restored if it had been previously enabled.
1. Connect to the switch and log in to an account that has admin permissions.
2. Disable CS_CTL mode:
portcfgqos --disable [slot/]port csctl_mode

Alternatively, you can disable CS_CTL mode and set QoS to auto-enable (AE):
portcfgqos --default [slot/]port csctl_mode

Using CS_CTL auto mode at the chassis level
You can use the CS_CTL QoS mode options for the configureChassis CLI command to change the
chassis-wide default mode (refer to Table 72 on page 417), as in the following example.
switch:admin> configurechassis
Configure...
cfgload attributes (yes, y, no, n): [no] y
Enforce secure config Upload/Download (yes, y, no, n): [no]
Enforce signature validation for firmware (yes, y, no, n): [no]
Add Suffix to the uploaded file name (yes, y, no, n): [no]
Custom attributes (yes, y, no, n): [no] y
Config Index (0 to ignore): (0..1000) [0]
system attributes (yes, y, no, n): [no] y
system.blade.bladeFaultOnHwErrMsk: (0x0..0x7fffffff) [0x0]
system.cpuLoad: (10..121) [121]
system.i2cTurboCnfg: (0..2) [1]
fos attributes (yes, y, no, n): [no] y
CSCTL QoS Mode (0 = default; 1 = auto mode): (0..1) [1]

Set CSCTL QoS Mode to 1 to enable auto mode, establishing the settings shown in Table 73 on
page 417. Set CSCTL QoS Mode to 0 to disable auto mode and revert to default settings, shown in
Table 72 on page 417.

NOTE

As noted previously, this is a chassis-level configuration. It does not provide options to enable
CS_CTL QoS on the ports.

Considerations for using CS_CTL-based frame prioritization
To use CS_CTL for QoS on a given port for a given flow, proceed with the following steps.
1. Determine whether to use the default mode (refer to Table 72 on page 417) or the auto mode
(refer to Table 73 on page 417). No choice results in the default mode.
2. In either case, ensure that the switch port connected to the initiator host and the switch port
connected to the target host have csctl_mode enabled, as in “Enabling CS_CTL-based frame
prioritization on ports” on page 417.

418

Fabric OS Administrator’s Guide
53-1002920-02

QoS zone-based traffic prioritization

14

QoS zone-based traffic prioritization
QoS zone-based traffic prioritization allows you to categorize the traffic flow between a host and a
target as having a high, medium, or low priority, depending on the type of zone.
For example, you could assign online transaction processing (OLTP) to high priority and backup
traffic to low priority.
All flows without QoS prioritization are considered medium priority.
High-, medium-, and low-priority flows are allocated to different virtual channels (VCs). High-priority
flows receive more fabric resources than medium-priority flows, which receive more resources than
low-priority flows.

NOTE

If there is a single low-priority flow to a destination ID (DID) and several medium-priority flows to that
same DID, then it is possible that the medium-priority flows would have less bandwidth. This is
because they have to share the medium-priority fabric resources, whereas the low-priority flow would
have a separate set of fabric resources for its exclusive use.
For new switches, QoS zone-based traffic prioritization is automatically enabled on the E_Ports,
except for long-distance E_Ports. For long-distance E_Ports, you must manually enable QoS
zone-based traffic prioritization.
If you upgrade to Fabric OS 7.2.0 from Fabric OS 7.1.x or earlier, and you did not previously have an
Adaptive Networking license, then all ports that had QoS mode set to AE (automatically enabled)
and were not using the QoS feature are automatically set to OFF after the upgrade. You must
manually configure these ports for QoS.

QoS zones
You assign high, medium, or low priority (QoS level) by configuring a QoS zone. A QoS zone is a
special zone that indicates the priority of the traffic flow between a given host/target pair. By
default, traffic in non-QoS zones is treated as medium priority.
The members of a QoS zone are the host/target pairs. QoS zones can contain WWN members
(WWNN or WWPN) or domain,index (D,I) members. If you use D,I notation in your QoS zones, refer
to “Limitations and restrictions for QoS zone-based traffic prioritization” on page 424 for some
considerations.
A QoS zone has a special name to differentiate it from a regular zone. The name of the zone
determines the priority of the traffic flow. The format of the QoS zone name is as follows:
For high priority:

QOSHid_xxxxx

For medium priority: QOSMid_xxxxx
For low priority:

QOSLid_xxxxx

In the QoS zone name format, id is a flow identifier that designates a specific virtual channel (VC)
for the traffic flow and xxxxx is the user-defined portion of the name. For example, the following are
valid QoS zone names:
QOSH3_HighPriorityTraffic
QOSL1_LowPriorityZone

Fabric OS Administrator’s Guide
53-1002920-02

419

14

QoS zones

The switch automatically sets the priority for the “host,target” pairs specified in the zones
according to the priority level (H, M, or L) in the zone name.
For high and low priority traffic, the flow id allows you to have control over the VC assignment and
control over balancing the flows throughout the fabric. The id range is as follows:

• 1 through 5 for high-priority traffic, which corresponds to VCs 10 through 14.
• 1 through 4 for medium-priority traffic, which corresponds to VCs 2 through 5. Note, however,
that the virtual channels for medium-priority traffic are always allocated by a round-robin
scheme, regardless of the id value.

• 1 through 2 for low-priority traffic, which corresponds to VCs 8 and 9.
The id is optional; if it is not specified, the virtual channels are allocated by means of a round-robin
scheme.

NOTE
If a QoS zone name prefix is specified in an LSAN zone (a zone beginning with the prefix "LSAN_"),
the QoS tag is ignored. Only the first prefix in a zone name is recognized. For example, a zone with
the name "LSAN_QOSH_zone1" is recognized as an LSAN zone and not a QoS zone.
Refer to “QoS over FC routers” on page 421 for additional considerations when using QoS to
prioritize traffic between device pairs in different edge fabrics.
For example, Figure 53 shows a fabric with two hosts (H1, H2) and three targets (S1, S2, S3). The
traffic prioritization is as follows:

• Traffic between H1 and S1 is high priority.
• Traffic between H1 and S3 and between H2 and S3 is low priority.
• All other traffic is medium priority, which is the default.
Domain 1

H1

Domain 3
1

9
14

H2
3

= Low priority
= Medium priority
= High priority

15

13

12

8

7

S2

S3

16

Domain 2

FIGURE 53

S1

Domain 4

QoS traffic prioritization

For this fabric, you could set up the following QoS zones:

420

QOSH_Zone1

Members: H1, S1

QOSL_Zone3

Members: H1, H2, S3

Fabric OS Administrator’s Guide
53-1002920-02

QoS zones

14

QoS on E_Ports
In addition to configuring the hosts and targets in a zone, you must also enable QoS on individual
E_Ports that might carry traffic between the host and target pairs. Path selection between the
“host,target” pairs is governed by FSPF rules and is not affected by QoS priorities. For example, in
Figure 54, QoS should be enabled on the encircled E_Ports.

NOTE
By default, QoS is enabled on 8-Gbps or higher ports, except for long-distance 8-Gbps ports. QoS is
disabled by default on all 4-Gbps ports and long-distance 8-Gbps ports.
Domain 1

H1

Domain 3
1

9
14

H2
3

15
= Low priority
= Medium priority
= High priority
= E_Ports with
QoS enabled

FIGURE 54

S1

13

12

8

7
S3

16

Domain 2

S2

Domain 4

QoS with E_Ports enabled

You must enable QoS on the E_Ports on both ISLs between domain 3 and domain 4, because
either path might be selected to carry the traffic.
You do not need to enable QoS on the E_Ports on the ISLs between domain 1 and domain 2 and
between domain 2 and domain 3, because these are not the shortest paths between the hosts and
the targets. However, if the ISL between domain 1 and domain 3 is broken, then the path through
domain 2 would be used.
To guarantee traffic priority, you should enable QoS on all possible E_Ports. Alternatively, you could
use a TI zone to limit the E_Ports that carry the traffic between a “host,target” pair and enable QoS
on only those E_Ports.
If QoS is not enabled on an E_Port, the traffic prioritization stops at that point. For example, in
Figure 54 if you disabled QoS on E_Ports “3,12” and “3,13,” then the traffic from H1 and H2 to S3
would be low priority from the hosts to domain 3, but would switch to the default (medium) priority
from domain 3 to the target S3.

QoS over FC routers
QoS over FC routers uses QoS traffic prioritization between devices in edge fabrics over an FC
router. Refer to Chapter 26, “Using FC-FC Routing to Connect Fabrics,” for information about FC
routers, phantom switches, and the FC-FC Routing Service.
To establish QoS over FC routers, you must perform the following tasks:

• Define QoS zones in each edge fabric.
Fabric OS Administrator’s Guide
53-1002920-02

421

14

QoS zones

• Define LSAN zones in each edge fabric.
• Enable QoS on the E_Ports in each edge fabric.
• Enable QoS on the EX_Ports in the backbone fabric.
Refer to “Setting QoS zone-based traffic prioritization over FC routers” on page 426 for detailed
instructions.
The following are requirements for establishing QoS over FC routers:

• QoS over FC routers is supported in Brocade native mode only. It is not supported in
interopmode 2 or interopmode 3.

• QoS over FC routers is supported for the following configurations:
- Edge-to-edge fabric configuration: Supported on all platforms.
- Backbone-to-edge fabric configuration: Supported on 16-Gbps-capable platforms only
(Brocade 6505, 6510, 6520, M6505, 6547, and Brocade DCX 8510 Backbone family),
and only if no other platforms are used. For all other platforms, you cannot prioritize the
flow between a device in an edge fabric and a device in the backbone fabric.

• QoS over FC routers is supported only if Virtual Fabrics is disabled in the backbone fabric. QoS
over FC routers cannot be enabled if Virtual Fabrics is also enabled in the backbone fabric.

• The port WWN of the host or target and the port WWN of the proxy device must be in both an
LSAN zone and a QoS zone.

• QoS over FC routers is supported on both EX_Ports and VEX_Ports.
• The EX_Ports (or VEX_Ports) in the path between the QoS devices must be on switches running
Fabric OS v6.3.0 or later.

• QoS zones must use WWN notation only; D,I notation is not supported for QoS over FCRs.

Virtual Fabrics considerations for QoS zone-based traffic prioritization
You can prioritize flows between devices in a logical fabric. The priority is retained for traffic going
across ISLs and through the base fabric XISLs.
For example, Figure 55 shows a logical fabric that includes H1 and S1. To set the traffic between
H1 and S1 to high priority, create a QoS zone in the logical fabric with H1 and S1 as members. Then
enable QoS on all of the E_Ports shown circled in the figure, including all of the E_Ports in the XISLs
(ports 10, 11, 12, 13, 14, 15, 16, and 17).

High-availability considerations for QoS zone-based traffic prioritization
If the standby control processor (CP) is running a Fabric OS version earlier than 6.3.0 and is
synchronized with the active CP, then QoS zones using D,I notation cannot be created. If the
standby CP is not synchronized or if no standby CP exists, then the QoS zone creation succeeds.
If QoS zones using D,I notation exist in either the defined or active configuration and the standby
CP tries to synchronize with the active CP, the synchronization fails if the standby CP is running a
Fabric OS version earlier than 6.3.0. Synchronization can succeed only if the QoS D,I zones are
removed.

422

Fabric OS Administrator’s Guide
53-1002920-02

QoS zones

Domain 1

14

Domain 3

8

9

H1

S1
1

2

5

6

3

4

8

7

LS3, FID1
Domain 7
Chassis 1

LS4, FID3
Domain 8

LS1, FID1
Domain 5

Domain 2

10

12

14

16

Base switch
Domain 10
11

13

LS2, FID3
Domain 6

Chassis 2

Base switch
Domain 9
15

17

= High priority
= E_Ports with QoS enabled

FIGURE 55

Traffic prioritization in a logical fabric

Supported configurations for QoS zone-based traffic prioritization
The following configuration rules apply to QoS zone-based traffic prioritization:

• All switches in the fabric must be running Fabric OS v6.0.0 or later.
ATTENTION
If QoS traffic crosses an ISL for a switch running a firmware version earlier than Fabric OS
v6.0.0, the frames are dropped.

• By default, all devices are assigned medium priority.
- To be assigned high or low priority, hosts and targets must be connected to a Brocade
8-Gbps or 16-Gbps switch or port blade.

-

To preserve the priority level across ISLs, the switches must be running Fabric OS v6.0.0 or
later.

• QoS is enabled by default on 8-Gbps and higher ports. QoS is disabled by default on all 4-Gbps
ports and long-distance ports.

Fabric OS Administrator’s Guide
53-1002920-02

423

14

Setting QoS zone-based traffic prioritization

Limitations and restrictions for QoS zone-based traffic prioritization
• Enabling and disabling QoS is potentially disruptive to the I/O on the affected port.
• If a host and target are included in two or more QoS zones with different priorities, the
following priorities take precedence:

-

High and medium zones = High priority
High and low zones = Low priority
Medium and low zones = Low priority
High, medium, and low zones = Low priority

For example, if an effective zone configuration has QOSH_z1 (H,T) and QOSL_z2 (H,T), the
traffic flow between H and T will be of low QoS priority.

• If QOSH_z1 (H,T) overlaps with a D,I (domain,index) zone at the H port, the traffic flow between
H and T is dropped to medium priority and the H port is marked as a session-based zoning
port.

•
•
•
•

Traffic prioritization is enforced on the egress ports only, not on the ingress ports.
Traffic prioritization is not supported on 10-Gbps ISLs.
Traffic prioritization is not supported on mirrored ports.
Traffic prioritization is not supported over LSAN zones. The traffic is always medium priority in
the ingress edge fabric, the backbone fabric, and the egress edge fabric.

• Traffic prioritization is not supported on a CryptoTarget container (redirection zone). Refer to
the Fabric OS Encryption Administrator’s Guide for information about redirection zones.

• Traffic prioritization is not supported in McDATA Fabric Mode (interopmode 2) or Open Fabric
Mode (interopmode 3).

•
•
•
•

You must be running Fabric OS v6.3.0 or later to create QoS zones that use D,I notation.
QoS zones that use D,I notation are not supported for QoS over FCR.
QoS zones that use D,I notation should not be used for loop or NPIV ports.
If QoS is enabled, an additional 16 buffer credits are allocated per port for 8-Gbps ports in
Extended Mode (LE). Refer to Chapter 25, “Managing Long-Distance Fabrics,” for information
about buffer credit allocation in extended fabrics.

• If some ports in a trunk group have QoS enabled and some ports have QoS disabled, then two
different trunks are formed, one with QoS enabled and one with QoS disabled.

Setting QoS zone-based traffic prioritization
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the zoneCreate command to create zones for high- and low-priority traffic.

• For high-priority traffic, use the following syntax:
zonecreate "QOSHid_zonename", "member[; member...]"

• For low-priority traffic, use the following syntax:
zonecreate "QOSLid_zonename", "member[; member...]"

424

Fabric OS Administrator’s Guide
53-1002920-02

Setting QoS zone-based traffic prioritization

14

The id range is from 1 through 5 for high-priority traffic, which corresponds to VCs 10 through
14. For low-priority traffic, the id range is from 1 through 2, which corresponds to VCs 8 and 9.
The id is optional; if it is not specified, the virtual channels are allocated by means of a
round-robin scheme.
3. Enter the cfgAdd command to add the QoS zone to the zone configuration, by using the
following syntax:
cfgadd "cfgname", "QOSzonename"

4. Enter the cfgSave command to save the change to the defined configuration.
5. Enter the cfgEnable command for the appropriate zone configuration to make the change
effective.
cfgenable "cfgname"

6. Enter the portCfgQos command to enable QoS on the E_Ports, by using the following syntax:
portcfgqos --enable [slot/]port

The portCfgQos command does not affect QoS prioritization. It only enables or disables the link
to pass QoS priority traffic.

NOTE

QoS is enabled by default on all ports (except long-distance ports). If you use the portCfgQos
command to enable QoS on a specific port, the port is toggled to apply this configuration, even
though the port already has QoS enabled. The port is toggled because the user configuration
changed, even though the actual configuration of the port did not change.
If you later use the portCfgQos command to enable QoS on the port again, the port is not toggled,
because the configuration did not change.
Example
sw0:admin> zonecreate "QOSH1_zone", "10:00:00:00:10:00:00:00;
10:00:00:00:20:00:00:00"
sw0:admin> zonecreate "QOSL2_zone", "10:00:00:00:30:00:00:00;
10:00:00:00:40:00:00:00"
sw0:admin> zoneshow
sw0:admin> cfgadd "cfg1", "QOSH1_zone"
sw0:admin> cfgadd "cfg1", "QOSL2_zone"
sw0:admin> cfgshow
Defined configuration:
cfg:
cfg1
zone1; QOSH1_zone; QOSL2_zone
zone: QOSH1_zone
10:00:00:00:10:00:00:00; 10:00:00:00:20:00:00:00
zone: QOSL2_zone
10:00:00:00:30:00:00:00; 10:00:00:00:40:00:00:00
zone: zone1
10:00:00:00:10:00:00:00; 10:00:00:00:20:00:00:00;
10:00:00:00:30:00:00:00; 10:00:00:00:40:00:00:00
Effective configuration:
No Effective configuration: (No Access)
sw0:admin> cfgsave
You are about to save the Defined zoning configuration. This
action will only save the changes on Defined configuration.
Any changes made on the Effective configuration will not

Fabric OS Administrator’s Guide
53-1002920-02

425

14

Setting QoS zone-based traffic prioritization over FC routers

take effect until it is re-enabled. Until the Effective
configuration is re-enabled, merging new switches into the
fabric is not recommended and may cause unpredictable
results with the potential of mismatched Effective Zoning
configurations.
Do you want to save Defined zoning configuration only? (yes, y, no, n): [no] y
Updating flash ...
sw0:admin> cfgenable "cfg1"
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected. If the update includes changes
to one or more traffic isolation zones, the update may result in
localized disruption to traffic on ports associated with
the traffic isolation zone changes
Do you want to enable 'cfg1' configuration (yes, y, no, n): [no] y
zone config "cfg1" is in effect
Updating flash ...
sw0:admin> portcfgqos --enable 3

Setting QoS zone-based traffic prioritization over FC routers
1. Connect to the switch in the edge fabric and log in using an account with admin permissions.
2. Create QoS zones in the edge fabric.
The QoS zones must have WWN members only, and not D,I members. Refer to “Setting QoS
zone-based traffic prioritization” on page 424 for instructions.
3. Create LSAN zones in the edge fabric.
Refer to “Controlling device communication with the LSAN” on page 621 for instructions.
4. Enter the portCfgQos command to enable QoS on the E_Ports.
5. Repeat step 1 through step 3 to create QoS zones and LSAN zones on the other edge fabric.
6. Connect to the FC router in the backbone fabric and log in using an account with admin
permissions.
7.

Enter the portCfgQos command to enable QoS on the EX_Ports.

Disabling QoS zone-based traffic prioritization
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgRemove command to remove the QoS zones from the current zone configuration.
3. Enter the portCfgQos command to disable QoS on the E_Ports.

426

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

15

Bottleneck Detection

In this chapter
• Bottleneck detection overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Supported configurations for bottleneck detection . . . . . . . . . . . . . . . . . .
• Enabling bottleneck detection on a switch . . . . . . . . . . . . . . . . . . . . . . . . .
• Displaying bottleneck detection configuration details . . . . . . . . . . . . . . . .
• Setting bottleneck detection alerts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Changing bottleneck detection parameters . . . . . . . . . . . . . . . . . . . . . . . .
• Advanced bottleneck detection settings . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Excluding a port from bottleneck detection. . . . . . . . . . . . . . . . . . . . . . . . .
• Displaying bottleneck statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Disabling bottleneck detection on a switch . . . . . . . . . . . . . . . . . . . . . . . . .

427
429
431
431
433
435
439
440
442
442

Bottleneck detection overview
A bottleneck is a port in the fabric where frames cannot get through as fast as they should. In other
words, a bottleneck is a port where the offered load is greater than the achieved egress
throughput. Bottlenecks can cause undesirable degradation in throughput on various links. When a
bottleneck occurs at one place, other points in the fabric can experience bottlenecks as the traffic
backs up.
Bottleneck detection is configured on a per-switch basis, with optional per-port exclusions.
Bottleneck detection is disabled by default. The best practice is to enable bottleneck detection on
all switches in the fabric, and leave it on to continuously gather statistics. Bottleneck detection
does not require a license.
The bottleneck detection feature enables you to do the following:

• Prevent degradation of throughput in the fabric.
The bottleneck detection feature alerts you to the existence and locations of devices that are
causing latency. If you receive alerts for one or more F_Ports, use the CLI to check whether
these F_Ports have a history of bottlenecks.

• Reduce the time it takes to troubleshoot network problems.
If you notice one or more applications slowing down, you can determine whether any latency
devices are attached to the fabric and where. You can use the CLI to display a history of
bottleneck conditions on a port. If the CLI shows above-threshold bottleneck severity, you can
narrow the problem down to device latency rather than problems in the fabric.

Fabric OS Administrator’s Guide
53-1002920-02

427

15

Bottleneck detection overview

You can use the bottleneck detection feature with other Adaptive Networking features to optimize
the performance of your fabric. For example, you can do the following:

• If the bottleneck detection feature detects a latency bottleneck, you can use TI zones or QoS
SID/DID traffic prioritization to isolate latency device traffic from high priority application
traffic.

• If the bottleneck detection feature detects ISL congestion, you can use ingress rate limiting to
slow down low priority application traffic, if it is contributing to the congestion.

Types of bottlenecks
The bottleneck detection feature detects two types of bottlenecks:

• Latency bottleneck
• Congestion bottleneck
A latency bottleneck is a port where the offered load exceeds the rate at which the other end of the
link can continuously accept traffic, but does not exceed the physical capacity of the link. This
condition can be caused by a device attached to the fabric that is slow to process received frames
and send back credit returns. A latency bottleneck caused by such a device can spread through the
fabric and can slow down unrelated flows that share links with the slow flow.
By default, bottleneck detection detects latency bottlenecks that are severe enough that they
cause 98 percent loss of throughput. This default value can be modified to a different percentage.
A congestion bottleneck is a port that is unable to transmit frames at the offered rate because the
offered rate is greater than the physical data rate of the line. For example, this condition can be
caused by trying to transfer data at 8 Gbps over a 4 Gbps ISL.
You can use the bottleneckMon command to configure separate alert thresholds for congestion
and latency bottlenecks.
Advanced settings allow you to refine the criterion for defining latency bottleneck conditions to
allow for more (or less) sensitive monitoring at the sub-second level. For example, you would use
the advanced settings to change the default value of 98 percent for loss of throughput. Refer to
“Advanced bottleneck detection settings” on page 439 for specific details.
If a bottleneck is reported, you can investigate and optimize the resource allocation for the fabric.
Using the zone setup and Top Talkers, you can also determine which flows are destined to any
affected F_Ports.

How bottlenecks are reported
Bottleneck detection uses the concept of an affected second when determining whether a
bottleneck exists on a port. Each second is marked as being affected or unaffected by a latency or
congestion bottleneck, based on certain criteria.
The bottleneck detection feature maintains two histories of affected seconds for each port—one
history for latency bottlenecks and another for congestion bottlenecks. A history is maintained for a
maximum of three hours for each port. You can view the history using the bottleneckmon --show
command, as described in “Displaying bottleneck statistics” on page 442.
Bottlenecks are also reported through RASlog alerts and SNMP traps. These two alerting
mechanisms cannot be turned on and off independently.

428

Fabric OS Administrator’s Guide
53-1002920-02

Supported configurations for bottleneck detection

15

You can use the bottleneckMon command to specify the following alerting parameters:

•
•
•
•
•

Whether alerts are to be sent when a bottleneck condition is detected
The size of the time window to look at when determining whether to alert
How many affected seconds are needed to generate the alert
How long to stay quiet after an alert
If an enabled alert is for congestion, for latency, or for both

NOTE
Changing alerting parameters affects RASlog alerting as well as SNMP traps.
For more detailed information on the bottleneckMon command, refer to the Fabric OS Command
Reference.

Supported configurations for bottleneck detection
The following configuration rules apply to bottleneck detection:

• Bottleneck detection is supported only on Fibre Channel ports and FCoE F_Ports.
• Bottleneck detection is supported only on the following port types:
- E_Ports
- EX_Ports
- F_Ports
- FL_Ports
• F_Port and E_Port trunks are supported.
• Long distance E_Ports are supported.
• SIM ports are not supported.
• Bottleneck detection is supported on 4-Gbps, 8-Gbps, and 16-Gbps platforms, including
10-Gbps speeds.

• Bottleneck detection is supported in Access Gateway mode.
• Bottleneck detection is supported whether Virtual Fabrics is enabled or disabled. In VF mode,
bottleneck detection is supported on all fabrics, including the base fabric. Refer to “Virtual
Fabrics considerations for bottleneck detection” on page 430 for additional information on
using bottleneck detection in VF mode.

Limitations of bottleneck detection
The bottleneck detection feature detects latency bottlenecks only at the point of egress, not
ingress. For example, for an E_Port, only the traffic egressing the port is monitored. For FCoE ports
at the FC-DCB (Data Center Bridging) boundary, traffic going from FC to DCB is monitored, but
traffic going from DCB to FC is not monitored.

ATTENTION
Latency bottleneck detection is not recommended for link utilizations above 85 percent.

Fabric OS Administrator’s Guide
53-1002920-02

429

15

Supported configurations for bottleneck detection

High availability considerations for bottleneck detection
The bottleneck detection configuration is maintained across a failover or reboot; however,
bottleneck statistics collected are lost.

Upgrade and downgrade considerations for bottleneck detection
The bottleneck detection configuration is persistent across firmware upgrades and downgrades.
The sub-second latency criterion parameter settings are not preserved on a downgrade to firmware
versions earlier than Fabric OS 7.0.0. If you downgrade and then upgrade back to Fabric OS 7.0.0,
the settings revert to their default values.

Trunking considerations for bottleneck detection
A trunk behaves like a single port. Both latency and congestion bottlenecks are reported on the
master port only, but apply to the entire trunk.
For masterless trunking, if the master port goes offline, the new master acquires all the
configurations and bottleneck history of the old master and continues with bottleneck detection on
the trunk.

Virtual Fabrics considerations for bottleneck detection
Bottleneck detection is supported in both VF and non-VF modes.
In VF mode, if a port on which bottleneck detection is enabled is moved out of a logical switch, any
per-port configurations are retained by the logical switch. The per-port configuration does not
propagate outside of the logical switch. If the port is returned to the logical switch, the previous
per-port configurations are automatically set for the port. Refer to “Changing bottleneck detection
parameters” on page 435 for more information about changing per-port configurations.
In logical fabrics, bottleneck detection is not performed on logical ISLs.
Because a base fabric carries traffic from multiple logical fabrics, bottlenecks reported in the base
fabric can be caused by a mixture of traffic from multiple logical fabrics or by traffic from a single
logical fabric. It is not possible to attribute a base fabric bottleneck to the exact logical fabric
causing it. Dedicated ISLs are exclusive to one logical fabric, and any bottleneck on a dedicated ISL
E_Port pertains entirely to the traffic of that logical fabric.

Access Gateway considerations for bottleneck detection
If bottleneck detection is enabled on a logical switch with some F_Ports connected to an Access
Gateway, you do not get information about which device is causing a bottleneck, because devices
are not directly connected to the Access Gateway. To detect bottlenecks on an Access Gateway,
enable bottleneck detection on the Access Gateway to which the devices are actually connected.

430

Fabric OS Administrator’s Guide
53-1002920-02

Enabling bottleneck detection on a switch

15

Enabling bottleneck detection on a switch
Enabling bottleneck detection permits both latency and congestion detection.
Bottleneck detection is enabled on a switch basis. It is recommended that you enable bottleneck
detection on every switch in the fabric. If you later add additional switches, including logical
switches, to the fabric, be sure to enable bottleneck detection on those switches as well.
When you enable bottleneck detection on a switch, the settings are applied to all eligible ports on
that switch. If ineligible ports later become eligible or, in the case of a logical switch, if ports are
moved to the logical switch, bottleneck detection is automatically applied to those ports.
You can later override these settings on a per-port basis, as described in “Changing bottleneck
detection parameters” on page 435.
Use the following procedure to enable bottleneck detection.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the bottleneckmon --enable command to enable bottleneck detection on all eligible
ports on the switch.
By default, alerts are not sent unless you specify the alert parameter; however, you can view a
history of bottleneck conditions for the port as described in “Displaying bottleneck statistics”
on page 442.
3. Repeat step 1 and step 2 on every switch in the fabric.

NOTE

A best practice is to use the default values for the alerting and sub-second latency criterion
parameters.
The following example enables bottleneck detection on the switch with alerts using default values
for thresholds and time, and is the recommended manner of enabling bottleneck detection
switch:admin> bottleneckmon --enable -alert

The following example enables bottleneck detection on the switch without alerts. In this case, even
though alerts are not delivered, you can still view the bottleneck history using either the CLI or BNA.
switch:admin> bottleneckmon --enable

Displaying bottleneck detection configuration details
Use the following procedure to display the bottleneck detection configuration details:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the bottleneckmon --status command to display the details of bottleneck detection
configuration for the switch, which includes the following details:

•
•
•
•

Fabric OS Administrator’s Guide
53-1002920-02

Whether the feature is enabled
Switch-wide parameters
Per-port overrides, if any
Excluded ports

431

15

Displaying bottleneck detection configuration details

The following initials in the “Per-port overrides for alert parameters,” section of the output indicate
which alerts have been set:

•
•
•
•

C indicates a congestion alert has been set.
L indicates a latency alert has been set.
Y indicates both alerts are set.
N indicates no alerts are set.

The following examples show the status of different bottleneck alerts.
Example of status showing that bottleneck detection is not enabled
switch:admin> bottleneckmon --status
Bottleneck detection - Disabled

Refer to “Enabling bottleneck detection on a switch” on page 431 for instructions on enabling
bottleneck detection.
Example of status output showing that congestion and latency alerts are enabled
Example
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000

Switch-wide alerting parameters:
============================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Yes
0.100
0.800
300 seconds
300 seconds

Example of status output showing that only a congestion alert at the switch level has been set
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
============================
Alerts
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Congestion only
0.800
300 seconds
300 seconds

Per-port overrides for alert parameters:
========================================

432

Fabric OS Administrator’s Guide
53-1002920-02

Setting bottleneck detection alerts

15

Port
Alerts? LatencyThresh
CongestionThresh
Time (s)
QTime(s)
================================================================================
1
Y
0.100
0.800
300
300
2
C
-0.800
600
600
3
L
0.100
-300
300
4
N
-----

NOTE

If there are no per-port overrides, “Per-port overrides for alert parameters” section is not displayed.

Setting bottleneck detection alerts
You can configure Fabric OS to log per-port alerts based on the latency and congestion history of
the port. Alerts are generated based on the number of affected seconds over a specified period of
time. If the number of affected seconds is higher than the threshold, an alert is generated.
This evaluation is done independently for latency and congestion.

NOTE
A congestion bottleneck detection alert is generated whenever a frame timeout occurs irrespective
of the number of affected seconds in the observation window.
The following bottleneckmon -alert parameters determine whether an alert is generated and the
reason for the alert.
The -time parameter specifies the time window. For this example, -time equals 12 seconds.
The -cthresh and -lthresh parameters specify the thresholds on number of affected seconds that
trigger alerts for congestion and latency bottlenecks, respectively.
For example, Figure 56 shows an interval of 12 seconds, in which 6 seconds are affected by a
congestion bottleneck and 3 seconds are affected by a latency bottleneck. This example uses the
default values for these parameters, where -cthresh = 0.8 (80%) and -lthresh = 0.1 (10%).
The following command results in the example illustrated in Figure 56:
bottleneckmon -alert -time 12 -cthresh 0.8 -lthresh 0.1

FIGURE 56

Fabric OS Administrator’s Guide
53-1002920-02

Affected seconds for bottleneck detection

433

15

Setting bottleneck detection alerts

For this time window, 50 percent of the seconds (6 out of 12 seconds) are affected by congestion.
This is below the threshold of 80 percent, so an alert would not be generated for a congestion
bottleneck.
For the same time window, 25 percent of the seconds (3 out of 12 seconds) are affected by
latency. This exceeds the threshold of 10 percent, so an alert would be generated for a latency
bottleneck.

Setting both a congestion alert and a latency alert
Entering the bottleneckmon --enable -alert command enables both alerts using the default alert
values.
This example enables both alerts and shows their values.
switch:admin> bottleneckmon --enable -alert
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Yes
0.100
0.800
300 seconds
300 seconds

Setting a congestion alert only
Entering the bottleneckmon --enable -alert=congestion command enables a congestion alert. This
example enables a congestion alert and shows its values.
switch:admin> bottleneckmon --enable -alert=congestion
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

434

Congestion only
0.800
300 seconds
300 seconds

Fabric OS Administrator’s Guide
53-1002920-02

Changing bottleneck detection parameters

15

Setting a latency alert only
Entering the bottleneckmon --enable -alert=latency command enables a congestion alert. This
example enables a latency alert and shows its values.
switch:admin> bottleneckmon --enable -alert=latency
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Averaging time for alert
Quiet time for alert
-

Latency only
0.100
300 seconds
300 seconds

Changing bottleneck detection parameters
When you enable bottleneck detection, you can configure switch-wide or port-specific alerting
parameters. The alerting parameters indicate whether alerts are sent, and the threshold, time, and
quiet-time options, as well as the sub-second latency criterion for ports.
After you enable bottleneck detection, you can change the alerting parameters for the entire switch
or only for individual ports. For example, you can change the latency threshold for only port 47
without affecting any other port. You can also change the parameters on ports that have been
excluded from bottleneck detection. For a trunk, you can change the parameters only on the
master port.
Alert-related parameters can only be specified with --config when -alert is specified.
This is because -noalert is assumed if -alert is not specified, and -noalert cancels all alert-related
parameters. As long as you want alerts, you must include the exact form of alert (-alert,
-alert=congestion, or -alert=latency) in every --config operation, even if alerts are already enabled.
The retention of settings applies only to the --config command, not to --enable.
An --enable operation behaves as if there is no pre-existing user configuration. If the --enable
command does not include -alert, but does specify alert-related parameters, that command will
fail.

NOTE

Entering the --config command changes only those settings specified in the command; all others are
left alone. The only exceptions are the -alert (restores alerts using recorded values) or -noalert
(disables all alerts) switches. If you want alerts, you must specify what you want as the -alert value
for every bottleneckmon - -config -alert command.
Use the following procedure to configure the bottleneck detection parameters.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the bottleneckmon --config command to set the alerting and sub-second latency
criterion parameters.

Fabric OS Administrator’s Guide
53-1002920-02

435

15

Changing bottleneck detection parameters

Use the -alert parameter to enable congestion and latency alerts. Use the -cthresh parameter to
specify the severity threshold for congestion that triggers an alert. Use the -lthresh parameter to
specify the severity threshold for latency that triggers an alert. Use the -time parameter to specify
the time window in seconds over which the percentage of seconds affected by bottleneck
conditions is computed and compared with the threshold. Use the -qtime parameter to specify the
minimum number of seconds between consecutive alerts. Refer to the Fabric OS Command
Reference for more information.
To remove any port-specific alerting and sub-second latency criterion parameters and revert to the
switch-wide parameters, enter the bottleneckmon --configclear command. To remove and erase all
bottleneck alerts and their criteria, enter bottleneckmon --disable. Refer to “Disabling bottleneck
detection on a switch” on page 442 for more details.

Examples of applying and changing bottleneck detection parameters
The following examples show how to change various bottleneck detection parameters, and how the
changes made are retained when the next set of changes is made. For each example, after the
configuration command is run, the bottleneckmon --status command is run to show the new
settings, which are bolded just for the examples.
Example 1: Setting time window, quiet time, and threshold values for an entire switch

This example sets the time window to 150 seconds, the quiet time to 150 seconds, the congestion
threshold to 0.7 (70%) and the latency threshold to 0.2 (20%) for the entire switch.
switch:admin> bottleneckmon --config -alert -time 150 -qtime 150 -cthresh 0.7
-lthresh 0.2
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Yes
0.200
0.700
150 seconds
150 seconds

Example 2: Changing time window value for an entire switch

This example changes the time window value to 200 seconds for the entire switch.
switch:admin> bottleneckmon --config -alert -time 200
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000

436

Fabric OS Administrator’s Guide
53-1002920-02

Changing bottleneck detection parameters

Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

15

Yes
0.200
0.700
200 seconds
150 seconds

Example 3: Disabling bottleneck detection alerts for a port

This example disables bottleneck detection alerts for port 46 only.
switch:admin> bottleneckmon --config -noalert 46
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Yes
0.200
0.700
200 seconds
150 seconds

Per-port overrides for alert parameters:
========================================
Port
Alerts? LatencyThresh
CongestionThresh
Time (s)
QTime (s)
=================================================================================
46
N
-----

Example 4: Selecting latency-only alerts and changing the latency threshold value for a port

This example changes the alerts to latency-only and the latency threshold value to 75 percent, both
on port 47 only.
switch:admin> bottleneckmon --config -alert=latency -lthresh 0.75 47
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Fabric OS Administrator’s Guide
53-1002920-02

Yes
0.200
0.700
200 seconds
150 seconds

437

15

Changing bottleneck detection parameters

Per-port overrides for alert parameters:
========================================
Port
Alerts? LatencyThresh
CongestionThresh
Time (s)
QTime (s)
=================================================================================
46
N
----47
L
0.750
-200
150

Example 5: Changing the latency time value for a port

This example changes the time value to 250 seconds for port 47 only. Notice that the command
must include -alert=latency to preserve the latency-only alerts configured in the previous example.
In general, -alert must be specified (with =latency or =congestion if desired) on every --config
command when alerts are desired.
switch:admin> bottleneckmon --config -alert=latency -time 250 47
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Yes
0.200
0.700
200 seconds
150 seconds

Per-port overrides for alert parameters:
========================================
Port
Alerts? LatencyThresh
CongestionThresh
Time (s)
QTime (s)
=================================================================================
46
N
----47
L
0.750
-250
150

Example 6: Clearing bottleneck detection override values from ports

This example removes any changed bottleneck detection parameter values from ports 46 and 47.
Notice that the “Per-port overrides for alert parameters” section of the output is not displayed
because there are no per-port overrides.
switch:admin> bottleneckmon --configclear 46-47
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000

438

Fabric OS Administrator’s Guide
53-1002920-02

Advanced bottleneck detection settings

Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

15

Yes
0.200
0.700
200 seconds
150 seconds

Adjusting the frequency of bottleneck alerts
Depending on the circumstances, a problematic switch or port may be triggering alerts more
frequently than desired. The -qtime parameter can be used to throttle alerts by specifying the
minimum number of seconds between consecutive alerts. Thresholds are configured separately for
each type of bottleneck and statistical data are collected independently for each condition.
This parameter applies individually to each type of bottleneck detection, so there can be one
latency alert and one congestion alert in one quiet time.
Example of setting quiet time

This example sets a latency threshold of 0.8 for a time window of 30 seconds, and specifies that an
alert should be sent when 80 percent (0.8) of the one-second samples over any period of 30
seconds is affected by latency bottleneck conditions; the system then waits 60 seconds before
issuing the next alert (assuming that there is one).
switch:admin> bottleneckmon --enable -lthresh 0.8 -time 30 -qtime 60
-alert=latency

Advanced bottleneck detection settings
You can use the sub-second latency criterion parameters to refine the criterion for determining
whether a second is marked as affected by latency bottlenecks. For example, you may want to use
the sub-second latency criterion parameters in the following cases:

• You notice an under-performing application, but do not see any latency bottlenecks detected.
You can temporarily increase the sub-second sensitivity of latency bottleneck detection on the
specific F_Ports for this application.

• You want greater-than-default (sub-second) latency sensitivity on your fabric, so you set
sub-second latency criterion parameters at the time you enable bottleneck detection.

• You want to reduce the number of alerts you are receiving about known latency bottlenecks in
the fabric, so you temporarily decrease the sub-second latency sensitivity on these ports.

• You have a latency bottleneck on an ISL that is not at the edge of the fabric.
The sub-second latency criterion parameters are always applicable. These parameters affect alerts
and, even if alerting is not enabled, they affect the history of bottleneck statistics.
The following sub-second latency criterion parameters are shown with the default values in
parentheses:

• -lsubsectimethresh (0.8) is similar to the -lthresh alerting parameter, except on a sub-second
level. The default value of 0.8 means that at least 80 percent of a second must be affected by
latency for the second to be marked as affected.

Fabric OS Administrator’s Guide
53-1002920-02

439

15

Excluding a port from bottleneck detection

• -lsubsecsevthresh (50) specifies the factor by which throughput must drop in a second for that
second to be considered affected by latency. The default value of 50 means that the observed
throughput in a second must be no more than 1/50th the capacity of the port for that second
to be counted as an affected second. 1/50th of capacity equals 2 percent of capacity, which
translates to 98 percent loss of throughput.
Sub-second latency criterion parameters apply only to latency bottlenecks and not congestion
bottlenecks.
When you enable bottleneck detection, you can specify switch-wide sub-second latency criterion
parameters. After you enable bottleneck detection, you can change the sub-second latency
criterion parameters only on a per-port basis. You cannot change them on the entire switch, as you
can with alerting parameters, unless you disable and then re-enable bottleneck detection.
Changing the sub-second latency criterion parameters on specific ports causes an interruption in
the detection of bottlenecks on those ports, which means the history of bottlenecks is lost on these
ports. Also note the following behaviors if you change the sub-second latency criterion parameters:

• Traffic through these ports is not affected.
• History of latency bottlenecks and congestion bottlenecks is lost on these ports. Other ports
are not affected, however.

• The interruption occurs whether you set or clear per-port overrides on the sub-second latency
criterion parameters.

• Because of the interruption, you can never have an alert for a port such that the alert spans
periods of time with different sub-second latency criteria on that port.

Excluding a port from bottleneck detection
When you exclude a port from bottleneck detection, no data is collected from the port and no alerts
are generated for the port. All statistics history for the port is discarded.
Alerting parameters for the port are preserved, so if you later include the port for bottleneck
detection, the alerting parameters are restored.
Per-port exclusions may be needed if, for example, a long-distance port is known to be a bottleneck
because of credit insufficiency. In general, however, per-port exclusions are not recommended.
For trunking, if you exclude a slave port from bottleneck detection, the exclusion has no effect as
long as the port is a trunk slave. The exclusion takes effect only if the port becomes a trunk master
or leaves the trunk.
Use the following procedure to exclude a port from bottleneck detection.
1. Connect to the switch to which the target port belongs and log in using an account with admin
permissions.
2. Enter the bottleneckmon --exclude command to exclude the port from bottleneck detection.
To later include the port, enter the bottleneckmon --include command.

440

Fabric OS Administrator’s Guide
53-1002920-02

Excluding a port from bottleneck detection

15

Example showing how to exclude a single port from bottleneck detection

The following example excludes port 7 only from bottleneck detection. Refer to “Disabling
bottleneck detection on a switch” on page 442 for more information.

NOTE

Excluding the master port excludes the entire trunk, even if individual slave ports are not excluded.
switch:admin> bottleneckmon --exclude 7
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Yes
0.200
0.700
200 seconds
150 seconds

Per-port overrides for alert parameters:
========================================
Port
Alerts? LatencyThresh
CongestionThresh
Time (s)
QTime (s)
=================================================================================
46
N
----47
L
0.750
-250
150
Excluded ports:
===============
Port
====
7

Example showing how to re-include bottleneck detection for a port

This example restores bottleneck detection for port 7. Notice that the “Excluded ports” section of
the output is not displayed because there are no excluded ports.
switch:admin> bottleneckmon --include 7
switch:admin> bottleneckmon --status
Bottleneck detection - Enabled
==============================
Switch-wide sub-second latency bottleneck criterion:
====================================================
Time threshold
- 0.800
Severity threshold
- 50.000
Switch-wide alerting parameters:
================================
Alerts
Latency threshold for alert
Congestion threshold for alert Averaging time for alert
Quiet time for alert
-

Fabric OS Administrator’s Guide
53-1002920-02

Yes
0.200
0.700
200 seconds
150 seconds

441

15

Displaying bottleneck statistics

Per-port overrides for alert parameters:
========================================
Port
Alerts? LatencyThresh
CongestionThresh
Time (s)
QTime (s)
=================================================================================
46
N
----47
L
0.750
-250
150

Displaying bottleneck statistics
You can use the bottleneckmon --show command to display a history of bottleneck conditions, for
up to three hours. This command has several display options:

• Display only latency bottlenecks, only congestion bottlenecks, or both combined.
• Display bottleneck statistics for a single port, bottleneck statistics for all ports on the switch, or
a list of ports affected by bottleneck conditions.

• Continuously update the displayed data with fresh data.
Use the following procedure to display the bottleneck statistics.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the bottleneckmon --show command.
Example of displaying the bottleneck history in 5-second windows over a period of 30 seconds

In this example, the definition of bottlenecked ports is any port that had a bottleneck occur during
any second in the corresponding interval.
switch:admin> bottleneckmon --show -interval 5 -span 30
==================================================================
Wed Jan 13 18:54:35 UTC 2010
==================================================================
List of bottlenecked ports in most recent interval:
5
==================================================================
Number of
From
To
bottlenecked ports
==================================================================
Jan 13 18:54:05
Jan 13 18:54:10
1
Jan 13 18:54:10
Jan 13 18:54:15
2
Jan 13 18:54:15
Jan 13 18:54:20
1
Jan 13 18:54:20
Jan 13 18:54:25
1
Jan 13 18:54:25
Jan 13 18:54:30
0
Jan 13 18:54:30
Jan 13 18:54:35
0

Disabling bottleneck detection on a switch
When you disable bottleneck detection on a switch, all bottleneck configuration details are
discarded, including the list of excluded ports and non-default values of alerting parameters.
Use the following procedure to disable bottleneck detection.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the bottleneckmon --disable command to disable bottleneck detection on the switch.

442

Fabric OS Administrator’s Guide
53-1002920-02

Disabling bottleneck detection on a switch

15

Example of disabling bottleneck detection on a switch
switch:admin> bottleneckmon --disable
switch:admin> bottleneckmon --status
Bottleneck detection - Disabled

Fabric OS Administrator’s Guide
53-1002920-02

443

15

444

Disabling bottleneck detection on a switch

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

16

In-flight Encryption and Compression

In this chapter
• In-flight encryption and compression overview . . . . . . . . . . . . . . . . . . . . . .
• Configuring in-flight encryption and compression on an EX_Port . . . . . . .
• Configuring in-flight encryption and compression on an E_Port . . . . . . . .
• Viewing the encryption and compression configuration. . . . . . . . . . . . . . .
• Configuring and enabling authentication for in-flight encryption. . . . . . . .
• Enabling in-flight encryption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Enabling in-flight compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Disabling in-flight encryption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Disabling in-flight compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

445
450
451
452
453
455
456
456
457

In-flight encryption and compression overview
In-flight encryption provides security for frames while they are in flight between two switches.
In-flight compression provides better bandwidth use on the ISLs, especially over long distance.
The in-flight encryption and compression features allow frames to be encrypted or compressed at
the egress point of an ISL between two Brocade switches, and then to be decrypted or
decompressed at the ingress point of the ISL. Frames are never left in an encrypted or compressed
state when delivered to an end device.
These features use port-based encryption and compression. You can enable the in-flight encryption
and compression features for both E_Ports and EX_Ports on a per-port basis. By default, these
features are initially disabled for all ports on a switch.

NOTE
No license is required to configure and enable in-flight encryption or compression.
Both ends of the ISL must terminate in 16 Gbps-capable FC ports.
Encryption and compression can be enabled at the same time, or you can enable either encryption
or compression selectively. Figure 57 shows an example of 16 Gbps links connecting three
Brocade switches. One link is configured with encryption and compression, one with just
encryption, and one with just compression.

Fabric OS Administrator’s Guide
53-1002920-02

445

16

In-flight encryption and compression overview

En
cr
yp
tio

on

Compression/Encryption

si

es

pr

FIGURE 57

om

16G

C

n

16G

16G

Encryption and compression on 16 Gbps ISLs

Supported ports for in-flight encryption and compression
The in-flight encryption and compression features are supported only on E_Ports and EX_Ports,
and only on the Brocade 6510 and 6520 switches, 16 Gbps embedded switches, and the Brocade
DCX 8510 Backbone family.
The ports can run at any speed, but must be 16 Gbps-capable.
Encryption and compression are also compatible with the following features:

•
•
•
•

E_Ports or EX_Ports with trunking, QoS, or long distance features enabled.
Flow control modes: R_RDY, VC_RDY, and EXT_VC_RDY.
XISL ports in VF mode.
FCP data frames and non-FCP data frames except ELS and BLS frames.
FCP data frames are of Type = 0x8. For encryption, R_CTL = 0x1 and R_CTL = 0x4 are
supported. For compression, only R_CTL = 0x1 is supported.
Non-FCP data frames are of Type != 0x8.
Non-FCP frames with ELS/BLS (R_CTL == 0x2 || R_CTL == 0x8) are not supported.

In-flight encryption and compression restrictions
• Ports must be 16 Gbps-capable, although port speed can be any configurable value.
• Configuration is dynamic based on port speed. Refer to Table 74 on page 447 for specific
details about the number of ports supported for encryption and compression.

• The devices at either end of the ISL must run Fabric OS 7.0.0 or later software.
• Only E_Ports, EX_Ports, and XISL ports (in VF mode) support encryption or compression.
ICL ports do not support encryption or compression.

• Encryption is not supported in FIPS mode. In-flight encryption is not FIPS-compliant.
• The payload size of a frame is restricted to 2048 bytes.

446

Fabric OS Administrator’s Guide
53-1002920-02

In-flight encryption and compression overview

16

Bandwidth and port limits for in-flight encryption and compression
Fabric OS supports up to 32 Gbps of data encryption and 32 Gbps of data compression per
16 Gbps-capable FC platform. This limits the number of ports that can have these features enabled
at any one time.
The port speed affects the number of supported ports. The slower the speed, the more ports are
supported. In general, at 16 Gbps, the number of supported ports is 2 per ASIC or trunk.
The following table shows some examples of how port speed affects the number of supported ports
for different implementations.

TABLE 74

Number of ports supported for in-flight encryption and compression at various port speeds
Blades (FC16-32, FC16-48)1

Port speed

Encryption only

Compression only

Encryption and compression

16 Gbps

4 ports

4 ports

4 ports

10 Gbps

6 ports

6 ports

6 ports

8/4/2 Gbps

8 ports

8 ports

8 ports

Auto-negotiate (AN)

4 ports

4 ports

4 ports

6510 Fixed-port switches and 16 Gbps embedded switches2
16 Gbps

2 ports

2 ports

2 ports

10 Gbps

3 ports

3 ports

3 ports

8/4/2 Gbps

4 ports

4 ports

4 ports

Auto-negotiate (AN)

2 ports

2 ports

2 ports

6520 Fixed-port switches3
16 Gbps

8 ports

8 ports

8 ports

10 Gbps

12 ports

12 ports

12 ports

8/4/2 Gbps

16 ports

16 ports

16 ports

Auto-negotiate (AN)

8 ports

8 ports

8 ports

1.

The port blades have two ASICs; the per ASIC limit = numbers above/two

2.

The Brocade 6510 switch and 16 Gbps embedded switches have one ASIC; the per ASIC limit = numbers above

3.

The Brocade 6520 has four edge ASICs; the per ASIC limit = numbers above/four

This table does not show all the possible combinations of different speeds for the encryption and
compression ports; other combinations are also supported. The number of supported ports is
automatically calculated based on the speeds chosen.

Port speed on encryption- or compression-enabled ports
The port speed determines the maximum number of ports on a device that can support the in-flight
encryption and compression features.
If the port speed is configured as AUTO NEG, the speed of the port is taken as 16 Gbps for
calculation purposes. It is recommended that you configure the ports to a specific speed before
enabling encryption or compression.

Fabric OS Administrator’s Guide
53-1002920-02

447

16

In-flight encryption and compression overview

The port speed values can be displayed through several commands, including portEncCompShow,
portShow, and switchShow.
You can change the port speed on any port that has encryption or compression enabled with the
portCfgSpeed command. If the capacity is available, the port is configured with the new speed. If
there is not enough capacity available, you cannot change the port speed.
Refer to “Setting port speeds” on page 94 for more information.

How in-flight encryption and compression are enabled
Encryption and compression capabilities and configurations from each end of the ISL are
exchanged during E_Port or EX_Port initialization. Capabilities and configurations must match,
otherwise port segmentation or disablement occurs.
If the port was configured for compression, then the compression feature is enabled.
If the port was configured for encryption, authentication is performed and the keys needed for
encryption are generated. The encryption feature is enabled if authentication is successful. If
authentication fails, then the ports are segmented.

ATTENTION
Any mismatch in configuration at either end of the IFL or authentication failure results in
segmentation or, in rare cases, the port being disabled.
The most common reasons for E_Port or EX_Port segmentation include the following situations:

• Port authentication fails.
• Encryption or compression configurations do not match at both ends.
For example, if at one end there is a switch that does not support encryption or compression,
the port will be disabled.

• An encryption or compression configuration is enabled but resources are not available, or
there are other failures preventing encryption or compression from being enabled.

• The number of available ports has reached the bandwidth limitation.
NOTE

If trunking is enabled, be aware that the ports creating the bandwidth limitation will form a
trunk group, while the rest of the ports will be segmented.
You can also decommission any port that has in-flight encryption and compression enabled. Refer
to “Port decommissioning” on page 92 for details on decommissioning ports.

Authentication and key generation for encryption and compression
The following points apply to authentication and key generation on the supported devices:

• Authentication and key generation only apply to ports that are configured for encryption. They
do not apply to ports that are only configured for compression.

• The in-flight encryption protocol supports the AES-GCM authenticated encryption block cipher
mode. A key, Initial Vector (IV), segment number, and salt are required to encrypt the data
before it is transmitted, and to decode the data after it is received on the other end of the link.

448

Fabric OS Administrator’s Guide
53-1002920-02

In-flight encryption and compression overview

16

• The Diffie-Hellman Challenge Handshake Authentication Protocol (DH-CHAP) must be
configured along with the DH group 4 for port level authentication as a prerequisite for in-flight
encryption. Pre-shared secret keys must be configured on the devices on both ends of the ISL
to perform authentication. Authentication secrets greater than 32 characters are
recommended for stronger encryption keys. Once the link is authenticated, the keys are
generated and exchanged.

• In-flight encryption uses DH-CHAP authentication (SHA-1 algorithm) followed by Internet Key
Exchange (IKE) protocol (HMAC-SHA-512 algorithm) to generate the keys.

• The encryption keys never expire. While the port remains online, the keys generated for the
port remain the same. When a port is disabled, segmented, or taken offline, a new set of keys
is generated when the port is enabled again.

• All members of a trunk group use the same set of keys as the master port. Slave ports do not
exchange keys. If the master port goes offline causing an E_Port or EX_Port change, the trunk
continues to use the same set of keys.

Availability considerations for encryption and compression
To provide redundancy in the event of encryption or compression port failures, you should connect
each ISL or trunk group to different ASICs on the peer switch.
For FC16-32 or FC16-48 blades, if the two ports configured for encryption or compression within
the same ASIC are not configured for trunking, it is recommended to connect each ISL to a different
ASIC on the peer switch. Similarly, configure the two ports on the other ASIC of the blade. If the
ports are configured for trunking, it is recommended to connect each trunk group to different ASICs
on the peer switch.
For Brocade 6510 and 6520 switches, and 16 Gbps embedded switches, if the two ports are not
configured for trunking, it is recommended that you connect each ISL to different ASICs on the peer
switch.

NOTE

If any port on the ASIC with encryption or compression enabled encounters rare error conditions that
require error recovery to be performed on the encryption engine within that ASIC, all encryption or
compression-enabled ports on that ASIC go offline.

Virtual Fabrics considerations for encryption and compression
The E_Ports and EX_Ports in the user-created logical switch, base switch, or default switch, and the
EX_Ports on base switches can support encryption and compression, with some exceptions.
You can configure encryption on XISL ports, but not on LISL ports. However, frames from the LISL
ports are implicitly encrypted or compressed as they pass through encryption- or
compression-enabled XISL ports.
You cannot move a port from one logical switch to another logical switch if in-flight encryption or
compression is enabled on the port. You must disable the encryption and compression
configurations before moving the port, and then enable encryption and compression after the port
has moved.

Fabric OS Administrator’s Guide
53-1002920-02

449

16

Configuring in-flight encryption and compression on an EX_Port

In-flight compression on long-distance ports
When configuring in-flight compression on long-distance ports, it is recommended to configure the
long-distance ports with double the number of buffers.
Configure the port to use the long-distance LS mode and specify the number of buffers to allocate
to the port. You can see what the average compression ratio and the average frame size values are
and adjust the allocated credit accordingly using the portEncCompShow and portBufferShow
commands. You can then use the portBufferCalc command to estimate the assigned credit value to
optimize performance.

Compression ratios for compression-enabled ports
An average compression ratio of 2:1 is provided. The compression ratio value is recalculated every
five seconds, and is the ratio between the accumulated original length and the compressed length
of the data over the previous five seconds.
When a port is configured for compression, entering portStatsShow displays the port’s
compression ratio. The value shown by portStatsShow is a five-second average. Your results
depend on the pattern of the payload data.
The ASIC Compression Block can compress data only if there is at least 3 bytes of data.
The portBufferShow command shows the average frame size for both received (rx) and transmitted
(tx) frames. The rx values are after compression and the tx values are before compression.
Because encryption adds more payload to the port in addition to compression, the compression
ratio calculation is significantly affected on ports configured for both encryption and compression.
This is because the compressed length then also includes the encryption header. This overhead
affects the ratio calculation. To obtain accurate compression ratio data, it is recommended that you
enable ports for compression only.

Configuring in-flight encryption and compression on an EX_Port
When you configure in-flight encryption and compression across an IFL, first configure the EX_Port
and then configure the E_Port. The encryption and compression settings must match at either end
of the IFL.
The following steps summarize how to enable in-flight encryption or compression on an EX_Port.
Perform these steps on the FC router.
1. Determine which ports are available for encryption or compression.
Refer to “Viewing the encryption and compression configuration” on page 452 for instructions.
2. Obtain the WWN of the edge switch using the fcrEdgeShow command.
You need this WWN when you set up the secret key.
switch:admin> fcredgeshow
FID EX-port E-port Neighbor Switch (PWWN, SWWN )
Flags
--------------------------------------------------------------------------20
1
1
20:01:00:05:33:13:70:3e 10:00:00:05:33:13:70:3e

450

Fabric OS Administrator’s Guide
53-1002920-02

Configuring in-flight encryption and compression on an E_Port

16

3. If you are enabling encryption on the port, configure port level authentication for the port.
Omit this step if you want to enable only compression on the port.
Refer to “Configuring and enabling authentication for in-flight encryption” on page 453 for
instructions.
4. Enable encryption on the port.
Refer to “Enabling in-flight encryption” on page 455 for instructions.
5. Enable compression on the port.
Refer to “Enabling in-flight compression” on page 456 for instructions.
6. Obtain the WWN of the front phantom domain using the portCfgExPort command.
You need this WWN when you set up the secret key on the E_Port on the other end of the IFL.
FCR:admin> portcfgexport 1
Port 1 info
Admin:
State:
Pid format:
Operate mode:
Edge Fabric ID:
Front Domain ID:
Front WWN:
Principal Switch:
(output truncated)

enabled
OK
core(N)
Brocade Native
20
160
50:00:53:31:37:43:ee:14
8

Following successful port initialization, the configured features are enabled and active. You can use
the fcrEdgeShow command to check that the EX_Port has come online with encryption or
compression enabled.
Next, configure encryption and compression on the E_Port at the other end of the IFL.

Configuring in-flight encryption and compression on an E_Port
The following steps summarize how to enable encryption or compression on an E_Port.
To configure in-flight encryption and compression across an IFL, first configure encryption and
compression on the EX_Port in the FC router.
Perform the following steps to configure the E_Port in the switch.
1. Determine which ports are available for encryption or compression.
Refer to “Viewing the encryption and compression configuration” on page 452 for instructions.
2. If you are enabling encryption on the port, configure port level authentication for the port.
Omit this step if you want to enable only compression on the port.
Refer to “Configuring and enabling authentication for in-flight encryption” on page 453 for
instructions.
3. Enable encryption on the port.
Refer to “Enabling in-flight encryption” on page 455 for instructions.
4. Enable compression on the port.

Fabric OS Administrator’s Guide
53-1002920-02

451

16

Viewing the encryption and compression configuration

Refer to “Enabling in-flight compression” on page 456 for instructions.
Following successful port initialization, the configured features are enabled and active. You can use
the islShow command to check that the E_Port has come online with encryption or compression
enabled. Alternatively, you can use the portEncCompShow command to see which ports are active.
If port initialization is not successful, you can check for port segmentation errors with the
switchShow command. This command will tell you if the segmentation was due to mismatched
encryption or compression configurations on the ports at either end of the ISL, if port-level
authentication failed, or if a required resource was not available.

Viewing the encryption and compression configuration
Before enabling ports for in-flight encryption or compression, you should determine which ports are
available. Enabling encryption or compression fails if you try to exceed the number of allowable
ports available for encryption or compression on the ASIC.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the portEncCompShow command.
The following example shows the output for two ASICs.
ASIC 1 (below the line of dashes) already has compression configured and active on user ports 348
and 349. Given the limit of two ports per ASIC, ASIC 1 has no more ports available for encryption or
compression.
ASIC 0 (above the dashed line) has no ports configured for either encryption or compression and
therefore has any two ports available for this purpose.
switch:admin> portenccompshow
User
Encryption
Compression
Config
Port
Configured
Active
Configured
Active
Speed
-----------------------------------17
No
No
No
No
4G
18
No
No
No
No
4G
19
No
No
No
No
4G
(output truncated)
149
No
No
No
No
4G
150
No
No
No
No
4G
151
No
No
No
No
4G
---------------------------------------------------------------88
No
No
No
No
4G
89
No
No
No
No
4G
90
No
No
No
No
4G
(output truncated)
348
No
No
Yes
Yes
4G
349
No
No
Yes
Yes
4G
350
No
No
No
No
4G
351
No
No
No
No
4G

The output displays the user port number. For bladed switches, use the switchShow command to
determine the slot number of a specific user port.

452

Fabric OS Administrator’s Guide
53-1002920-02

Configuring and enabling authentication for in-flight encryption

16

Configuring and enabling authentication for in-flight encryption
Authentication and a secret key must be configured and established before configuring in-flight
encryption.
To enable authentication between an FC router and an edge fabric switch, you must first bring all
EX_Ports online without using authentication. After this, the front WWN of any online EX_Port
connected to the same switch can be used to configure the secret keys in the edge fabric switch.
You must obtain the WWN of the peer switch to configure the secret key. If you are configuring an
EX_Port on an FC router, you can use the fcrEdgeShow command to obtain the WWN of the switch
at the other end of the IFL.
1. Log in to the switch using an account with admin permissions, or an account with OM
permissions for the Authentication RBAC class of commands.

ATTENTION
When setting a secret key pair, you are entering the shared secrets in plain text. Use a secure
channel, such as SSH or the serial console, to connect to the switch on which you are setting
the secrets.
2. Configure DH-CHAP for authentication using the authUtil --set command with the -a option.
switch:admin> authutil --set -a dhchap
Authentication is set to dhchap.

You can specify either dhchap or all. The dhchap option explicitly specifies DH-CHAP. Although
all enables both FCAP and DH-CHAP, the active protocol defaults to DH-CHAP for all ports
configured for in-flight encryption.
If DH-CHAP is specified, then all switches in the fabric must enable DH-CHAP and establish
pre-shared secrets. If the protocol is set to all, you must establish pre-shared secrets or
certificates based on the encryption method selected (DH-CHAP or FCAP).
3. Set the DH group to group 4 using the authUtil --set command with the -g option.
switch:admin> authutil --set -g "4"
DH Group was set to 4.

You can specify either "4" or "*". The "4" option explicitly enables DH group 4. Although "*"
enables all DH groups (0 through 4), the DH group defaults to group 4 for all ports configured
for in-flight encryption.
4. Enter the secAuthSecret --set command to establish pre-shared secrets at each end of the
ISL.
It is recommended to use a 32-bit secret for an ISL carrying encrypted or compressed traffic.
switch:admin> secauthsecret --set

When prompted, enter the WWN for the remote switch and secret strings for the local switch
and the remote switch.
5. Activate DH-CHAP authentication using the authUtil --policy command to set the switch policy
mode to Active or On.
switch:admin> authutil --policy -sw active

If you are configuring authentication on an EX_Port, there is no need to set the authentication
policy to Active or On. EX_Ports can operate on any switch authentication policy.

Fabric OS Administrator’s Guide
53-1002920-02

453

16

Configuring and enabling authentication for in-flight encryption

6. Verify the authentication configuration using the authUtil --show command.
The following example sets up authentication in preparation for in-flight encryption. Specifically, it
configures DH-CHAP for authentication, sets the DH group to group 4, sets up a secret key, and
activates authentication.
switch:admin> authutil --set -a dhchap
Authentication is set to dhchap.

switch:admin> authutil --set -g "4"
DH Group was set to 4.

switch:admin> secauthsecret --set
This command is used to set up secret keys for the DH-CHAP authentication.
The minimum length of a secret key is 8 characters and maximum 40
characters. Setting up secret keys does not initiate DH-CHAP
authentication. If switch is configured to do DH-CHAP, it is performed
whenever a port or a switch is enabled.
Warning: Please use a secure channel for setting secrets. Using
an insecure channel is not safe and may compromise secrets.
Following inputs should be specified for each entry.
1. WWN for which secret is being set up.
2. Peer secret: The secret of the peer that authenticates to peer.
3. Local secret: The local secret that authenticates peer.
Press enter to start setting up secrets >
Enter peer WWN, Domain, or switch name (Leave blank when done):
10:00:00:05:1e:e5:cb:00
Enter peer secret:
Re-enter peer secret:
Enter local secret:
Re-enter local secret:
Enter peer WWN, Domain, or switch name (Leave blank when done):
Are you done? (yes, y, no, n): [no] y
Saving data to key store... Done.

switch:admin> secauthsecret --show
WWN
DId
Name
----------------------------------------------10:00:00:05:1e:e5:cb:00
150
dcx_150

switch:admin> authutil --policy -sw active
Warning: Activating the authentication policy requires either DH-CHAP secrets or
PKI certificates depending on the protocol selected. Otherwise, ISLs will be
segmented during next E-port bring-up.
ARE YOU SURE (yes, y, no, n): [no] y
Auth Policy is set to ACTIVE

switch:admin> authutil --show

454

Fabric OS Administrator’s Guide
53-1002920-02

Enabling in-flight encryption

16

AUTH TYPE
HASH TYPE
GROUP TYPE
-------------------------------------dhchap
md5
4
Switch Authentication Policy: ACTIVE
Device Authentication Policy: OFF

For additional information about establishing DH-CHAP secrets, refer to “Secret key pairs for
DH-CHAP” on page 249. For additional information about configuring DH-CHAP authentication,
refer to “Authentication policy for fabric elements” on page 243.

Enabling in-flight encryption
Enable in-flight encryption to provide security for frames while they are in flight between two
switches. Frames are encrypted at the egress point of an ISL and then decrypted at the ingress
point.
Enabling encryption is an offline event. Ports must be disabled first, and then re-enabled after.
Before performing this procedure, it is recommended that you check for port availability. Enabling
encryption fails if you try to exceed the number of allowable ports available for encryption or
compression on the ASIC. Refer to “Viewing the encryption and compression configuration” on
page 452 for details.
You must also authenticate the port as described in “Configuring and enabling authentication for
in-flight encryption” on page 453.
1. Connect to the switch and log in using an account with secure admin permissions, or an
account with OM permissions for the EncryptionConfiguration RBAC class of commands.
2. Enter the portDisable command to disable the port on which you want to configure encryption.
3. Enter the portCfgEncrypt --enable command.
The following example enables encryption on port 15 of an FC16-32 blade in slot 9 of an
enterprise class platform:
switch:admin> portcfgencrypt --enable 9/15

4. Enter the portEnable command to enable the port.
After manually enabling the port, the new configuration becomes active.
The following example enables in-flight encryption on port 0.
switch:admin> portdisable 0
switch:admin> portcfgencrypt --enable 0
switch:admin> portenable 0

You can verify the configuration using the portCfgShow command.
switch:admin> portcfgshow 0
Area Number:
0
Octet Speed Combo:
3(16G,10G)
(output truncated)
D-Port mode:
D-Port over DWDM
Compression:
Encryption:

Fabric OS Administrator’s Guide
53-1002920-02

OFF
..
OFF
ON

455

16

Enabling in-flight compression

Enabling in-flight compression
Enable in-flight compression to provide better bandwidth use on the ISLs, especially over long
distance. Frames are compressed at the egress point of an ISL and then decompressed at the
ingress point.
Enabling compression is an offline event. Ports must be disabled first, and then re-enabled after.
Before performing this procedure, it is recommended that you check for port availability. Enabling
compression fails if you try to exceed the number of allowable ports available for encryption or
compression on the ASIC. Refer to “Viewing the encryption and compression configuration” on
page 452 for details.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the SwitchPortConfiguration RBAC class of commands.
2. Enter the portDisable command to disable the port on which you want to configure
compression.
3. Enter the portCfgCompress --enable command to enable compression.
The following example enables compression on port 15 in slot 9 of an enterprise class
platform:
switch:admin> portcfgcompress --enable 9/15

4. Enter the portEnable command to enable the port.
After enabling the port, the new configuration becomes active.
The following example enables compression on port 0.
switch:admin> portdisable 0
switch:admin> portcfgcompress --enable 0
switch:admin> portenable 0

You can verify the configuration using the portCfgShow command.
switch:admin> portcfgshow 0
Area Number:
0
Octet Speed Combo:
3(16G,10G)
(output truncated)
D-Port mode:
D-Port over DWDM
Compression:
Encryption:

OFF
..
ON
ON

Disabling in-flight encryption
Disabling encryption is an offline event. Ports must be disabled first, and then re-enabled after.
1. Connect to the switch and log in using an account with secure admin permissions, or an
account with OM permissions for the EncryptionConfiguration RBAC class of commands.
2. Disable the port using the portDisable command.

456

Fabric OS Administrator’s Guide
53-1002920-02

Disabling in-flight compression

16

3. Disable encryption on the port using the portCfgEncrypt --disable command.
The following example disables encryption on port 15 in slot 9 of an enterprise class platform:
switch:admin> portcfgencrypt --disable 9/15

4. Enable the port using the portEnable command.
The following example disables encryption on port 0.
myswitch:admin> portdisable 0
myswitch:admin> portcfgencrypt --disable 0
myswitch:admin> portenable 0

You can verify the configuration using the portCfgShow command.
myswitch:admin> portcfgshow 0
Area Number:
0
Speed Level:
AUTO(SW)
(output truncated)
D-Port mode:
D-Port over DWDM
Compression:
Encryption:

OFF
..
OFF
OFF

Disabling in-flight compression
Disabling compression is an offline event. Ports must be disabled first, and then re-enabled after.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the SwitchPortConfiguration RBAC class of commands.
2. Disable the port using the portDisable command.
3. Disable compression on the port using the portCfgCompress --disable command.
The following example disables compression on port 15 in slot 9 of an enterprise class
platform:
switch:admin> portcfgcompress --disable 9/15

4. Enable the port using the portEnable command.
The following example disables compression on port 0.
myswitch:admin> portdisable 0
myswitch:admin> portcfgcompress --disable 0
myswitch:admin> portenable 0

You can verify the configuration using the portCfgShow command.
myswitch:admin> portcfgshow 0
Area Number:
0
Speed Level:
AUTO(SW)
(output truncated)
D-Port mode:
D-Port over DWDM
Compression:
Encryption:

Fabric OS Administrator’s Guide
53-1002920-02

OFF
..
OFF
OFF

457

16

458

Disabling in-flight compression

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

17

Diagnostic Port

In this chapter
• Diagnostic Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Supported platforms for D_Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Licensing requirements for D_Port. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Understanding D_Port. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Supported topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Using D_Port without HBAs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Using D_Port with HBAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Controlling testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Example test scenarios and output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

459
459
460
460
463
465
467
469
469

Diagnostic Port
Diagnostic Port (D_Port) mode allows you to convert a Fibre Channel port into a diagnostic port for
testing link traffic and running electrical loopback and optical loopback tests. The test results can
be very useful in diagnosing a variety of port and link problems.
You can run D_Port tests between the following:

•
•
•
•

A pair of switches
A pair of Access Gateways
An Access Gateway and a switch
A switch and a host bus adapter (HBA)
The HBAs can be Brocade or non-Brocade HBAs. The non-Brocade HBAs must have Brocade
HBA D_Port support.

Supported platforms for D_Port
D_Port functionality is supported only on 16 Gbps-capable platforms, running Fabric OS 7.0.0 or
later. The ports must use 10 Gbps or 16 Gbps Brocade-branded SFP transceivers.
Table 75 lists the Brocade switches and Fabric OS releases that support the D_Port feature.

Fabric OS Administrator’s Guide
53-1002920-02

459

17

Licensing requirements for D_Port

TABLE 75

Supported platforms for D_Port

Product

Fabric OS release and later

Brocade DCX 8510-4 Backbone

v7.0.0

Brocade DCX 8510-8 Backbone

v7.0.0

Brocade 6505 switch

v7.0.1

Brocade 6510 switch

v7.0.0

Brocade 6520 switch

v7.1.0

D_Port functionality is supported on the following HBAs:

• Brocade 16-Gbps HBA (Brocade Fabric Adapter 1860) ports operating in HBA mode with a
16-Gbps SFP+ on Brocade 16-Gbps switches running Fabric OS version 7.1 or later.
Brocade HBA v3.1 provides limited support for D_Port. Brocade HBA v3.2 provides extensive
support for D_Port, including dynamic D_Port mode.

• Non-Brocade 16-Gbps HBA (Must have HBA D_Port support.)
For applicable topologies, refer to “Supported topologies” on page 463.

Licensing requirements for D_Port
The D_Port feature does not require a license if you are running tests between a pair of Brocade
devices, whether the devices are switches, Access Gateways, or HBAs.
If you want to run D_Port tests between a switch and a non-Brocade HBA, the Fabric Vision license
is required. Also, the HBA vendor must have implemented the Brocade HBA D_Port support.

Understanding D_Port
A port in D_Port mode does not carry any user traffic, and is designed to run only specific
diagnostics tests for identifying link-level faults or failures.
To bring up a port in D_Port mode, you must configure both ends of the link between a given pair of
switches (or switches configured as Access Gateways), and you must disable the existing port
before you can configure it as a D_Port. Figure 58 illustrates an example D_Port connection
between a pair of switches through SFP transceivers (port assignments will vary). For all topologies
supported, refer to “Supported topologies” on page 463.

FIGURE 58

460

Example of a basic D_Port connection between switches

Fabric OS Administrator’s Guide
53-1002920-02

Understanding D_Port

17

Once the ports are configured and enabled as D_Ports, the following basic test suite is executed in
the following order, depending on the SFPs installed:
1. Electrical loopback (with 16 Gbps SFP+ only)
2. Optical loopback (with 16 Gbps SFP+ only)
3. Link traffic (with 10 Gbps SFPs, 16 Gbps SFP+, and QSFP+)
4. Link latency and distance measurement (with 10 Gbps SFPs, 16 Gbps SFP+, and QSFP+)

NOTE

Electrical and optical loopback tests are not supported for ICLs.
The following steps are the fundamental parts of D_Port testing:
1. The user configures the desired ports on both ends of the connection.
2. Once both sides are configured, a basic test suite is initiated automatically when the link
comes online, conducting diagnostic tests in the following order: (1) electrical loopback, (2)
optical loopback, (3) link traffic, and (4) link latency and distance measurement.
3. After the automatic test is complete, the user can view results through the CLI or a GUI and
rectify issues (if any) that are reported.
4. The user can also start (and restart) the test manually to verify the link.

Advantages of D_Port
Use the D_Port tests for the following situations:

• Testing a new link before adding it to the fabric
• Testing a trunk member before joining it with the trunk
• Testing long-distance cables and SFP transceivers

D_Port configuration mode and nature of test
D_Port has two modes: static and dynamic. In static mode, you explicitly configure the port to as a
D_Port. In dynamic mode, the port is automatically set to a D_Port based on an external request
from a remote port.
Dynamic D_Port mode is supported only on connections between a switch and an HBA.
By default, the switch has the capability to support dynamic D_Port mode. You can turn this
capability off using the configure command, as shown here.
switch:admin> configure
Not all options will be available on an enabled switch.
To disable the switch, use the "switchDisable" command.
Configure...
Fabric parameters (yes, y, no, n): [no] y
WWN Based persistent PID (yes, y, no, n): [no]
Dynamic D-Port (on, off): [on] off

Fabric OS Administrator’s Guide
53-1002920-02

461

17

Understanding D_Port

Table 76 summarizes D_Port test initiation modes and test start behavior.

TABLE 76

D_Port configuration mode and nature of test

D_Port mode/nature of test

Description

Mode

Static

User configures port explicitly. Port remains as D_Port until user removes
configuration.

Dynamic

No user configuration required. D_Port mode is initiated by external request from
remote port.

Nature of test

Automatic Test automatically starts when the port comes online.
Manual

User starts test from switch side (using the portdporttest - -start command) or from
HBA side (refer to “BCU D_Port commands” on page 468).

When the tests complete, the port behavior depends on the mode:

• For static D_Ports, you must remove the D_Port configuration at either or both ends of the link
to bring up the port as a regular E_Port or F_Port.

• For a switch port in dynamic D_Port mode, the port automatically reverts back to an F_Port or
an E_Port, if the port at the other end reverts to a regular port.

General limitations and considerations for D_Port
• The link to be tested must be marginally functional and able to carry a minimal number of
frames before it can become a D_Port link. The D_Port feature is useful for diagnosing
marginal faults only; it cannot detect the complete failure of any component.

• On QSFP+ ICL ports that are configured as D_Ports, only link traffic can be run. Electrical and
optical loopback tests are not supported.

• Brocade recommends that D_Port tests be limited to a maximum of eight D_Ports at once.
Otherwise, there is a possibility of false alarms.

• When a large number of D_Ports are configured, the test is run on one port per blade at a time,
and other ports wait until the test is completed. No test begins until the fabric is stable.

• Note the following High Availability (HA) considerations:
- There is no HA support for D_Port test options and results. Consequently, such information
is not synchronized with the standby side. Any information from a previous test is lost
following a failover.

-

During an HA failover and reboot on one side of the link, the link goes through
reinitialization and may restart the test. However, the test cannot proceed if the remote
port is not ready to proceed further (the remote port may already be done with the D_Port
test and in final state). In such a case, the test will eventually fail with “Remote port not
ready”. Restarting the test from either side will recover the port.

• D_Ports on an Access Gateway are supported only when there is no mapping between F_Ports
and N_Ports; this includes static mapping and preferred port mapping. In addition, device
(WWN) mapping is also not retained when a D_Port is used. If an Access Gateway port to be
tested is mapped, the port mapping (including static and preferred port mapping) must be
removed before the D_Port can be used. (Refer to “Saving port mappings on an Access
Gateway” on page 464.)

• Access Gateway does not support D_Port dynamic mode. If the port on the connected switch or
adapter is configured as a D_Port, the Access Gateway port is not automatically enabled as a
D_Port. The D_Port must be configured on the Access Gateway port in static mode.

462

Fabric OS Administrator’s Guide
53-1002920-02

Supported topologies

17

Also refer to “Limitations and considerations for D_Port with HBAs” on page 468.

Supported topologies
The following supported topologies illustrate at a high level how D_Port functionality can be used:

•
•
•
•

“Topology 1: ISLs” on page 463
“Topology 2: ICLs” on page 463
“Topology 3: Access Gateways” on page 464
“Topology 4: HBA to switch” on page 465

Topology 1: ISLs
Figure 59 illustrates ISLs that connect multiple switches through a pair of chassis. E represents
E_Ports to be configured as D_Ports.

FIGURE 59

ISLs connecting multiple switches and chassis

For configuration details, refer to “Using D_Port without HBAs” on page 465.

Topology 2: ICLs
Figure 60 illustrates ICLs between slots 5 and 8 in corresponding chassis. E represents E_Ports to
be configured as D_Ports.

FIGURE 60

ICLs connecting chassis blades

For configuration details, refer to “Using D_Port without HBAs” on page 465.

Fabric OS Administrator’s Guide
53-1002920-02

463

17

Supported topologies

Topology 3: Access Gateways
Figure 61 illustrates a switch configured as a single Access Gateway connected to a fabric switch. N
and F represent, respectively, an N_Port and an F_Port to be configured as D_Ports.
The Access Gateway must be either a Brocade 6505 or 6510.

FIGURE 61

Single Access Gateway to switch

Figure 62 illustrates multiple Access Gateways connected to a switch in a cascaded topology. N
and F represent, respectively, an N_Port and an F_Port to be configured as D_Ports.

FIGURE 62

Multiple Access Gateways cascaded to switch

For configuration details, refer to “Using D_Port without HBAs” on page 465.

NOTE
N_Port-to-F_Port and device (WWN) mappings must be removed from an Access Gateway port
before configuring the Access Gateway port as a D_Port. Refer to “Saving port mappings on an
Access Gateway” on page 464.
Figure 63 illustrates connectivity between an HBA and an Access Gateway. F represents an F_Port
to be configured as a D_Port.

FIGURE 63

Access Gateway to HBA

Saving port mappings on an Access Gateway
Before configuring ports as D_Ports on a switch configured as an Access Gateway, you must
remove N_Port-to-F_Port and device (WWN) mappings. Fabric OS commands are available to save
N_Port mappings. Once you save them, you can display the saved N_Port mappings to reconfigure
them after D_Port is disabled. A command is also available to delete saved N_Port mappings.
For more details, refer to Chapter 2, “Configuring Ports in Access Gateway Mode,” in the Access
Gateway Administrator’s Guide.

464

Fabric OS Administrator’s Guide
53-1002920-02

Using D_Port without HBAs

17

Topology 4: HBA to switch
Figure 64 illustrates connectivity between an HBA and a switch. F represents an F_Port to be
configured as a D_Port.
This topology supports dynamic D_Port mode on both the switch and the HBA. In dynamic mode,
the port does not need to be configured explicitly as a D_Port. It comes up in D_Port mode when it
receives a request from the remote port.

FIGURE 64

HBA to switch

For configuration details, refer to “Using D_Port with HBAs” on page 467.

Using D_Port without HBAs
The following sections apply to topologies 1, 2, and 3:

• “Enabling D_Port” on page 465
• “Disabling D_Port” on page 466
The following examples use the command-line interface. Refer also to the latest Brocade Host
Connectivity Manager (HCM) and Brocade Network Advisor documentation to use those graphical
user interface (GUI) applications to configure D_Port.

Enabling D_Port
Use this procedure to configure a basic D_Port diagnostics session between two switches, as
shown in Figure 58 on page 460.

NOTE

The automatic test might fail if you do not follow the sequence of steps exactly.
1. Disable Port 1 on Switch A by using the portDisable [slot/]port command.
switchA:admin> portdisable 1

2. Configure Port 1 on Switch A as a D_Port by using portCfgDport - -enable [slot/]port.
switchA:admin> portcfgdport --enable 1

3. Repeat steps 1 and 2 for the corresponding port (in this example, Port 2) on Switch B.
switchB:admin> portdisable 2
switchB:admin> portcfgdport --enable 2

4. Enable Port 1 on Switch A by using the portEnable [slot/]port command.
switchA:admin> portenable 1

Fabric OS Administrator’s Guide
53-1002920-02

465

17

Using D_Port without HBAs

5. Enable Port 2 on Switch B by using the portEnable [slot/]port command.
switchB:admin> portenable 2

The basic test suite starts as soon as both ports are enabled.
6. While the test is running, enter the portDportTest - -show [slot/]port command to view test
results. The following test is successful.
switch:admin> portdporttest --show 10/39
D-Port Information:
===============================================
Slot:
10
Port:
39
Remote WWNN:
10:00:00:05:33:7e:69:c4
Remote port:
24
Mode:
Manual
No. of test frames:
12 Million
Test frame size:
1024 Bytes
Pattern:
JTSPAT
FEC (enabled/option/active):
Yes/No/No
CR (enabled/option/active): No/No/No
Start time:
Mon Jan 16 05:57:51 2012
End time:
Mon Jan 16 05:58:56 2012
Status:
PASSED
=============================================================================
Test
Start time Result
EST(HH:MM:SS)
Comments
=============================================================================
Electrical loopback 05:57:52
PASSED
-----------------Optical loopback
05:58:06
PASSED
-----------------Link traffic test
05:58:13
PASSED
-----------------=============================================================================
Roundtrip link latency: 934 nano-seconds
Estimated cable distance:
1 meters
Buffers required:
1 (for 1024 byte frames at 16Gbps speed)

7.

To display a summary of the D_Port, use the portDportTest - -show all command.
switch:admin> portdporttest --show all
Port State
SFP Capabilities Test Result
==============================================
24
ONLINE E,O
PASSED
26
ONLINE E,O
PASSED
33
OFFLINE --FAILED

8. Optional: If one of the switches reboots, or if the test does not complete on one of the
switches, restart the test on both switches. Use the portDportTest - -stop command and restart
the test with the portDportTest - -start command on both switches.

Disabling D_Port
Use this procedure to disable the D_Port diagnostics session enabled in “Enabling D_Port.”
1. Disable Port 1 on Switch A by using the portDisable 1 [slot/]port command.
switchA:admin> portdisable 1

2. Disable the D_Port on Port 1 on Switch A by using portCfgDport - -disable 1.
switchA:admin> portcfgdport --disable 1

466

Fabric OS Administrator’s Guide
53-1002920-02

Using D_Port with HBAs

17

3. Repeat steps 1 and 2 for Port 2 on Switch B.
switchB:admin> portdisable 2
switchB:admin> portcfgdport --disable 2

4. Enable Port 1 on Switch A by using the portEnable [slot/]port command.
switchA:admin> portenable 1

5. Enable Port 2 on Switch B by using the portEnable [slot/]port command.
switchB:admin> portenable 2

Using D_Port with HBAs
When HBAs are used, D_Port mode initiates electrical loopback, optical loopback, and link traffic
diagnostic tests on the link between the HBA and the connected switch port. Results can be viewed
from the switch by means of Fabric OS commands and from the adapter by means of the Brocade
Command Line Utility (BCU) and Brocade Host Connectivity Manager (HCM) during or after the test.
Once in D_Port mode, the adapter port does not participate in fabric operations, log in to a remote
device, or run data traffic.
HBAs support testing in dynamic D_Port mode. If a D_Port is enabled on the switch only, it forces
the connected adapter port into D_Port mode. The switch initiates and stops tests on the adapter
port as specified by the switch configuration. Testing is started by means of BCU commands or
HCM options.
In dynamic D_Port mode, you can disable the physical port by using the bcu port --disable
command to exit D_Port mode. When you enable the port again, the switch will again force the
adapter port into D_Port mode if the switch port is still enabled as a D_Port.
The following sections apply to “Topology 4: HBA to switch” on page 465.

• “Automatic mode configuration” on page 467
• “Dynamic mode configuration” on page 468
• “BCU D_Port commands” on page 468

Automatic mode configuration
This procedure enables a D_Port diagnostic session from the connected switch. After the default
test suite is run automatically, you can run specific tests manually to obtain additional detail.
1. Disable the switch port by using the portDisable [slot/]port command.
2. Configure the switch port that you want to enable as a D_Port by using the portCfgDport
- -enable [slot/]port command.
3. Disable the adapter port by using the adapter bcu port - -disable command.
4. Enable the switch port by using the portEnable [slot/]port command.
5. Enable the adapter port as a D_Port by using the adapter bcu diag - -dportenable command
and configure test parameters.
For more details on adapter configuration, refer to the Brocade Fabric Adapters Administrator’s
Guide.

Fabric OS Administrator’s Guide
53-1002920-02

467

17

Using D_Port with HBAs

Dynamic mode configuration
This procedure enables a dynamic D_Port diagnostic session from the connected switch to an HBA.

NOTE
D_Port on HBAs is supported only on 16-Gbps SFP transceivers.
1. Disable the switch port by using the portDisable [slot/]port command.
2. Enable the switch port as a D_Port by using the portCfgDport - -enable [slot/]port command.
3. Enable the switch port by using the portEnable [slot/]port command.
To verify whether the port is a D_Port, use the bcu port - -list command and look for a “D” in the
listing.
For more details on adapter configuration, refer to the Brocade Fabric Adapters Administrator’s
Guide.

BCU D_Port commands
The following BCU commands can be used for D_Port configuration and control:

• bcu diag --dportenable — Enables D_Port on a specific port, sets the test pattern, and sets the
frame count for testing.

• bcu diag --dportdisable — Disables D_Port on a specific port and sets the port back to an
N_Port or NL_Port.

• bcu diag --dportshow — Displays test results for a test in progress on a specific port.
• bcu diag --dportstart — Restarts a test on a specific port when the test has completed.
• bcu port --list — Displays the D_Port enabled or disabled state on the adapter and connected
switch.

NOTE
If you stop the test from the switch side, you should disable D_Port on the HBA side.
Use bcu diag - -dportdisable in static D_Port mode or bcu port - -disable in dynamic D_Port mode.

Limitations and considerations for D_Port with HBAs
Keep in mind the following limitations and considerations for D_Port configurations:

• D_Port is supported only on Brocade 16-Gbps HBA (Brocade Fabric Adapter 1860) ports
operating in HBA mode with a 16-Gbps SFP+ on Brocade 16-Gbps switches running Fabric OS
version 7.1 or later. In addition, the Brocade adapter must be using driver version 3.2.0 or
higher.

• D_Port is supported on non-Brocade 16-Gbps HBAs if you have a Fabric Vision license present
on the switch and if the HBA vendor has implemented the Brocade HBA D_Port support.

• D_Ports do not support a loop topology.
• D_Ports are not supported in a configuration of an HBA to another HBA (in target mode).
• D_Ports on the HBA do not support forward error correction (FEC) and credit recovery (CR). If
these features are enabled on the switch side, the HBA ignores them.

• D_Port is not supported on adapter ports configured in CNA mode.

468

Fabric OS Administrator’s Guide
53-1002920-02

Controlling testing

17

• Toggling the port on either side of the link does not restart the test.
• Because of SFP electrical wrap (EWRAP) bleed-through, during the beginning of switch
electrical loopback testing, the HBA will receive some broken frames, which cause the port
statistic error counter to increase. Examples are “CRC err,” “bad EOF,” and “invalid order set.”
Similar results occur for the optical loopback test. You should ignore these port statistics on
the HBA.

• The following commands from the switch are not supported by the HBA, and the HBA will drop
or reject them:

-

portdporttest --stop
portdporttest --restart
portdporttest --setarg

Although the adapter supports portdporttest - -start, options for this command are ignored.
With the exception of -fec and -cr, the - -start suboptions will work for D_Port on an HBA.

• D_Port is useful for diagnosing marginal faults only. A complete failure of any component
cannot be detected.

• D_Port configuration is not supported on mezzanine cards.
• The maximum number of D_Ports on which the tests can run simultaneously depends on the
HBA firmware version.

TABLE 77

Limitation on number of D_Ports for simultaneous tests

HBA firmware version

Maximum number of D_Ports on which tests can be run simultaneously

HBA v3.2.0

4

HBA v3.2.3

8

• Sometimes a port may get stuck in G_Port state if one end of the link is in D_Port mode and
the other end is not. If this happens, make the D_Port mode compatible on both ends of the
link. Either re-configure both ends of the link as D_Ports, or remove the D_Port configuration
from both ends of the link.

• Powering off and on or plugging in and out slots containing ports in D_Port mode result in
those ports losing the dynamic D_Port state when the slot or port is back up. If this happens
you need to re-configure dynamic D_Port mode on the port.

Controlling testing
You can stop and start D_Port testing on a port by using the following commands:

• portdporttest - -stop [slot/]port
• portdporttest - -start [slot/]port

Example test scenarios and output
In addition to the examples shown in“Enabling D_Port” on page 465, other practical scenarios are
shown in the following sections.

Fabric OS Administrator’s Guide
53-1002920-02

469

17

Example test scenarios and output

Confirming SFP and link status with an HBA
The steps in the following example illustrate how the bcu diag - -dportenable command will fail with
an SFP installed but without a connection to the switch.
1. Confirm the initial port status.
switch:admin> bcu port --list
--------------------------------------------------------------------------Port# FN Type PWWN/MAC
FC Addr/ Media State
Spd
Eth dev
--------------------------------------------------------------------------1/0
fc
10:00:8c:7c:ff:1c:e9:00 160000
sw
Linkup
16G*
0
fc
10:00:8c:7c:ff:1c:e9:00 160000
sw
Linkup
16G*
1/1
fc
10:00:8c:7c:ff:1c:e9:01 -sw
Linkdown
--1
fc
10:00:8c:7c:ff:1c:e9:01 -sw
Linkdown
-----------------------------------------------------------------------------

2. Disable the port.
switch:admin> bcu port --disable 1/0
port disabled

3. Remove the connection to the switch and attempt to enable the D_Port.
switch:admin> bcu diag --dportenable 1/0
ERROR: Timer expired - Retry if persists contact support

4. Install an SFP and attempt to enable the D_Port.
switch:admin> bcu diag --dportenable 1/0
ERROR: Switch port is not D_Port capable or D_Port is disabled

5. Connect to the HBA without the SFP and disable the native port.
switch:admin> bcu port --disable 1/0
port disabled

6. Attempt to enable the D_Port.
switch:admin> bcu diag --dportenable 1/0
ERROR: SFP is not present.
D-port will be enabled but it will be operational only after inserting a valid
SFP.

Starting and stopping D_Port testing
Use the portDportTest command to start or stop D_Port testing or show test results.
You can display the complete results from either the responder or the initiator switch. If the initiator
switch is running Fabric OS v7.1.x or earlier, the responder displays only the local D_Port results,
and you must query the initiator to see the complete results.
The following example shows the D_Port results.
switch:admin> portdporttest --show 26
D-Port Information:
===================
Port:
26
Remote WWNN:
10:00:00:05:33:13:2f:b5

470

Fabric OS Administrator’s Guide
53-1002920-02

Example test scenarios and output

17

Remote port:
42
Mode:
Automatic
Start time:
Wed Feb 2 01:41:43 2011
End time:
Wed Feb 2 01:43:23 2011
Status:
PASSED
================================================================================
Test
Start time
Result
EST(secs) Comments
================================================================================
Electrical loopback
01:42:08
PASSED
----------Optical loopback
01:42:16
PASSED
----------Link traffic test
01:43:15
PASSED
----------================================================================================
Roundtrip link latency:
1108 nano-seconds
Estimated cable distance:
20 meters
Buffers required:
1 (for 1024 byte frames at 16Gbps speed)

The following example shows the portdporttest - -show output where the electrical and optical tests
pass but the link test fails.
switch:admin> portdporttest --show 10/39
D-Port Information:
===================
Slot:
10
Port:
39
Remote WWNN:
10:00:00:05:33:7e:69:c4
Remote port:
24
Mode:
Manual
No. of test frames:
12 Million
Test frame size:
1024 Bytes
Pattern:
JTSPAT
FEC (enabled/option/active):
Yes/No/No
CR (enabled/option/active):
No/No/No
Start time:
Mon Jan 16 05:57:51 2012
End time:
Mon Jan 16 05:58:56 2012
Status:
FAILED
================================================================================
Test
Start time Result
EST(HH:MM:SS)
Comments
================================================================================
Electrical loopback 05:57:52
PASSED
---------------Optical loopback
05:58:06
PASSED
---------------Link traffic test
05:58:13
FAILED
-------See failure report
================================================================================
Roundtrip link latency:
934 nano-seconds
Estimated cable distance:
1 meters
Buffers required:
1 (for 1024 byte frames at 16Gbps speed)
Failure report:
Errors detected (local):
Errors detected (remote):

CRC, Bad_EOF, Enc_out
CRC, Bad_EOF

Please use portstatsshow and porterrshow for more details on the above errors.
Refer to file /var/tmp/dport/slot10port39_stats.txt, for link statistics prior to
the port was set to D-Port

Use the portDportTest - -show all command to display the capabilities and test results of all the
D_Ports in a switch.
switch:admin> portdporttest --show all
Port State
SFP Capabilities Test Result

Fabric OS Administrator’s Guide
53-1002920-02

471

17

Example test scenarios and output

=============================================
24
ONLINE
E,O
PASSED
26
ONLINE
E,O
FAILED
33
ONLINE
E,O
PASSED

Use the switchShow command to see D_Port information.
switch:admin> switchshow
switchName:
switch_10
switchType:
109.1
switchState:
Online
switchMode:
Native
switchRole:
Principal
switchDomain:
1
switchId:
fffc01
switchWwn:
10:00:00:05:33:13:2f:b4
zoning:
OFF
switchBeacon:
OFF
FC Router:
OFF
Allow XISL Use: ON
LS Attributes: [FID: 10, Base Switch: No, Default Switch: No, Address Mode 0]
Index Port Address Media Speed State
Proto
==============================================
24 24
010000
id
N16
Online
FC
26 26
010200
id
N16
Online
FC
mismatch)
33 33
010300
id
N8
Online
FC

D-Port Loopback->Port 24
D-Port segmented,(D-Port mode
D-Port 10:00:00:05:33:13:2f:b5

Use the portCfgShow command to see which ports are D_Port-enabled.
switch:admin> portcfgshow
Ports of Slot 0
24 26 27
----------------------+---+---+--Octet Speed Combo
1
1
1
Speed
AN AN AN
AL_PA Offset 13
.. .. ..
Trunk Port
ON ON ON
Long Distance
.. .. ..
.....
.....
Port Auto Disable
.. .. ..
CSCTL mode
.. .. ..
D-Port mode
ON ON ON
D-Port over DWDM
.. .. ..
Compression
.. .. ..
Encryption
.. .. ..
FEC
ON ON ON
Fault Delay
0
0
0
where AE:QoSAutoEnable, AN:AutoNegotiate, ..:OFF, -:NotApplicable, ??:INVALID

472

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

18

NPIV

In this chapter
• NPIV overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Configuring NPIV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Enabling and disabling NPIV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Viewing NPIV port configuration information. . . . . . . . . . . . . . . . . . . . . . . .

473
475
476
476

NPIV overview
N_Port ID Virtualization (NPIV) enables a single Fibre Channel protocol port to appear as multiple,
distinct ports, providing separate port identification within the fabric for each operating system
image behind the port (as if each operating system image had its own unique physical port). NPIV
assigns a different virtual port ID to each Fibre Channel protocol device. NPIV is designed to enable
you to allocate virtual addresses without affecting your existing hardware implementation. The
virtual port has the same properties as an N_Port, and is therefore capable of registering with all
services of the fabric. This chapter does not discuss the Access Gateway feature. For more
information on the Access Gateway feature, refer to the Access Gateway Administrator’s Guide.
Each NPIV device has a unique device PID, Port WWN, and Node WWN, and behaves the same as
all other physical devices in the fabric. In other words, multiple virtual devices emulated by NPIV
appear no different than regular devices connected to a non-NPIV port.
The same zoning rules apply to NPIV devices as non-NPIV devices. Zones can be defined by
domain,port notation, by WWN zoning, or both. However, to perform zoning to the granularity of the
virtual N_Port IDs, you must use WWN-based zoning.
If you are using domain,port zoning for an NPIV port, and all the virtual PIDs associated with the
port are included in the zone, then a port login (PLOGI) to a non-existent virtual PID is not blocked
by the switch; rather, it is delivered to the device attached to the NPIV port. In cases where the
device is not capable of handling such unexpected PLOGIs, use WWN-based zoning.
The following example shows the number of NPIV devices in the output of the switchShow
command. The number of NPIV devices is equal to the sum of the base port plus the number of
NPIV public devices. The base port is the N_Port listed in the switchShow output. Based on the
formula, index 010000 shows only 1 NPIV device and index 010300 shows a total of 222 NPIV
devices (one N_Port FLOGI device and 221 NPIV devices).
Example of NPIV devices
switch:admin> switchshow
switchName:
5100
switchType:
71.2
switchState:
Online
switchMode:
Access Gateway Mode
switchWwn:
10:00:00:05:1e:41:49:3d
switchBeacon:
OFF

Fabric OS Administrator’s Guide
53-1002920-02

473

18

NPIV overview

Index Port Address Media Speed State Proto
==============================================
0
0
010000 id
N4
Online FC F-Port
1
1
010100 id
N4
Online FC F-Port
2
2
010200 id
N4
Online FC F-Port
3
3
010300 id
N4
Online FC F-Port

20:0c:00:05:1e:05:de:e4 0xa06601
1 N Port + 4 NPIV public
1 N Port + 119 NPIV public
1 N Port + 221 NPIV public

On the Brocade DCX and DCX-4S with the FC8-64 blade, the base port is not included in the NPIV
device count. The following example shows 63 NPIV devices total.
Index Slot Port Address Media Speed State
Proto
==================================================
127
12
15
a07f40 id
N4
Online FC F-Port
(AoQ)

1 N Port + 63 NPIV public

Upgrade considerations
The maximum logins per switch has decreased with Fabric OS v6.4.0. When upgrading from a
release previous to Fabric OS v6.4.0 and later, the configured maximum is carried forward and may
exceed the Fabric OS v6.4.0 limit. It is recommended to reconfigure this parameter to be within the
range permitted in Fabric OS v6.4.0 and later.

Fixed addressing mode
Fixed addressing mode is the default addressing mode used in all platforms that do not have
Virtual Fabrics enabled. When Virtual Fabrics is enabled on the Brocade DCX and DCX-4S, fixed
addressing mode is used only on the default logical switch. The number of NPIV devices supported
on shared area ports (48-port blades) is reduced to 64 from 128 when Virtual Fabrics mode is
enabled.

10-bit addressing mode
The 10-bit addressing mode is the default mode for all the logical switches created in the Brocade
DCX and DCX-4S Backbones. The number of NPIV or loop devices supported on a port is 64.
Table 78 shows the number of NPIV devices supported on the Brocade DCX and DCX-4S
Backbones.

TABLE 78

474

Number of supported NPIV devices

Platform

Virtual Fabrics

Logical switch type

NPIV support

DCX

Disabled

N/A

Yes, 127 virtual device limit.1

DCX

Enabled

Default switch

Yes, 63 virtual device limit.1

DCX

Enabled

Logical switch

Yes, 255 virtual device limit.2, 3

DCX

Enabled

Base switch

No.

DCX-4S

Disabled

N/A

Yes, 255 virtual device limit.

DCX-4S

Enabled

Default switch

Yes, 255 virtual device limit.

Fabric OS Administrator’s Guide
53-1002920-02

Configuring NPIV

TABLE 78

18

Number of supported NPIV devices (Continued)

Platform

Virtual Fabrics

Logical switch type

NPIV support

DCX-4S

Enabled

Logical switch

Yes, 255 virtual device limit.3

DCX-4S

Enabled

Base switch

No.

1. Maximum limit support takes precedence if user-configured maximum limit is greater. This applies to shared
areas on the FC4-48, FC8-48, and FC8-64 port blades.
2. The first 112 physical NPIV-capable devices connected to a logical switch using 10-bit addressing can log in 255
logical devices. The physical NPIV-capable devices after 112, 113, and higher, are limited to 63 logical devices.
3.

Maximum limit of 63 for 10-bit areas connected to third-party (non-Brocade) NPIV HBAs.

Configuring NPIV
The NPIV feature is enabled by default. You can set the number of virtual N_Port_IDs per port to a
value from 1 through 255 per port. The default setting is 126.
The portCfgNpivPort command is used to specify the maximum number of virtual N_port_IDs per
port on a switch. It can also be used to enable or disable NPIV. Once NPIV is enabled on the port,
you can specify the number of logins per port. If the NPIV feature has been disabled, then the NPIV
port configuration does not work.
The addressing mode can limit the maximum number of NPIV logins to 127 or 63 depending on the
mode. The portCfgNPIVPort command can set the maximum number of NPIV logins limit to
anything from 1 through 255, regardless of the addressing mode. Whichever of these two
(addressing mode or the value configured through portCfgNPIVPort) is lower will be the maximum
number that can be logged in.

CAUTION
The portDisable command disables the port and stops all traffic flowing to and from the port.
Perform this command during a scheduled maintenance.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portDisable command.
3. Enter the portCfgNPIVPort --setloginlimit command with the port number and the number of
logins per port.
4. Press Enter.
5. Enter the portEnable command to enable the port.
Example of setting the login limit
switch:adnin> portcfgnpivport --setloginlimit 7/0 128
NPIV Limit Set to 128 for Port 128
switch:adnin> portcfgshow
Area Number:
Octet Speed Combo:
Speed Level:
AL_PA Offset 13:
Trunk Port
Long Distance

Fabric OS Administrator’s Guide
53-1002920-02

7/0
128
1(16G|8G|4G|2G)
AUTO(SW)
OFF
ON
OFF

475

18

Enabling and disabling NPIV

VC Link Init
Locked L_Port
Locked G_Port
Disabled E_Port
Locked E_Port
ISL R_RDY Mode
RSCN Suppressed
Persistent Disable
LOS TOV enable
NPIV capability
QOS E_Port
Port Auto Disable:
Rate Limit
EX Port
Mirror Port
Credit Recovery
F_Port Buffers
Fault Delay:
NPIV PP Limit:
CSCTL mode:
Frame Shooter Port
D-Port mode:
D-Port over DWDM
Compression:
Encryption:
FEC:

OFF
OFF
OFF
OFF
OFF
OFF
OFF
OFF
OFF
ON
AE
OFF
OFF
OFF
OFF
ON
OFF
0(R_A_TOV)
128
OFF
OFF
OFF
..
OFF
OFF
ON

Enabling and disabling NPIV
NPIV is enabled for every port.

NOTE

NPIV is a requirement for FCoE.
1. Connect to the switch and log in using an account assigned to the admin role.
2. To enable or disable NPIV on a port, enter the portCfgNPIVPort command with either the
--enable or --disable option.
The following example shows NPIV being enabled on port 10 of a Brocade 5100:
switch:admin> portCfgNPIVPort --enable 10

NOTE
If the NPIV feature is disabled, the port is toggled if NPIV devices are logged in from that F_Port (a
true NPIV port). Otherwise, the firmware considers that port as an F_Port even though the NPIV
feature was enabled.

Viewing NPIV port configuration information
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgShow command to view the switch ports information.

476

Fabric OS Administrator’s Guide
53-1002920-02

Viewing NPIV port configuration information

18

The following example shows whether a port is configured for NPIV:
switch:admin> portcfgshow
Ports of Slot 0
0 1 2 3
4 5 6 7
8 9 10 11
12 13 14 15
-----------------+--+--+--+--+----+--+--+--+----+--+--+--+----+--+--+-Speed
AN AN AN AN
AN AN AN AN
AN AN AN AN
AN AN AN AN
Trunk Port
ON ON ON ON
ON ON ON ON
ON ON ON ON
ON ON ON ON
Long Distance
.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
VC Link Init
.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
Locked L_Port
.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
Locked G_Port
.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
Disabled E_Port
.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
ISL R_RDY Mode
.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
RSCN Suppressed
.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
Persistent Disable.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
NPIV capability
ON ON ON ON
ON ON ON ON
ON ON ON ON
ON ON ON ON

3. Use the switchShow and portShow commands to view NPIV information for a given port. If a
port is an F_Port, and you enter the switchShow command, then the port WWN of the N_Port is
returned. For an NPIV F_Port, there are multiple N_Ports, each with a different port WWN. The
switchShow command output indicates whether or not a port is an NPIV F_Port, and identifies
the number of virtual N_Ports behind it. The following example is sample output from the
switchShow command:
switch:admin> switchshow
switchName:switch
switchType:66.1
switchState:Online
switchMode:Native
switchRole:Principal
switchDomain:1
switchId:fffc01
switchWwn:10:00:00:05:1e:82:3c:2a
zoning:OFF
switchBeacon:OFF
FC Router:OFF
FC Router BB Fabric ID:128
Area Port Media Speed State
Proto
=====================================
0
0
id
N1
Online
F-Port
1
1
id
N4
No_Light
2
2
id
N4
Online
F-Port
3
3
id
N4
No_Light
4
4
id
N4
No_Light
...


1 Nport + 1 NPIV devices.
20:0e:00:05:1e:0a:16:59

4. Use the portShow command to view the NPIV attributes and all the N_Port (physical and
virtual) port WWNs that are listed under portWwn of device(s) connected. The following
example is sample output for the portShow command:
switch:admin> portshow 2
portName: 02
portHealth: HEALTHY
Authentication: None
portDisableReason: None
portCFlags: 0x1

Fabric OS Administrator’s Guide
53-1002920-02

477

18

Viewing NPIV port configuration information

portFlags: 0x24b03 PRESENT ACTIVE F_PORT G_PORT NPIV LOGICAL_ONLINE LOGIN
NOELP LED ACCEPT
portType: 10.0
portState: 1Online
portPhys: 6In_Sync
portScn:
32F_Port
port generation number:
148
portId:
630200
portIfId:
43020005
portWwn:
20:02:00:05:1e:35:37:40
portWwn of device(s) connected:
c0:50:76:ff:fb:00:16:fc
c0:50:76:ff:fb:00:16:f8
...

...
c0:50:76:ff:fb:00:16:80
50:05:07:64:01:a0:73:b8
Distance: normal
portSpeed: N2Gbps
Interrupts:
Unknown:
Lli:
Proc_rqrd:
Timed_out:
Rx_flushed:
Tx_unavail:
Free_buffer:
Overrun:
Suspended:
Parity_err:
2_parity_err:
CMI_bus_err:

0
0
294803
0
0
0
0
0
0
0
0
0
0

Link_failure: 16
Loss_of_sync: 422
Loss_of_sig: 808
Protocol_err: 0
Invalid_word: 0
Invalid_crc: 0
Delim_err:
0
Address_err: 1458
Lr_in:
15
Lr_out:
17
Ols_in:
16
Ols_out:
15

Frjt:
Fbsy:

0
0

Viewing virtual PID login information
Use the portLoginShow command to display the login information for the virtual PIDs of a port. The
following example is sample output from the portLoginShow command:
switch:admin> portloginshow 2
Type PID
World Wide Name
credit df_sz cos
=====================================================
fe 630240 c0:50:76:ff:fb:00:16:fc
101 2048
c
fe 63023f c0:50:76:ff:fb:00:16:f8
101 2048
c
fe 63023e c0:50:76:ff:fb:00:17:ec
101 2048
c
...

...
ff 630202 c0:50:76:ff:fb:00:17:70
192 2048
c
ff 630201 c0:50:76:ff:fb:00:16:80
192 2048
c

478

scr=3
scr=3
scr=3

d_id=FFFFFC
d_id=FFFFFC

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

19

Fabric-Assigned PWWN

In this chapter
• Fabric-Assigned PWWN overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• User- and auto-assigned FA-PWWN behavior . . . . . . . . . . . . . . . . . . . . . . .
• Configuring an FA-PWWN for an HBA connected to an Access Gateway . .
• Configuring an FA-PWWN for an HBA connected to an edge switch . . . . .
• Supported switches and configurations for FA-PWWN . . . . . . . . . . . . . . . .
• Configuration upload and download considerations for FA-PWWN . . . . . .
• Security considerations for FA-PWWN . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Restrictions of FA-PWWN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Access Gateway N_Port failover with FA-PWWN . . . . . . . . . . . . . . . . . . . . .

479
480
481
482
483
483
483
484
484

Fabric-Assigned PWWN overview
Fabric-Assigned PWWN simplifies server deployment in a Fibre Channel SAN (FC SAN) environment
by using a virtual port World Wide Name (PWWN) instead of a physical PWWN to configure zoning
and LUN mapping and masking.
Server deployment typically requires that multiple administrative teams (for example, server and
storage teams) coordinate with each other to perform configuration tasks such as zone creation in
the fabric and LUN mapping and masking on the storage device. These tasks must be completed
before the server is deployed. Before you can configure WWN zones and LUN masks, you must find
out the physical PWWN of the server. This means that administrative teams cannot start their
configuration tasks until the physical server arrives (and its physical PWWN is known). Because the
configuration tasks are sequential and interdependent across various administrative teams, it may
take several days before the server gets deployed in an FC SAN.
You can use a fabric-assigned PWWN (FA-PWWN) to configure services such as zoning and LUN
masking before you have a physical server. An FA-PWWN is a “virtual” port WWN that can be used
instead of the physical PWWN. When the server is later attached to the SAN, the FA-PWWN is
assigned to the server.
For example, you can use the FA-PWWN feature to perform the following tasks:

• Replace one server with another server, or replace failed HBAs or adapters within a server,
without having to change any zoning or LUN mapping and masking configurations.

• Easily move servers across ports or Access Gateways by way of reassigning the FA-PWWN to
another port.

• Use the FA-PWWN to represent a server in boot LUN zone configurations so that any physical
server that is mapped to this FA-PWWN can boot from that LUN, thus simplifying boot over SAN
configuration.

Fabric OS Administrator’s Guide
53-1002920-02

479

19

User- and auto-assigned FA-PWWN behavior

NOTE

The server must use a Brocade HBA or adapter to use the FA-PWWN feature. Refer to the release
notes for the HBA or adapter versions that support this feature.
You can configure an FA-PWWN for the following topologies:

• An FA-PWWN for an HBA device that is connected to an Access Gateway switch
• An FA-PWWN for an HBA device that is connected directly to an edge switch

FIGURE 65

Fabric-assigned port World Wide Name provisioning scenarios

User- and auto-assigned FA-PWWN behavior
Each switch port and Access Gateway port can have up to two FA-PWWNs, one assigned
automatically and one assigned by the user. FA-PWWNs must be unique, and only one FA-PWWN
can be active at any given time.
The automatically assigned FA-PWWN is created by default if you enable the feature without
explicitly providing a virtual PWWN.
The user-assigned FA-PWWN takes precedence over the automatically assigned FA-PWWN. This
means the switch will bind the user-assigned FA-PWWN to the port if both a user-assigned and an
automatically assigned FA-PWWN are available. If you want to select the automatically assigned
FA-PWWN over the user-assigned FA-PWWN, you must delete the user-assigned FA-PWWN from the
port to which it has been assigned.
The switch ensures that automatically assigned FA-PWWNs are unique in a fabric. However, it is
your responsibility to ensure that user-assigned FA-PWWNs are also unique throughout the fabric.

ATTENTION
You must ensure that the same user-assigned FA-PWWN is not used in multiple chassis. There is no
fabric-wide database, so adding the same FA-PWWNs in multiple chassis causes duplicate PWWNs.

480

Fabric OS Administrator’s Guide
53-1002920-02

Configuring an FA-PWWN for an HBA connected to an Access Gateway

19

Configuring an FA-PWWN for an HBA connected to an Access Gateway
To configure an FA-PWWN, assign the FA-PWWN on the Access Gateway switch. The FA-PWWN
feature is enabled by default on the HBA. Refer to the Brocade Adapters Administrator’s Guide for a
list of supported HBAs.
1. Log in to the edge switch to which the Access Gateway is directly connected.
2. Assign the FA-PWWN.

• If you are manually assigning a WWN, enter the following command:
fapwwn --assign -ag AG_WWN -port AG_port -v Virtual_PWWN

• If you want the WWN to be automatically assigned, enter the following command:
fapwwn --assign -ag AG_WWN -port AG_port

3. Display the FA-PWWN.
fapwwn --show -ag all

You should see output similar to the following sample. The FA-PWWNs are in the Virtual Port
WWN column. (In this example, long lines of output are shown split across two lines, for better
readability.)
----------------------------------------------------------AG Port
Port
Device Port WWN
\
----------------------------------------------------------10:00:00:05:1e:65:8a:d5/16 ---:--:--:--:--:--:--:-- \
10:00:00:05:1e:d7:3d:dc/8
20
20:08:00:05:1e:d7:2b:74 \
10:00:00:05:1e:d7:3d:dc/9
20
20:09:00:05:1e:d7:2b:73 \
10:00:00:05:1e:d7:3d:dc/16 ---:--:--:--:--:--:--:-- \

-----------------------------------------------------------Virtual Port WWN
PID
Enable MapType
-----------------------------------------------------------52:00:10:00:00:0f:50:30
-Yes
AG/Auto
11:22:33:44:55:66:77:88
11403 Yes
AG/User
52:00:10:00:00:0f:50:33
11404 Yes
AG/Auto
52:00:10:00:00:0f:50:38
-Yes
AG/Auto

4. Display the FA-PWWN on the HBA.
The following steps are to be executed on the server and not the switch.
a.

Log in to the server as root.

b.

Enter the following command to display the FA-PWWN.
# bcu port --faa_query port_id
FAA state: Enabled
PWWN: 11:22:33:44:55:66:77:88
PWWN source: Fabric

The HBA retains the FA-PWWN until rebooted. This means you cannot unplug and plug the cable
into a different port on the Access Gateway. You must reboot the HBA before moving the HBA to a
different port.

Fabric OS Administrator’s Guide
53-1002920-02

481

19

Configuring an FA-PWWN for an HBA connected to an edge switch

If you move an HBA to a different port on a switch running Fabric OS v7.0.0 or later, the HBA will
disable its port. The port remains disabled even if you then move the HBA to a port on a switch
running a version of Fabric OS earlier than 7.0.0.

Configuring an FA-PWWN for an HBA connected to an edge switch
To configure an FA-PWWN, assign the FA-PWWN on the edge switch. The FA-PWWN feature is
enabled by default on the HBA. Refer to the Brocade Adapters Administrator’s Guide for a list of
supported HBAs.
1. Log in to the edge switch to which the device is connected.
2. Assign the FA-PWWN.

• If you are manually assigning a WWN, enter the following command:
fapwwn --assign -port [slot/]port -v Virtual_PWWN

• If you want the WWN to be automatically assigned, enter the following command:
fapwwn --assign -port [slot/]port

3. Display the FA-PWWN.
fapwwn --show -port all

You should see output similar to the following sample. The FA-PWWNs are in the VPWWN
column.
----------------------------------------------------------------------Port
PPWWN
VPWWN
PID Enable MapType
----------------------------------------------------------------------0 --:--:--:--:--:--:--:-- 52:00:10:00:00:0f:50:30 10101 Yes Port/Auto
1 --:--:--:--:--:--:--:-- 11:22:33:44:33:22:11:22 -Yes Port/User
52:00:10:00:00:0f:50:44
10 --:--:--:--:--:--:--:-- 52:00:10:00:00:0f:50:45 -Yes Port/Auto

4. Display the FA-PWWN on the HBA.
The following steps are to be executed on the server and not the switch.
a.

Log in to the server as root.

b.

Enter the following command to display the FA-PWWN.
# bcu port --faa_query port_id
FAA state: Enabled
PWWN: 11:22:33:44:33:22:11:22
PWWN source: Fabric

The HBA retains the FA-PWWN until it is rebooted. This means you cannot unplug and plug the
cable into a different port on the switch. You must reboot the HBA before moving the HBA to a
different port.
If you move an HBA to a different port on a switch running Fabric OS v7.0.0 or later, the HBA will
disable its port. The port remains disabled even if you then move the HBA to a port on a switch
running a version of Fabric OS earlier than 7.0.0.

482

Fabric OS Administrator’s Guide
53-1002920-02

Supported switches and configurations for FA-PWWN

19

Supported switches and configurations for FA-PWWN
The FA-PWWN feature is supported only on switches running Fabric OS 7.0.0 or later and only on
Brocade HBAs and adapters. The HBA can be connected to an edge switch or to an Access Gateway
switch.
The FA-PWWN feature is supported on the following platforms:

• Switch platforms running Fabric OS v7.0.0 or later:
- Brocade DCX, DCX-4S, and DCX 8510 family
- Brocade 300
- Brocade 5100
- Brocade 5300
- Brocade 6505
- Brocade 6510
- Brocade 6520
- Brocade 7800
- Brocade VA-40FC
• Access Gateway platforms running Fabric OS v7.0.0 or later:
- Brocade 300
- Brocade 5100
- Brocade 6505
- Brocade 6510
Refer to the release notes for the supported Brocade HBA or adapter versions.

Configuration upload and download considerations for FA-PWWN
The configuration upload and download utilities can be used to import and export the FA-PWWN
configuration.

ATTENTION
Brocade recommends you delete all FA-PWWNs from the switch with the configuration being
replaced before you upload or download a modified configuration. This is to ensure no duplicate
FA-PWWNs in the fabric.

Security considerations for FA-PWWN
If security is a concern, ensure that only authorized users can configure FA-PWWNs. Device
authentication and DCC policies provide additional security between the switch and the server.
The FA-PWWN feature can be enabled only by authorized administrators. Thus, existing user-level
authentication and authorization mechanisms should be used to ensure only authorized users can
configure this feature.

Fabric OS Administrator’s Guide
53-1002920-02

483

19

Restrictions of FA-PWWN

If you are concerned about security for FA-PWWNs, you should configure device authentication. You
can use authentication at the device level to ensure security between the switch and the server.
Refer to “Device authentication policy” on page 246 for information about configuring device
authentication.
You can also use the Device Connection Control (DCC) policy to ensure that only an authorized
physical server can connect to a specific switch port.

NOTE

When creating the DCC policy, use the physical device WWN and not the FA-PWWN.
If you use DCC, a policy check is done on the physical PWWN on the servers. In the case of an HBA,
the FA-PWWN is assigned to the HBA only after the DCC check is successful. Refer to “DCC policy
behavior with Fabric-Assigned PWWNs” on page 241 for additional information.

Restrictions of FA-PWWN
The FA-PWWN feature is not supported with some Fibre Channel fabric features.
FA-PWWN is not supported for the following:

•
•
•
•
•
•

FCoE devices
FL_Ports
Swapped ports (using the portswap command)
Cascaded Access Gateway topologies
FICON/FMS mode
With F_Port trunking on directly attached Brocade HBAs or adapters

NOTE

FA-PWWN is supported with F_Port trunking on the supported Access Gateway platforms.

Access Gateway N_Port failover with FA-PWWN
If an Access Gateway is connected to multiple switches, you should configure the same FA-PWWNs
on both switches to avoid having to reboot the host in case of failover.
If the same FA-PWWNs are not configured on the switches, and if an FA-PWWN F_Port on an Access
Gateway fails over to an N_Port that is connected to a different switch, the FA-PWWN assigned to
the Access Gateway F_Port following the failover will be different than it was before the failover
occurred. This situation may require the host to reboot to bring it back online. Even after the reboot,
the host may potentially go into a different zone because the FA-PWWN is different.

484

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

Managing Administrative Domains

20

In this chapter
• Administrative Domains overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485
• Admin Domain management for physical fabric administrators . . . . . . . . 494
• SAN management with Admin Domains . . . . . . . . . . . . . . . . . . . . . . . . . . . 506

Administrative Domains overview
An Administrative Domain (Admin Domain or AD) is a logical grouping of fabric elements that
defines which switches, ports, and devices you can view and modify. An Admin Domain is a filtered
administrative view of the fabric.

NOTE
If you do not implement Admin Domains, the feature has no impact on users and you can ignore this
chapter.
Admin Domains permit access to a configured set of users. Using Admin Domains, you can
partition the fabric into logical groups and allocate administration of these groups to different user
accounts. These accounts can manage only the Admin Domains assigned to them and cannot
make changes to the rest of the fabric.
For example, you can put all the devices in a particular department in the same Admin Domain for
ease of managing those devices. If you have remote sites, you could put the resources in the
remote site in an Admin Domain and assign the remote site administrator to manage those
resources.
Admin Domains and Virtual Fabrics are mutually exclusive and are not supported at the same time
on a switch.
Do not confuse Admin Domains with zones:

• Zones define which devices and hosts can communicate with each other.
• Admin Domains define which users can manage which devices, hosts, and switches.
You can have up to 256 Admin Domains in a fabric (254 user-defined and 2 system-defined),
numbered from 0 through 255.
Admin Domains are designated by a name and a number. This document refers to specific Admin
Domains using the format “ADn” where n is a number between 0 and 255.

ATTENTION
The Admin Domain administrator can define up to 254 ADs (AD1 through AD254) in the AD
database; however, it is recommended that no more than 16 active Admin Domains run
concurrently. More than 16 active Admin Domains might cause performance degradation and
unpredictable system behavior.

Fabric OS Administrator’s Guide
53-1002920-02

485

20

Administrative Domains overview

NOTE

Do not confuse an Admin Domain number with the domain ID of a switch. They are two different
identifiers. The Admin Domain number identifies the Admin Domain and has a range from 0 through
255. The domain ID identifies a switch in the fabric and has a range from 1 through 239.
Figure 66 shows a fabric with two Admin Domains: AD1 and AD2.
AD1

AD2

FIGURE 66

Fabric with two Admin Domains

Figure 67 shows how users get a filtered view of this fabric, depending on which Admin Domain
they are in. As shown in Figure 67, users can see all switches and E_Ports in the fabric, regardless
of their Admin Domain; however, the switch ports and end devices are filtered based on Admin
Domain membership.

FIGURE 67

486

Filtered fabric views when using Admin Domains

Fabric OS Administrator’s Guide
53-1002920-02

Administrative Domains overview

20

Admin Domain features
Admin Domains allow you to do the following:

• Define the scope of an Admin Domain to encompass ports and devices within a switch or a
fabric.

• Share resources across multiple Admin Domains. For example, you can share array ports and
tape drives between multiple departments. In Figure 66 on page 486, one of the storage
devices is shared between AD1 and AD2.

• Have a separate zone database for each Admin Domain. Refer to “Admin Domains, zones, and
zone databases” on page 510 for more information.

• Move devices from one Admin Domain to another without traffic disruption, cable reconnects,
or discontinuity in zone enforcement.

• Provide strong fault and event isolation between Admin Domains.
• Have visibility of all physical fabric resources. All switches, E_Ports, and FRUs (including blade
information) are visible.

• Continue to run existing third-party management applications. Prior and existing versions of
third-party management applications continue to work with admin IDs and user IDs.

Requirements for Admin Domains
Implementing Admin Domains in a fabric has the following requirements:

• The default zone mode setting must be set to No Access before you create Admin Domains
(refer to “Setting the default zoning mode for Admin Domains” on page 495 for instructions).

• Virtual Fabrics must be disabled before you create Admin Domains (refer to “Disabling
Virtual Fabrics mode” on page 325 for instructions).

• Gigabit Ethernet (GbE) ports cannot be members of an Admin Domain.
• Traffic Isolation Zoning is supported within Admin Domains, with some restrictions, as
described in “Admin Domain considerations for Traffic Isolation Zoning” on page 398.

• If the fabric includes LSAN zones:
- The LSAN zone names must not end with “_ADn”.
- The LSAN zone names must not be longer than 57 characters.
Refer to Chapter 26, “Using FC-FC Routing to Connect Fabrics,” for information about the FC-FC
Routing Service and LSAN zones.

Admin Domain access levels
Admin Domains offer a hierarchy of administrative access. To manage Admin Domains, you must
be a physical fabric administrator. A physical fabric administrator is a user with admin permissions
and access to all Admin Domains (AD0 through AD255). Only a physical fabric administrator can
perform Admin Domain configuration and management.
Other administrative access is determined by your defined Role-Based Access Control (RBAC) role
and AD membership. Your role determines your access level and permission to perform an
operation. Your AD membership determines the fabric resources on which you can operate.
Table 79 lists each Admin Domain user type and describes its administrative access and
capabilities.

Fabric OS Administrator’s Guide
53-1002920-02

487

20

Administrative Domains overview

TABLE 79

AD user types

User type

Description

Physical fabric
administrator

User account with admin permissions and with access to all Admin Domains (AD0 through
AD255).
Creates and manages all Admin Domains.
Assigns other administrators or users to each Admin Domain.
The default admin account is the first physical fabric administrator.
Only a physical fabric administrator can create other physical fabric administrators.

Administrative
Domain users

Can be assigned to one or more Admin Domains.
Manage the resources within their Admin Domains.
If their role permits, can create user accounts and assign them to Admin Domains in their list.
Cannot view other Admin Domain definitions. They can view only members of their own Admin
Domains.

User-defined Admin Domains
AD1 through AD254 are user-defined Admin Domains. These user-defined Admin Domains can be
created only by a physical fabric administrator (refer to “Admin Domain access levels” on page 487
for more information).
In Figure 66 on page 486, AD1 and AD2 are user-defined Admin Domains.

System-defined Admin Domains
AD0 and AD255 are system-defined Admin Domains. AD0 and AD255 always exist and cannot be
deleted or renamed. They are reserved for use in creation and management of Admin Domains.

AD0
AD0 is a system-defined Admin Domain. Unlike user-defined Admin Domains, AD0 has an implicit
and an explicit membership list. User-defined Admin Domains have only an explicit membership
list.

• The implicit membership list contains all devices, switch ports, and switches that have not
been assigned to any other Admin Domain.
Initially, the AD0 implicit membership list contains all devices, switch ports, and switches in the
fabric. When you create AD1 through AD254, the devices, switch ports, and switches used to
create these user-defined Admin Domains disappear from the AD0 implicit membership list.

• The explicit membership list contains all devices, switch ports, and switches that you explicitly
add to AD0 and can be used to force device and switch sharing between AD0 and other Admin
Domains.
AD0 is managed like any user-defined Admin Domain. The only difference between AD0 and
user-defined Admin Domains is the implicit membership list.
The implicit members of AD0 change dynamically as the membership of other Admin Domains
changes. The explicit members of AD0 are not deleted unless you explicitly remove them.
For example, if DeviceA is not a member of any user-defined Admin Domain, then it is an implicit
member of AD0.

488

Fabric OS Administrator’s Guide
53-1002920-02

Administrative Domains overview

20

If you explicitly add DeviceA to AD0, then DeviceA is both an implicit and an explicit member of AD0.
AD0 implicit members
DeviceA

AD0 explicit members
DeviceA

AD2 members
none

If you add DeviceA to AD2, then DeviceA is deleted from the AD0 implicit membership list, but is not
deleted from the AD0 explicit membership list.
AD0 implicit members
none

AD0 explicit members
DeviceA

AD2 members
DeviceA

If you then remove DeviceA from AD2, DeviceA is added back to the AD0 implicit membership list
(assuming DeviceA is not in any other Admin Domain).
AD0 implicit members
DeviceA

AD0 explicit members
DeviceA

AD2 members
none

When a new device is added to the fabric, it automatically becomes an implicit member of AD0
until it is explicitly added to an Admin Domain.
AD0 is useful when you create Admin Domains because you can see which devices, switch ports,
and switches are not yet assigned to any Admin Domains.
AD0 owns the root zone database (legacy zone database).

AD255
AD255 is a system-defined Admin Domain that is used for Admin Domain management. AD255
always contains all of the devices in the entire physical fabric. You can use AD255 to get an
unfiltered view of the fabric and to view the hierarchical zone databases of AD0 through AD254. All
Admin Domain management is done in the AD255 context.
AD255 does not have a zone database associated with it; you cannot use AD255 to perform any
zoning management tasks (non-read operations such as creating or modifying zones).
Figure 68 on page 490 shows the same fabric from Figure 66 on page 486, but with AD0 and
AD255 shown. AD0 contains the two devices that are not in any of the user-defined Admin
Domains (AD1 and AD2). AD255 always encompasses the entire physical fabric.

Fabric OS Administrator’s Guide
53-1002920-02

489

20

Administrative Domains overview

FIGURE 68

Fabric with AD0 and AD255

Home Admin Domains and login
You are always logged in to an Admin Domain, and you can view and modify only the devices in that
Admin Domain.
If you have access to more than one Admin Domain, one of them is designated as your home
Admin Domain, the one you are automatically logged in to. If your home Admin Domain is deleted
or deactivated, then by default you are logged in to the lowest-numbered active Admin Domain in
your Admin Domain list. The home Admin Domain, like the Admin Domain list, is a configurable
property of a non-default user account. Here is some additional information about AD accounts:

• You can log in to only one Admin Domain at a time. You can later switch to a different Admin
Domain (refer to “Switching to a different Admin Domain context” on page 508 for
instructions).

• For default accounts such as admin and user, the home Admin Domain defaults to AD0 and
cannot be changed.

• The Admin Domain list for the default admin account is 0 through 255, which gives this
account automatic access to any Admin Domain as soon as the domain is created, and makes
this account a physical fabric administrator.

• The Admin Domain list for the default user account is AD0 only.

490

Fabric OS Administrator’s Guide
53-1002920-02

Administrative Domains overview

20

• For user-defined accounts, the home Admin Domain defaults to AD0 but an administrator can
set the home Admin Domain to any Admin Domain to which the account is given access.

• If you are in any Admin Domain context other than AD0, the Admin Domain number is included
in the system prompt displayed during your session. The following are example prompts for
when you are in the AD0, AD1, and AD255 contexts, respectively:
switch:admin>
switch:AD1:admin>
switch:AD255:admin>

Admin Domain member types
You define an Admin Domain by identifying members of that domain. Admin Domain members can
be devices, switch ports, or switches. Defining these member types is similar to defining a
traditional zone member type. An Admin Domain does not require or have a new domain ID or
management IP address linked to it.

Device members
Device members are defined by the device World Wide Name (WWN) and have the following
properties:

• A device member can be either a device port WWN or a device node WWN.
• A device member grants view access to the device and zoning rights. View rights are also
granted to the switch port to which the device is attached.

• A device member provides a pure virtual view. The cabling and switch port diagnostics and
control are done by the physical fabric administrator.
Port control is provided only through switch port membership and is not provided for device
members. When you create an Admin Domain, the end device members do not need to be online,
even though their WWNs are used in the Admin Domain definition.
You can share device members across multiple Admin Domains. You can also zone shared devices
differently in each Admin Domain. A device WWN member does not automatically grant usage of
corresponding domain,index members in the zone configuration. If you specify a device WWN
member in the Admin Domain member list, zone enforcement ignores zones with the
corresponding port (the port to which the device is connected) member usage.

Switch port members
Switch port members are defined by switch domain,index and have the following properties:

• A switch port member grants port control rights and zoning rights for that switch port.
• A switch port member grants view access and zoning rights to the device connected to that
switch port.

• A switch port member allows you to share domain,index members across multiple Admin
Domains. In each Admin Domain, you can also zone shared devices differently.

• A switch port member implicitly includes all devices connected to the specified domain,index
members in the Admin Domain membership.

• A switch port member allows you to specify a range of indices as Admin Domain members, for
example: 100,5-8. The index range arguments are expanded and stored in the Admin Domain
member list.

Fabric OS Administrator’s Guide
53-1002920-02

491

20

Administrative Domains overview

If a device is a member of an Admin Domain, the switch port to which the device is connected
becomes an indirect member of that Admin Domain and the domain,index is removed from the
AD0 implicit membership list.

NOTE

If the switch domain ID changes, the domain,index members are invalid (they are not automatically
changed). You must then reconfigure the Admin Domain with the current domain,index members.

Switch members
Switch members are defined by the switch WWN or domain ID, and have the following properties:

• A switch member grants administrative control to the switch.
• A switch member grants port control for all ports in that switch.
• A switch member allows switch administrative operations such as disabling and enabling a
switch, rebooting, and firmware downloads.

• A switch member does not provide zoning rights for the switch ports or devices.
To allow devices to be zoned within Admin Domains, you must specify the port members using
domain,index or device WWN members.
E_Ports (including VE_Ports, EX_Ports, and VEX_Ports) are implicitly shared across all Admin
Domains. An administrator can perform port control operations only if the domain,index of the
E_Port is part of the Admin Domain.

NOTE

Only the WWN of the switch is saved in the Admin Domain. If you change the domain ID of the switch,
the Admin Domain ownership of the switch is not changed.

Admin Domains and switch WWNs
Admin Domains are treated as fabrics. Because switches cannot belong to more than one fabric,
switch WWNs are converted so that they appear as unique entities in different Admin Domains
(fabrics). This WWN conversion is done only in the AD1 through AD254 context. AD0 and AD255
use unconverted switch WWNs.
The switch WWN has the following format:
10:00:nn:nn:nn:nn:nn:nn
In an Admin Domain context, the switch WWN is converted from NAA=1 to NAA=5 format, with the
Admin Domain number added, using the following syntax:
5n:nn:nn:nn:nn:nn:n9:xx
In the syntax, xx is the Admin Domain number.
For example, the following switch WWN is in NAA=1 format:
10:00:00:60:69:e4:24:e0
The following switch WWN is the converted WWN for the previous example in AD1:
50:06:06:9e:42:4e:09:01

492

Fabric OS Administrator’s Guide
53-1002920-02

Administrative Domains overview

20

Figure 69 on page 493 shows an unfiltered view of a fabric with two switches, three devices, and
two Admin Domains. The devices are labeled with device WWNs and the switches are labeled with
domain IDs and switch WWNs.

FIGURE 69

Fabric showing switch and device WWNs

Figure 70 shows the filtered view of the fabric as seen from AD3 and AD4. The switch WWNs are
converted to the NAA=5 syntax; the device WWNs and domain IDs remain the same.

Fabric Visible to AD3 User
WWN = 10:00:00:00:c2:37:2b:a3

WWN = 10:00:00:00:c7:2b:fd:a3

Domain ID = 1
WWN = 50:00:51:f0:52:36:f9:03

Domain ID = 2
WWN = 50:00:52:e0:63:46:e9:03

WWN = 10:00:00:00:c2:37:2b:a3

Fabric Visible to AD4 User

Domain ID = 1
WWN = 50:00:51:f0:52:36:f9:04

Domain ID = 2
WWN = 50:00:52:e0:63:46:e9:04

WWN = 10:00:00:00:c8:3a:fe:a2

FIGURE 70

Fabric OS Administrator’s Guide
53-1002920-02

Filtered fabric views showing converted switch WWNs

493

20

Admin Domain management for physical fabric administrators

Admin Domain compatibility, availability, and merging
Admin Domains maintain continuity of service for Fabric OS features and operate in mixed-release
Fabric OS environments. High availability is supported with some backward compatibility.
When an E_Port comes online, the adjacent switches merge their AD databases. The receiving
switch accepts an AD database from the neighboring switch only if the local AD database is empty
or if the new AD database exactly matches both the defined and effective configurations of the
local AD database. If the AD database merge fails, the E_Port is segmented with an “AD conflict”
error code.

Admin Domain management for physical fabric administrators
NOTE
This section is for physical fabric administrators who are managing Admin Domains.
The ad command follows a batched-transaction model, which means that changes to the Admin
Domain configuration occur in the transaction buffer.
An Admin Domain configuration can exist in several places:

• Effective configuration — The Admin Domain configuration that is currently in effect.
• Defined configuration — The Admin Domain configuration that is saved in flash memory. There
might be differences between the effective configuration and the defined configuration.

• Transaction buffer — The Admin Domain configuration that is in the current transaction buffer
and has not yet been saved or canceled.
How you end the transaction determines the disposition of the Admin Domain configuration in the
transaction buffer. The following commands end the Admin Domain transaction:
ad --save

Saves the changes in the transaction buffer to the defined configuration in
persistent storage and propagates the defined configuration to all switches
in the fabric. Note that for delete and clear operations, if one or more of the
deleted Admin Domains are in the effective configuration, you cannot use
--save, but must use --apply instead.

ad --apply

Saves the changes to the defined configuration in persistent storage and
enforces the defined configuration on all switches in the fabric, replacing the
effective configuration.

ad --transabort Aborts the transaction and clears the transaction buffer. The effective and
defined configurations remain unchanged.
You can enter the ad --transshow command at any time to display the ID of the current Admin
Domain transaction.

494

Fabric OS Administrator’s Guide
53-1002920-02

Admin Domain management for physical fabric administrators

20

Setting the default zoning mode for Admin Domains
To begin implementing an Admin Domain structure within your SAN, you must first set the default
zoning mode to No Access. You must be in AD0 to change the default zoning mode.
1. Log in to the switch with the appropriate RBAC role.
2. Ensure you are in the AD0 context by entering the ad --show command to determine the
current Admin Domain.
If necessary, switch to the AD0 context by entering the ad --select 0 command.
3. Set the default zoning mode to No Access, as described in “Setting the default zoning mode”
on page 361.

Creating an Admin Domain
To create an Admin Domain, you must specify an Admin Domain name, number, or both:

• If you create an Admin Domain using only a number, the Admin Domain name is automatically
assigned to be “ADn”, where n is the number you specified.
For example, if you specify AD number = 4, then AD name is set to “AD4”.

• If you create an Admin Domain using only a name, the Admin Domain number is automatically
assigned and is the lowest available AD number, except if you specify a name in the format
“ADn”, in which case the Admin Domain number is assigned to be n.
For example, if you specify AD name = “blueAD” and the lowest available AD number is 5, then
AD name is “blueAD” and AD number is 5.
If you specify AD name = “AD15” and the lowest available AD number is 6, then AD name is
“AD15” and AD number is 15. Because the specified name is in the format “ADn”, the AD
number is assigned to be n and not the lowest available AD number.
When you create an Admin Domain, you must specify at least one member (switch, switch port, or
device). You cannot create an empty Admin Domain. For more information about these member
types, refer to “Admin Domain member types” on page 491.
A newly created Admin Domain has no zoning defined and the default access mode is No Access.
This means the devices in the Admin Domain cannot communicate with each other. You must set
up zones in the newly created Admin Domain to allow devices to access each other, even if the
devices were already zoned together prior to your moving them to the Admin Domain. Refer to
“Admin Domains, zones, and zone databases” on page 510 for additional information about how
zones work with Admin Domains.
You create Admin Domains in the transaction buffer. You can either save the newly created Admin
Domain to a defined configuration or make it the effective configuration directly.
The following procedure describes the steps for creating Admin Domains.
1. Log in to the switch as the physical fabric administrator.
2. Disable Virtual Fabrics, if necessary, as described in “Disabling Virtual Fabrics mode” on
page 325. Admin Domains and Virtual Fabrics cannot co-exist.
3. Set the default zone mode to No Access, if you have not already done so. Refer to “Setting the
default zoning mode” on page 361 for instructions.

Fabric OS Administrator’s Guide
53-1002920-02

495

20

Admin Domain management for physical fabric administrators

4. Switch to the AD255 context, if you are not already in that context:
ad --select 255

5. Enter the ad --create command using the -d option to specify device and switch port members
and the -s option to specify switch members:
ad --create ad_id -d "dev_list" -s "switch_list"

6. Enter the appropriate command based on whether you want to save or activate the Admin
Domain definition:

• To save the Admin Domain definition, enter ad --save.
• To save the Admin Domain definition and directly apply the definition to the fabric, enter ad
--apply.

7.

Set up zones in the newly created Admin Domain. Refer to Chapter 12, “Administering
Advanced Zoning,” for instructions.

Example of creating Admin Domains

The following example creates Admin Domain AD1, consisting of two switches, which are
designated by domain ID and switch WWN.
switch:AD255:admin> ad --create AD1 -s "97; 10:00:00:60:69:80:59:13"

The following example creates Admin Domain “blue_ad,” consisting of two switch ports
(designated by domain,index), one device (designated by device WWN), and two switches
(designated by domain ID and switch WWN).
switch:AD255:admin> ad --create blue_ad –d "100,5; 1,3;
21:00:00:e0:8b:05:4d:05" –s "97; 10:00:00:60:69:80:59:13"

User assignments to Admin Domains
After you create an Admin Domain, you can specify one or more user accounts as the valid
accounts that can use that Admin Domain. User accounts have the following characteristics with
regard to Admin Domains:

• A user account can have only a single role.
• You can configure a user account to have access to the physical fabric through AD255 and to a
list of Admin Domains (AD0 through AD254).

• You can configure a user account to have access to only a subset of your own Admin Domain
list. Only a physical fabric administrator can create another physical fabric administrator user
account.

• Users capable of using multiple Admin Domains can designate one of these Admin Domains as
the home Admin Domain, which is the default Admin Domain context after login.

• If you do not specify one, the home Admin Domain is the lowest valid Admin Domain in the
numerically-sorted AD list.

• Users can log in to their Admin Domains and create their own Admin Domain-specific zones
and zone configurations.

496

Fabric OS Administrator’s Guide
53-1002920-02

Admin Domain management for physical fabric administrators

20

Creating a new user account for managing Admin Domains
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the userConfig --add command using the -r option to set the role, the -a option to provide
access to Admin Domains, and the -h option to specify the home Admin Domain.
userconfig --add username -r role -h home_AD -a "AD_list"

Example of creating new user accounts

The following example creates new user account ad1admin with an admin role and assigns
one Admin Domain, blue_ad1, to it. This example also assigns blue_ad1 as the user’s home
Admin Domain.
switch:admin> userconfig --add ad1admin -r admin -h blue_ad1 -a "blue_ad1"

The following example creates new user account ad2admin with an admin role, access to
Admin Domains 1 and 2, and home Admin Domain set to 2.
switch:admin> userconfig --add ad2admin -r admin -h 2 -a "1,2"

Assigning Admin Domains to an existing user account
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the userConfig --addad command using the -a option to provide access to Admin
Domains and the -h option to specify the home Admin Domain.
userconfig --addad username -h home_AD -a "AD_list"

Example

The following example assigns Admin Domain green_ad2 to the existing user account
ad1admin.
switch:admin> userconfig --addad ad1admin -a "green_ad2"

Creating a physical fabric administrator user account
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the userConfig --add command using the -r option to set the role to admin and the -a
option to provide access to Admin Domains 0 through 255.
userconfig --add username -r admin -h home_AD -a "0-255"

Example

The following example creates new user account pfa_admin1 with an admin role, access to all
Admin Domains (AD0 through AD255), and home Admin Domain set to 255. This user account
is now a physical fabric administrator.
switch:admin> userconfig --add pfa_admin1 -r admin -h 255 -a "0-255"

Fabric OS Administrator’s Guide
53-1002920-02

497

20

Admin Domain management for physical fabric administrators

Removing an Admin Domain from a user account
When you remove an Admin Domain from an account, all of the currently active sessions for that
account are logged out.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the userConfig --deletead command:
userconfig --deletead username [-h admindomain_ID] [-a admindomain_ID_list]

If the –h argument is not specified, the home Admin Domain either remains as it was or
becomes the lowest Admin Domain ID in the remaining list.
Example of removing Admin Domain green_ad2 from the user account adm1
switch:admin> userconfig --deletead adm1 -a "green_ad2"
Broadcast message from root (pts/0) Wed Jan 27 20:57:14 2010...
Security Policy, Password or Account Attribute Change: adm1 will be logged out
Ads for account adm1 has been successfully deleted.

Activating an Admin Domain
An Admin Domain can be in either an active or inactive state. When you create an Admin Domain, it
is automatically in the active state.
1. Connect to the switch and log in using an account with admin permissions.
2. Switch to the AD255 context, if you are not already in that context.
ad --select 255

3. Enter the ad --activate command.
ad --activate ad_id

You are prompted for confirmation.
By default, after the Admin Domain is activated, the devices specified under that AD are not
able to see each other until they are zoned together.
4. Enter the appropriate command based on whether you want to save or activate the Admin
Domain definition:

• To save the Admin Domain definition, enter ad --save.
• To save the Admin Domain definition and directly apply the definition to the fabric, enter ad
--apply.

Example

The following example activates Admin Domain AD_B5.
switch:AD255:admin> ad --activate AD_B5
You are about to activate a new admin domain.
Do you want to activate ’AD_B5’ admin domain (yes, y, no, n): [no]: y
switch:AD255:admin>

498

Fabric OS Administrator’s Guide
53-1002920-02

Admin Domain management for physical fabric administrators

20

Deactivating an Admin Domain
If you deactivate an Admin Domain, the members assigned to the Admin Domain can no longer
access their hosts or storage unless those members are part of another Admin Domain.
You cannot log in to an Admin Domain that has been deactivated. You must activate an Admin
Domain before you can log in to it.
1. Connect to the switch and log in using an account with admin permissions.
2. Disable the zone configuration under the Admin Domain you want to deactivate.
cfgdisable

3. Switch to the AD255 context, if you are not already in that context.
ad --select 255

4. Enter the ad --deactivate command.
ad --deactivate ad_id

You are prompted for confirmation.
5. Enter the appropriate command based on whether you want to save or activate the Admin
Domain definition:

• To save the Admin Domain definition, enter ad --save.
• To save the Admin Domain definition and directly apply the definition to the fabric, enter ad
--apply.

All active user sessions associated with the Admin Domain are terminated. The ad --deactivate
command does not disable ports.
Example of deactivating Admin Domain AD_B4
switch:AD255:admin> ad --deactivate AD_B4
You are about to deactivate an AD.
This operation will fail if an effective zone configuration exists in the AD
Do you want to deactivate ’AD_B5’ admin domain (yes, y, no, n): [no] y
switch:AD255:admin>

Adding members to an existing Admin Domain
1. Connect to the switch and log in using an account with admin permissions.
2. Switch to the AD255 context, if you are not already in that context.
ad --select 255

3. Enter the ad --add command using the -d option to specify device and switch port members
and the -s option to specify switch members.
ad --add ad_id -d "dev_list" -s "switch_list"

In the syntax, ad_id is the Admin Domain name or number, dev_list is a list of device WWNs or
domain,index members, and switch_list is a list of switch WWNs or domain IDs.

Fabric OS Administrator’s Guide
53-1002920-02

499

20

Admin Domain management for physical fabric administrators

4. Enter the appropriate command based on whether you want to save or activate the Admin
Domain definition:

• To save the Admin Domain definition, enter ad --save.
• To save the Admin Domain definition and directly apply the definition to the fabric, enter ad
--apply.

Example of adding two switch ports, designated by domain,index, to AD1
switch:AD255:admin> ad --add AD1 -d "100,5; 4,1"

Removing members from an Admin Domain
If you remove the last member of an Admin Domain, that Admin Domain is automatically deleted.
1. Connect to the switch and log in using an account with admin permissions.
2. Switch to the AD255 context, if you are not already in that context.
ad --select 255

3. Enter the ad --remove command using the -d option to specify device and switch port members
and the -s option to specify switch members.
ad --remove ad_id -d "dev_list" -s "switch_list"

Removing the last member element of an Admin Domain deletes the Admin Domain.
4. Enter the appropriate command based on whether you want to save or activate the Admin
Domain definition:

• To save the Admin Domain definition, enter ad --save.
• To save the Admin Domain definition and directly apply the definition to the fabric, enter ad
--apply.

Example 1

The following example removes port 5 of domain 100 and port 3 of domain 1 from AD1.
switch:AD255:admin> ad --remove AD1 –d "100,5; 1,3"

Example 2

The following example removes switch 100 from the membership list of AD4.
switch:AD255:admin> ad --remove AD4 –s "100"

Renaming an Admin Domain
Use this procedure if you want to change the name of an Admin Domain. You can also change
auto-assigned names (ADn).
The rename operation does not take effect if the Admin Domain you want to rename is part of the
effective configuration.
1. Connect to the switch and log in using an account with admin permissions.
2. Switch to the AD255 context, if you are not already in that context.
ad --select 255

500

Fabric OS Administrator’s Guide
53-1002920-02

Admin Domain management for physical fabric administrators

20

3. Enter the ad --rename command with the present name and the new name.
ad --rename present_name new_name

4. Enter the appropriate command based on whether you want to save or activate the Admin
Domain definition:

• To save the Admin Domain definition, enter ad --save.
• To save the Admin Domain definition and directly apply the definition to the fabric, enter ad
--apply.

The Admin Domain numbers remain unchanged after the operation.
Example of changing the name of Admin Domain Eng_AD to Eng_AD2
switch:AD255:admin> ad --rename Eng_AD Eng_AD2

Deleting an Admin Domain
When you delete an Admin Domain, its devices no longer have access to the members of the zones
with which it was associated.
1. Connect to the switch and log in using an account with admin permissions.
2. Switch to the Admin Domain that you want to delete.
ad --select ad_id

3. Enter the appropriate command to clear the zone database under the Admin Domain you want
to delete.

• To remove the effective configuration, enter cfgdisable.
• To remove the defined configuration, enter cfgclear.
• To save the changes to nonvolatile memory, enter cfgsave.
4. Switch to the AD255 context.
ad --select 255

5. Enter the ad --delete command.
ad --delete ad_id

The ad --delete command prompts you for confirmation before triggering the deletion. The
command succeeds whether the Admin Domain is in an activated or deactivated state.
6. Enter the ad --apply command to save the Admin Domain definition and directly apply the
definition to the fabric.
Example of deleting Admin Domain AD_B3
switch:AD255:admin> ad --delete AD_B3
You are about to delete an AD.
This operation will fail if zone configuration exists in the AD
Do you want to delete ’AD_B3’ admin domain (yes, y, no, n): [no] y
switch:AD255:admin>

Fabric OS Administrator’s Guide
53-1002920-02

501

20

Admin Domain management for physical fabric administrators

Deleting all user-defined Admin Domains
When you clear the Admin Domain configuration, all user-defined Admin Domains are deleted, the
explicit membership list of AD0 is cleared, and all fabric resources (switches, ports, and devices)
are returned to the implicit membership list of AD0.
You cannot clear the Admin Domain configuration if zone configurations exist in any of the
user-defined Admin Domains.
If you want to remove all Admin Domains while retaining device connectivity (for example, if you
want to enable Virtual Fabrics), use the procedure described in “Deleting all user-defined Admin
Domains non-disruptively.”
1. Clear all individual AD zone databases, in separate transactions, before proceeding with this
operation. Refer to “Clearing all zone configurations” on page 367 for instructions.
2. Connect to the switch and log in using an account with admin permissions.
3. Switch to the AD255 context, if you are not already in that context.
ad --select 255

4. Enter the ad --clear command.
This option prompts you for confirmation before triggering the deletion of all Admin Domains.
5. Enter the ad --apply command to save the Admin Domain definition and directly apply the
definitions to the fabric.
Example
switch:AD255:admin> ad --clear
You are about to delete all ADs definitions.
This operations will fail if zone configurations exists in AD1-AD254
Do you want to clear all admin domains (yes, y, no, n): [no] y
switch:AD255:admin>

Deleting all user-defined Admin Domains non-disruptively
To disable Admin Domains non-disruptively, you must do the following before you clear the
user-defined ADs:

• Create and activate zone configurations in AD0 that are equivalent to the zone configurations
in each of the user-defined ADs.

• Define all of the members that are currently in user-defined ADs in AD0.
This will ensure that the devices are able to communicate when they are removed from the
user-defined ADs.
You can use this procedure to remove all Admin Domains before enabling Virtual Fabrics.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the cfgshow command in the AD255 context to display the zone configurations for all
Admin Domains.
ad --exec 255 "cfgshow"

502

Fabric OS Administrator’s Guide
53-1002920-02

Admin Domain management for physical fabric administrators

20

3. Enter the zone --copy command to copy the zones from all user-defined Admin Domains to
AD0.
zone --copy source_AD.source_name dest_name

In this syntax, source_AD is the name of the user-defined AD from which you are copying the
zone, source_name is the name of the zone to be copied, and dest_name is the name to give
to the zone after it is copied to AD0.
4. Copy the newly added zones in AD0 to the zone configuration.
cfgadd "cfgName", "member[;member]"

5. Enable the configuration to complete the transaction.
cfgenable cfgName

6. Switch to the AD255 context.
ad --select 255

7.

Explicitly add devices that are present in the user-defined ADs to AD0.
ad --add AD0 -d "dev_list"

8. Enter the ad --apply command to save the Admin Domain definition and directly apply the
definitions to the fabric.
ad --apply

At this point, all of the devices in the user-defined ADs are also defined and zoned in AD0.
9. Clear the user-defined ADs.
ad --clear -f

10. Enter the ad --apply command to save the Admin Domain definition and directly apply the
definitions to the fabric.
ad --apply

All user-defined Admin Domains have now been removed, but all device communication that was
allowed with the original Admin Domain configuration is still permitted in the context of AD0.
Example

The following example assumes the configuration shown in Figure 71 on page 504:

•
•
•
•
•

Three Admin Domains: AD0, plus two user-defined Admin Domains (AD1 and AD2).
AD0 has two devices, WWN1 and WWN2, in the AD0_RedZone.
AD1 has two devices, WWN2 and WWN3, in the AD1_BlueZone.
AD2 has two devices, WWN4 and WWN5, in the AD2_GreenZone.
The device WWN2 is in both AD0 and AD1.

Fabric OS Administrator’s Guide
53-1002920-02

503

20

Admin Domain management for physical fabric administrators

FIGURE 71

AD0 and two user-defined Admin Domains, AD1 and AD2

At the conclusion of the procedure, all devices and zones are moved to AD0, and the user-defined
Admin Domains are deleted, as shown in Figure 72.

FIGURE 72

AD0 with three zones

sw0:admin> ad --exec 255 "cfgshow"
Zone CFG Info for AD_ID: 0

(AD Name: AD0, State: Active) :

Defined configuration:
cfg:
AD0_cfg AD0_RedZone
zone: AD0_RedZone
10:00:00:00:01:00:00:00; 10:00:00:00:02:00:00:00
Effective configuration:
cfg:
AD0_cfg
zone: AD0_RedZone
10:00:00:00:01:00:00:00
10:00:00:00:02:00:00:00
Zone CFG Info for AD_ID: 1

(AD Name: AD1, State: Active) :

Defined configuration:
cfg:
AD1_cfg AD1_BlueZone
zone: AD1_BlueZone

504

Fabric OS Administrator’s Guide
53-1002920-02

Admin Domain management for physical fabric administrators

20

10:00:00:00:02:00:00:00; 10:00:00:00:03:00:00:00
Effective configuration:
cfg:
AD1_cfg
zone: AD1_BlueZone
10:00:00:00:02:00:00:00
10:00:00:00:03:00:00:00
Zone CFG Info for AD_ID: 2

(AD Name: AD2, State: Active) :

Defined configuration:
cfg:
AD2_cfg AD2_GreenZone
zone: AD2_GreenZone
10:00:00:00:04:00:00:00; 10:00:00:00:05:00:00:00
Effective configuration:
cfg:
AD2_cfg
zone: AD2_GreenZone
10:00:00:00:04:00:00:00
10:00:00:00:05:00:00:00

sw0:admin> zone --copy AD1.AD1_BlueZone AD0_BlueZone
sw0:admin> zone --copy AD2.AD2_GreenZone AD0_GreenZone
sw0:admin> cfgadd "AD0_cfg", "AD0_BlueZone; AD0_GreenZone"
sw0:admin> cfgenable AD0_cfg
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected. If the update includes changes
to one or more traffic isolation zones, the update may result in
localized disruption to traffic on ports associated with
the traffic isolation zone changes
Do you want to enable 'AD0_cfg' configuration (yes, y, no, n): [no] y
zone config "AD0_cfg" is in effect
Updating flash ...

sw0:admin> ad --select 255
sw0:AD255:admin> ad --add AD0 -d "10:00:00:00:03:00:00:00;
10:00:00:00:04:00:00:00; 10:00:00:00:05:00:00:00"
sw0:AD255:admin> ad --apply
You are about to enforce the saved AD configuration.
This action will trigger AD apply to all switches in the fabric
Do you want to apply all admin domains (yes, y, no, n): [no] y

sw0:AD255:admin> ad --clear -f
You are about to delete all ADs definitions and zone databases under them.
This could involve multiple independent zone transactions and
no auto recovery will be done in case of failure in the middle.
Do you want to clear all admin domains (yes, y, no, n): [no] y
sw0:AD255:admin> ad --apply
You are about to enforce the saved AD configuration.
This action will trigger AD apply to all switches in the fabric
Do you want to apply all admin domains (yes, y, no, n): [no] y

Fabric OS Administrator’s Guide
53-1002920-02

505

20

SAN management with Admin Domains

Validating an Admin Domain member list
You can validate the device and switch member list. You can list non-existing or offline Admin
Domain members. You can also identify misconfigurations of the Admin Domain.
The Admin Domain validation process is not applicable for AD0, because AD0 implicitly contains all
unassigned online switches and their devices.
1. Connect to the switch and log in using an account with admin permissions.
2. Switch to the AD255 context, if you are not already in that context.
ad --select 255

3. Enter the ad --validate command.
ad --validate ad_id -m mode

If you do not specify any parameters, the entire AD database (transaction buffer, defined
configuration, and effective configuration) is displayed.
If you do not specify an Admin Domain, information about all existing Admin Domains is
displayed.
The -m mode option can be used with the following values:

• 0 to display the Admin Domain configuration in the current transaction buffer.
• 1 to display the Admin Domain configuration stored in the persistent memory (defined
configuration).

• 2 to display the currently enforced Admin Domain configuration (effective configuration).
Example of validating the member list of Admin Domain 10 in the current transaction buffer
switch:AD255:admin> ad --validate 10 –m 0
Current AD Number: 255 AD Name: AD255
Transaction buffer configuration:
--------------------------------AD Number:
2
AD Name: ad2
State: Active
Switch port members:
1,1; 1,3; 2,5+; 3,6;
---------------------------* - Member does not exist
+ - Member is AD Unaware

SAN management with Admin Domains
This section is for both users and administrators and describes how Admin Domains affect
commands and other Fabric OS features. If you are a physical fabric administrator and you want to
create, modify, or otherwise manage Admin Domains, refer to “Admin Domain management for
physical fabric administrators” on page 494.
The Admin Domain looks like a virtual switch or fabric to a user. However, based on the user role
and type (User_ID), users are presented with only their relevant AD-based views (refer to Figure 66
on page 486 and Figure 67 on page 486). Any devices and switch ports that are not defined as
part of the Admin Domain are not shown and are not available to that AD user.
Each Admin Domain can also have its own zone configurations (defined and effective) with zones
and aliases under them.

506

Fabric OS Administrator’s Guide
53-1002920-02

SAN management with Admin Domains

20

CLI commands in an AD context
The CLI command input arguments are validated against the AD member list; they do not work with
input arguments that specify resources that are not members of the current Admin Domain. All
commands present filtered output, showing only the members of the current Admin Domain.
For example, switchShow displays details for the list of AD members present in that switch. Note
the following about the switchShow output:

• Because all E_Ports and EX_Ports are shared across all Admin Domains, they are shown under
all Admin Domains.

• Other ports are displayed without any attribute details (with an explanation that they are not
part of the current Admin Domain).
A port or device appears in CLI command output or other management tool outputs if any one of
the conditions listed in Table 80 is met.

TABLE 80

Ports and devices in CLI output

For

Condition

domain,index

•
•
•
•

Device WWN

The port is specified in the domain,index member list of the Admin Domain.
One or more WWNs specified in the AD member list is attached to the domain,index.
The device WWN is specified in the AD WWN member list.
The device WWN is attached to one of the domain,index members specified in the AD
member list.

RASlog and syslog output is not filtered based on AD membership.
Refer to the Fabric OS Command Reference for more detailed information about command syntax
and usage and to understand how existing commands behave in an AD context.

Executing a command in a different AD context
You can execute a command in an Admin Domain that is different from your current AD context.
The Admin Domain must be one that you can access. This option creates a new shell with the
current User_ID, switches to the specified Admin Domain, performs the specified command, and
exits the shell.
1. Connect to the switch and log in.
2. Enter the ad --exec command, specifying the Admin Domain and the command you want to
execute.
ad --exec ad_id "command"

Example of executing the switchShow command in the AD7 context
switch:AD255:admin> ad --exec 7 "switchshow"

Fabric OS Administrator’s Guide
53-1002920-02

507

20

SAN management with Admin Domains

Displaying an Admin Domain configuration
You can display the membership information and zone database information of a specified Admin
Domain. Notice the following differences in the information displayed based on the Admin Domain:

• AD255: If you do not specify the AD name or number, all information about all existing Admin
Domains is displayed.

• AD0–AD254: The membership of the current Admin Domain is displayed.
• AD0: The device and switch list members are categorized into implicit and explicit member lists.
1. Connect to the switch and log in as any user type.
2. Enter the ad --show command.
ad --show

If you are in the AD0 context, you can use the -i option to display the implicit membership list of
AD0; otherwise, only the explicit membership list is displayed.
ad --show -i

If you are in the AD255 context, all Admin Domain configurations from the transaction buffer,
defined configuration, and effective configuration are displayed, unless you use the -m option:
ad --show ad_id -m mode

In the syntax, ad_id is the Admin Domain for which you want to display information and mode is
one of the following values:

• 0 to display the Admin Domain configuration in the current transaction buffer.
• 1 to display the Admin Domain configuration stored in the persistent memory (defined
configuration).

• 2 to display the currently enforced Admin Domain configuration (effective configuration).
Example of displaying membership information about AD1
switch:AD1:admin> ad --show
Current AD Number: 1 AD Name: TheSwitches
Effective configuration:
-----------------------AD Number: 1 AD Name:

TheSwitches

Switch WWN members:

State: Active
50:06:06:99:00:2a:e9:01;
50:00:51:e0:23:36:f9:01;
50:06:06:98:05:be:99:01;

Switching to a different Admin Domain context
You can switch between different Admin Domain contexts. This option creates a new shell with a new
Admin Domain context. If the corresponding Admin Domain is not activated, the operation fails.
1. Connect to the switch and log in as any user type.
2. Enter the ad --select command and the Admin Domain to which you want to switch.
3. Leave the new Admin Domain context by exiting from the shell.
logout

508

Fabric OS Administrator’s Guide
53-1002920-02

SAN management with Admin Domains

20

You cannot switch to another Admin Domain context from within the shell created by ad
--select. You must first exit the shell, and then issue the ad --select command again.
Example of switching to a different Admin Domain context

The following example switches to the AD12 context and back. Note that the prompt changes
to display the Admin Domain.
switch:admin> ad --select 12
switch:AD12:admin> logout
switch:admin>

Admin Domain interactions with other Fabric OS features
The Admin Domain feature provides interaction with other Fabric OS features and across
third-party applications. You can manage Admin Domains with Web Tools as well as the CLI. If the
current Admin Domain owns the switch, you can perform Fabric Watch operations.
Admin Domain interactions do not extend to user session tunneling across switches. A user logged
in to a switch can control only the local switch ports as specified in the Admin Domain.
When the fabric is in secure mode, the following restrictions apply:

• There is no support for ACL configuration under each Administrative Domain.
• ACL configuration commands are allowed only in AD0 and AD255. None of the policy
configurations are validated with AD membership.
Table 81 lists some of the Fabric OS features and considerations that apply when using Admin
Domains.

TABLE 81

Admin Domain interaction with Fabric OS features

Fabric OS feature

Admin Domain interaction

ACLs

If no user-defined Admin Domains exist, you can run ACL configuration commands in only
AD0 and AD255. If any user-defined Admin Domains exist, you can run ACL configuration
commands only in AD255.
You cannot use ACL configuration commands or validate ACL policy configurations
against AD membership under each Admin Domain.

Advanced Performance
Monitoring (APM)

All APM-related filter setup and statistics viewing is allowed only if the local switch is part
of the current Admin Domain.

Configuration upload
and download

Refer to “Configuration upload and download in an AD context” on page 512 for details.

Fabric Watch

Fabric Watch configuration operations are allowed only if the local switch is part of the
current Admin Domain.

FC-FC Routing Service

You can create LSAN zones as a physical fabric administrator or as an individual AD
administrator. The LSAN zone can be part of the root zone database or the AD zone
database.
FCR collects the LSAN zones from all ADs. If both edge fabrics have matching LSAN
zones and both devices are online, FCR triggers a device import.
LSAN zone enforcement in the local fabric occurs only if the AD member list contains
both of the devices (local and imported devices) specified in the LSAN zone.
To support legacy applications, WWNs are reported based on the AD context using
NAA=5. As a result, you cannot use the NAA=5 field alone in the WWN to detect an FC
router.

Fabric OS Administrator’s Guide
53-1002920-02

509

20

SAN management with Admin Domains

TABLE 81

Admin Domain interaction with Fabric OS features (Continued)

Fabric OS feature

Admin Domain interaction

FDMI

FDMI operations are allowed only in AD0 and AD255.

FICON

Admin Domains support FICON. However, you must perform additional steps because
FICON management requires additional physical control of the ports. You must set up the
switch as a physical member of the FICON AD.
Device Connection Control (DCC) and Switch Connection Control (SCC) policies are
supported only in AD0 and AD255, because ACL configurations are supported only in
AD0 and AD255.

iSCSI

iSCSI operations are supported only in AD0.

LSAN zoning

Refer to “Admin Domains and LSAN zones” on page 511 for details.

Management
applications

Management interfaces that access the fabric without a user’s credentials continue to
get the physical fabric view. Examples include SNMPv1, Web Tools, HTTP access,
unzoned management server query, FAL in-band CT requests from FAL Proxy to FAL
Target, and FC-CT-based management applications.
Access from applications or hosts using management server calls can be controlled
using the management server ACL support provided by the msConfigure command. Note
that this is a switch-specific setting and not a fabric-wide setting.

Port swapping and PID
formats

Admin Domain port members are specified in domain,index format. Based on the PID
format, a domain,index member indicates a slot and port in the switch. The
domain,index member is effectively a member of that AD.
Port swapping has no effect on AD support as port swapping swaps only the area
numbers of two ports and Admin Domains are specified using domain,index members.
For detailed information about configuring the PID format, refer to Chapter 3,
“Performing Advanced Configuration Tasks”.

RSCN

Admin Domains do not introduce any RSCN changes to devices or hosts.

Virtual Fabrics

Virtual Fabrics and Admin Domains are mutually exclusive and are not supported at the
same time on a switch. To use Admin Domains, you must first disable Virtual Fabrics; to
use Virtual Fabrics, you must first delete all Admin Domains.
If you connect a switch with Admin Domains to a Virtual Fabrics-enabled switch, the link
is segmented with the reason “VF AD conflict.”

Zoning

Refer to “Admin Domains, zones, and zone databases” on page 510 for details.

Admin Domains, zones, and zone databases
Admin Domains introduce two types of zone database nomenclature and behavior:

• Root zone database
If you do not use Admin Domains, there is only one zone database. This legacy zone database
is known as the root zone database. If you create Admin Domains, several zone databases
exist: the root zone database, which is owned by AD0, and other zone databases, one for each
user-defined Admin Domain.
AD-level zone information is merged with the root zone configuration and enforced.

• AD zone databases
Each AD (AD1 through AD254) has its own zone database, with the defined and effective zone
configurations and all related zone objects (zones, zone aliases, and zone members). Each AD
has its own zone transaction buffer. Within an Admin Domain, you can configure zoning only
with the devices that are present in that Admin Domain.

510

Fabric OS Administrator’s Guide
53-1002920-02

SAN management with Admin Domains

20

The AD zone database also has the following characteristics:

-

Each zone database has its own name space. For example, you can define a zone name of
test_z1 in more than one Admin Domain.

-

There is no zone database linked to the physical fabric (AD255) and no support for zone
database updates. In the physical fabric context (AD255), you can only view the complete
hierarchical zone database, which is all of the zone databases in AD0 through AD254.

-

You can concurrently edit the separate zone databases.
With AD support, zoning updates are supported selectively at each AD level. For example,
a zone change in AD1 results in an update request only for the AD1 zone database.

Zoning operations ignore any resources not in the Admin Domain, even if they are specified in the
zone. The behavior functions similarly to specifying offline devices in a zone. All zones from each
AD zone configuration are enforced. The enforcement policy encompasses zones in the effective
zone configuration of the root zone database and the effective zone configurations of each AD.

NOTE

You must define the Admin Domain members in the same way that they are defined as zone
members. That is, if an object is defined by WWN in a zone, it must be defined by WWN in an Admin
Domain. If it is defined by domain,index in a zone, it must be defined by domain,index in the Admin
Domain. If both zoning schemes are used, then objects must be defined in the Admin Domain by
both WWN and domain,index.
Using the zone --validate command, you can see all zone members that are not part of the current
zone enforcement table but are part of the zoning database. A member might not be part of the
zone enforcement table for the following reasons:

• The device is offline.
• The device is online but is not part of the current Admin Domain.
Refer to “Validating a zone” on page 358 for instructions on using the zone --validate command.

NOTE

AD zone databases do not have an enforced size limit. The zone database size is calculated by the
upper limit of the AD membership definition and the sum of all the zone databases for each AD.
Admin Domains support the default zone mode of No Access only. Before configuring any Admin
Domain, you must set the default zone to No Access mode. Admin Domains without effective zone
configurations are presented with No Access. Refer to “Default zoning mode” on page 360 for more
information.
If the administrative domain feature is not active (AD1 through AD254 are not configured and no
explicit members are added to AD0), AD0 supports both All Access and No Access default zone
modes.

Admin Domains and LSAN zones
Logical storage area networks (LSANs) under each Admin Domain are collated into a single name
space and sent out to FCR phantom domains using the following format:
_AD

For example, a zone with name lsan_for_linux_farm in AD5 is internally converted to
lsan_for_linux_farm_AD005.

Fabric OS Administrator’s Guide
53-1002920-02

511

20

SAN management with Admin Domains

LSAN zone names in AD0 are never converted for backward-compatibility reasons.
The auto-converted LSAN zone names might collide with LSAN zone names in AD0 (in the example,
if AD0 contains lsan_for_linux_farm_AD005, this causes a name collision). Fabric OS does not
detect or report such name clashes.
LSAN zone names greater than 57 characters are not converted or sent to the FCR phantom
domain.
LSAN zones defined within an Admin Domain must contain devices that are applicable to that
Admin Domain only. A device must not be included in more than one LSAN zone across multiple
Admin Domains. Device discovery problems might occur if LSAN zones in one Admin Domain
contain devices that belong to another Admin Domain.
Refer to Chapter 26, “Using FC-FC Routing to Connect Fabrics,” for information about LSAN zones.

Configuration upload and download in an AD context
The behavior of the configUpload and configDownload commands varies depending on the AD
context and whether the switch is a member of the current Admin Domain. In the AD context, these
commands include only the zone configuration of the current Admin Domain. If the switch is a
member of the Admin Domain, all switch configuration parameters are saved and the zone
database for that Admin Domain is also saved.
Table 82 lists the sections in the configuration file and the Admin Domain contexts in which you
can upload and download these sections. Refer to Chapter 9, “Maintaining the Switch
Configuration File,” for additional information about uploading and downloading configurations.

NOTE

You cannot use configDownload to restore a single Admin Domain. To restore a single Admin
Domain, you must first delete all Admin Domains and then issue configDownload to restore them.

TABLE 82

Configuration upload and download scenarios in an AD context
Configuration file sections

AD contexts

iSCSI

ACL

Zone

AD headers

Switch configuration
and other parameters

AD255:

Yes

Yes

Yes1

Yes

Yes

Yes

Yes

1

Yes

Yes

Yes

2

No

Yes

Yes

2

No

No

Yes

2

No

Yes

Yes

3

No

Yes

Yes

3

No

No

With ADs
Without ADs

AD0:

With ADs and switch membership
With ADs and without switch membership
Without ADs

AD1 – AD254: With switch membership
Without switch membership
1.

Yes
Yes
Yes
Yes
No
No

No
No
Yes
No
No

Zone databases for AD0 through AD254.

2.

Only zone database for AD0.

3.

Only zone database for current AD.

The configDefault command does not clear zone or Admin Domain database information.
This command is allowed only if the switch is a member of the current Admin Domain.

512

Fabric OS Administrator’s Guide
53-1002920-02

Section

Licensed Features

II

This section describes optionally licensed Brocade Fabric OS features and includes the following
chapters:

•
•
•
•
•
•

Chapter 21, “Administering Licensing”
Chapter 22, “Inter-chassis Links”
Chapter 23, “Monitoring Fabric Performance”
Chapter 24, “Managing Trunking Connections”
Chapter 25, “Managing Long-Distance Fabrics”
Chapter 26, “Using FC-FC Routing to Connect Fabrics”

Fabric OS Administrator’s Guide
53-1002920-02

513

514

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

21

Administering Licensing

In this chapter
• Licensing overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Brocade 7800 Upgrade license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• ICL licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• 8G licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Slot-based licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• 10G licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Temporary licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Viewing installed licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Activating a license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Adding a licensed feature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Removing a licensed feature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Ports on Demand. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

515
523
523
525
526
527
530
532
533
533
534
535

Licensing overview
Feature licenses are often part of the licensed paper pack supplied with your switch software; if
not, they can be purchased separately from your switch vendor, who provides the transaction keys
to activate the associated feature or features. Each product, each feature, and each individual
switch within a fabric requires its own license key.
Licenses may be associated with a feature version. If a feature has a version-based license, that
license is valid only for a particular version of the feature. If you want a newer version of the
feature, you must purchase a new license. If a license is not version-based, then it is valid for all
versions of the feature. Likewise, if you downgrade Fabric OS to an earlier version, some licenses
associated with specific features of the version you are downgrading may not work.

NOTE
To preserve licenses and the functioning of features associated with the licenses installed on your
switch, use the configUpload command before you upgrade or downgrade Fabric OS.
You can use the licenseShow command to display the license keys on a switch. Save the output to
a text file in a secure location. If licenses are lost or removed from the switch, you can use the saved
output to recover or add the lost licenses.
Fabric OS includes basic switch and fabric support software, and support for optionally licensed
software that is enabled using license keys.

Fabric OS Administrator’s Guide
53-1002920-02

515

21

Licensing overview

Some licenses may display with the text “Obsolete license.” This happens because of changes in
licensing requirements of some features that no longer require a license key, yet are still installed
on a switch.

ATTENTION
The Adaptive Networking and Server Application Optimization (SAO) licenses are no longer required
to be explicitly installed in Fabric OS 7.2.0 and later. These licenses are automatically enabled for
new switches and for existing switches that are upgraded to Fabric OS 7.2.0 or later.
If these licenses were installed on the switch prior to upgrading to Fabric OS 7.2.0, they will display
as “Obsolete license”. Do not remove these licenses, as they are required if you downgrade and want
to maintain QoS functionality.
Some Brocade HBA configuration settings require the Adaptive Networking or SAO license to be
installed on the switch. Starting in Fabric OS 7.2.0, these licenses are always installed.
Table 83 lists the optionally licensed features that are available in Fabric OS 7.2.

TABLE 83

Available Brocade licenses

License

Description

10 Gigabit FCIP/Fibre Channel
(10G license)

•
•
•

•
7800 Upgrade

•
•
•

Allows 10 Gbps operation of FC ports on the Brocade 6510 or
6520 switches or the FC ports of FC16-32 or FC16-48 port
blades installed on a Brocade DCX 8510 Backbone.
Enables the two 10-GbE ports on the FX8-24 extension blade
when installed on the Brocade DCX, DCX-4S, DCX 8510-4, or
DCX 8510-8 Backbone.
Allows selection of the following operational modes on the
FX8-24 blade:
- 10 1-GbE ports and 1 10-GbE port, or
- 2 10-GbE ports
License is slot-based when applied to a Brocade Backbone. It is
chassis-based when applied to a Brocade 6510 or 6520 switch.
Enables full hardware capabilities on the Brocade 7800 base
switch, increasing the number of Fibre Channel ports from four
to sixteen and the number of GbE ports from two to six.
Supports up to eight FCIP tunnels instead of two.
Supports advanced capabilities such as tape read/write
pipelining.

NOTE: The Brocade 7800 switch must have the 7800 Upgrade
license to add FICON Management Server (CUP) or Advanced
FICON Acceleration licenses. Refer to “Brocade 7800 Upgrade
license” on page 523 for details.

516

Fabric OS Administrator’s Guide
53-1002920-02

Licensing overview

TABLE 83

21

Available Brocade licenses (Continued)

License

Description

Advanced Extension

•
•

•
•
Advanced FICON Acceleration

•

•

Enables two advanced extension features: FCIP Trunking and
Adaptive Rate Limiting.
FCIP Trunking feature allows all of the following:
- Multiple (up to 4) IP source and destination address pairs
(defined as FCIP Circuits) using multiple (up to 4) 1-GbE or
10-GbE interfaces to provide a high bandwidth FCIP tunnel
and failover resiliency.
- Support for up to 4 of the following QoS classes: Class-F,
high, medium and low priority, each as a TCP connection.
The Adaptive Rate Limiting feature provides a minimum
bandwidth guarantee for each tunnel with full usage of available
network bandwidth without any negative impact to throughput
performance under high traffic load.
Available on the Brocade 7800 switch, and the Brocade DCX and
DCX-4S and the Brocade DCX 8510 family for the FX8-24 on an
individual slot basis.
Allows use of specialized data management techniques and
automated intelligence to accelerate FICON tape read and write
and IBM Global Mirror data replication operations over distance,
while maintaining the integrity of command and
acknowledgement sequences.
Available on the Brocade 7800 switch, and the Brocade DCX and
DCX-4S and the Brocade DCX 8510 family for the FX8-24 on an
individual slot basis.

Brocade Advanced Performance
Monitoring

•

Brocade Extended Fabrics

Provides greater than 10 km of switched fabric connectivity at full
bandwidth over long distances (depending on the platform, this can
be up to 3000 km).

•

Enables performance monitoring of networked storage
resources.
Includes the Top Talkers feature.

NOTE: This license is not required for long distance connectivity using
licensed 10G ports.
Brocade Fabric Watch
Brocade ISL Trunking

•
•
•
•

Brocade Ports on Demand

Monitors mission-critical switch operations.
Includes Port Fencing capabilities.
Provides the ability to aggregate multiple physical links into one
logical link for enhanced network performance and fault
tolerance.
Includes Access Gateway ISL Trunking on those products that
support Access Gateway deployment.

Allows you to instantly scale the fabric by provisioning additional ports
using license key upgrades.
NOTE: Applies to the Brocade 300, 5100, 5300, M6505, 6505,
6510, 6520, 6547, and VA-40FC switches.

DataFort Compatibility

Provides ability to read, write, decrypt, and encrypt the NetApp
DataFort-encrypted Disk LUNs and Tapes to all of the following:
• Brocade Encryption Switch
• Brocade enterprise platforms with FS8-18 blade
Includes metadata, encryption and compression algorithms.
NOTE: Availability is limited. Contact your vendor for details.

Fabric OS Administrator’s Guide
53-1002920-02

517

21

Licensing overview

TABLE 83

Available Brocade licenses (Continued)

License

Description

Encryption Performance Upgrade

Provides additional encryption bandwidth on encryption platforms.
For the Brocade Encryption Switch, two Encryption Performance
Upgrade licenses can be installed to enable the full available
bandwidth. On a Brocade enterprise platform, a single Performance
License can be installed to enable full bandwidth on all FS8-18 blades
installed in the chassis.

Enhanced Group Management

Enables full management of the device in a data center fabric with
deeper element management functionality and greater management
task aggregation throughout the environment. This license is used in
conjunction with Brocade Network Advisor application software. This
license is applicable to all of the Brocade 8G and 16G FC platforms.
NOTE: This license is enabled by default on all 16G FC platforms, and
on DCX and DCX-4S platforms that are running Fabric OS
v7.0.0 or later.
This license is not included by default on 8G FC fixed port
switches (300, 5100, 5300, VA-40FC, and 8G FC embedded
switches).

Enterprise ICL

Allows you to connect more than four chassis in a fabric using ICLs.
You can connect up to four Brocade DCX 8510 Backbones via ICLs
without this license. If the number of interconnected chassis using
ICLs exceeds four, then all of the chassis using ICLs require the
Enterprise ICL license.
You must also have an ICL POD license on each Brocade DCX 8510 to
activate the ICL ports The Enterprise ICL license only allows
connection of more than four chassis using ICLs in a fabric; it does not
enable the ICL ports on a chassis.
NOTE: Applies to the Brocade DCX 8510 Backbone family only.

518

Fabric Vision (FV)

Allows you to activate the following features:
Monitoring and Alerting Policy Suite (MAPS)
Flow Vision
Run D_Port tests between a switch and non-Brocade HBAs
This license replaces the Advanced Performance Monitor (APM) and
Fabric Watch (FW) licenses. If you have the Fabric Vision license, you
can use Advanced Performance Monitoring and Fabric Watch features
without the APM and FW licenses.

FCoE

Enables Fibre Channel over Ethernet (FCoE) functions.

FICON Management Server
(Also known as Control Unit Port or “CUP”)

Enables host-control of switches in mainframe environments.

High Performance Extension over FCIP/FC
(formerly known as “FC-IP Services”)

Includes the IPsec capabilities.

ICL 8-Link

Activates all eight links on ICL ports on a Brocade DCX-4S or half of
the ICL bandwidth for each ICL port on the Brocade DCX platform by
enabling only eight links out of the sixteen links available. This allows
you to purchase half the bandwidth of DCX ICL ports initially and
upgrade with an additional ICL 8-Link license to utilize the full ICL
bandwidth at a later time. This license is also useful for environments
that want to create ICL connections between a DCX and a DCX-4S; the
latter cannot support more than eight links on an ICL port.
Available on the Brocade DCX and DCX-4S Backbones only.

•
•
•

Fabric OS Administrator’s Guide
53-1002920-02

Licensing overview

TABLE 83

21

Available Brocade licenses (Continued)

License

Description

ICL 16-Link

Activates all 16 links on ICL ports on a Brocade DCX chassis. Each
chassis must have the ICL 16-Link license installed in order to enable
the full 16-link ICL connections.
Available on the Brocade DCX only.

Integrated Routing

Allows any ports in Brocade 5100, 5300, 6510, 6520, 6547, and
VA-40FC switches, the Brocade Encryption Switch, or the Brocade
DCX, DCX-4S, and DCX 8510 family platforms to be configured as an
EX_Port supporting FC-FC routing.

Inter-Chassis Link (1st POD)

Activates half of the ICL bandwidth on a DCX 8510-8, or all the ICL
bandwidth on a DCX 8510-4, allowing you to enable only the
bandwidth needed, and upgrade to additional bandwidth at a later
time. This license is also useful for environments that wish to create
ICL connections between a DCX 8510-8 and a DCX 8510-4; the latter
platform supports only half the number of ICL links that the former
platform supports.
Available on the Brocade DCX 8510 Backbones only.

Inter-Chassis Link (2nd POD)

Activates the remaining ICL bandwidth on the Brocade DCX 8510-8
chassis. Each chassis must have this ICL license installed in order to
enable all available ICL connections.
Available on the Brocade DCX 8510-8 only.

Table 84 lists licensed features, each feature’s associated license name, and, if applicable, the
location on the local or any connecting switch on which the license must be installed.

TABLE 84

License requirements and location name by feature

Feature

License

Where license should be installed

Adaptive Rate Limiting

Advanced Extension

Local switch.

Administrative Domains

No license required.

N/A

Bottleneck Detection

No license required.

N/A

Brocade Network Advisor

No license required for base use.

See the Brocade Network Advisor
User Manual.

Configuration
up/download

No license required.

N/A

D_Port

No license required for D_Port tests between
two switches or between a switch and a
Brocade HBA.
Fabric Vision license is required for D_Port tests
between a switch and a non-Brocade HBA.

If Fabric Vision license is required, it
should be installed on the local
switch.

Diagnostic tools

No license required.

N/A

Distributed Management
Server

No license required.

N/A

Enterprise ICL

Enterprise ICL

Each ICL-connected Brocade DCX
8510 chassis in the fabric when
there are five or more such chassis
in the fabric.

Fabric OS Administrator’s Guide
53-1002920-02

NOTE: The configUpload and configDownload
commands are provided automatically
with Fabric OS on the switch.

519

21

Licensing overview

TABLE 84

License requirements and location name by feature (Continued)

Feature

License

Where license should be installed

Extended Fabrics

Extended Fabrics

Local switch and any attached
switches.

Fabric Watch

No license required for baseline monitoring
capabilities.
Fabric Watch license or Fabric Vision license
required for full functionality.

See the Fabric Watch
Administrator’s Guide.

FCIP

High Performance Extension over FCIP/FC

NOTE: Local and attached
switches. License is needed
on both sides of tunnel.

FCIP Trunking

Advanced Extension

Local and attached switches.

Fibre Channel
Routing/EX_Ports

Integrated Routing

Local switch.

FICON

No license required.

N/A

FICON-CUP

FICON Management Server

Local switch.

FICON Tape Read and
Write Emulation over an
FCIP Tunnel

•
•

FICON Tape
High Performance Extension over FCIP/FC
license or Advanced FICON Acceleration
license on Brocade 7800

Local and attached switches.

FICON XRC Sequence
Emulation over an FCIP
Tunnel

•
•

FICON XRC
High Performance Extension over FCIP/FC
license or Advanced FICON Acceleration
license on Brocade 7800

Local and attached switches.

FIPS

No license required.

N/A

Firmware download

No license required.

N/A

NOTE: The firmwareDownload command is
provided automatically with Fabric OS
on the switch.
Flow Vision:
Flow Generator
Flow Performance
Monitor
• Flow Mirror

Fabric Vision

Full fabric connectivity

Full Fabric

•
•

NOTE: Also called the Fabric license (visible in
licenseShow output) and the E_Port
Upgrade license.

520

Local switch.

NOTE: If you have both the Advanced
Performance Monitoring and the Fabric
Watch licenses installed, you do not
need the Fabric Vision license.
Local switch. May be required on
attached switches.

In-flight encryption and
compression

No license required.

N/A

Inband Management

No license required.

N/A

Ingress rate limiting

No license required.
Adaptive Networking with QoS license is
required for switches running Fabric OS
versions earlier than 7.2.0.
The Brocade 6520 does not require a license
regardless of Fabric OS version.

N/A for local switches running
Fabric OS 7.2.0 or later.
License required on local switches
running Fabric OS versions earlier
than 7.2.0.

Fabric OS Administrator’s Guide
53-1002920-02

Licensing overview

TABLE 84

21

License requirements and location name by feature (Continued)

Feature

License

Where license should be installed

Inter-chassis link (ICL)

•

Local and attached platforms.

•
•
•
•

ICL 1st POD (Ports on Demand) on the
Brocade DCX 8510 Backbone family only.
ICL 2nd POD on the Brocade DCX 8510-8
only.
ICL 8-link on the Brocade DCX and DCX-4S
only.
ICL 16-link on the Brocade DCX only.
Enterprise ICL on the Brocade DCX 8510
Backbone family only, for topologies with
more than four chassis with ICLs.

IPSec

No license required.

N/A

IPsec for FCIP tunnels

High Performance Extension over FCIP/FC

NOTE: Local and attached
switches. License is needed
on both sides of tunnel.

LDAP

No license required.

N/A

Logical fabric

No license required.

N/A

Logical switch

No license required.

N/A

Long distance

Extended Fabrics

Local and attached switches.
NOTE: License is needed on both
sides of connection.
Local switch.

Monitoring and Alerting
Policy Suite (MAPS)

Fabric Vision

NPIV

No license required.

N/A

OpenSSH public key

No license required.

N/A

Performance monitoring

Advanced Performance Monitoring or Fabric
Vision license for advanced features. No license
required for basic features.

Local switch.

Port fencing

Fabric Watch

Local switch.

Ports

•

Local switch.

NOTE: If you have both the Advanced
Performance Monitoring and the Fabric
Watch licenses installed, you do not
need the Fabric Vision license.

•
•
•

Fabric OS Administrator’s Guide
53-1002920-02

Ports on Demand licenses required,
applicable to a select set of switches only.
7800 Upgrade license for the 7800
switches to use all ports.
10 Gigabit FCIP/Fibre Channel license to
use 10Gb FC ports on FC16-32 blades,
FC16-48 blades, and the Brocade 6510
and 6520.
10 Gigabit FCIP/Fibre Channel license to
enable 10Gb Ethernet ports on the FX8-24
extension blades.

521

21

Licensing overview

TABLE 84

License requirements and location name by feature (Continued)

Feature

License

Where license should be installed

QoS

No license required.
Adaptive Networking with QoS license is
required for switches running Fabric OS
versions earlier than 7.2.0.
The Brocade 6520 does not require a license
regardless of Fabric OS version.

N/A for local switches running
Fabric OS 7.2.0 or later.
License required on local and
attached switches running Fabric OS
versions earlier than 7.2.0.

QoS on HBA

No license required.
Adaptive Networking with QoS license is
required for switches running Fabric OS
versions earlier than 7.2.0.
The Brocade 6520 does not require a license
regardless of Fabric OS version.

N/A for local switches running
Fabric OS 7.2.0 or later.
License required on local switches
running Fabric OS versions earlier
than 7.2.0.

RADIUS

No license required.

N/A

RBAC

No license required.

N/A

No license required.

N/A

Routing traffic

NOTE: Port-based or exchanged-based routing,
static routes, frame-order deliver, and
dynamic routes all included.
Security

No license required.

N/A

NOTE: DCC, SCC, FCS, IP Filter, and
authentication policies all included.
SNMP

No license required.

N/A

Speed

8 Gbps license needed to support 8 Gbps on
the Brocade 300, 5100, 5300, and VA-40FC
switches and embedded switches only.

Local switch

NOTE: The 8 Gbps license is installed by
default, and you should not remove it.
A 10-Gb FCIP/Fibre Channel license is needed
to support 10Gb FC ports on FC16-32 blades,
FC16-48 blades, and the Brocade 6510 and
6520, as well as to support the 10Gb Ethernet
ports on FX8-24 blades. (See the Ports feature
above for more information.)

522

SSH public key

No license required.

N/A

TACACS+

No license required.

N/A

Top Talkers

Advanced Performance Monitoring

Local switch and attached switches.

Traffic Isolation

No license required.

N/A

Trunking

•
•

ISL Trunking or
ISL Trunking Over Extended Fabrics
For ICL trunking, no license is required.

Local and attached switches.

Two-to-four domains in a
fabric

Value Line (Two/Four)

Local switch. May be required on
attached switches.

USB usage

No license required.

N/A

Virtual Fabrics

No license required.

N/A

Fabric OS Administrator’s Guide
53-1002920-02

Brocade 7800 Upgrade license

TABLE 84

21

License requirements and location name by feature (Continued)

Feature

License

Where license should be installed

Web Tools

No license required.

Local and any switch you will be
managing using Web Tools.

Zoning

No license required.

N/A

Brocade 7800 Upgrade license
The Brocade 7800 has four Fibre Channel (FC) ports and two GbE ports active by default. The
number of physical ports active on the Brocade 7800 is fixed. There is one upgrade license to
activate the rest of the FC and GbE ports for a total of 16 FC ports and 6 GbE ports. The 7800
Upgrade license activates FC and GbE ports, and also activates additional features outlined in
Table 85.

NOTE
You must reboot the Brocade 7800 switch after installing the 7800 Upgrade license.

TABLE 85

Base to Upgrade license comparison

Feature

Base model

7800 Upgrade license

Number of Fibre Channel (FC) ports

4

16

Number of GbE ports

2

6

Number of 10-GbE ports

0

0

Number of FCIP Tunnels

2

8

Tape Pipelining over FCIP Tunnel

No

Yes

ICL licensing
Brocade ICL links operate between the core blades of the DCX 8510 Backbone family, or between
the core blades of the DCX and DCX-4S Backbones. Typically, if both core blades are installed, then
they are active on the DCX and DCX-4S (or DCX 8510 family) Backbones.
ICL ports on core blades of a DCX 8510-8 can be used only with an ICL (1st or 2nd) POD license.
ICL ports on core blades of a DCX 8510-4 can be used only with an ICL 1st POD license.
ICL ports on core blades of a DCX can be used only with an ICL 16-link or ICL 8-link license. ICL
ports on core blades of a DCX-4S can be used only with an ICL 8-link license.
After the addition or removal of a license, the license enforcement is performed on the ICL ports
only when the portDisable and portEnable commands are issued on the ports. An ICL license must
be installed on the enterprise platforms at both ends of the ICL connection.

ICL 1st POD license
The ICL 1st POD license activates half of the ICL bandwidth on the Brocade DCX 8510-8 platform or
all of the ICL bandwidth on the Brocade DCX 8510-4.

Fabric OS Administrator’s Guide
53-1002920-02

523

21

ICL licensing

On the Brocade DCX 8510-8, this license enables QSFP ports 0–7; QSFP ports 8–15 are disabled.
(QSFP ports 0–7 correspond to core blade port numbers 0–31, and QSFP ports 8–15 correspond
to core blade port numbers 32–63, as observed in switchShow output.)
This license allows you to purchase half the bandwidth of the Brocade DCX 8510-8 ICL ports
initially and upgrade with an additional ICL license to use the full ICL bandwidth later. This license is
also useful for environments with ICL connections between a Brocade DCX 8510-8 and a DCX
8510-4, as the latter supports half the bandwidth of the DCX 8510-8 on each ICL port.
This license is available on the Brocade DCX 8510-8 and DCX 8510-4 platforms only.

ICL 2nd POD license
The ICL 2nd POD license provides dedicated high-bandwidth links between two Brocade DCX
8510-8 platforms without consuming valuable front-end ports. Each Brocade DCX 8510-8 platform
must have the ICL 2nd POD license installed to enable the full number of ICL connections possible.
This license is available for the Brocade DCX 8510-8 only.

ICL 8-link license
The ICL 8-link license activates half of the ICL bandwidth for each ICL port on the Brocade DCX
platform by enabling only half of the ICL links available. This allows you to purchase half the
bandwidth of the Brocade DCX ICL ports initially and upgrade with an additional ICL license to use
the full ICL bandwidth later. This license is also useful for environments with ICL connections
between a Brocade DCX and a DCX-4S, as the latter cannot support more than eight links on an ICL
port.
This license is available on the Brocade DCX-4S and DCX platforms only.

ICL 16-link license
The ICL 16-link license provides dedicated high-bandwidth links between two Brocade DCX chassis,
without consuming valuable front-end ports. Each Brocade DCX chassis must have the ICL 16-link
license installed in order to enable the full number of ICL connections possible (16 links in the case
of a DCX chassis).
This license is available for the Brocade DCX only.

Enterprise ICL license
The Enterprise ICL (EICL) license allows you to connect more than four Brocade DCX 8510
Backbones through ICLs. This license is available on the Brocade DCX 8510-8 and DCX 8510-4
platforms only.
The EICL license is required in addition to the ICL POD license.
The following requirements apply:

• Connection of four or fewer DCX 8510 Backbones with ICLs does not require the EICL license.
However, if you add additional ICL-connected chassis, then all ICL-connected chassis require
the EICL license.

• With the EICL license installed, a maximum of 10 chassis are allowed to be connected together
via ICLs.

524

Fabric OS Administrator’s Guide
53-1002920-02

8G licensing

21

• When Virtual Fabrics are used, the limit on the number of chassis connected together via ICLs
depends only on the physical chassis and not on the logical switches.

• If the maximum number of ICL-connected chassis exceeds the allowed limit with or without the
EICL license, additional links may either be disabled or segmented. The disabling or
segmenting reason code depends on whether the EICL license is installed.

• If ICL links to a chassis become segmented for non-EICL-related reasons, these links are part
of the fabric, and the chassis containing these segmented links is included in the maximum
chassis count. If the maximum chassis count (with or without the EICL license) is reached with
these segmented links, then any additional links will become segmented. Therefore, to add
additional links, you first must disable the links that became segmented due to non-EICL
reasons. This should reduce the maximum chassis count and allow the new links to join.
Example switchShow output if no Enterprise ICL license is installed

A message such as the following is displayed if a required EICL license is not installed:
440
8
24
-----id
16G
Online
FC
segmented,10:00:00:05:33:0d:52:00 (No EICL License)(Trunk
441
8
25
-----id
16G
Online
FC
segmented,10:00:00:05:33:0d:52:00 (No EICL License)(Trunk

E-Port
master)
E-Port
master)

Example switchShow output if maximum number of chassis is reached

A message such as the following is displayed if the maximum number of supported chassis is
reached:
384
5
0
-----id
16G
Online
FC E-Port
segmented,10:00:00:05:1e:39:bf:9a (EICL License Limited)(Trunk master)
385
5
1
-----id
16G
Online
FC E-Port
segmented,10:00:00:05:1e:39:bf:9a (EICL License Limited)(Trunk master)

8G licensing
ATTENTION
The 8G license is installed by default and you should not remove it. Port operation may become
disrupted, and ports may be prevented from operating at 8 Gbps when the license is removed.
The 8G license applies to the Brocade 300, 5100, 5300, and VA-40FC switches and the 8 Gbps
embedded switches; this license does not apply to the Brocade 6505, 6510, or 6520.
The following list describes the basic rules of using, adding, or removing 8G licenses:

• Without an 8G license, even if there is an 8 Gbps SFP plugged into a port in an applicable
platform, the port would be enabled to run at a maximum speed of 4 Gbps.

• To obtain an 8G license, only the license ID from the switch is required. When you add the 8G
license, you must enter either the portDisable and portEnable commands on each individual
port on the switch, or the switchDisable and switchEnable commands on the switch, to enable
8 Gbps features.

• When you remove the 8G license, the ports that are online and already running at 8 Gbps are
not disturbed until the port goes offline or the switch is rebooted. The switch ports return to
their pre-licensed state maximum speed of 4 Gbps.

Fabric OS Administrator’s Guide
53-1002920-02

525

21

Slot-based licensing

Slot-based licensing
Slot-based licensing is used on the Brocade DCX and DCX 8510 Backbone families to support the
FX8-24 blade, and on the Brocade DCX 8510 Backbone family to support the 16-Gbps FC port
blades (FC16-24 and FC16-48). License capacity is equal to the number of slots. These licenses
allow you to select the slots that the license will enable up to the capacity purchased and to
increase the capacity without disrupting slots that already have licensed features running. Each
slot-based license key is for a single feature.
Features utilizing slot-based licenses on the FX8-24 blade include:

• 10 Gigabit FCIP/Fibre Channel
• Advanced Extension
• Advanced FICON Acceleration
NOTE
The 10 GbE feature on the FX8-24 blade and the 10 Gbps FC feature on the 16-Gbps FC blades are
both enabled by the same 10 Gigabit FCIP/Fibre Channel license (10G license). This license can also
enable the 10 Gbps FC feature on a Brocade 6510 or 6520 switch as a chassis-based license.
Any unassigned slot-based license will be automatically assigned to applicable blades that are
detected in the chassis when the license is installed. If you have more applicable blades than
available license capacity, then you can manually assign or re-assign the licenses as necessary.
Once a license is assigned to a slot, whether it has been automatically assigned or manually
assigned, the assignment will remain until you manually reassign the license to another slot. This
design allows for various maintenance operations to occur without having the license move around
to other slots.
Use the following procedure to activate a slot-based licensed feature:
1. Install a slot-based license on the platform with sufficient slot count for the number of slots
upon which you plan to activate the feature.
2. Configure slots so that the licensed feature is assigned to slots. No more slots can be
configured than specified in the license.
3. Configure the application that uses the licensed feature on the blade in the slot. This operation
verifies that the previous two steps have been successfully completed.
Once these steps are complete, the feature will work on the blade.

Upgrade and downgrade considerations
When a slot-based license is present on the switch, firmware downgrade to pre-Fabric OS v6.3.0 is
allowed, but the slot-based features that were licensed will not be functional.
On upgrade to Fabric OS v7.0.0 or later, any slot-based license that displayed the 10-GbE feature
name in the earlier release now appears as “10 Gigabit FCIP/Fibre Channel (FTR_10G) license.”

Assigning a license to a slot
Use the following procedure to assign a license to a slot.

526

Fabric OS Administrator’s Guide
53-1002920-02

10G licensing

21

1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions in the license class of RBAC commands.
2. Enter the licenseSlotCfg -add command to add the license to the appropriate slot.

Removing a license from a slot
Use the following procedure to remove a slot-based license from a blade slot.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions in the license class of RBAC commands.
2. Deconfigure the application that uses the licensed feature on the blade slot.
3. Enter the licenseSlotCfg -remove command to remove the license from the slot.

10G licensing
The 10 Gbps FCIP/Fibre Channel license (10G license) enables the following features:

• 10 Gbps access on the 16-Gbps FC ports on Brocade 6510 or 6520 switches, and FC16-32
and FC16-48 port blades.

• The two 10-GbE ports on the FX8-24 extension blade.
This 10G license is applied as a slot-based license on the FC16-32 and FC16-48 port blades and
on the FX8-24 extension blade; generic rules for adding slot-based licenses apply, as described in
“Slot-based licensing” on page 526. When this license is applied to a Brocade 6510 or 6520
switch, it is applied to the whole chassis.
Whether you have a bladed (DCX, DCX-4S, DCX 8510-8, or DCX 8510-4) platform or nonbladed
(Brocade 6510 or 6520) switch, you add the 10G license to the chassis using the licenseAdd
command, as for any license.
For the bladed platforms, you can either allow automatic license assignment, or choose the blades
you want the licenses assigned to manually, as for any slot-based license. Automatic assignment is
done sequentially by slot number, beginning with the lowest numbered slot with an enabled blade
that supports this feature (FX8-24, FC16-32, or FC16-48 blade), and that does not already have the
license applied. If the automatic license assignment does not match your needs, you can use the
licenseSlotCfg --remove and licenseSlotCfg --add commands to remove the license manually from
a slot and assign it to a different slot with an FX8-24, FC16-32, or FC16-48 blade.
The same multiple slot-based 10G license can be applied to a mixture of 16-Gbps blades and
FX8-24 blades. For example, if you have a 10G license for two-slot capacity, and you have an
FX8-24 blade in one slot and an FC16-48 blade in a second slot, then the same license can
activate the 10GE ports on the FX8-24 blade and enable 10 Gbps operation on the 10-Gbps FC
ports on the FC16-48 blade.
After applying a 10G license to the Brocade 6510 or 6520 chassis or to a 16-Gbps FC blade, you
must also configure the port octet (portCfgOctetSpeedCombo command) with the correct port octet
speed group and configure each port to operate at 10 Gbps (portCfgSpeed command). It is
necessary to configure the port octet because only certain combinations of port speeds are
allowed within the port octet. No license is required for the octet group. If the speed configuration
operation succeeds and a 10G-capable SFP is inserted in the port connector, the port will allow
operation at 10 Gbps when the link becomes active at that speed.

Fabric OS Administrator’s Guide
53-1002920-02

527

21

10G licensing

Before removing a 10G license from an entire platform (licenseRemove command) or from a
specific blade (licenseSlotCfg --remove command), you must first deconfigure all affected FC ports
to no longer operate at 10 Gbps.

NOTE

An FC port that is operating at 10 Gbps FC speed on a 16-Gbps FC blade or 16-Gbps FC switch does
not need an Extended Fabrics license to be used for FC long distance connectivity.
FC ports licensed and configured to operate at 10 Gbps on a Brocade 6510 or 6520 switch or
16-Gbps FC port blade cannot interoperate with 10-Gbps FC ports on the M-6140 platform or the
FC10-6 blade. The new FC ports use different protocols and physical connections.

Enabling 10 Gbps operation on an FC port
Use the following procedure to enable 10 Gbps operation on an FC port on a Brocade 6510 or
6520 switch or an FC16-32 or FC16-48 blade:
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the license and SwitchPortConfiguration classes of RBAC commands.
2. Use the licenseAdd command to add the 10G license.
3. Use the licenseShow command to verify the license.
Bladed platforms only: If the results of the automatic license assignment are not what you
intended, use the licenseSlotCfg command to reassign the license to the desired blades.
4. Use the portCfgOctetSpeedCombo command to set the combination speed for the port octet to
a setting that supports 10 Gbps operations. Valid settings for 10 Gbps operations include:

• 2—Auto-negotiated or fixed port speeds of 10 Gbps, 8 Gbps, 4 Gbps, and 2 Gbps
• 3—Auto-negotiated or fixed port speeds of 16 Gbps and 10 Gbps
5. Use the portCfgSpeed command to set the port speed on each port you want to operate at 10
Gbps.
Example of assigning a 10G license on an FC port blade and enabling 10 Gbps operation on a port

This example assigns a license to slot 4 on a DCX 8510-8 Backbone and enables 10 Gbps
operation on port 2 of the port blade in that slot. In this example, the 10G license was first
automatically assigned to slot 1.
8510-8switch:admin> licenseadd aTFPNFXGLmABANMGtT4LfSBJSDLWTYD3EFrr4WGAEMBA
8510-8switch:admin> licenseshow
aTFPNFXGLmABANMGtT4LfSBJSDLWTYD3EFrr4WGAEMBA
10 Gigabit FCIP/Fibre Channel (FTR_10G) license
Capacity 1
Consumed 1
Configured Blade Slots 1
8510-8switch:admin> licenseslotcfg -remove FTR_10G 1
8510-8switch:admin> licenseslotcfg -add FTR_10G 4
8510-8switch:admin> licenseshow
aTFPNFXGLmABANMGtT4LfSBJSDLWTYD3EFrr4WGAEMBA
10 Gigabit FCIP/Fibre Channel (FTR_10G) license
Capacity 1
Consumed 1
Configured Blade Slots 4
8510-8switch:admin> portcfgoctetspeedcombo 4/2 2

528

Fabric OS Administrator’s Guide
53-1002920-02

10G licensing

21

8510-8switch:admin> portcfgspeed 4/2 10
8510-8switch:admin>

Example of assigning a 10G license on a Brocade 6510 and enabling 10 Gbps operation on a port

This example assigns a license to a Brocade 6510 switch and enables 10 Gbps operation on
port 2.
6510-switch:admin> licenseadd aTFPNFXGLmABANMGtT4LfSBJSDLWTYD3EFrr4WGAEMBA
6510-switch:admin> licenseshow
aTFPNFXGLmABANMGtT4LfSBJSDLWTYD3EFrr4WGAEMBA
10 Gigabit FCIP/Fibre Channel (FTR_10G) license
Capacity 1
Consumed 1
6510-switch:admin> portcfgoctetspeedcombo 2 2
6510-switch:admin> portcfgspeed 2 10

Enabling the 10-GbE ports on an FX8-24 blade
Use the following procedure to enable the 10-GbE ports on an FX8-24 blade:
1. Connect to the Brocade Backbone and log in using an account with admin permissions, or an
account with OM permissions for the license class of RBAC commands.
2. Use the licenseAdd command to add the 10G license.
3. Use the licenseShow command to check the results of automatic license assignment. If the
results are not what you intended, use the licenseSlotCfg command to reassign the license to
the desired FX8-24 blades.
4. Use the licenseShow command to verify the license.
5. Use the bladeCfgGeMode --set mode command to configure the GbE port mode for the
FX8-24 blade. To enable the 10-GbE ports, set the mode parameter to one of the following:

• 10g—Enables both 10-GbE ports, disables all ten 1-GbE ports.
• dual—Enables the xge0 port (but not xge1) and also enables all ten 1-GbE ports.
Example of assigning a 10G license on an FX8-24 extension blade and enabling both 10-GbE ports

This example assigns a license to slot 7 on a DCX 8510-4 Backbone and enables both 10-GbE
ports on the FX8-24 blade in that slot. In this example, the license was first automatically assigned
to slot 1.
8510-4switch:admin> licenseadd aTFPNFXGLmABANMGtT4LfSBJSDLWTYD3EFrr4WGAEMBA
8510-4switch:admin> licenseshow
aTFPNFXGLmABANMGtT4LfSBJSDLWTYD3EFrr4WGAEMBA
10 Gigabit FCIP/Fibre Channel (FTR_10G) license
Capacity 1
Consumed 1
Configured Blade Slots 1
8510-4switch:admin> licenseslotcfg -remove FTR_10G 1
8510-4switch:admin> licenseslotcfg -add FTR_10G 7
8510-4switch:admin> licenseshow
aTFPNFXGLmABANMGtT4LfSBJSDLWTYD3EFrr4WGAEMBA
10 Gigabit FCIP/Fibre Channel (FTR_10G) license
Capacity 1
Consumed 1
Configured Blade Slots 7
8510-4switch:admin> bladecfggemode --set 10G -slot 7

Fabric OS Administrator’s Guide
53-1002920-02

529

21

Temporary licenses

8510-4switch:admin> switchshow
…
158
7
30
019e00
-159
7
31
019f00
-7 ge0
-7 ge1
-7 ge2
-7 ge3
-7 ge4
-7 ge5
-7 ge6
-7 ge7
-7 ge8
-7 ge9
-7 xge0
-7 xge1
--

-slot 7
--1G
1G
1G
1G
1G
1G
1G
1G
1G
1G
10G
10G

Offline
Offline
No_Module
No_Module
No_Module
No_Module
No_Module
No_Module
No_Module
No_Module
No_Module
No_Module
No_Module
No_Module

VE
VE
FCIP
FCIP
FCIP
FCIP
FCIP
FCIP
FCIP
FCIP
FCIP
FCIP
FCIP
FCIP

Disabled
Disabled
Disabled
Disabled
Disabled
Disabled
Disabled
Disabled
Disabled
Disabled

(10G
(10G
(10G
(10G
(10G
(10G
(10G
(10G
(10G
(10G

Mode)
Mode)
Mode)
Mode)
Mode)
Mode)
Mode)
Mode)
Mode)
Mode)

Temporary licenses
A temporary license applies a “try-before-you-buy” approach to certain features so that you can
experience the feature and its capabilities prior to buying the license. Once you have installed the
license, you are given a time limit to use the feature. A temporary license can be either a regular
temporary license or a universal temporary license.

• A regular temporary license is available on a per-switch basis.
• A universal temporary license can be installed on a switch, but can be applied to multiple
switches.
The following licenses are available as temporary or universal temporary licenses:

•
•
•
•
•
•
•
•
•
•
•
•

530

10 Gigabit FCIP/Fibre Channel license (slot-based)
Advanced Extension license (slot-based)
Advanced FICON Acceleration license (slot-based)
Advanced Performance Monitoring license
Enterprise ICL license
Fabric (E_Port) license
Fabric Watch license
FICON Management Server (CUP) license
Extended Fabrics license
High Performance Extension over FCIP/FC license
Integrated Routing license
ISL Trunking license

Fabric OS Administrator’s Guide
53-1002920-02

Temporary licenses

21

Restrictions on upgrading temporary slot-based licenses
If the capacity of the permanent license is equal to or greater than the capacity of the temporary
license and you use the same slot assignments, then replacing the temporary license with a
permanent license is non-disruptive. If either condition changes, however, then the process is
disruptive.
If the permanent license is for fewer slots than the temporary license, you must do the following:
1. Remove the temporary license. The removal process disables the feature.
2. Install the permanent license on the appropriate slots.
If the permanent license is for different slots than the temporary license, you must do the following:
1. Install the permanent license. The temporary license is automatically replaced on the original
slots.
2. Deconfigure the application that uses the licensed feature on the original slots.
3. Remove the license from the original slots using the licenseSlotCfg -remove command.
4. Add the license to the new slots using the licenseSlotCfg -add command.

Date change restriction
Once the temporary license is installed, you cannot change the time of the switch until the
temporary license is removed. To change the time, you must remove the license, change the date
and time, and then re-install the license on the switch.

CAUTION
If you are using NTP to synchronize the time between your network devices, including switches or
Backbones, then do not attempt to change the system date and time when a temporary license is
installed.

Configupload and download considerations
The configDownload and configUpload commands download the legacy, enhanced, consumed
capacities, and temporary licenses.

Expired licenses
Once a temporary license has expired, you can view it through the licenseShow command. Expired
licenses have an output string of “License has expired”. RASlog warning messages are generated
every hour for licenses present in the database which have expired or are going to expire in the next
five days. An expired license may become unusable after a reboot, failover, firmware download, or a
port or switch disable or enable operation.

Fabric OS Administrator’s Guide
53-1002920-02

531

21

Viewing installed licenses

Removing an expired license
CAUTION
This procedure is disruptive to the switch.
Use the following procedure to remove an expired license.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the reboot command for the expiry to take affect.

Universal temporary licenses
Universal temporary license keys include a duration period. Once installed on a switch, an
expiration date is calculated and the duration is decremented until there is no remaining time, at
which point it is expired. Because of this, universal temporary licenses should not be installed on a
switch until you are ready to use or test the feature, so as not to unnecessarily consume a portion
of the temporary-use duration.
The expiration date is based on the system time at the installation of the license plus the number
of days for which the universal temporary license is valid. Universal temporary licenses cannot be
removed and reinstalled on the same switch.
Universal temporary licenses are always retained in the license database on the product even
though they can be explicitly deleted from any user interface.

Extending a universal temporary license
Extending a universal temporary license is done by adding a temporary license with an expiry date
after the universal temporary license expiry date, or by adding a permanent license. Re-applying an
existing universal temporary license is not allowed.

Universal temporary license shelf life
All universal temporary licenses are encoded with a “shelf life” expiration date. Once this date is
reached, the temporary licensed feature can no longer be used on the switch.

Viewing installed licenses
Use the following procedure to view all installed licenses.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the licenseShow command.

532

Fabric OS Administrator’s Guide
53-1002920-02

Activating a license

21

Activating a license
The transaction key is case-sensitive; it must be entered exactly as it appears in the paperpack.
To lessen the chance of error, copy and paste the transaction key. The quotation marks are
optional.
Use the following procedure to activate a license.
1. Take the appropriate action based on whether you have a license key:

• If you have a license key, go to “Adding a licensed feature”.
• If you do not have a license key and are using a transaction key, launch an Internet
browser and go to the Brocade website at http://www.brocade.com.
2. Select Products > Software License Keys.
The Software License Keys instruction page appears.
3. Enter the requested information in the required fields and click Next.
A verification screen appears.
4. Verify the information appears correctly.
Click Submit if the information displayed is correct. If the information is incorrect, click
Previous, correct the information, and click Submit.
An information screen displays the license keys and you will receive an e-mail with the software
license keys and installation instructions.

Adding a licensed feature
To enable a feature, go to the feature’s appropriate section in this manual. Enabling a feature on a
switch may be a separate task from adding the license.
For the Brocade Backbones, licenses are effective on both control processor (CP) blades, but are
valid only when the CP blade is inserted into a Backbone that has an appropriate license ID stored
in the WWN card. If a CP is moved from one Backbone to another, the license works in the new
Backbone only if the WWN card is the same in the new Backbone. Otherwise, you must transfer
licenses from the old platform to the new platform by obtaining new licenses for the previously
licensed features using the new license ID.
For example, if you swap one CP blade at a time, or replace a single CP blade, then the existing CP
blade (the active CP blade) propagates the licenses to the new CP blade if the WWN card has been
moved to the new platform.
If you move a standby CP from one Backbone to another, then the active CP will propagate its
configuration (including license keys) onto that standby CP.
Use the following procedure to add a licensed feature.
1. Connect to the switch and log in using an account with admin permissions.
2. Activate the license using the licenseAdd command.
3. Verify the license was added by entering the licenseShow command. The licensed features
currently installed on the switch are listed. If the feature is not listed, enter the licenseAdd
command again.

Fabric OS Administrator’s Guide
53-1002920-02

533

21

Removing a licensed feature

Some features may require additional configuration, or you may need to disable and re-enable
the switch to make them operational; see the feature documentation for details.
switch:admin> licenseshow
aAYtMJg7tmMZrTZ9JTWBC4SXWLJMY3QfBJYHG:
Fabric license
Remote Switch license
Remote Fabric license
Extended Fabric license
Entry Fabric license
Fabric Watch license
Performance Monitor license
Trunking license
4 Domain Fabric license
FICON_CUP license
High-Performance Extension over FCIP/FC license
Full Ports on Demand license - additional 16 port upgrade license
2 Domain Fabric license
Integrated Routing license
Storage Application Services license
FICON Tape license
FICON XRC license
Inter Chassis Link license
Enhanced Group Management license
8 Gig FC license
DataFort Compatibility license

Removing a licensed feature
Use the following procedure to remove a licensed feature.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the licenseShow command to display the active licenses.
3. Remove the license key using the licenseRemove command.
The license key is case-sensitive and must be entered exactly as given. The quotation marks
are optional. After removing a license key, the licensed feature is disabled when the switch is
rebooted or when a switch disable and enable is performed.
4. Enter the licenseShow command to verify the license is disabled.
switch:admin> licenseshow
bQebzbRdScRfc0iK:
Entry Fabric license
Fabric Watch license
SybbzQQ9edTzcc0X:
Fabric license
switch:admin> licenseremove "bQebzbRdScRfc0iK"
removing license key "bQebzbRdScRfc0iK"

Entering the licenseShow command after the licenseRemove command displays the remaining
licenses.
switch:admin> licenseshow
SybbzQQ9edTzcc0X:
Fabric license

If there are no license keys, licenseShow displays “No licenses.”

534

Fabric OS Administrator’s Guide
53-1002920-02

Ports on Demand

21

Ports on Demand
The Brocade models in the following list can be purchased with the number of licensed ports
indicated. As your needs increase, you can activate unlicensed ports up to a device-constrained
maximum by purchasing and installing the optional Ports on Demand licensed product.

• Brocade 300—Can be purchased with 8 ports and no E_Port, 8 ports with full fabric access, or
16 ports with full fabric access. A maximum of 16 ports is allowed; 8-port systems can be
upgraded in 4-port increments. An E_Port license upgrade is also available for purchase.

• Brocade 5100—Can be purchased with 24, 32, or 40 licensed ports. A maximum of 40 ports is
allowed.

• Brocade 5300—Can be purchased with 48, 64, or 80 licensed ports. A maximum of 80 ports is
allowed.

• Brocade M6505—Can be purchased with 12 or 24 licensed ports. A maximum of 24 ports is
allowed.

• Brocade 6505—Can be purchased with 12 or 24 licensed ports. A maximum of 24 ports is
allowed.

• Brocade 6510—Can be purchased with 24, 36, or 48 licensed ports. A maximum of 48 ports is
allowed.

• Brocade 6520—Can be purchased with 48, 72, or 96 licensed ports. A maximum of 96 ports is
allowed.

• Brocade 6547—Can be purchased with 12, 24, or 48 licensed ports. A maximum of 48 ports is
allowed.

• Brocade VA-40FC—Can be purchased with 24, 32, or 40 licensed ports. A maximum of 40
ports is allowed.

ATTENTION
Licenses are not interchangeable between units. For example, if you bought a POD license for a
Brocade 300, you cannot use that license on a Brocade 5100 or VA-40FC. The licenses are based
on the switch License Identifiers and are not interchangeable.
Table 86 shows the ports that are enabled by default and the ports that can be enabled after you
install the first and second Ports on Demand licenses for each switch type.

TABLE 86

List of available user ports when implementing PODs

Platform

Available user ports,
No POD license

Available user ports,
POD1 or POD2 present

Available user ports,
Both POD licenses present

Brocade 300

0-7

0-15

0-23

Brocade 5100

0-23

0-31

0-39

Brocade 5300

0-47

0-63

0-79

Brocade 5410

0-11

N/A

N/A

Brocade 5424

1-8 and 17-20

POD1: 0, 9-16, and 21-23

0-23

Brocade 5450

1-10 and 19-22

POD1: 0, 11-18, and 23-25

0-25

Brocade 5480

1-8 and 17-20

POD1: 9-12 and 21-22
POD2: 0, 13-16, and 23

0-23

Brocade M6505

1–8 and 17–20

0–23

N/A

Fabric OS Administrator’s Guide
53-1002920-02

535

21

Ports on Demand

TABLE 86

List of available user ports when implementing PODs (Continued)

Platform

Available user ports,
No POD license

Available user ports,
POD1 or POD2 present

Available user ports,
Both POD licenses present

Brocade 6505

0–11

POD 1: 0–23

N/A

Brocade 6510

0-23

0-35

0-47

Brocade 6520

0–47

0–71

0–95

Brocade 6547

0–8 and 29–31

POD1: 0–14 and 29–37
POD2: 0–8, 15–31, and 39–47

0–47

Brocade VA-40FC

0-23

0-31

0-39

Ports on Demand is ready to be unlocked in the switch firmware. Its license key may be part of the
licensed paperpack supplied with switch software, or you can purchase the license key separately
from your switch vendor. You may need to generate a license key from a transaction key supplied
with your purchase. If so, launch an Internet browser and go to the Brocade website at
http://www.brocade.com. Click Products > Software Products > Software License Keys and follow
the instructions to generate the key.
Each Ports on Demand license activates the next group of ports in numerical order in either 4-port
or 8- or 12-port increments, depending on the model. Before installing a license key, you must
insert transceivers in the ports to be activated. Remember to insert the transceivers in the lowest
group of inactive port numbers first. For example, if only 16 ports are currently active and you are
installing one Ports on Demand license key, make sure to insert the transceivers in ports 16
through 23. If you later install a second license key, insert the transceivers in ports 24 through 31.
For details on inserting transceivers, see the switch’s hardware reference manual.

Displaying installed licenses
If a single license is installed that enables all Ports on Demand, the license will display as
“Full Ports on Demand license - additional X port upgrade license.” If there are other individual
Ports on Demand licenses installed, these will also be displayed when listing the licenses for a
switch, and you will see either “First Ports on Demand license - additional Y port upgrade license”
or “Second Ports on Demand license - additional Z port upgrade license.” In cases where there are
multiple Ports on Demand licenses, the total additional allowed ports will not exceed the total
displayed for the “Full Ports on Demand” license.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the licenseshow command.
switch:admin> licenseshow
SdSSc9SyRSTuTTdz:
First Ports on Demand license - additional 16 port upgrade license
SdSSc9SyRSTeXTdn:
Second Ports on Demand license - additional 16 port upgrade license
SdSSc9SyRSTuXTd3:
Full Ports on Demand license - additional 32 port upgrade license

536

Fabric OS Administrator’s Guide
53-1002920-02

Ports on Demand

21

ATTENTION
If you enable or disable an active port, you will disrupt any traffic and potentially lose data flowing
on that port.
If the port is connected to another switch, you will segment the switch from the fabric and all traffic
flowing between the disabled port and the fabric will be lost.
If you remove a Ports on Demand license, the licensed ports will become disabled after the next
platform reboot or the next port deactivation.

Activating Ports on Demand
Use the following procedure to activate Ports on Demand.
1. Connect to the switch and log in using an account with admin permissions.
2. Verify the current states of the ports using the portShow command.
In the portShow output, the Licensed field indicates whether the port is licensed.
3. Install the Brocade Ports on Demand license.
For instructions on how to install a license, see “Adding a licensed feature” on page 533.
4. Use the portEnable command to enable the ports.
Alternatively, you can disable and re-enable the switch to activate ports.
5. Use the portShow command to check the newly activated ports.

Dynamic Ports on Demand
The Dynamic Ports on Demand (POD) feature automatically assigns POD licenses from a pool of
available licenses based on the server blade or switch installation.
The following platforms support Dynamic POD:

• Switches:
- Brocade 6505
- Brocade 6510
- Brocade 6520
• Embedded switch modules for bladed servers:
- Brocade 5410
- Brocade 5424
- Brocade 5450
- Brocade 5460
- Brocade 5470
- Brocade 5480
- Brocade M6505
- Brocade 6547

Fabric OS Administrator’s Guide
53-1002920-02

537

21

Ports on Demand

For the embedded switch modules, the Dynamic POD feature detects and assigns ports to a POD
license only if the server blade is installed with an HBA present. A server blade that does not have a
functioning HBA is treated as an inactive link during initial POD port assignment. For the non-server
blade switches, the dynamic assignment occurs when an attached Fibre Channel link transitions to
the “link active” state.
The Dynamic POD feature assigns the ports to the POD license as they come online. Typically,
assignments are sequential, starting with the lowest port number. However, variations in the
equipment attached to the ports can cause the ports to take different amounts of time to come
online. This means that the port assignment order is not guaranteed.
If the switch detects more active links than allowed by the current POD licenses, then some ports
will not be assigned a POD license. Ports that do not receive a POD assignment have a state of
No Sync or In Sync; these ports are not allowed to progress to the online state. Ports that cannot be
brought online because of insufficient POD licenses have a state of (No POD License) Disabled. You
can use the switchShow command to display the port states.

Displaying the port license assignments
When you display the available licenses, you can also view the current port assignment of those
licenses.
Use the following procedure to display the port license assignments.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the licensePort --show command.
Example showing manually assigned POD licenses
switch:admin> licenseport --show
24 ports are available in this switch
Full POD license is installed
Dynamic POD method is in use
24 port assignments are provisioned for use in this switch:
12 port assignments are provisioned by the base switch license
12 port assignments are provisioned by a full POD license
24 ports are assigned to installed licenses:
12 ports are assigned to the base switch license
12 ports are assigned to the full POD license
Ports assigned to the base switch license:
1, 2, 3, 4, 5, 6, 7, 8, 17, 18, 19, 20
Ports assigned to the full POD license:
0, 9, 10, 11, 12, 13, 14, 15, 16, 21, 22, 23

Enabling Dynamic Ports on Demand
If the switch is in the static POD mode, then activating the Dynamic POD will erase any prior port
license assignments the next time the switch is rebooted. The static POD assignments become the
initial Dynamic POD assignments. After the Dynamic POD feature is enabled, you can customize
the POD license associations.
Use the following procedure to enable Dynamic Ports on Demand.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the licensePort --method command with the dynamic option to change the license
assignment method to dynamic.

538

Fabric OS Administrator’s Guide
53-1002920-02

Ports on Demand

21

switch:admin> licenseport --method dynamic
The POD method has been changed to dynamic.
Please reboot the switch now for this change to take effect.

3. Enter the reboot command to restart the switch.
switch:admin> reboot

4. Enter the licensePort --show command to verify the switch started the Dynamic POD feature.
switch:admin> licenseport --show
24 ports are available in this switch
Full POD license is installed
Dynamic POD method is in use
24 port assignments are provisioned for use in this switch:
12 port assignments are provisioned by the base switch license
12 port assignments are provisioned by a full POD license
8 ports are assigned to installed licenses:
8 ports are assigned to the base switch license
0 ports are assigned to the full POD license
Ports assigned to the base switch license:
1, 2, 5, 6, 8*, 21, 22, 23
Ports assigned to the full POD license:
None
Ports not assigned to a license:
0, 3, 4, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20
16 license reservations are still available for use by unassigned ports
1 license assignment is held by an offline port (indicated by *)

Disabling Dynamic Ports on Demand
Disabling the Dynamic POD feature changes the POD method to static and erases any prior port
license associations or assignments the next time the switch is rebooted.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the licensePort --method command with the static option to change the license
assignment method to static.
switch:admin> licenseport --method static
The POD method has been changed to static.
Please reboot the switch now for this change to take effect.

3. Enter the reboot command to restart the switch.
4. Enter the licensePort --show command to verify the switch changed to static POD.
switch:admin> licenseport --show
24 ports are available in this switch
Full POD license is installed
Dynamic POD method is in use
24 port assignments are provisioned for use in this switch:
12 port assignments are provisioned by the base switch license
12 port assignments are provisioned by a full POD license
24 ports are assigned to installed licenses:
12 ports are assigned to the base switch license
12 ports are assigned to the full POD license
Ports assigned to the base switch license:
1, 2, 3, 4, 5, 6, 7, 8, 17, 18, 19, 20
Ports assigned to the full POD license:
0, 9, 10, 11, 12, 13, 14, 15, 16, 21, 22, 23

Fabric OS Administrator’s Guide
53-1002920-02

539

21

Ports on Demand

Reserving a port license
You can allocate licenses by reserving and releasing POD assignments to specific ports. Disabled
ports are not candidates for automatic license assignment by the Dynamic POD feature.
Persistently disable an otherwise viable port to prevent it from coming online, and thereby preserve
a license assignment for another port.
Reserving a license for a port assigns a POD license to that port whether the port is online or
offline. That license will not be available to other ports that come online before the specified port.
To allocate licenses to a specific port instead of automatically assigning them as the ports come
online, reserve a license for the port. The port receives a POD assignment if any are available.
Use the following procedure to reserve Dynamic Ports on Demand licenses.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the licensePort --show command to verify there are port reservations available.
switch:admin> licenseport --show
24 ports are available in this switch
Full POD license is installed
Dynamic POD method is in use
24 port assignments are provisioned for use in this switch:
12 port assignments are provisioned by the base switch license
12 port assignments are provisioned by a full POD license
10 ports are assigned to installed licenses:
10 ports are assigned to the base switch license
0 ports are assigned to the full POD license
Ports assigned to the base switch license:
1*, 2*, 3*, 4*, 5*, 6*, 8*, 21, 22, 23
Ports assigned to the full POD license:
None
Ports not assigned to a license:
0, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20

3. Take the following appropriate action based on whether port reservations are available:

• If a port reservation is available, then issue the licensePort --reserve command to reserve
a license for the port.
switch:admin> licenseport -reserve 0

• If all port reservations are assigned, select a port to release its POD license. Follow the
instructions in “Releasing a port from a POD set” to release a port from its POD
assignment. Once the port is released, you can reserve it.

Releasing a port from a POD set
Releasing a port removes it from the POD set; the port then appears as “unassigned” until it comes
back online. Persistently disabling the port ensures that the port cannot come back online and be
automatically assigned to a POD assignment. Before you can re-assign a license, you must disable
the port and release the license.
After a port is assigned to the POD set, the port is licensed until it is manually removed from the
POD port set. When a port is released from its POD port set (Base, Single, or Double), it creates a
vacancy in that port set.

540

Fabric OS Administrator’s Guide
53-1002920-02

Ports on Demand

21

Use the following procedure to release a port from a POD set:
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the switchDisable command to take the switch offline.
switch:admin> switchdisable

3. Enter the switchShow command to verify the switch state is offline.
4. Enter the licensePort --release command to remove the port from the POD license.
switch:admin> licenseport --release 0

5. Enter the licensePort --show command to verify the port is no longer assigned to a POD set.
switch:admin> licenseport --show
24 ports are available in this switch
Full POD license is installed
Dynamic POD method is in use
24 port assignments are provisioned for use in this switch:
12 port assignments are provisioned by the base switch license
12 port assignments are provisioned by a full POD license
10 ports are assigned to installed licenses:
10 ports are assigned to the base switch license
0 ports are assigned to the full POD license
Ports assigned to the base switch license:
1*, 2*, 3*, 4*, 5*, 6*, 8*, 21, 22, 23
Ports assigned to the full POD license:
None
Ports not assigned to a license:
0, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20

6. Enter the switchEnable command to bring the switch back online.
7.

Enter the switchShow command to verify the switch state is now online.

Fabric OS Administrator’s Guide
53-1002920-02

541

21

542

Ports on Demand

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

22

Inter-chassis Links

In this chapter
• Inter-chassis links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• ICLs for the Brocade DCX 8510 Backbone family . . . . . . . . . . . . . . . . . . . .
• ICLs for the Brocade DCX Backbone family . . . . . . . . . . . . . . . . . . . . . . . . .
• Virtual Fabrics considerations for ICLs . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Supported topologies for ICL connections. . . . . . . . . . . . . . . . . . . . . . . . . .

543
544
546
547
547

Inter-chassis links
An inter-chassis link (ICL) is a licensed feature used to interconnect two Brocade DCX or DCX 8510
Backbones. ICL ports in the core blades are used to interconnect the Backbones, potentially
increasing the number of usable ports in the Backbone chassis.
The Brocade Backbones support two types of ICLs:

• The Brocade DCX 8510 Backbone family supports optical ICL QSFPs.
• The Brocade DCX Backbone family supports proprietary copper ICL connectors.
When two Brocade Backbones are interconnected by ICLs, each chassis requires a unique domain
and is managed as a separate switch.

NOTE
You cannot interconnect a Brocade DCX Backbone family chassis to a Brocade DCX 8510 Backbone
family chassis.
The ICL ports appear as regular ports, with some restrictions. All port parameters associated with
ICL ports are static, and all portCfg commands are blocked from changing any of the ICL port
parameters, except for EX_Port and FEC port parameters. The only management associated with
ICL ports and cables is monitoring the status of the LEDs on the ICL ports and any maintenance if
the Attention LED is blinking yellow.
The ICL ports are managed as E_Ports. For the Brocade DCX_8510 Backbone family, you can also
configure EX_Ports on the ICLs. Refer to Chapter 26, “Using FC-FC Routing to Connect Fabrics,” for
instructions.
When you connect two Brocade Backbones, the following features are supported:

• Trunking
• Buffer-to-buffer credit sharing
• QoS

Fabric OS Administrator’s Guide
53-1002920-02

543

22

ICLs for the Brocade DCX 8510 Backbone family

NOTE

A Brocade trunking license is not required for trunking on ICL connections.
Refer to the specific hardware reference manuals for additional information about LED status
meanings and ICL connections, including instructions on how to cable ICLs.

License requirements for ICLs
ICL ports can be used only with an ICL license. An ICL license must be installed on both platforms
forming the ICL connection.
All ICL ports must be disabled and then re-enabled for the license to take effect. After the addition
or removal of an ICL license, the license enforcement is performed on the ICL ports only when you
issue the portDisable and portEnable commands on the switch for the ports or the bladeDisable
and bladeEnable commands for the core blade.
For more information on how license enforcement occurs, refer to Chapter 21, “Administering
Licensing”.

ICLs for the Brocade DCX 8510 Backbone family
Each ICL connects the core blades of two Brocade DCX 8510 chassis and provides up to 64 Gbps
of throughput within a single cable.
You can have up to 32 QSFP ports in a Brocade DCX 8510-8 chassis or 16 QSFP ports in a Brocade
DCX 8510-4 chassis, with up to 2 Tbps ICL bandwidth and support for up to 100 meters on
universal optical cables.
The 100-meter ICL is supported beginning in Fabric OS 7.1.0, when using 100-meter-capable
QSFPs over OM4 cable only.
The Brocade DCX 8510-8 has four port groups on the CR16-8 core blade. The Brocade DCX 8510-4
has two port groups on the CR16-4 core blade. Each port group has four QSFP connectors, and
each QSFP connector maps to four user ports. Refer to the hardware reference manuals for details
about the port groups.
Following are ICL configuration guidelines for trunking bandwidth and High Availability:

• ICLs must be installed in groups of two. Each pair of ICLs must be in the same port group.
• The recommended minimum number of ICLs between two Brocade DCX 8510 chassis is four.
Additional ICLs should be added in increments of two—one on each core blade.

544

Fabric OS Administrator’s Guide
53-1002920-02

ICLs for the Brocade DCX 8510 Backbone family

22

• For High Availability, you should have at least two ICLs from each core blade. Figure 73 shows
two Brocade DCX 8510-8 chassis connected with full redundancy using four ICL connections.

Domain 1
DCX 8510-8

FIGURE 73

Domain 2
DCX 8510-8

Minimum configuration for 64 Gbps ICLs

• The maximum number of ICLs between two Brocade DCX 8510-4 chassis or between a
Brocade DCX 8510-8 and a Brocade DCX 8510-4 is 16.
The maximum number of ICLs between two Brocade DCX 8510-8 chassis is 32.
Because the FSPF routing logic uses only the first 16 paths to come online, only 16 ICLs are
utilized. With Virtual Fabrics, however, you can define two logical switches on the chassis and
have 16 ICLs in each.

NOTE

Brocade recommends that you have a maximum of eight ICLs connected to the same
neighboring domain, with a maximum of four ICLs from each core blade.

• The ICLs can connect to either core blade in the neighboring chassis. Unlike the copper ICLs,
the QSFP ICLs do not need to be cross-connected.

NOTE
QSFP ICLs and ISLs in the same logical switch and connected to the same neighboring switch are
not supported. This is a topology restriction with 16 Gbps ICLs and any ISLs that are E_Ports or
VE_Ports.
If Virtual Fabrics is enabled, you can have ICLs and ISLs between a pair of Brocade DCX 8510
chassis if the ICLs are in a different logical switch than the ISLs.

ICL trunking on the Brocade DCX 8510-8 and DCX 8510-4
ICL trunks form automatically but additional licenses may be required for enabling all ICL ports or
for larger ICL configurations. For more information about ICL licensing options, refer to Chapter 21,
“Administering Licensing”.
Each Quad Small Form-Factor Pluggable (QSFP) cable has four ports, each terminating on a
different application-specific integrated circuit (ASIC). These ports cannot form a trunk with each
other, but can form trunks only with corresponding ports on another QSFP.
Each core blade in the Brocade DCX 8510-8 contains 16 ICL trunk groups. Each core blade in the
Brocade DCX 8510-4 contains 8 ICL trunk groups. Each ICL trunk group contains 4 user ports, one
from each QSFP.

Fabric OS Administrator’s Guide
53-1002920-02

545

22

ICLs for the Brocade DCX Backbone family

To establish ICL trunking between platforms in the Brocade DCX 8510 Backbone family, the QSFP
cables must be in the same trunk group, as illustrated in Figure 73.
Refer to the specific hardware reference manuals for information about port numbering and
connecting the ICL cables.

ICLs for the Brocade DCX Backbone family
The Brocade DCX has two ICL connectors at ports ICL0 and ICL1 on each core blade, each
aggregating a set of 16 ports. Thus, each core blade provides 32 ICL ports and there are 64 ICL
ports available for the entire Brocade DCX chassis. All the ICL connector ports must be connected
to the same two Brocade DCX or DCX-4S chassis.
The Brocade DCX-4S has two ICL connector ports at ICL0 and ICL1, each aggregating a set of 8
ports. Thus, each core blade provides 16 ICL ports and there are 32 ICL ports available for the
entire Brocade DCX-4S chassis. All the ICL connector ports must be connected to the same two
Brocade DCX or DCX-4S chassis.
Only the following cross-ICL group connections are allowed, as illustrated in Figure 74:

• ICL0 ports on the first chassis connect to ICL1 ports on the second chassis.
• ICL1 ports on the first chassis connect to ICL0 ports on the second chassis.

FIGURE 74

DCX-4S allowed ICL connections

The following ICL connections are not allowed:

• ICL0 ports to ICL0 ports
• ICL1 ports to ICL1 ports

546

Fabric OS Administrator’s Guide
53-1002920-02

Virtual Fabrics considerations for ICLs

22

ICL trunking on the Brocade DCX and DCX-4S
ICL trunks form automatically but additional licenses may be required for enabling all ICL ports or
for larger ICL configurations. For more information about ICL licensing options, refer to Chapter 21,
“Administering Licensing”. The ICLs are managed the same as ISL trunks.

• On the Brocade DCX, each ICL is managed as two 8-port ISL trunks.
• On the Brocade DCX-4S, each ICL is managed as one 8-port ISL trunk.
Follow the guidelines in the specific hardware reference manuals for connecting the ICL cables.

Virtual Fabrics considerations for ICLs
In Virtual Fabrics, the ICL ports can be split across the logical switch, base switch, and default
switch. The triangular topology requirement must be met for each fabric individually.
The following restrictions apply:

• ICL ports cannot be in a logical switch that is using XISLs. The “Allow XISL Use” attribute for the
switch must be off.

• All of the user ports in an ICL cable must be in the same logical switch. Distributing the user
ports within the same cable across multiple logical switches is not supported.

Supported topologies for ICL connections
You can connect the Brocade Backbones in a mesh topology and a core-edge topology. A brief
description of each follows. (You can also connect two DCX 8510 chassis point-to-point.)
The illustrations in this section show sample topologies. Refer to the Brocade SAN Scalability
Guidelines for details about maximum topology configurations.

Mesh topology
You can connect the Brocade Backbones in a mesh topology, in which every chassis is connected
to every other chassis.
A simple form of the mesh topology is the triangular topology (shown in Figure 75). The triangular
topology is supported by three Brocade Backbone chassis. The chassis for each topology must all
be from the same family:

• Brocade DCX Backbone family (DCX or DCX-4S)
• Brocade DCX 8510 Backbone family (DCX 8510-8 or DCX 8510-4)

Fabric OS Administrator’s Guide
53-1002920-02

547

22

Supported topologies for ICL connections

FIGURE 75

ICL triangular topology with Brocade DCX 8510-8 chassis

During an ICL break in the triangular topology, the chassis that has the connections of the other
two is the main chassis. Any error messages relating to a break in the topology appear in the
RASlog of the main chassis.
For the Brocade DCX Backbone family only: If one ICL is broken but there is a regular ISL, the
triangular topology holds given that the ISL cost is lower than the total cost through the ICL linear
topology. If a direct ICL link between two chassis is broken, the triangular topology is considered
broken when the ISL path between the two switches is a multiple hop. In this case, the triangular
topology broken message is posted independently of the cost of the ISL path being lesser or
greater than the ICL path between the two switches.
Another form is the full nine-mesh topology shown in Figure 76. This topology is supported by DCX
8510-8 Backbones only. (You can use DCX 8510-4 Backbones for a five-mesh topology.)

548

Fabric OS Administrator’s Guide
53-1002920-02

Supported topologies for ICL connections

FIGURE 76

22

Full nine-mesh topology

Core-edge topology
You can also connect the Brocade DCX 8510 Backbones in a core-edge topology. For example,
Figure 77 shows six chassis connected in a core-edge topology (four edges and two cores).
Although Figure 77 shows only the Brocade DCX 8510-8, each chassis can be either a
Brocade DCX 8510-4 or a DCX 8510-8. You can have up to eight edges with DCX 8510-8 cores or
up to four edges with DCX 8510-4 cores.
Each line in Figure 77 represents four QSFP cables. The cabling scheme should follow the parallel
example shown in Figure 73.

Fabric OS Administrator’s Guide
53-1002920-02

549

22

Supported topologies for ICL connections

FIGURE 77

550

64 Gbps ICL core-edge topology

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

23

Monitoring Fabric Performance

In this chapter
• Advanced Performance Monitoring overview . . . . . . . . . . . . . . . . . . . . . . .
• End-to-end performance monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Frame monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Top Talker monitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Trunk monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Saving and restoring monitor configurations. . . . . . . . . . . . . . . . . . . . . . . .
• Performance data collection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

551
553
558
562
567
567
568

Advanced Performance Monitoring overview
Advanced Performance Monitoring is a licensed feature that provides a comprehensive tool for
monitoring the performance of networked storage resources.
Additional performance monitoring features, such as CRC error reports, are provided through Web
Tools and Brocade Network Advisor. Refer to the Web Tools Administrator’s Guide and Brocade
Network Advisor User Manual for information about monitoring performance using a graphical
interface.
Advanced Performance Monitoring requires either the Brocade Advanced Performance Monitoring
license or the Fabric Vision license.

NOTE

Advanced Performance Monitoring features cannot be used if any Flow Vision flow (active or defined)
is present on the switch.
Advanced Performance Monitoring commands are available only to users with admin permissions.
Use the perfhelp command to display a list of commands associated with Advanced Performance
Monitoring.

NOTE
The command examples in this chapter use the slot/port syntax required by Brocade Backbones.
For fixed-port switches, use only the port number where needed in the commands.

Types of monitors
Advanced Performance Monitoring provides the following monitors:

• End-to-end monitors (EE monitors) measure the traffic between a host and target pair.

Fabric OS Administrator’s Guide
53-1002920-02

551

23

Advanced Performance Monitoring overview

• Frame monitors measure the traffic transmitted through a port with specific values in the first
64 bytes of the frame.

• Top Talker monitors measure the flows that are major consumers of bandwidth on a switch or
port.

Restrictions for installing monitors
• Advanced Performance Monitoring is not supported on VE_Ports and EX_Ports. If you issue
commands for Advanced Performance Monitoring on VE_Ports or EX_Ports, you will receive
error messages.

• All monitor types are allowed only on physical ports.
• Top Talker monitors and EE monitors on E_Ports should be installed only in the ingress
direction.

Virtual Fabrics considerations for Advanced Performance Monitoring
In a fabric with Virtual Fabrics enabled, the number of logical switches that can be configured with
monitors is restricted. Table 87 lists the platforms that support logical switches and, for each
platform, the maximum number of logical switches that can support performance monitors.

TABLE 87

Number of logical switches that support performance monitors

Platform

Maximum number of logical switches supported

Maximum number of logical switches on which
monitors are supported

Brocade DCX
Brocade DCX-4S
Brocade 8510 family

8

4

Brocade 6510

4

4

Brocade 6520

4

4

Brocade 5100
Brocade VA-40FC

3

3

Brocade 5300

4

3

Each logical switch can have its own set of performance monitors. The installation of monitors is
restricted to the ports that are present in the respective logical switch.

• Top Talker monitors and EE monitors are supported on the default logical switch, the base
switch, and user-defined logical switches.

• Frame monitors are not supported on logical ISLs (LISLs) in user-defined logical switches.
If a port is moved from one logical switch to another, the behavior of monitors installed on that port
is as follows:

• Frame monitor: Any frame monitors on the port are deleted. To keep the frame monitor, the
monitor must be manually installed on the port after the move.

• Top Talker (fabric mode): If the fabric mode Top Talkers feature is enabled on the logical switch,
a fabric mode Top Talker monitor is automatically installed on the port after it is moved to the
logical switch.

552

Fabric OS Administrator’s Guide
53-1002920-02

End-to-end performance monitoring

23

• Top Talker (port mode): Any port mode Top Talker monitors on the port are deleted. To keep the
port mode Top Talker monitor, the monitor must be manually installed on the port after the
move.

Access Gateway considerations for Advanced Performance Monitoring
EE monitors and frame monitors are supported on switches in Access Gateway mode. Top Talker
monitors are not supported on these switches.
EE monitors must be installed on F_Ports. Frame monitors can be installed on F_Ports or N_Ports.
Refer to the Access Gateway Administrator’s Guide for additional information.

End-to-end performance monitoring
Use end-to-end (EE) monitoring when you want to monitor throughput between a pair of devices.
End-to-end performance monitoring counts the number of words in Fibre Channel frames for a
specified Source ID (SID) and Destination ID (DID) pair.
To enable EE performance monitoring, you must configure an EE monitor on a port, specifying the
SID-DID pair (in hexadecimal). The monitor counts only those frames with a matching SID and DID.
Each SID or DID has the following three fields:

• Domain ID (DD)
• Area ID (AA)
• AL_PA
For example, the SID 0x118a0f denotes DD 0x11, AA 0x8a, and AL_PA 0x0f.
An EE monitor includes these counts:

• RX_COUNT - Words in frames received at the port
For frames received at the port with the EE monitor installed, the RX_COUNT is updated if the
frame SID is the same as the SID in the monitor and the frame DID is the same as the DID in
the monitor.

• TX_COUNT - Words in frames transmitted from the port
For frames transmitted from the port with the EE monitor installed, TX_COUNT is updated if the
frame DID is the same as the SID in the monitor and the frame SID is the same as the DID in
the monitor.

Maximum number of EE monitors
The maximum number of end-to-end monitors supported varies depending on the switch model:

• The Brocade DCX 8510, 6505, 6510, 6520, M6505, and 6547 models allow up to 512
end-to-end monitors shared by all ports in the same ASIC. Also, these models allow up to 256
end-to-end monitors per port.

• The Brocade DCX, DCX-4S, 5100, VA-40FC, and Brocade Encryption Switch models allow up to
1024 end-to-end monitors shared by all ports in the same ASIC. Also, these models allow up to
256 end-to-end monitors per port.

Fabric OS Administrator’s Guide
53-1002920-02

553

23

End-to-end performance monitoring

• The Brocade 300, 5300, 5410, 5424, 5430, 5450, 5460, 5470, 5480, and 7800 models
allow up to 768 end-to-end monitors shared by all ports in the same ASIC. Also, these models
allow up to 192 end-to-end monitors per port.
The number of interswitch links (ISLs) configured on the switch affects the amount of resources
available for end-to-end monitors.
Virtual Fabrics considerations: If Virtual Fabrics is enabled, the Brocade DCX, DCX-4S, DCX 8510,
and 5300 models allow up to 256 end-to-end monitors on one logical switch. The Brocade 5100,
6510, 6520, and VA-40FC allow up to 341 end-to-end monitors on one logical switch.

Supported port configurations for EE monitors
You can configure EE monitors on F_Ports and, depending on the switch model, on E_Ports. The
following platforms support EE monitors on E_Ports:

•
•
•
•
•
•

Brocade 6505
Brocade 6510
Brocade 6520
Brocade M6505
Brocade 6547
Brocade DCX 8510 family

Identical EE monitors cannot be added to the same port. Two EE monitors are considered identical
if they have the same SID and DID values after applying the end-to-end mask.
An EE monitor and a port Top Talker monitor cannot coexist on the same port.
Coexistence of EE monitors and Top Talker monitors on ports belonging to the same ASIC is not
recommended because the statistics for the same flow going through ports on the same ASIC may
be inaccurate.

Adding EE monitors
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the following command:
perfaddeemonitor [slotnumber/]portnumber sourceID destID

When you add an EE monitor to a port, specify the sourceID and destID in the ingress direction. For
example, Figure 78 shows two devices:

• Host A is connected to domain 1 (0x01), switch area ID 18 (0x12), AL_PA 0x00.
• Dev B is a storage device connected to domain 2 (0x02), switch area ID 30 (0x1e), AL_PA
0x00.
Host A

F_Port
2

E_Port
Domain 1

E_Port
13

3

F_Port
Domain 2

14
DID = 0x021e00

SID = 0x011200
Monitor 1

FIGURE 78

554

Dev B

Monitor 2

Monitor 3

Monitor 4

Setting end-to-end monitors on a port

Fabric OS Administrator’s Guide
53-1002920-02

End-to-end performance monitoring

23

End-to-end performance monitoring looks at traffic on SID and DID pairs in any direction. That is,
even if the SID is for a remote device, the traffic is monitored in both directions (the Tx and Rx
counters are reversed).
Example of monitoring the traffic from Host A to Dev B

On Domain 1, add a monitor to the F_Port, as follows:
switch:admin> perfaddeemonitor 2/2 "0x011200" "0x021e00"

This monitor (Monitor 1) counts the frames that have an SID of 0x011200 and a DID of 0x021e00.
For Monitor 1, RX_COUNT is the number of words from Host A to Dev B, and TX_COUNT is the
number of words from Dev B to Host A.
Example of monitoring the traffic from Dev B to Host A

On Domain 2, add a monitor to the F_Port as follows:
switch:admin> perfaddeemonitor 2/14 "0x021e00" "0x011200"

This monitor (Monitor 4) counts the frames that have an SID of 0x021e00 and a DID of 0x011200.
For Monitor 4, RX_COUNT is the number of words from Dev B to Host A, and TX_COUNT is the
number of words from Host A to Dev B.
The E_Port monitors are configured similar to the F_Port monitors, but the ingress and egress
directions are reversed.
For Monitor 2:
switch:admin> perfaddeemonitor 2/3 "0x021e00" "0x011200"

For Monitor 3:
switch:admin> perfaddeemonitor 2/13 "0x011200" "0x021e00"

Setting a mask for an EE monitor
End-to-end monitors count the number of words in Fibre Channel frames that match a specific SID
and DID pair. If you want to match only part of the SID or DID, you can set a mask on the port to
compare only certain parts of the SID or DID. By default, the frame must match the entire SID and
DID to trigger the monitor. By setting a mask, you can choose to have the frame match only one or
two of the three fields (domain ID, area ID, and AL_PA) to trigger the monitor.
You specify the masks in the form dd:aa:pp, where dd is the domain ID mask, aa is the area ID
mask, and pp is the AL_PA mask. The values for dd, aa, and pp are either ff (the field must match)
or 00 (the field is ignored). The default EE mask value is ff:ff:ff.

NOTE

Only one mask per port can be set. When you set a mask, all existing end-to-end monitors are
deleted.

ATTENTION
End-to-end masks are supported only on the Brocade Encryption Switch.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the perfSetPortEEMask command.
perfsetporteemask [slotnumber/]portnumber "TxSIDMsk" "TxDIDMsk" "RxSIDMsk"
"RxDIDMsk"

Fabric OS Administrator’s Guide
53-1002920-02

555

23

End-to-end performance monitoring

The perfSetPortEEMask command sets the mask for all end-to-end monitors of a port. If any
end-to-end monitors are programmed on a port when the perfSetPortEEMask command is issued,
then a message displays similar to the following example:
switch:admin> perfsetporteemask 1/2, "00:ff:ff"
Changing EE mask for this port will cause ALL EE monitors on this port to be
deleted.
Continue? (yes, y, no, n): [no] y
The EE mask on port 2 is set and EE monitors on this port are deleted

The perfSetPortEEMask command sets a mask for the domain ID, area ID, and AL_PA of the SIDs
and DIDs for frames transmitted from and received by the port.
Figure 79 shows the mask positions in the command. A mask (“ff”) is set on slot 1, port 2 to
compare the AL_PA fields on the SID and DID in all frames (transmitted and received) on port 2.
The frame SID and DID must match only the AL_PA portion of the specified SID and DID pair. Each
port can have only one EE mask. The mask is applied to all end-to-end monitors on the port.
Individual masks for each monitor on the port cannot be specified.
Transmitted from port
SID mask

DID mask

Received by port
SID mask

DID mask

perfsetporteemask 1/2, "00:ff:ff" "00:ff:ff" "00:ff:ff" "00:ff:ff"
AL_PA mask
Area ID mask
Domain ID mask

FIGURE 79

Mask positions for end-to-end monitors

Deleting EE monitors
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the perfMonitorShow command to list the valid end-to-end monitor numbers for a port.
3. Enter the perfDelEEMonitor command to delete a specific monitor.
If you do not specify which monitor number to delete, you are asked if you want to delete all
entries.
Example

The following example displays the end-to-end monitors on port 0 (the monitor numbers are listed
in the KEY column) and deletes monitor number 2 on port 0:
switch:admin> perfmonitorshow --class EE 0
There are 4 end-to-end monitor(s) defined on port

0.

KEY
SID
DID
OWNER_APP
TX_COUNT
RX_COUNT
OWNER_IP_ADDR
-------------------------------------------------------------------------------------0 0x000024 0x000016 WEB_TOOLS
0x0000000000000000 0x0000000000000000 10.106.7.179
1 0x000022 0x000033 WEB_TOOLS
0x0000000000000000 0x0000000000000000 10.106.7.179
2 0x000123 0x000789 WEB_TOOLS
0x0000000000000000 0x0000000000000000 10.106.7.179
3 0x001212 0x003434 WEB_TOOLS
0x0000000000000000 0x0000000000000000 10.106.7.179
switch:admin> perfdeleemonitor 0, 2
End-to-End monitor number 2 deleted

556

Fabric OS Administrator’s Guide
53-1002920-02

End-to-end performance monitoring

23

Displaying EE monitor counters
You can use this procedure display the end-to-end monitors on a specified port. You can display
either the cumulative count of the traffic detected by the monitors or a snapshot of the traffic at
specified intervals.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the perfMonitorShow command.
perfmonitorshow --class monitor_class [slotnumber/]portnumber [interval]

Example of displaying an end-to-end monitor on a port at 10-second intervals
switch:admin> perfMonitorShow --class EE 4/5 10
Showing EE monitors 4/5 10: Tx/Rx are # of bytes
0
1
2
3
4
--------- --------- --------- --------- --------Tx
Rx
Tx
Rx
Tx
Rx
Tx
Rx
Tx
Rx
========= ========= ========= ========= =========
0
0
0
0
0
0
0
0
0
0
53m 4.9m 53m 4.9m 53m 4.9m 53m 4.9m 53m 0
53m 4.4m 53m 4.4m 53m 4.4m 53m 4.4m 53m 0
53m 4.8m 53m 4.8m 53m 4.8m 53m 4.8m 53m 0
53m 4.6m 53m 4.6m 53m 4.6m 53m 4.6m 53m 0
53m 5.0m 53m 5.0m 53m 5.0m 53m 5.0m 53m 0
53m 4.5m 53m 4.5m 53m 4.5m 53m 4.5m 53m 0

Example of displaying EE monitors on a port
switch:admin> perfMonitorShow --class EE 4/5
There are 7 end-to-end monitor(s) defined on port 53.
KEY
SID
DID
OWNER_APP TX_COUNT
RX_COUNT
OWNER_IP_ADDR
-----------------------------------------------------------------------------------------0 0x58e0f 0x1182ef
TELNET
0x0000000000000000 0x0000000000000000
N/A
0 0x21300 0x21dda
TELNET
0x00000004d0ba9915 0x0000000067229e65
N/A
1 0x21300 0x21ddc
TELNET
0x00000004d0baa754 0x0000000067229e65
N/A
2 0x21300 0x21de0
TELNET
0x00000004d0bab3a5 0x0000000067229e87
N/A
3 0x21300 0x21de1
TELNET
0x00000004d0bac1e4 0x0000000067229e87
N/A
4 0x21300 0x21de2
TELNET
0x00000004d0bad086 0x0000000067229e87
N/A
5 0x11000 0x21fd6 WEB_TOOLS 0x00000004d0bade54 0x0000000067229e87 192.168.169.40
6 0x11000 0x21fe0 WEB_TOOLS 0x00000004d0baed41 0x0000000067229e98 192.168.169.40

Clearing EE monitor counters
The following example clears statistics counters for an end-to-end monitor:
switch:admin> perfMonitorClear --class EE 1/2 5
End-to-End monitor number 5 counters are cleared

The following example clears statistics counters for all end-to-end monitors on a specific port:
switch:admin> perfMonitorClear --class EE 1/2
This will clear ALL EE monitors' counters on port 2, continue?
(yes, y, no, n): [no] y

Fabric OS Administrator’s Guide
53-1002920-02

557

23

Frame monitoring

Frame monitoring
Frame monitoring counts the number of times a frame with a particular pattern is transmitted by a
port, and generates alerts when thresholds are crossed. Frame monitoring is achieved by defining
a filter, or frame type, for a particular purpose. The frame type can be a standard type (for example,
an SCSI read command filter that counts the number of SCSI read commands that have been
transmitted by the port) or a user-defined frame type customized for your particular use. For a
complete list of the standard, predefined frame types, refer to the fmMonitor command description
in the Fabric OS Command Reference.
The maximum number of frame monitors and offsets per port depends on the platform. Table 88
shows the maximum number of frame monitors, in any combination of standard and user-defined
frame types, and the maximum number of offsets per port.

TABLE 88

Maximum number of frame monitors and offsets per port

Platform

Maximum number of frame monitors
per port

Maximum number of offsets
per port

Brocade 300, 5300, 5410, 5424, 5450,
5460, 5470, 5480, and 7800

8

131

Brocade 5100, 6505, 6510, 6520, M6505,
6547, VA-40FC, DCX, DCX-4S, DCX 8510, and
Brocade Encryption Switch

12

252

1.

For switches in Access Gateway mode, the maximum number of offsets per port is 7.

2.

For switches in Access Gateway mode, the maximum number of offsets per port is 15.

The actual number of frame monitors that can be configured on a port depends on the complexity
of the frame types. For trunked ports, the frame monitor is configured on the trunk master.
Static offsets are preset with offset and value combinations. Brocade also supports additional
dynamic offsets. When a user-specified offset and value combination matches that already
allocated by a Brocade application, resources are shared. The maximum number of configurable
offsets by fmMonitor may increase or decrease as a result of resource sharing.
Virtual Fabrics considerations: Frame monitors are not supported on logical ISLs (LISLs), but are
supported on ISLs and extended ISLs (XISLs).

License requirements for frame monitoring
The Fabric Vision license provides full monitoring capability. If you have this license, you do not
need the Advanced Performance Monitoring or the Fabric Watch licenses.
If you do not have the Fabric Vision license, you need the following licenses:

• Advanced Performance Monitoring license is required to use the fmMonitor command.
• Fabric Watch license is required to use the monitoring functionality through Fabric Watch.
When you configure actions and alerts through the fmMonitor command, Fabric Watch uses these
values and generates alerts based on the configuration. If you do not have a Fabric Watch or Fabric
Vision license, these values are ignored. Refer to the Fabric Watch Administrator’s Guide for more
information about using Fabric Watch.

558

Fabric OS Administrator’s Guide
53-1002920-02

Frame monitoring

23

Creating frame types to be monitored
In addition to the standard frame types, you can create custom frame types to gather statistics that
fit your needs. To define a custom frame type, you must specify a series of offsets, bitmasks, and
values. For all transmitted frames, the switch performs the following tasks:

•
•
•
•

Locates the byte found in the frame at the specified offset.
Applies the bitmask to the byte found in the frame.
Compares the new value with the given value.
Increments the filter counter if a match is found.

You can specify up to four values to compare against each offset. If more than one offset is
required to properly define a filter, the bytes found at each offset must match one of the given
values for the filter to increment its counter. If one or more of the given offsets does not match any
of the given values, the counter does not increment.
The value of the offset must be between 0 and 63, in decimal format. Byte 0 indicates the first byte
of the Start of Frame (SOF), byte 4 is the first byte of the frame header, and byte 28 is the first byte
of the payload. Thus, only the SOF, frame header, and first 36 bytes of payload can be selected as
part of a filter definition. Offset 0 is a special case, which can be used to monitor the first 4 bytes of
the frame (SOF). When the offset is set to 0, the values 0 through 7 that are checked against that
offset are predefined, as shown in Table 89.

TABLE 89

Predefined values at offset 0

Value

SOF

0

SOFf

1

SOFc1

2

SOFi1

3

SOFn1

4

SOFi2

5

SOFn2

6

SOFi3

7

SOFn3

Creating a frame monitor
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fmMonitor --create command to create a user-defined frame.
Complete details of the fmMonitor command parameters are provided in the Fabric OS Command
Reference. The highth and action options set values and actions for Fabric Watch, but do not apply
to monitoring. To apply the custom values, use the thconfig --apply command. Refer to the Fabric
Watch Administrator’s Guide for more information about using this command.
Example of creating a user-defined frame type
switch:admin> fmmonitor --create myframemonitor -pat
"17,0xFF,0x07;7,0x4F,0x01;" -action email

Fabric OS Administrator’s Guide
53-1002920-02

559

23

Frame monitoring

Example of creating a user-defined frame type and applying frame monitors to ports 3, 4, and 5
switch:admin> fmmonitor --create myframemonitor -pat
"17,0xFF,0x007;7,0x4F,0x01;" -port 3-5

Deleting frame types
Deleting a frame type removes the entire configuration, including configured thresholds and
associated actions. It also removes any frame monitors of the specified type from all ports.
You can delete only user-defined frame types; you cannot delete the predefined frame types.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fmMonitor --delete command to delete a specific frame type.
Example
switch:admin> fmmonitor --delete myframemonitor

Adding frame monitors to a port
If the switch does not have enough resources to add a frame monitor to a port, then other frame
monitors on that port may have to be deleted to free resources.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fmMonitor --addmonitor command to add a frame monitor to one or more ports.
The set of ports to be removed from monitoring is automatically saved to the persistent
configuration unless you specify the -nosave option on the command.
Example

The following example adds a standard SCSI frame type monitor to ports 3 through 12.
switch:admin> fmmonitor --addmonitor SCSI -port 3-12

Removing frame monitors from a port
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fmMonitor --delmonitor command to remove a specific monitor from one or more
ports.
The set of ports to be removed from monitoring is automatically saved to the persistent
configuration unless you specify the -nosave option on the command.
Example

The following example removes the user-defined frame monitor, myframemonitor, from all ports.
switch:admin> fmmonitor --delmonitor myframemonitor

Saving a frame monitor configuration
When you assign or remove frame monitors on ports, the list of ports to be monitored is
automatically saved persistently, unless you specified the -nosave option.

560

Fabric OS Administrator’s Guide
53-1002920-02

Frame monitoring

23

1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fmMonitor --save command to save the set of ports on which the frame type is
monitored to the persistent configuration.
Example

In the following example, the first command adds a standard SCSI frame type monitor to ports 3
through 12, but does not save the port configuration. The second command saves the port
configuration persistently.
switch:admin> fmmonitor --addmonitor SCSI -port 3-12 -nosave
switch:admin> fmmonitor --save SCSI

Displaying frame monitors
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fmMonitor --show command.
Example

The following example displays the existing frame types and associated bit patterns on the switch.
switch:admin> fmmonitor --show
FRAME_TYPE
BIT PATTERN
---------------------------------------scsi
12,0xFF,0x08;
scsiread
12,0xFF,0x08;4,0xFF,0x06;40,0xFF,0x08,0x28;
scsiwrite
12,0xFF,0x08;4,0xFF,0x06;40,0xFF,0x08,0x28,0x0A,0x2A;
scsirw
12,0xFF,0x08;4,0xFF,0x06;40,0xFF,0x08,0x28,0x0A,0x2A;
scsi2reserve
12,0xFF,0x08;4,0xFF,0x06;40,0xFF,0x16,0x56;
scsi3reserve
12,0xFF,0x08;4,0xFF,0x06;40,0xFF,0x5F;41,0xFF,0x01
ip
12,0xFF,0x05;
abts
4,0xFF,0x81;40,0xFF,0x81;12,0xFF,0x0;17,0xFF,0x0;
baacc
4,0xff,0x84;12,0xff,0x00;17,0xff,00;

The following example displays configuration details for the predefined SCSI frame monitor. Notice
that in the last entry, the “-” in the Count column indicates that the monitor is configured, but is not
installed on the port.
switch:admin> fmmonitor --show SCSI
Port|Frame Type |Count
|HIGH Thres|Actions
|TIMEBASE |CFG
----------------------------------------------------------------------------000001|scsi
|0x0000000000000123|1000
|Email
|None
|saved
000002|scsi
|0x0000000000000125|1000
|Email
|None
|saved
000003|scsi
|0x0000000000000143|1000
|Email
|None
|saved
000022|scsi
||0
|None
|None
|saved

The following example displays values for the predefined SCSI frame monitor on port 5 every 5
seconds.
switch:admin> fmmonitor --show scsi -port 5 -timeinterval 5
Port|Count
|
---------------------2011-03-21 00:59:50
000005|

Fabric OS Administrator’s Guide
53-1002920-02

48.3k

561

23

Top Talker monitors

2011-03-21 00:59:55
000005|

48.6k

(output truncated)

Clearing frame monitor counters
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the fmMonitor --clear command to clear the counters on the ports on which the
specified frame type is monitored.
Example

The following example clears the counters for the ABTS monitor from ports 7 through 10.
switch:admin> fmmonitor --clear ABTS -port 7-10

Top Talker monitors
Top Talker monitors determine the flows (SID and DID pairs) that are the major users of bandwidth
(after initial stabilization). Top Talker monitors measure bandwidth usage data in real time and
relative to the port on which the monitor is installed.

NOTE

Initial stabilization is the time taken by a flow to reach the maximum bandwidth. This time varies
depending on the number of flows in the fabric and other factors. This time can be up to 14 seconds
in the Backbones, and up to 82 seconds in the fixed-port switches.
Applications can use Top Talker monitors data to do the following:

• Re-route the traffic through different ports that are less busy, so as not to overload a given
port.

• Alert you to the major users of bandwidth (top talking flows) on a port if the total traffic on the
port exceeds the acceptable bandwidth consumption.
You can use Top Talker monitors to identify the SID and DID pairs that consume the most
bandwidth and can then configure them with certain Quality of Service (QoS) attributes so they get
proper priority. Refer to Chapter 14, “Optimizing Fabric Behavior,” for information on QoS.
The Top Talker monitors are based on SID and DID pairs and not WWNs. Once Top Talker monitors
are installed on a switch or port, they remains installed across power cycles.
Top Talker monitors support two modes, port mode and fabric mode:

• Port mode Top Talker monitor
A Top Talker monitor can be installed on a port to measure the traffic originating from the port
and flowing to different destinations.

562

Fabric OS Administrator’s Guide
53-1002920-02

Top Talker monitors

23

You can configure Top Talker monitors on F_Ports and, depending on the switch model, on
E_Ports. The following platforms support Top Talker monitors on E_Ports:

-

Brocade 6505
Brocade 6510
Brocade 6520
Brocade M6505
Brocade 6547
Brocade DCX 8510 family

• Fabric mode Top Talker monitor
In fabric mode, Top Talker monitors are installed on all E_Ports in the fabric and measure the
data rate of all the possible flows in the fabric (ingress E_Port traffic only). In fabric mode, Top
Talker monitors can determine the top n bandwidth users on a given switch.
You can install Top Talker monitors in either port mode or fabric mode, but not both.

ATTENTION
A fabric mode Top Talker monitor and an EE monitor cannot be configured on the same fabric. You
must delete the EE monitor before you configure the fabric mode Top Talker monitor.
How do Top Talker monitors differ from EE monitors? EE monitors provide counter statistics for
traffic flowing between a given SID and DID pair. Top Talker monitors identify all possible SID and
DID flow combinations that are possible on a given port and provide a sorted output of the top
talking flows. Also, if the number of flows exceeds the hardware resources, existing EE monitors fail
to get real-time data for all of them; however, Top Talker monitors can monitor all flows for a given
E_Port or F_Port.
Virtual Fabric considerations: All logical switches in the same chassis can use either fabric mode
Top Talker monitors or port mode Top Talker monitors and EE monitors. You cannot use fabric mode
Top Talker monitors and EE monitors together on the same logical switch.
Admin Domain considerations: Top Talker monitors are always installed in AD255.
NPIV considerations: Top Talker monitors take NPIV devices into consideration when calculating the
top talking flows.
Top Talker monitors are not supported on the embedded platforms: Brocade 5410, 5424, 5450,
5460, 5470, and 5480.

Top Talker monitors and FC-FC routing
You can enable Top Talker monitors on a platform that is configured to be an FC router. Top Talker
monitors and FC routers are concurrently supported on the following platforms:

•
•
•
•

Brocade 6505
Brocade 6510
Brocade 6520
Brocade DCX 8510 Backbone family, with the following blades only: FC16-32, FC16-48

On all other platforms, you can have either Top Talker monitors or FC-FC routing, but not both.
Top Talker monitors are supported on an FC router in both backbone-to-edge and edge-to-edge
configurations.

Fabric OS Administrator’s Guide
53-1002920-02

563

23

Top Talker monitors

Note the following restrictions:

• An E_Port-attached switch must be connected and merged with the backbone FC router before
you can enable Top Talker monitors on the FC router.

• Fabric mode Top Talker monitors do not support requests for domains (either front port domain
or xlate domain).

• Fabric mode Top Talker monitors do not monitor flows over EX_Ports.
For example, if a host is connected directly to an FC router and the target is on the edge switch
(refer to Figure 80 on page 564), no flows are monitored because none of the flows traverse
an E_Port on the FC router.
In Figure 81 on page 564, however, the flows across the E_Port on the FC router are
monitored.

Edge fabric

E_Port

FC router

EX_Port

Backbone fabric

FIGURE 80

Fabric mode Top Talker monitors on FC router do not monitor any flows

Edge fabric

E_Port
E_Port
E_Port

FC router

EX_Port

Backbone fabric

FIGURE 81

564

Fabric mode Top Talker monitors on FC router monitor flows over the E_Port

Fabric OS Administrator’s Guide
53-1002920-02

Top Talker monitors

23

Limitations of Top Talker monitors
Be aware of the following when using Top Talker monitors:

•
•
•
•
•

Top Talker monitors cannot detect transient surges in traffic through a given flow.
You cannot install a Top Talker monitor on a mirrored port.
Top Talker monitors can monitor only 10,000 flows at a time.
Top Talker monitors are not supported on VE_Ports, EX_Ports, and VEX_Ports.
The maximum number of all port mode Top Talker monitors on an ASIC is 16. If Virtual Fabrics
is enabled, the maximum number of all port mode Top Talker monitors on an ASIC is 8.

• If the ingress and egress monitor ports are configured on the same ASIC, port mode Top Talker
monitors on F_Ports show the flow from only one of the ports, either the ingress or the egress
port, but not both.

Adding a Top Talker monitor to a port (port mode)
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the perfTTmon --add command.
perfttmon --add [egress | ingress] [slotnumber/]port

The following example monitors the incoming traffic on port 7.
perfttmon --add ingress 7

The following example monitors the outgoing traffic on slot 2, port 4 on a Backbone.
perfttmon --add egress 2/4

Adding Top Talker monitors on all switches in the fabric (fabric mode)
When fabric mode is enabled, you can no longer install Top Talker monitors on an F_Port unless you
disable fabric mode.
1. Connect to the switch and log in using an account with admin permissions.
2. Remove any EE monitors in the fabric, as described in “Deleting EE monitors” on page 556.
Fabric mode Top Talker monitors and EE monitors cannot both exist in the fabric.
3. Enter the perfTTmon --add fabricmode command.
perfttmon --add fabricmode

The system responds with the following message:
Before enabling fabric mode, please remove all EE monitors in the fabric
continue? (yes, y, no, n):

4. Enter y at the prompt to continue.
Top Talker monitors are added to E_Ports in the fabric and fabric mode is enabled. Any Top
Talker monitors that were already installed on F_Ports are automatically uninstalled.
If EE monitors are present on the local switch, the command fails with the following message:
Cannot install Fabric Mode Top Talker because EE monitor is already present

Fabric OS Administrator’s Guide
53-1002920-02

565

23

Top Talker monitors

If EE monitors are present on remote switches, the command succeeds; however, on the
remote switches, fabric mode fails and a RASlog message is displayed on those switches.
If a new switch joins the fabric, you must run the perfTTmon --add fabricmode command on
that switch. The Top Talker monitor configuration information is not automatically propagated
to the new switch.

Displaying the top n bandwidth-using flows on a port (port mode)
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the perfTTmon --show command.
perfttmon --show [slotnumber/]port [n] [wwn | pid]

The output is sorted based on the data rate of each flow. If you do not specify the number of
flows to display, then the command displays the top 8 flows or the total number of flows,
whichever is less.
The following example displays the top 5 flows on port 7 in WWN (default) format:
perfttmon --show 7 5

The following example displays the top flows on slot 2, port 4 on a Backbone in PID format:
switch:admin> perfttmon --show 2/4 pid
========================================
Src_PID
Dst_PID
MB/sec
========================================
0xa90800
0xa05200
6.926
0xa90800
0xa908ef
6.872

Displaying top talking flows for a given domain ID (fabric mode)
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the perfTTmon --show dom command.
perfttmon --show dom domainid [n] [wwn | pid]

Fabric mode must be enabled for this option.
The output is sorted based on the data rate of each flow. If you do not specify the number of
flows to display, then the command displays the top 8 flows or the total number of flows,
whichever is less. The command can display a maximum of 32 flows.
The following example displays the top 5 flows on domain 1 in WWN (default) format:
perfttmon --show dom 1 5

The following example displays the top flows on domain 2 in PID format:
switch:admin> perfttmon --show dom 2 pid
=================================================================
Src_PID
Dst_PID
MB/sec
Potential E-Ports
=================================================================
0x03f600
0x011300
121.748
2/0,2/2,2/3
0x03f600
0x011300
121.748
3/14,3/15

566

Fabric OS Administrator’s Guide
53-1002920-02

Trunk monitoring

23

Deleting a Top Talker monitor on a port (port mode)
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the perfTTmon --delete command.
perfttmon --delete [slotnumber/]port

The following example deletes the monitor on port 7:
perfttmon --delete 7

The following example deletes the monitor on slot 2, port 4 on a Backbone:
perfttmon --delete 2/4

Deleting all fabric mode Top Talker monitors
1. Connect to the switch and log in using an account with admin permissions.
2. Enter the perfTTmon --delete fabricmode command.
perfttmon --delete fabricmode

All Top Talker monitors are deleted.

Trunk monitoring
To monitor E_Port (ISL) and F_Port trunks, you can set monitors only on the master port of the
trunk. If the master changes, the monitor automatically moves to the new master port.
If a monitor is installed on a port that later becomes a slave port when a trunk comes up, the
monitor automatically moves to the master port of the trunk.

Trunk monitoring considerations
• End-to-end (EE) monitors are supported for ISLs only on the Brocade 6505, 6510, 6520,
M6505, 6547, and DCX 8510 family.

• If an EE monitor is installed on a trunk group and you disable the trunk, the EE monitor will be
installed only on the last master port of that trunk group, which may not be the actual port on
which the EE monitor was installed when the trunk was enabled.

• For F_Port trunks, end-to-end masks are allowed only on the F_Port trunk master. Unlike the
monitors, if the master changes, the mask does not automatically move to the new master
port.

• All platforms support 12 frame monitors for trunks, except for the Brocade 300, which
supports 8 frame monitors for trunks.

Saving and restoring monitor configurations
To prevent the switch configuration flash from running out of memory, the number of monitors
saved to flash memory is limited as follows:

• The total number of EE monitors per port is limited to 16.
Fabric OS Administrator’s Guide
53-1002920-02

567

23

Performance data collection

• The total number of frame monitors per port is limited to 16.
• The total number of monitors per switch is limited to 512.
When there are more than 512 monitors in the system, monitors are saved to flash memory in the
following order:

• The EE monitors for each port (from 0 to MAX_PORT)
• The frame monitors for each port
EE monitors get preference saving to flash memory when the total number of monitors in a switch
exceeds 512. If the total number of monitors per port or switch exceeds the limit, then you will
receive an error message indicating the count has been exceeded and that some monitors have
been discarded.
1. Connect to the switch and log in using an account with admin permissions.
2. Enter one of the following commands, depending on the action you want to perform:

• To save the current EE monitor and frame monitor configuration settings into nonvolatile
memory, use the perfCfgSave command.
switch:admin> perfcfgsave
This will overwrite previously saved Performance Monitoring
settings in FLASH. Do you want to continue? (yes, y, no, n): [no] y
Please wait ...
Performance monitoring configuration saved in FLASH.

• To restore a saved monitor configuration, use the perfCfgRestore command, for example,
to restore the original performance monitor configuration after making several changes.
switch:admin> perfcfgrestore
This will overwrite current Performance Monitoring settings in RAM. Do you
want to continue? (yes, y, no, n): [no] y
Please wait... Performance monitoring configuration restored from FLASH
ROM.

• To clear the previously saved performance monitoring configuration settings from
nonvolatile memory, use the perfCfgClear command.
switch:admin> perfcfgclear
This will clear Performance Monitoring settings in FLASH. The RAM settings
won’t change. Do you want to continue? (yes, y, no, n): [no] y
Please wait... Committing configuration...done.
Performance Monitoring configuration cleared from FLASH.

Performance data collection
Data collected through Advanced Performance Monitoring is deleted when the switch is rebooted.
Using the Brocade Network Advisor Enterprise Edition, you can store performance data persistently.
For details on this feature, refer to the Brocade Network Advisor User Manual.

568

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

24

Managing Trunking Connections

In this chapter
• Trunking overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Supported platforms for trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Supported configurations for trunking. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Requirements for trunk groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Recommendations for trunk groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Configuring trunk groups. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Enabling trunking. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Disabling trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Displaying trunking information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Trunk Area and Admin Domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• ISL trunking over long-distance fabrics . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• EX_Port trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• F_Port trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Displaying F_Port trunking information . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Disabling F_Port trunking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Enabling the DCC policy on a trunk area . . . . . . . . . . . . . . . . . . . . . . . . . . .

569
571
571
572
572
573
574
574
574
576
576
577
579
585
585
586

Trunking overview
Trunking optimizes the use of bandwidth by allowing a group of links to merge into a single logical
link, called a trunk group. Traffic is distributed dynamically and in order over this trunk group,
achieving greater performance with fewer links. Within the trunk group, multiple physical ports
appear as a single port, thus simplifying management. Trunking also improves system reliability by
maintaining in-order delivery of data and avoiding I/O retries if one link within the trunk group fails.
Trunking is frame-based instead of exchange-based. Because a frame is much smaller than an
exchange, this means that frame-based trunks are more granular and better balanced than
exchange-based trunks and provide maximum utilization of links.
The Trunking license is required for any type of trunking, and must be installed on each switch that
participates in trunking. For details on obtaining and installing licensed features, refer to Chapter
21, “Administering Licensing”.

Fabric OS Administrator’s Guide
53-1002920-02

569

24

Trunking overview

Types of trunking
Trunking can be between two switches, between a switch and an Access Gateway module, or
between a switch and a Brocade adapter. The types of trunking are as follows:

• ISL trunking, or E_Port trunking, is configured on an inter-switch link (ISL) between two
Fabric OS switches and is applicable only to E_Ports.

• ICL trunking is configured on an inter-chassis link (ICL) between two Brocade DCX or DCX 8510
Backbones and is applicable only to ports on the core blades.
Refer to Chapter 22, “Inter-chassis Links” for detailed information about ICL trunking.

• EX_Port trunking is configured on an inter-fabric link (IFL) between an FC router (EX_Port) and
an edge fabric (E_Port). The trunk ports are EX_Ports connected to E_Ports.
Refer to “EX_Port frame trunking configuration” on page 619 for additional information about
EX_Port trunking.

• F_Port trunking is configured on a link between a switch and either an Access Gateway module
or a Brocade adapter. The trunk ports are F_Ports (on the switch) connected to N_Ports (on the
Access Gateway or adapter).

• N_Port trunking is configured on a link between a switch and either an Access Gateway module
or a Brocade adapter. It is similar to F_Port trunking. The trunk ports are N_Ports (on the
Access Gateway or adapter) connected to F_Ports (on the switch).
For more information, refer to “Configuring F_Port trunking for a Brocade adapter” on
page 581, the Access Gateway Administrator’s Guide, and the Brocade Adapters
Administrators Guide.

NOTE

This chapter uses the term F_Port trunking to refer to a trunk between the F_Ports on a switch and
the N_Ports on either an Access Gateway module or a Brocade adapter. This type of trunk might be
referred to as N_Port trunking in the Access Gateway Administrator’s Guide or Brocade Adapters
Administrator’s Guide.

Masterless trunking
Masterless trunking means that if the master port goes offline, one of the slave ports automatically
becomes the new master port, thus avoiding traffic disruption. The new master port uses the old
master port area and the old master port is assigned a new, unused area. In this way, the port
identifier (PID) of the trunk does not change if the master port goes offline.
If trunking is not masterless, and if the master port goes offline, traffic disruption can occur
because the slave ports in the trunk group go offline to select the new master port and then come
back online.
Masterless trunking is supported for most platforms and trunking types:

• All F_Port trunking is masterless.
• ISL and ICL trunking are masterless.
• EX_Port trunking is masterless, except on Brocade DCX or DCX 8510 Backbones with Virtual
Fabrics disabled.

570

Fabric OS Administrator’s Guide
53-1002920-02

Supported platforms for trunking

24

License requirements for trunking
Trunking of non-ICL ports (E_Ports, EX_Ports, and F_Ports) requires the Trunking license. This
license must be installed on each switch that participates in trunking.
Trunking of ICL ports (E_Ports and EX_Ports) does not require a Trunking license.

ATTENTION
After you add the Trunking license, to enable trunking functionality, you must disable and then
re-enable each port to be used in trunking, or disable and re-enable the switch.
Refer to Chapter 21, “Administering Licensing,” for information about activating licenses.

Port groups for trunking
For trunk groups to form, several conditions must be met. One of the conditions is that all of the
ports in a trunk group must belong to the same port group. A port group is a group of eight ports,
based on the user port number, such as 0–7, 8–15, 16–23, and up to the number of ports on the
switch. The maximum number of port groups is platform-specific.
Figure 82 shows the port groups for the Brocade 5100.
Ports in a port group are usually contiguous, but they may not be. Refer to the hardware reference
manual for your switch for information about which ports can be used in the same port group for
trunking.

FIGURE 82

Port group configuration for the Brocade 5100

Supported platforms for trunking
Trunking is supported on the FC ports of all Brocade platforms and blades supported in Fabric OS
v7.0.0 and later.
EX_Port trunking is supported only on those platforms that support EX_Ports. Refer to “Supported
platforms for FC-FC routing” on page 594 for more information.

Supported configurations for trunking
• Trunk links can be 2 Gbps, 4 Gbps, 8 Gbps, 10 Gbps, or 16 Gbps, depending on the Brocade
platform.

• The maximum number of ports per trunk and trunks per switch depends on the Brocade
platform.

• You can have up to eight ports in one trunk group to create high-performance ISL trunks
between switches, providing up to 128 Gbps (based on a 16-Gbps port speed).

Fabric OS Administrator’s Guide
53-1002920-02

571

24

Requirements for trunk groups

• If in-flight encryption or compression is enabled, you can have a maximum of only two ports per
trunk.

• An E_Port or EX_Port trunk can be up to eight ports wide. All the ports must be adjacent to
each other, in the clearly marked groups on the front of the switch.
Trunks operate best when the cable length of each trunked link is roughly equal to the length of the
others in the trunk. For optimal performance, no more than 30 meters difference is recommended.
Trunks are compatible with both short-wavelength (SWL) and long-wavelength (LWL) fiber-optic
cables and transceivers.
Trunking is performed according to the Quality of Service (QoS) configuration on the master and the
slave ports. That is, in a given trunk group, if there are some ports with QoS enabled and some with
QoS disabled, they form two different trunks, one with QoS enabled and the other with QoS
disabled. For more information on QoS, refer to “QoS zones” on page 419.

High Availability support for trunking
Trunking is a High Availability (HA) supported feature. The HA protocol for trunking is as follows:

• If trunking is disabled prior to the HA failover, it remains disabled after the HA failover.
• If trunking is enabled prior to the HA failover, it remains enabled after the HA failover.

Requirements for trunk groups
The following requirements apply to all types of trunking:

• All of the ports in a trunk group must belong to the same port group.
• All of the ports in a trunk group must meet the following conditions:
- They must be running at the same speed.
- They must be configured for the same distance.
- They must have the same encryption, compression, QoS, and FEC settings.
• Trunk groups must be between Brocade switches (or Brocade adapters in the case of F_Port
trunking). Brocade trunking is proprietary and is not supported on M-EOS or third-party switches.

• There must be a direct connection between participating switches.
• Trunking cannot be done if ports are in ISL R_RDY mode. (You can disable this mode by using
the portCfgIslMode command.)

• Trunking is supported only on FC ports. Virtual FC ports (VE_Ports or VEX_Ports) do not support
trunking.

Recommendations for trunk groups
To identify the most useful trunk groups, consider the following recommendations along with the
standard guidelines for SAN design:

• Evaluate the traffic patterns within the fabric.

572

Fabric OS Administrator’s Guide
53-1002920-02

Configuring trunk groups

24

• Place trunking-capable switches adjacent to each other.
This maximizes the number of trunk groups that can form. If you are using a core and edge
topology, place trunking-capable switches at the core of the fabric and any switches that are
not trunking-capable at the edge of the fabric.

• When connecting two switches with two or more ISLs, ensure that all trunking requirements
are met to allow a trunk group to form.

• Determine the optimal number of trunk groups between each set of linked switches,
depending on traffic patterns and port availability.
The goal is to avoid traffic congestion without unnecessarily using ports that could be used to
attach other switches or devices. Consider these points:

-

Each physical ISL uses two ports that could otherwise be used to attach node devices or
other switches.

-

Trunk groups can be used to resolve ISL oversubscription if the total capability of the trunk
group is not exceeded.

• Consider how the addition of a new path will affect existing traffic patterns:
- A trunk group has the same link cost as the master ISL of the group, regardless of the
number of ISLs in the group. This allows slave ISLs to be added or removed without
causing data to be rerouted, because the link cost remains constant.

-

The addition of a path that is shorter than existing paths causes traffic to be rerouted
through that path.

-

The addition of a path that is longer than existing paths may not be useful, because the
traffic will choose the shorter paths first.

• Plan for future bandwidth addition to accommodate increased traffic.
For trunk groups over which traffic is likely to increase as business requirements grow,
consider leaving one or two ports in the group available for the future nondisruptive addition of
bandwidth.

• Consider creating redundant trunk groups where additional ports are available or paths are
particularly critical.
This helps to protect against oversubscription of trunk groups, multiple ISL failures in the same
group, and the rare occurrence of an ASIC failure.

• To provide the highest level of reliability, deploy trunk groups in redundant fabrics to help
ensure that ISL failures do not disrupt business operations.

Configuring trunk groups
After you install the Trunking license, you must re-initialize the ports that are to be used in trunk
groups so that they recognize that trunking is enabled. This procedure needs to be performed only
once, and is required for all types of trunking.
To re-initialize the ports, you can either disable and then re-enable the switch, or disable and then
re-enable the affected ports.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the islShow command to determine which ports are used for ISLs.

Fabric OS Administrator’s Guide
53-1002920-02

573

24

Enabling trunking

3. Enter the portDisable command for each port to be used in a trunk group.
Alternatively, you can enter the switchDisable command to disable all ports on the switch.
4. Enter the portEnable command for each port that you disabled in step 3, or enter the
switchEnable command to enable all of the ports on the switch.

NOTE

F_Port trunking requires additional steps to configure the Trunk Area (TA). Refer to “Configuring
F_Port trunking for an Access Gateway” on page 580 or “Configuring F_Port trunking for a Brocade
adapter” on page 581 for information.

Enabling trunking
You can enable trunking for a single port or for an entire switch. Because trunking is automatically
enabled when you install the Trunking license, you need to use this procedure only if trunking has
been subsequently disabled on a port or switch. Enabling trunking disables and re-enables the
affected ports. As a result, traffic through these ports may be temporarily disrupted.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgTrunkPort port mode command to enable trunking on a port.
Enter the switchCfgTrunk mode command to enable trunking on all ports on the switch.
Mode 1 enables trunking.
In the following example, trunking is being enabled on slot 1, port 3.
switch:admin> portcfgtrunkport 1/3 1

Disabling trunking
You can disable trunking for a single port or for an entire switch. Disabling trunking disables and
re-enables the affected ports. As a result, traffic through these ports may be temporarily disrupted.
Trunking on ICLs is always enabled and cannot be disabled.
Disabling trunking fails if a Trunk Area (TA) is enabled on the port. Use the portTrunkArea command
to remove the TA before disabling trunking.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgTrunkPort port mode command to disable trunking on a port.
Enter the switchCfgTrunk mode command to disable trunking on all ports on the switch.
Mode 0 disables trunking.
switch:admin> switchcfgtrunk 0

Displaying trunking information
You can use the trunkShow command to view the following information:

• All the trunks and members of a trunk.

574

Fabric OS Administrator’s Guide
53-1002920-02

Displaying trunking information

24

• Whether the trunking port connection is the master port connection for the trunk group.
• Whether trunks are formed correctly.
• Trunking information for a switch that is part of an FC router backbone fabric interlinking
several edge fabrics.

• Trunking information, including bandwidth and throughput for all the trunk groups in a switch.
Use the portPerfShow command to monitor problem areas where there are congested paths or
dropped links to determine whether you need to adjust the fabric design by adding, removing, or
reconfiguring ISLs and trunking groups. For additional information on using the Brocade Advanced
Performance Monitor to monitor traffic, refer to Chapter 23, “Monitoring Fabric Performance”.
To view detailed information about F_Port trunking, refer to “Displaying F_Port trunking
information” on page 585.
Use the following procedure to view trunking information:
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the trunkShow command.
The following example shows trunking groups 1, 2, and 3; ports 4, 13, and 14 are masters.
switch:admin> trunkshow
1: 6-> 4 10:00:00:60:69:51:43:04

99 deskew 15 MASTER

2: 15-> 13 10:00:00:60:69:51:43:04 99 deskew 16 MASTER
12-> 12 10:00:00:60:69:51:43:04 99 deskew 15
14-> 14 10:00:00:60:69:51:43:04 99 deskew 17
13-> 15 10:00:00:60:69:51:43:04 99 deskew 16
3: 24-> 14 10:00:00:60:69:51:42:dd
2 deskew 15 MASTER

The following example shows trunking information along with the bandwidth and throughput
for all the trunk groups in a switch.
switch:admin> trunkshow -perf
1: 2-> 2 10:00:00:05:1e:81:56:8b
1 deskew 15 MASTER
3-> 3 10:00:00:05:1e:81:56:8b
1 deskew 17
Tx: Bandwidth 4.00Gbps, Throughput 1.66Gbps (48.45%)
Rx: Bandwidth 4.00Gbps, Throughput 1.66Gbps (48.44%)
Tx+Rx: Bandwidth 8.00Gbps, Throughput 3.33Gbps (48.44%)
2:

5->113 10:00:00:05:1e:46:42:01
3 deskew 15 MASTER
4->112 10:00:00:05:1e:46:42:01
3 deskew 15
Tx: Bandwidth 16.00Gbps, Throughput 1.67Gbps (12.12%)
Rx: Bandwidth 16.00Gbps, Throughput 1.67Gbps (12.12%)
Tx+Rx: Bandwidth 32.00Gbps, Throughput 3.33Gbps (12.12%)

3: 10-> 10 10:00:00:05:1e:81:56:8b
1 deskew 15 MASTER
11-> 11 10:00:00:05:1e:81:56:8b
1 deskew 15
Tx: Bandwidth 4.00Gbps, Throughput 1.66Gbps (48.45%)
Rx: Bandwidth 4.00Gbps, Throughput 1.67Gbps (48.48%)
Tx+Rx: Bandwidth 8.00Gbps, Throughput 3.33Gbps (48.46%)
4: 12->892 10:00:00:05:1e:46:42:01
3 deskew 15 MASTER
13->893 10:00:00:05:1e:46:42:01
3 deskew 15
Tx: Bandwidth 16.00Gbps, Throughput 1.67Gbps (12.12%)
Rx: Bandwidth 16.00Gbps, Throughput 1.66Gbps (12.11%)
Tx+Rx: Bandwidth 32.00Gbps, Throughput 3.33Gbps (12.11%)

Fabric OS Administrator’s Guide
53-1002920-02

575

24

Trunk Area and Admin Domains

Trunk Area and Admin Domains
Ports from different Admin Domains (ADs) are not allowed to join the same Trunk Area (TA) group.
The portTrunkArea command prevents the different ADs from joining the TA group.
When you assign a TA, the ports within the TA group have the same index. The index that was
assigned to the ports is no longer part of the switch. Any domain,index (D,I) AD that was assumed
to be part of the domain may no longer exist for that domain because it was removed from the
switch.

Example of Trunk Area assignment on port domain,index
If you have AD1: 3,8; 3,9; 4,13; 4,14 and AD2: 3,10; 3,11, and then create a TA with index 8 with
ports that have index 8, 9, 10, and 11, then index 9, 10, and 11 are no longer with domain 3. This
means that AD2 does not have access to any ports because index 10 and 11 no longer exist on
domain 3. This also means that AD1 no longer has 3,9 in effect, because index 9 no longer exists
for domain 3. Port 3,8, which is the TA group, can still be seen by AD1 along with 4,13 and 4,14.
If a port within a TA is removed, the index is added back to the switch. For example, the same AD1
and AD2 with TA 8 holds true. If you remove port 9 from the TA, it adds index 9 back to the switch.
That means port 3,9 can be seen by AD1 along with 3,8, 4,13, and 4,14.

ISL trunking over long-distance fabrics
In long-distance fabrics, if a port speed is set to autonegotiate, then the maximum speed, which is
16 Gbps, is assumed for reserving buffers for the port. If the port is running at only 2 Gbps, this
wastes buffers. For long-distance ports, you should specify the port speed instead of setting it to
autonegotiate.
In addition to the criteria listed in “Supported configurations for trunking” on page 571, observe the
following criteria for trunking over long-distance fabrics:

• Trunking over long-distance fabrics is supported only on switches running Fabric OS v6.1.0 and
later.

• Extended Fabrics and Trunking licenses are required on all participating switches.
• When configuring long distance, you must configure the portCfgLongDistance
--vc_translation_link_init parameter to be the same on all ports in a long-distance fabric.
For additional information on configuring long distance, refer to “Configuring an extended ISL” on
page 589.
Table 90 summarizes support for trunking over long distance for the Brocade DCX and DCX 8510
Backbones and supported blades.

TABLE 90

576

Trunking over long distance for the Brocade Backbones and blades

Long-distance mode

Distance

Number of 2-Gbps ports

Number of 4-Gbps ports

LE

10 km

48 (six 8-port trunks)

48 (six 8-port trunks)

L0

Normal

See note below

48 (six 8-port trunks)

LD

200 km

4 (one 2-port trunk per switch)

0

LD

250 km

4 (one 2-port trunk per switch)

0

Fabric OS Administrator’s Guide
53-1002920-02

EX_Port trunking

TABLE 90

24

Trunking over long distance for the Brocade Backbones and blades (Continued)

Long-distance mode

Distance

Number of 2-Gbps ports

Number of 4-Gbps ports

LD

500 km

0

0

LS

Static

See note below

NOTE
The L0 mode supports up to 5 km at 2 Gbps, up to 2 km at 4 Gbps, and up to 1 km at 8 Gbps.
The distance for the LS mode is static. You can specify any distance greater than 10 km.
The distance supported depends on the available buffers, the number of back-end ports, and the
number of ports that are offline. For more information on setting port speeds, refer to Chapter 3,
“Performing Advanced Configuration Tasks”.

EX_Port trunking
You can configure EX_Ports to use trunking just as you do regular E_Ports. EX_Port trunking
support is designed to provide the best utilization and balance of frames transmitted on each link
between the FC router and the edge fabric. You should trunk all ports connected to the same edge
fabrics.
The FC router front domain has a higher node WWN—derived from the FC router—than that of the
edge fabric. Therefore, the FC router front domain initiates the trunking protocol on the EX_Port.
After initiation, the first port from the trunk group that comes online is designated as the master
port. The other ports that come online on the trunk group are considered to be the slave ports.
Adding or removing a slave port does not cause frame drop; however, removing a slave port causes
the loss of frames in transit.
If router port cost is used with EX_Port trunking, the master port and slave ports share the router
port cost of the master port.
The restrictions for EX_Port frame trunking are the same as for E_Ports—all the ports must be
adjacent to each other, in the clearly marked groups on the front of the switch.

ATTENTION
EX_Port trunking should be enabled only if the entire configuration is running Fabric OS v5.2.0 or
later.
Refer to Chapter 26, “Using FC-FC Routing to Connect Fabrics,” for more information about
EX_Ports and the FC router.

Masterless EX_Port trunking
EX_Port trunking is masterless except for EX_Ports on Brocade DCX and DCX 8510 Backbones.
For the Backbones, Virtual Fabrics must be enabled for masterless EX_Port trunking to take effect.
For the fixed-port switches, Virtual Fabrics can be enabled or disabled.
If masterless EX_Port trunking is not in effect and the master port goes offline, the entire
EX_Port-based trunk re-forms and is taken offline for a short period of time. If there are no other
links to the edge fabric from the Backbone, the master port going offline may cause a traffic
disruption in the backbone fabric.

Fabric OS Administrator’s Guide
53-1002920-02

577

24

EX_Port trunking

Supported configurations and platforms for EX_Port trunking
EX_Port trunking is a Fiber Channel Routing (FCR) software feature and requires that you have a
Trunking license installed on the FC router and on the edge fabric connected to the other side of
the trunked EX_Ports. The Trunking license is not required for EX_Ports on an ICL.
EX_Port trunking is supported only with Brocade edge fabrics.
You can use EX_Port frame trunking in the following configurations and cases:

• For ports with speeds of 2 Gbps up to a maximum speed of 16 Gbps and trunking over long
distance.

• In the edge fabric, when the FC router is connected to a switch that supports eight ports from
the trunkable group.

• When the FC router is connected to an edge fabric through a mix of trunked and nontrunked
EX_Ports; all will share the same front domain.

• In edge-to-edge, backbone-to-edge, and dual-backbone configurations.
Masterless EX_Port trunking has additional configuration requirements. Refer to “Masterless
EX_Port trunking” for these additional requirements.

NOTE

QoS and EX_Port trunking can coexist. However, if some ports in the trunk group have QoS enabled
and some have QoS disabled, then two trunk groups will form: one with QoS enabled and one with
QoS disabled.

Backward compatibility support
For backward compatibility, an FC router that supports EX_Port trunking can continue to
interoperate with older FC routers and all previously supported Brocade switches in the backbone
fabric or Brocade edge fabric.

Configuring EX_Port trunking
With EX_Port trunking, you use the same CLI commands as you do for E_Port trunking. Refer to
“Configuring trunk groups” on page 573 for instructions.

Displaying EX_Port trunking information
1. Log in as an admin and connect to the switch.
2. Enter the switchShow command to display trunking information for the EX_Ports.
The following is an example of a master EX_Port and a slave EX_Port displayed in switchShow
output.
switch:admin> switchshow
Index Slot Port Address Media Speed State
==============================================
16
2
0
ee1000
id
N4
No_Light
17
2
1
ee1100
id
N4
Online
EX_Port
18
2
2
ee1200
id
N4
Online
EX_Port
(fabric id = 2 )(Trunk master)

578

(Trunk port, master is Slot 2 Port 2 )
10:00:00:05:1e:35:bb:32 "MtOlympus_82"

Fabric OS Administrator’s Guide
53-1002920-02

F_Port trunking

19
2
20
2
21
2
22
2
23
2
(fabric id

3
ee1300
id
4
ee1400
id
5
ee1500
id
6
ee1600
id
7
ee1700
id
= 2 )(Trunk master)

N4
N4
N4
N4
N4

No_Light
Online
Online
Online
Online

EX_Port
EX_Port
EX_Port
EX_Port

24

(Trunk port, master is Slot 2 Port 7 )
(Trunk port, master is Slot 2 Port 7 )
(Trunk port, master is Slot 2 Port 7 )
10:00:00:60:69:80:1d:bc "MtOlympus_72"

F_Port trunking
You can configure F_Port trunking in the following scenarios:

• Between F_Ports on a Fabric OS switch and N_Ports on an Access Gateway module
• Between F_Ports on a Fabric OS switch and N_Ports on a Brocade adapter
For F_Port trunking, you must create a Trunk Area (TA) within the trunk group. When you assign an
area within a trunk group, that group is enabled for F_Port trunking. The TA that you assign must be
within the 8-port trunk group beginning with port 0 (zero). After you assign a TA to a port, the port
immediately acquires the TA as the area of its PID. Likewise, after you remove a TA from a port, the
port immediately acquires the default area as its PID. F_Port trunking prevents reassignments of
the Port ID (also referred to as the Address Identifier) when F_Ports go offline, and it increases
F_Port bandwidth.
Refer to the Access Gateway Administrator’s Guide and the Brocade Adapters Administrator’s
Guide for information about configuring the corresponding N_Port trunking on the Access Gateway
and the Brocade adapter.

F_Port trunking for Access Gateway
You can configure trunking between the F_Ports on an edge switch and the N_Ports on an Access
Gateway module.

NOTE
You cannot configure F_Port trunking on the F_Ports of an Access Gateway module.
F_Port trunking keeps F_Ports from becoming disabled when they are mapped to an N_Port on a
switch in Access Gateway (AG) mode. With F_Port trunking, any link within a trunk can go offline or
become disabled, but the trunk remains fully functional and there are no reconfiguration
requirements.
Figure 83 shows a switch in AG mode without F_Port masterless trunking. Figure 84 shows a
switch in AG mode with F_Port masterless trunking.

Fabric OS Administrator’s Guide
53-1002920-02

579

24

F_Port trunking

FIGURE 83

Switch in Access Gateway mode without F_Port masterless trunking

FIGURE 84

Switch in Access Gateway mode with F_Port masterless trunking

NOTE

You do not need to map the host to the master port manually because the Access Gateway will
perform a cold failover to the master port.
Refer to “Configuring F_Port trunking for an Access Gateway” on page 580 for instructions on
configuring F_Port trunking.

Requirements for F_Port trunking on an Access Gateway
In addition to the requirements listed in “Requirements for trunk groups” on page 572, refer to the
Access Gateway Administrator’s Guide for additional requirements that are specific to F_Port
trunking on an Access Gateway.

Configuring F_Port trunking for an Access Gateway
Access Gateway trunking configuration is mostly on the edge switch. On the Access Gateway
module, you only need to ensure that the Trunking license is applied and enabled.

580

Fabric OS Administrator’s Guide
53-1002920-02

F_Port trunking

24

Use the following procedure on the edge switch connected to the Access Gateway module to
configure F_Port trunking.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgShow command to ensure that the ports have trunking enabled. If trunking is
not enabled, enter the portCfgTrunkPort port 1 command.
3. Enter the portDisable command for each port to be included in the TA.
4. Enter the portTrunkArea --enable command to enable the trunk area.
For example, the following command creates a TA for ports 36–39 with index number 37.
switch:admin> porttrunkarea --enable 36-39 -index 37
Trunk index 37 enabled for ports 36, 37, 38 and 39.

When you assign a trunk area on a port, trunking is automatically enabled on the F_Ports. The
portTrunkArea command does not unassign a TA if its previously assigned Area_ID is the same
Address Identifier (Area_ID) of the TA unless all the ports in the trunk group are specified to be
unassigned.
5. Enter the portEnable command to re-enable the ports in the TA.

F_Port trunking for Brocade adapters
You can configure trunking between the F_Ports on an edge switch and the Brocade adapters.
In addition to the requirements listed in “Requirements for trunk groups” on page 572, note the
following requirements, which are specific to F_Port trunking for Brocade adapters:

• The edge switch must be running in Native mode. You cannot configure trunking between the
Brocade adapters and the F_Ports of an Access Gateway module.

• You can configure only two F_Ports in one trunk group.
Refer to the Brocade Adapters Administrator’s Guide for information about configuring the
corresponding N_Port trunking on the Access Gateway and the Brocade adapter.

Configuring F_Port trunking for a Brocade adapter
F_Port trunking for Brocade adapters requires configuration on the FC switch as well as on the
Brocade HBAs. This section describes the configuration steps you perform on the switch. Refer to
the Brocade Adapters Administrator’s Guide for a detailed description and requirements of N_Port
trunking on the adapters.
1. On the switch side, perform the following steps:
a.

Configure both ports for trunking by using the portCfgTrunkPort command.
switch:admin> portcfgtrunkport 3/40 1
switch:admin> portcfgtrunkport 3/41 1

b.

Disable the ports to be used for trunking by using the portDisable command.
switch:admin> portdisable 3/40
switch:admin> portdisable 3/41

c.

Enable the trunk on the ports by using the portTrunkArea command.
switch:admin> porttrunkarea --enable 3/40-41 -index 296
Trunk index 296 enabled for ports 3/40 and 3/41.

Fabric OS Administrator’s Guide
53-1002920-02

581

24

F_Port trunking

2. On the host side, enable trunking as described in the Brocade Adapters Administrator’s Guide.
3. On the switch side, enable the ports by using the portEnable command.
switch:admin> portenable 3/40
switch:admin> portenable 3/41

F_Port trunking considerations
Table 91 describes the F_Port masterless trunking considerations.

TABLE 91

F_Port masterless trunking considerations

Category

Description

AD

You cannot create a Trunk Area on ports with different Admin Domains. You cannot create a
Trunk Area in AD255.

Area assignment

You statically assign the area within the trunk group on the edge switch. That group is the
F_Port trunk.
The static trunk area you assign must fall within the ASIC's trunk group of the switch or
blade starting from port 0, and must be one of the port’s default areas of the trunk group.
10-bit addressing is the default mode for all dynamically created partitions in the Brocade
DCX and DCX 8510-8 platforms.

Authentication

Authentication occurs only on the F_Port trunk master port and only once per the entire
trunk. This behavior is the same as E_Port trunk master authentication. Because only one
port in the trunk does FLOGI to the switch, and authentication follows FLOGI on that port,
only that port displays the authentication details when you issue the portShow command.
NOTE: Switches in Access Gateway mode do not perform authentication.

configdownload

If you issue the configDownload command for a port configuration that is not compatible
with F_Port trunking, and the port is Trunk Area-enabled, then the port will be persistently
disabled. F_Port trunks will never be restored through configDownload.
NOTE: Long distance, port mirroring, non-CORE_PID, and FastWrite are not compatible
with F_Port trunking.

domain,index (D,I)

Creating a Trunk Area may remove the Index ("I") from the switch to be grouped to the Trunk
Area. All ports in a Trunk Area share the same index. This means that a domain,index (D,I),
which refers to an index that might have been removed, will no longer be part of the switch.
NOTE: Be sure to include AD, zoning, and DCC when creating a Trunk Area.
You can remove the port from the Trunk Area to place the index back in effect. D,I behaves
as normal, but you may see the effects of grouping ports into a single index.
Also, D,I continues to work for Trunk Area groups. The index can be used in D,I if it was the
index for the Trunk Area group.

DCC Policy

582

DCC policy enforcement for the F_Port trunk is based on the Trunk Area; the FDISC requests
to a trunk port are accepted only if the WWN of the attached device is part of the DCC policy
against the TA. The PWWN of the FLOGI sent from the AG will be dynamic for the F_Port
trunk master. Because you do not know ahead of time what PWWN the AG will use, the
PWWN of the FLOGI will not go through DCC policy check on an F_Port trunk master.
However, the PWWN of the FDISC will continue to go through DCC policy check.

Fabric OS Administrator’s Guide
53-1002920-02

F_Port trunking

TABLE 91

24

F_Port masterless trunking considerations (Continued)

Category

Description

Default Area

Port X is a port that has its Default Area the same as its Trunk Area. The only time you can
remove port X from the trunk group is when the entire trunk group has the Trunk Area
disabled.

Downgrade

You can have trunking on, but you must disable the trunk ports before performing a
firmware downgrade.

ATTENTION
Removing a Trunk Area on ports running traffic is disruptive because you must disable
the port to disable the Trunk Area on the port. Use caution before assigning a Trunk
Area if you need to downgrade to a firmware version earlier than Fabric OS v6.2.0.
FastWrite

When you assign a Trunk Area to a trunk group, the trunk group cannot have FastWrite
enabled on those ports. If a port is FastWrite-enabled, the port cannot be assigned a Trunk
Area.

FICON

FICON is not supported on F_Port trunk ports. However, FICON can still run on ports that are
not F_Port trunked within the same switch.

HA Sync

If you plug in a standby CP with a firmware version earlier than Fabric OS v6.2.0 and a
Trunk Area is present on the switch, the CP blades will become out of sync.

Long Distance

Long distance is not allowed on F_Port trunks, which means that a Trunk Area is not
allowed on long-distance ports. You cannot enable long distance on ports that have a Trunk
Area assigned to them.

Management Server Registered Node ID (RNID), Link Incident Record Registration (LIRR), and Query Security
Attribute (QSA) ELSs are not supported on F_Port trunks.
NPIV

N_Port ID virtualization (NPIV) is supported on the F_Port master trunk.

PID format

F_Port trunking is supported only in the CORE PID format.

Port mirroring

Port mirroring is not supported on Trunk Area ports or on the PID of an F_Port trunk port.
Port mirroring is not supported on the Brocade Encryption Switch.

Port Swap

When you assign a Trunk Area to a trunk group, the Trunk Area cannot be port swapped. If a
port is swapped, then you cannot assign a Trunk Area to that port.

Port Types

Only F_Port trunk ports are allowed on a Trunk Area port. All other port types are
persistently disabled.

PWWN

The entire Trunk Area trunk group shares the same Port WWN within the trunk group. The
PWWN is the same across the F_Port trunk that has 0x2f or 0x25 as the first byte of the
PWWN. The TA is part of the PWWN in the format listed in Table 92 on page 584.

QoS

QoS is supported.

Routing

Routing will route against the F_Port trunk master. Bandwidth information will be modified
accordingly as the F_Port trunk forms.

Trunk Master

No more than one trunk master is allowed in a trunk group. The second trunk master will be
persistently disabled with the reason "Area has been acquired”.

Upgrade

There are no limitations on upgrading to Fabric OS v7.0.0 and later if the F_Port is present
on the switch. Upgrading is not disruptive.

Table 92 describes the PWWN format for F_Port and N_Port trunk ports.

Fabric OS Administrator’s Guide
53-1002920-02

583

24

F_Port trunking

TABLE 92

PWWN format for F_Port and N_Port trunk ports

NAA = 2

2f:xx:nn:nn:nn:nn:nn:nn
(1)

Port WWNs for:
switch’s Fx_Ports.

The valid range of xx is [0–FF],
for a maximum of 256.

NAA = 2

25:xx:nn:nn:nn:nn:nn:nn
(1)

Port WWNs for:
switch's FX_Ports

The valid range of xx is [0–FF],
for a maximum of 256.

F_Port trunking in Virtual Fabrics
F_Port trunking functionality performs the same in Virtual Fabrics as it does in non-Virtual Fabrics
platforms except for the Brocade DCX and DCX 8510-8. Fabric OS uses a 10-bit addressing model,
which is the default mode for all dynamically created logical switches in the DCX platform.
In the DCX and DCX 8510 platforms, F_Port trunk ports dynamically receive an 8-bit area address
that remains persistent. After F_Port trunking configurations are removed from a port in a logical
switch, that port returns to the default 10-bit area addressing model, which supports up to 1024
F_Ports in a logical switch.

NOTE
Because the DCX and DCX 8510-8 platforms have a maximum of 576 ports, out of the 1024 10-bit
address range, addresses 448–1023 are reserved for the 10-bit address space. Addresses 0–447
are reserved for assigning to NPIV/Loop ports to support 112 (448/4) NPIV/Loop ports in a logical
switch with 256 devices each.
The following are the F_Port trunking considerations for Virtual Fabrics:

• If a port is enabled for F_Port trunking, you must disable the configuration before you can
move a port from the logical switch.

• If the user-bound area for a port is configured by means of the portAddress command, the port
cannot be configured as an F_Port trunk port. You must explicitly remove the user-bound area
before enabling F_Port trunking.

• If you swap a port by using the portSwap command, you must undo the port swap before
enabling F_Port trunking.

• The Port WWN format in a Virtual Fabric is 2z:zz:xx:xx:xx:xx:xx:xx. The z:zz is the logical port
number; for example, the logical port 450 will be 1:c2. The xx:xx:xx:xx:xx:xx is based on the
logical fabric WWN.
For example, if the logical fabric WWN is 10:00:00:05:1e:39:fa:f3, and logical port number is
450, then the Port WWN of the F_Port trunk will be 21:c2: 00:05:1e:39:fa:f3.

• F_Port trunks are not allowed on the base switch because you cannot have F_Ports on the
base switch.

• If F_Port trunking is enabled on some ports in the default switch, and you disable Virtual
Fabrics, all of the F_Port trunking information is lost.

• All of the ports in an F_Port trunk must belong to a single trunk group of ports on the platform
and must also belong to the same logical switch.
Refer to Chapter 11, “Managing Virtual Fabrics,” for detailed information about Virtual Fabrics.

584

Fabric OS Administrator’s Guide
53-1002920-02

Displaying F_Port trunking information

24

Displaying F_Port trunking information
Use the following commands on the edge switch to verify the F_Port trunking configuration.

• Enter the switchShow command to display the switch and port information.
• Enter the portTrunkArea --show enabled command to display the TA-enabled port
configuration.
switch:admin> porttrunkarea --show enabled
Port Type
State
Master TI DI
------------------------------------36
F-port Master 36
37 36
37
F-port Slave
36
37 37
38
F-port Slave
36
37 38
39
F-port Slave
36
37 39

• Enter the portTrunkArea --show trunk command to display the trunking information.
switch:admin> porttrunkarea --show trunk
Trunk Index 37: 39->0
sp: 8.000G bw: 16.000G deskew 15 MASTER
Tx: Bandwidth 16.00Gbps, Throughput 1.63Gbps (11.84%)
Rx: Bandwidth 16.00Gbps, Throughput 1.62Gbps (11.76%)
Tx+Rx: Bandwidth 32.00Gbps, Throughput 3.24Gbps (11.80%)
38->1
sp: 8.000G bw: 8.000G deskew 15
Tx: Bandwidth 16.00Gbps, Throughput 1.63Gbps (11.84%)
Rx: Bandwidth 16.00Gbps, Throughput 1.62Gbps (11.76%)
Tx+Rx: Bandwidth 32.00Gbps, Throughput 3.24Gbps (11.80%)
37->1
sp: 8.000G bw: 8.000G deskew 15
Tx: Bandwidth 16.00Gbps, Throughput 1.63Gbps (11.84%)
Rx: Bandwidth 16.00Gbps, Throughput 1.62Gbps (11.76%)
Tx+Rx: Bandwidth 32.00Gbps, Throughput 3.24Gbps (11.80%)
36->1
sp: 8.000G bw: 8.000G deskew 15
Tx: Bandwidth 16.00Gbps, Throughput 1.63Gbps (11.84%)
Rx: Bandwidth 16.00Gbps, Throughput 1.62Gbps (11.76%)
Tx+Rx: Bandwidth 32.00Gbps, Throughput 3.24Gbps (11.80%)

Disabling F_Port trunking
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portDisable command to disable the ports that are to be removed from the trunk
area.
3. Enter the portTrunkArea --disable command to remove ports from the trunk area.
This command does not unassign a TA if its previously assigned Area_ID is the same address
identifier (Area_ID) of the TA unless all the ports in the trunk group are specified to be
unassigned.
switch:admin> portdisable 0-2
switch:admin> porttrunkarea --disable 0-2
Trunk index 2 disabled for ports 0, 1, and 2.

Fabric OS Administrator’s Guide
53-1002920-02

585

24

Enabling the DCC policy on a trunk area

Enabling the DCC policy on a trunk area
After you assign a trunk area, the portTrunkArea command checks whether there are any active
DCC policies on the port with the index TA, and then issues a warning to add all the device WWNs to
the existing DCC policy with index as TA.
All DCC policies that refer to an index that no longer exists will not be in effect.
1. Add the WWN of all the devices to the DCC policy against the TA.
2. Enter the secPolicyActivate command to activate the DCC policy.
In order for security to enforce the DCC policy on the trunk ports, you must enable the TA
before issuing the secPolicyActivate command.
3. Turn on the trunk ports.
Turn on trunk ports after issuing the secPolicyActivate command to prevent the ports from
becoming disabled in case there is a DCC security policy violation.
You can configure authentication on all Brocade trunking configurations. For more information on
authentication, refer to Chapter 8, “Configuring Security Policies”.

586

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

25

Managing Long-Distance Fabrics

In this chapter
• Long-distance fabrics overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Extended Fabrics device limitations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Long-distance link modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Configuring an extended ISL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Forward error correction on long-distance links . . . . . . . . . . . . . . . . . . . . .

587
588
588
589
591

Long-distance fabrics overview
The most effective configuration for implementing long-distance SAN fabrics is to deploy Fibre
Channel switches at each location in the SAN. Each switch handles local interconnectivity and
multiplexes traffic across long-distance dark fiber or wave-length division multiplexing (WDM) links,
while the Brocade Extended Fabrics software enables SAN management over long distances.
Brocade Extended Fabrics is an optional licensed feature for Brocade SAN deployment over
distances beyond 10 km. A Brocade Extended Fabrics license is required before you can implement
long-distance dynamic (LD) and long-distance static (LS) distance levels. The LD and LS settings
are necessary to achieve maximum performance results over inter-switch links (ISLs) that are
greater than 10 km.
For details about obtaining and installing licensed features, refer to Chapter 21, “Administering
Licensing”.
The Extended Fabrics feature enables the following functionality:

• Fabric interconnectivity over Fibre Channel at longer distances
ISLs can use long-distance dark fiber connections to transfer data. Wavelength-division
multiplexing, such as dense wavelength-division multiplexing (DWDM), coarse
wavelength-division multiplexing (CWDM), and time-division multiplexing (TDM), can be used to
increase the capacity of the links. As Fibre Channel speeds increase, the maximum distance
decreases for each switch.
The Extended Fabrics feature extends the distance the ISLs can reach over an extended fiber.
This extension is accomplished by providing enough buffer credits on each side of the link to
compensate for latency introduced by the extended distance.

• Simplified management over distance
Each device attached to the SAN appears as a local device, which simplifies deployment and
administration.

Fabric OS Administrator’s Guide
53-1002920-02

587

25

Extended Fabrics device limitations

• Optimized switch buffering
When Extended Fabrics is installed on gateway switches (with E_Port connectivity from one
switch to another), the ISLs (E_Ports) are configured with a large pool of buffer credits. The
enhanced switch buffers help ensure that data transfer can occur at near-full bandwidth to use
the connection over the extended links efficiently. This efficiency ensures the highest possible
performance on ISLs.

Extended Fabrics device limitations
Brocade recommends that you do not use the FC8-64 port blade for long distance because of its
limited buffers. This blade does not support long-wavelength (LWL) fiber optics and supports
limited distance. However, you can use the portCfgLongDistance command to reserve frame
buffers for the ports intended to be used in long-distance mode through DWDM.
There is a limited number of reserved buffers used for long distance for each blade. If some ports
are configured in long-distance mode and have buffers reserved for them, insufficient buffers may
remain for the other ports. In this case, some of the remaining ports may come up in degraded
mode.

Long-distance link modes
Use the portCfgLongDistance command to support long-distance links and to allocate sufficient
numbers of full-size frame buffers on a specific port. Changes made by this command are
persistent across switch reboots and power cycles.
The portCfgLongDistance command supports the following long-distance link modes:

• Normal Mode (LO) — L0 is the normal (default) mode for an E_Port. It configures the E_Port as
a standard (not long-distance) ISL. A total of 20 full-size frame buffers are reserved for data
traffic, regardless of the E_Port’s operating speed; therefore, the maximum supported link
distance is up to 5 km at 2 Gbps, up to 2 km at 4 Gbps, and up to 1 km at 8, 10, and 16 Gbps.

• Extended Mode (LE) — LE configures the distance for an E_Port when that distance is greater
than 5 km and up to 10 km. LE does not require an Extended Fabrics license. The baseline for
the buffer credit calculation is one buffer credit per km at 2 Gbps. This allocation yields the
following values for 10 km:

-

10 buffer credits per port at 2 Gbps
20 buffer credits per port at 4 Gbps
40 buffer credits per port at 8 Gbps
50 buffer credits per port at 10 Gbps
80 buffer credits per port at 16 Gbps

• Dynamic Mode (LD) — LD calculates buffer credits based on the distance measured during
port initialization. Brocade switches use a proprietary algorithm to estimate distance across an
ISL. The estimated distance is used to determine the buffer credits required in LD (dynamic)
extended link mode based on a maximum Fibre Channel payload size of 2,112 bytes. You can
place an upper limit on the calculation by providing a desired_distance value. Fabric OS
confines user entries to no larger than what it has estimated the distance to be. When the
measured distance is more than the specified desired distance, the desired distance (the
smaller value) is used in the calculation.

588

Fabric OS Administrator’s Guide
53-1002920-02

Configuring an extended ISL

25

• Static Mode (LS) — LS calculates a static number of buffer credits based only on a user-defined
desired_distance value. LS mode also assumes that all FC payloads are 2,112 bytes. Specify
LS mode to configure a static long-distance link with a fixed buffer allocation greater than 10
km.

Configuring an extended ISL
Before configuring an extended ISL, ensure that the following conditions are met:

• The ports on both ends of the ISL are operating at the same port speed, and can be configured
for the same distance_level without compromising local switch performance.

NOTE

A long-distance link also can be configured to be part of a trunk group. Two or more
long-distance links in a port group form a trunk group when they are configured for the same
speed and distance, and their link distances are nearly equal. For information on trunking
concepts and configurations, refer to Chapter 24, “Managing Trunking Connections”.

• Only qualified Brocade SFP transceivers are used. Only Brocade-branded or certain
Brocade-qualified SFP transceivers are supported.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the switchDisable command.
3. Enter the configure command to set the switch fabric-wide configurations.
Table 93 shows the fabric-wide settings that can be set:

TABLE 93

Fabric-wide settings

Field

Type

Default

Range

Domain

Number

1

Varies

R_A_TOV

Number

10000

E_D_TOV * 2 to 120000

E_D_TOV

Number

2000

1000 to R_A_TOV/2

WAN_TOV

Number

0

0 to R_A_TOV/4

MAX_HOPS

Number

7

7 to 19

4. For 8-Gbps platforms only, enter the portCfgFillword command to set ARB as the fill word. Refer
to the Fabric OS Command Reference for more for more information on configuring the fill
word for a single 8G FC port.
portcfgfillword [slot/]port, mode

The mode parameter in this command should be set to 3 if the vc_translation_link_init
parameter in the portCfgLongDistance command (in the next step) is set to 1.
5. Enter the portCfgLongDistance command.
portcfglongdistance [slot/]port [distance_level] [vc_translation_link_init]
[-distance desired_distance]

6. Repeat step 4 and step 5 for the remote extended ISL port. Both the local and remote
extended ISL ports must be configured to the same distance_level. When the connection is
initiated, the fabric will reconfigure.

Fabric OS Administrator’s Guide
53-1002920-02

589

25

Configuring an extended ISL

The following example configures slot 1, port 2 to support a 100-km link in LS mode and to use the
extended link initialization sequence. This example is for an 8-Gbps platform.
switch:admin> portcfgfillword 1/2 3
switch:admin> portcfglongdistance 1/2 LS 1 -distance 100
Reserved Buffers =
406
Warning: port may be reserving more credits depending on port speed.
switch:admin> portshow 1/2
portName:
portHealth: OFFLINE
Authentication: None
portDisableReason: None
portCFlags: 0x1
portFlags: 0x1
PRESENT U_PORT
portType: 17.0
portState: 2
Offline
Protocol: FC
portPhys: 2
No_Module
portScn:
0
port generation number:
0
portId:
010200
portIfId:
4312003b
portWwn:
20:02:00:05:1e:94:0f:00
portWwn of device(s) connected:
Distance: static (desired = 100 Km)
portSpeed: N8Gbps
LE domain: 0
FC Fastwrite: OFF
Interrupts:
Unknown:
Lli:
Proc_rqrd:
Timed_out:
Rx_flushed:
Tx_unavail:
Free_buffer:
Overrun:
Suspended:
Parity_err:
2_parity_err:
CMI_bus_err:

0
0
0
5
0
0
0
0
0
0
0
0
0

Link_failure:
Loss_of_sync:
Loss_of_sig:
Protocol_err:
Invalid_word:
Invalid_crc:
Delim_err:
Address_err:
Lr_in:
Lr_out:
Ols_in:
Ols_out:

0
0
3
0
0
0
0
0
0
0
0
0

Frjt:
Fbsy:

0
0

Enabling long distance when connecting to TDM devices
Use this procedure when connecting to time-division multiplexing (TDM) devices and your Brocade
switch has QoS and buffer credit recovery enabled.
1. Connect to the switch and log in using an account assigned to the admin role.
2. Disable QoS.
switch:admin> portcfgqos --disable [slot/]port

If you do not disable QoS, after the second or third link reset (LR), ARB fill words display. Refer
to the Fabric OS Command Reference for more for more information on setting fill words.

590

Fabric OS Administrator’s Guide
53-1002920-02

Forward error correction on long-distance links

25

3. Disable buffer credit recovery; buffer credit recovery is not compatible with the IDLE mode. If
you do not disable buffer credit recovery, it continues to perform a link reset.
switch:admin> portcfgcreditrecovery --disable [slot/]port

4. Configure the port to support long-distance links.
switch:admin> portcfglongdistance [slot/]port,LS,0,-distance 100

Forward error correction on long-distance links
Forward error correction (FEC) on user ports is supported for LD and LS long-distance modes. Use
the portCfgLongDistance command with the -fecEnable or -fecDisable options to enable or disable
FEC, respectively, on a user port. Alternatively, you can use the portCfgFec command with the
--enable or --disable option as you would for any regular port.
For additional details about FEC, refer to “Enabling forward error correction” on page 111.

Enabling FEC on a long-distance link
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgLongDistance command and include the -fecEnable option, or issue the
portCfgFec command with the --enable option.
3. Enter the portCfgFec --show command to verify the configuration.
Example
switch:admin> portcfglongdistance 1/20 LS 1 -distance 122 –fecenable
FEC has been enabled.
Reserved Buffers =
982
Warning: port (132) may be reserving more credits depending on port speed.
switch:admin> portcfgfec --show 1/20
Forward Error Correction capable:
Forward Error Correction configured:

YES
ON

Disabling FEC on a long-distance link
1. Connect to the switch and log in using an account assigned to the admin role.
2. Enter the portCfgLongDistance command and include the -fecDisable option, or issue the
portCfgFec command with the --disable option.
3. Enter the portCfgFec --show command to verify the configuration.
Example
switch:admin> portcfglongdistance 1/20 LS 1 -buffers 500 –fecdisable
FEC has been disabled.
Reserved Buffers =
982
Warning: port (132) may be reserving more credits depending on port speed.
switch:admin> portcfgfec --show 1/20
Forward Error Correction capable:
Forward Error Correction configured:

Fabric OS Administrator’s Guide
53-1002920-02

YES
OFF

591

25

592

Forward error correction on long-distance links

Fabric OS Administrator’s Guide
53-1002920-02

Chapter

26

Using FC-FC Routing to Connect Fabrics

In this chapter
• FC-FC routing overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Fibre Channel routing concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Setting up FC-FC routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Backbone fabric IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• FCIP tunnel configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Inter-fabric link configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• FC router port cost configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Shortest IFL cost configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• EX_Port frame trunking configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• LSAN zone configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Proxy PID configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Fabric parameter considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Inter-fabric broadcast frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Resource monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• FC-FC routing and Virtual Fabrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Upgrade and downgrade considerations for FC-FC routing . . . . . . . . . . . .
• Displaying the range of output ports connected to xlate domains . . . . . .

593
596
603
605
606
607
613
615
619
620
633
633
634
634
636
639
639

FC-FC routing overview
The FC-FC routing service provides Fibre Channel routing between two or more fabrics without
merging those fabrics. For example, using FC-FC routing, you can share tape drives across multiple
fabrics without the administrative problems, such as change management, network management,
scalability, reliability, availability, and serviceability, that might result from merging the fabrics.
Be aware that there are different routing terminologies in use:

• FC routing is only in a single fabric (Layer 2 routing). This type of routing is discussed in
Chapter 4, “Routing Traffic”.

• FC-FC routing is routing between two fabrics (Layer 3 routing) and is discussed in this chapter.
FC-FC routing supports connectivity between the following types of fabrics:

• Fabric OS and Fabric OS
• Fabric OS and Brocade Network OS
• Fabric OS and M-EOS

Fabric OS Administrator’s Guide
53-1002920-02

593

26

FC-FC routing overview

A Fibre Channel router (FC router) is a switch running the FC-FC routing service. The FC-FC routing
service can be simultaneously used as an FC router and as a SAN extension over wide area
networks (WANs) using FCIP.
You can set up QoS traffic prioritization over FC routers. Refer to “QoS” on page 415 for information
about QoS and instructions for setting traffic prioritization over an FC router.

ATTENTION
FC-FC routing is not supported on a Brocade 7800 that has been enabled for logical switches.

License requirements for FC-FC routing
The Integrated Routing license is required for FC-FC routing between Fabric OS fabrics and between
Fabric OS and M-EOS fabrics.

NOTE
The Integrated Routing license is not required for connectivity between Fabric OS and Brocade
Network OS fabrics or between Brocade Network OS fabrics connected by an FC router.
The Integrated Routing license allows 8-Gbps and 16-Gbps FC ports to be configured as EX_Ports
(or VEX_Ports) supporting FC-FC routing.
Enabling the Integrated Routing license and capability does not require a switch reboot.

NOTE

Brocade recommends that all FC routers in a backbone fabric either have the Integrated Routing
license or not. It is not recommended to mix licensed and unlicensed FC routers in the backbone
fabric.

Supported platforms for FC-FC routing
FC-FC routing is supported on the following platforms:

•
•
•
•
•
•
•
•

Brocade 5100 switch
Brocade 5300 switch
Brocade 6510 switch
Brocade 6520 switch
Brocade VA-40FC switch
Brocade 7800 Extension Switch
Brocade Encryption Switch
Brocade DCX and DCX 8510 Backbone families:

-

594

8-Gbps port blades (FC8-16, FC8-32, FC8-32E, FC8-48, FC8-48E, FC8-64)
16-Gbps port blades (FC16-32, FC16-48)
FX8-24 DCX Extension Blade

Fabric OS Administrator’s Guide
53-1002920-02

FC-FC routing overview

-

26

ICL ports on the core blades. EX-Port on ICL is supported only in DCX 8510-8 and DCX
8510-4 when all the port blades in the chassis belong to one of these blade types:
FC16-32, FC16-48, FC8-32E, FC8-48E).

NOTE

Device discovery will not occur properly with ICL EX-Port connected edge fabrics if the FC
Router has unsupported blades which are not mentioned above.

For the Brocade Backbone families, the backbones have a limit of 128 EX_Ports for each chassis.
Refer to the Network OS Administrator’s Guide for supported Network OS platforms.

Supported configurations for FC-FC routing
FC-FC routing supports the following configurations:

•
•
•
•
•
•

FC router connected to a Fabric OS nonsecured edge fabric.
FC router connected to a Fabric OS secured edge fabric.
FC router connected to a Brocade Network OS edge fabric (Network OS v2.1.1 or later).
FC router connected to an M-EOS Open Mode edge fabric.
FC router connected to an M-EOS Fabric Mode edge fabric.
FC router interoperating with legacy FC routers (Brocade 7500 switch).

In configurations with two backbone fabrics connected to the same edge fabric, routing is not
supported between edge fabrics that are not directly attached to the same backbone fabric.
Routing over multiple backbone fabrics is a multi-hop topology and is not allowed.
In an edge fabric that contains a mix of administrative domain (AD)-capable switches and switches
that are not aware of AD, the FC router must be connected directly to an AD-capable switch. For
more information, refer to “Use of Admin Domains with LSAN zones and FC-FC routing” on
page 620.
VEX edge to VEX edge device sharing is not supported.

Network OS connectivity limitations
You should be aware of the following configuration limitations for Network OS connectivity:

• All of the platforms listed in “Supported platforms for FC-FC routing” on page 594 support
FC-FC routing to a Brocade Network OS fabric, except for the Brocade Encryption Switch.

• VEX_Ports do not support Network OS connectivity.
• FCoE devices connected to an FCOE10-24 blade cannot communicate with FCoE devices in the
Network OS fabric.

• If Admin Domains are enabled, connectivity between the Fabric OS fabric and a Brocade
Network OS fabric is not supported.

• Connectivity between a Brocade Network OS fabric and an M-EOS fabric is not supported.

Fabric OS Administrator’s Guide
53-1002920-02

595

26

Fibre Channel routing concepts

Fibre Channel routing concepts
Fibre Channel routing introduces the following concepts:

• Fibre Channel router (FC router)
A switch running the FC-FC routing service. Refer to “Supported platforms for FC-FC routing” on
page 594 for a list of platforms that can be FC routers.

• EX_Port and VEX_Port
An EX_Port and VEX_Port function similarly to an E_Port and VE_Port respectively, but
terminate at the switch and do not propagate fabric services or routing topology information
from one edge fabric to another. Refer to the Fibre Channel over IP Administrator’s Guide for
details about VE_Ports.

• Edge fabric
An edge fabric is a Fibre Channel fabric with targets and initiators connected through the
supported platforms by using an EX_Port or VEX_Port.

• Backbone fabric
A backbone fabric is an intermediate network that connects one or more edge fabrics.
In a SAN, the backbone fabric consists of at least one FC router and possibly a number of
Fabric OS-based Fibre Channel switches (refer to Figure 87 on page 599).

• Inter-fabric link (IFL)
The link between an E_Port and EX_Port, or VE_Port and VEX_Port, is called an inter-fabric link
(IFL). You can configure multiple IFLs from an FC router to an edge fabric.
Figure 85 shows a metaSAN consisting of three edge fabrics connected through a Brocade
DCX with inter-fabric links.
Host
Edge
fabric 1

Edge
fabric 2

E_Port

Edge
fabric 3
E_Port

E_Port

Fibre
Channel
switch

Target

Target

IFL
IFL

Long distance IFL

Fibre
Channel
switch

EX_Ports

FC router

FIGURE 85

596

A metaSAN with inter-fabric links

Fabric OS Administrator’s Guide
53-1002920-02

Fibre Channel routing concepts

26

• Logical SANs (LSANs)
An LSAN is defined by zones in two or more edge or backbone fabrics that contain the same
devices. You can create LSANs that span fabrics. These LSANs enable Fibre Channel zones to
cross physical SAN boundaries without merging the fabrics while maintaining the access
controls of zones.
An LSAN device can be a physical device, meaning that it physically exists in the fabric, or it can
be a proxy device.
Figure 86 shows a metaSAN with a backbone consisting of one FC router connecting hosts in
edge fabrics 1 and 3 with storage in edge fabric 2 and the backbone fabric through the use of
LSANs. Three LSAN zones allow device sharing between the backbone fabric and edge fabric 1,
between edge fabric 1 and edge fabric 2, and between edge fabric 2 and edge fabric 3.
VE_Port

Edge fabric 2
IP cloud

Edge fabric 1
Edge fabric 3
E_Port

E_Port

IFL

IFL

IFL

VEX_Port

FC router
EX_Port (2)
= LSAN

Backbone fabric

FIGURE 86

A metaSAN with edge-to-edge and backbone fabrics and LSAN zones

• Proxy device
A proxy device is a virtual device imported into a fabric by a Fibre Channel router, and
represents a real device on another fabric. It has a name server entry and is assigned a valid
port ID. When a proxy device is created in a fabric, the real Fibre Channel device is considered
to be imported into this fabric. The presence of a proxy device is required for inter-fabric device
communication. Refer to “Proxy devices” on page 599 for additional information about proxy
devices.

• Proxy PID
A proxy PID is the port ID (PID) of the proxy device. The proxy device appears to the fabric as a
real Fibre Channel device, has a name server entry, and is assigned a valid port ID. The port ID
is relevant only on the fabric in which the proxy device has been created.

Fabric OS Administrator’s Guide
53-1002920-02

597

26

Fibre Channel routing concepts

• Fabric ID (FID)
Every EX_Port and VEX_Port uses the fabric ID (FID) to identify the fabric at the opposite end of
the inter-fabric link. The FID for every edge fabric must be unique from the perspective of each
backbone fabric.

-

If multiple EX_Ports (or multiple VEX_Ports) are attached to the same edge fabric, they
must be configured with the same FID.

-

If EX_Ports and VEX_Ports are attached to different edge fabrics, they must be configured
with a unique FID for each edge fabric.

If two different backbone fabrics are connected to the same edge fabric, the backbone fabric
IDs must be different, but the edge fabric IDs must be the same. If you configure the same
fabric ID for two backbone fabrics that are connected to the same edge fabric, a RASLog
message displays a warning about fabric ID overlap.

NOTE

Backbone fabrics that share connections to the same edge fabrics must have unique
backbone fabric IDs.

• MetaSAN
A metaSAN is the collection of all SANs interconnected with Fibre Channel routers.
A simple metaSAN can be constructed using an FC router to connect two or more separate
fabrics. Additional FC routers can be used to increase the available bandwidth between fabrics
and to provide redundancy.
Figure 87 shows a metaSAN consisting of a host in Edge SAN 1 connected to storage in Edge
SAN 2 through a backbone fabric connecting two FC routers.

598

Fabric OS Administrator’s Guide
53-1002920-02

Fibre Channel routing concepts

26

ISL

FC router

FC router

EX_Port

EX_Port

Backbone
fabric
IFL

IFL
E_Port

E_Port

Edge SAN 1

Edge SAN 2

= LSAN

FIGURE 87

Edge SANs connected through a backbone fabric

• Phantom domains
A phantom domain is a domain emulated by the Fibre Channel router. The FC router can
emulate two types of phantom domains: front phantom domains and translate phantom
domains. For detailed information about phantom domains, refer to “Phantom domains” on
page 601.

Proxy devices
An FC router achieves inter-fabric device connectivity by creating proxy devices (hosts and targets)
in attached fabrics that represent real devices in other fabrics. For example, a host in Fabric 1 can
communicate with a target in Fabric 2 as follows:

• A proxy target in Fabric 1 represents the real target in Fabric 2.
• Likewise, a proxy host in Fabric 2 represents the real host in Fabric 1.
The host discovers and sends Fibre Channel frames to the proxy target. The FC router receives
these frames, translates them appropriately, and then delivers them to the destination fabric for
delivery to the target.
The target responds by sending frames to the proxy host. Hosts and targets are exported from the
edge SAN to which they are attached and, correspondingly, imported into the edge SAN reached
through Fibre Channel routing. Figure 88 illustrates this concept.

Fabric OS Administrator’s Guide
53-1002920-02

599

26

Fibre Channel routing concepts

Proxy host
(imported device)

Host

Proxy target
(imported device)

Target
Fabric 1

Fabric 2
E_Port
IFL
E_Port

EX_Port

IFL

FC router

FIGURE 88

MetaSAN with imported devices

FC-FC routing topologies
The FC-FC routing service provides two types of routing:

• Edge-to-edge
Occurs when devices in one edge fabric communicate with devices in another edge fabric
through one or more FC routers.

• Backbone-to-edge
Occurs when FC routers connect to a common fabric—known as a backbone fabric—through
E_Ports. A backbone fabric can be used as a transport fabric that interconnects edge fabrics.
FC routers also enable hosts and targets in edge fabrics to communicate with devices in the
backbone fabric, known as backbone-to-edge routing. From the perspective of the edge fabric,
the backbone fabric is like any other edge fabric. For the edge fabric and backbone fabric
devices to communicate, the shared devices must be presented to each other's native fabric.
To do so, at least one translate phantom domain is created in the backbone fabric. This
translate phantom domain represents the entire edge fabric. The shared physical devices in
the edge have corresponding proxy devices on the translate phantom domain.
Each edge fabric has one and only one translate phantom domain to the backbone fabric. The
backbone fabric device communicates with the proxy devices whenever it needs to contact the
shared physical devices in the edge. The FC-FC routing service receives the frames from the
backbone switches destined to the proxy devices, and redirects the frames to the actual
physical devices. When connected to edge fabrics, the translate phantom domain can never be
the principal switch of the backbone fabric. Front domains are not created; rather, only
translate phantom domains are created in the backbone fabric.
Devices are exported from the backbone fabric to one or more edge fabrics using LSANs. Refer to
“LSAN zone configuration” on page 620 for more information.

600

Fabric OS Administrator’s Guide
53-1002920-02

Fibre Channel routing concepts

26

Phantom domains
A phantom domain is a domain created by the Fibre Channel router. The FC router creates two
types of phantom domains: front phantom domains and translate phantom domains.
A front phantom domain, or front domain, is a domain that is projected from the FC router to the
edge fabric. There is one front phantom domain from each FC router to an edge fabric, regardless
of the number of EX_Ports connected from that router to the edge fabric. Another FC router
connected to the same edge fabric projects a different front phantom domain.
A translate phantom domain, also referred to as translate domain or xlate domain, is a router
virtual domain that represents an entire fabric. The EX_Ports present xlate domains in edge fabrics
as being topologically behind the front domains; if the xlate domain is in a backbone fabric, then it
is topologically present behind the FC router because there is no front domain in a backbone fabric.
If an FC router is attached to an edge fabric using an EX_Port, it creates xlate domains in the fabric
corresponding to the imported edge fabrics with active LSANs defined. If you import devices into
the backbone fabric, then an xlate domain is created in the backbone device in addition to the one
in the edge fabric.
Figure 89 shows a sample physical topology. This figure shows four FC routers in a backbone fabric
and four edge fabrics connected to the FC routers.
Host

Target 1

E

Target 3

Fabric 2

Fabric 1
E

Target 2

E

E

E
EX

EX

EX

FC router 1

FIGURE 89

EX

EX

FC router 2

Fabric 4

Fabric 3

FC router 3

E

E

EX

EX

FC router 4

Sample topology (physical topology)

Figure 90 shows a phantom topology for the physical topology shown in Figure 89. In this figure,
the dashed lines and shapes represent the phantom topology from the perspective of Fabric 1.
Fabrics 2 and 3 also see phantom topologies, but they are not shown in this example. In this figure,
note the following:

• Front domain 1 and Front domain 2 are front domains for EX_Ports connecting to Fabric 1.
There is one front domain for each FC router that is connected to Fabric 1.

• Xlate domain 1 and Xlate domain 2 represent Fabrics 2 and 3, respectively. No xlate domain is
created for Fabric 4 because there are no LSAN devices in Fabric 4.

• Target 1’, Target 2’, and Target 3’ are proxy devices for Target 1, Target 2, and Target 3,
respectively.

Fabric OS Administrator’s Guide
53-1002920-02

601

26

Fibre Channel routing concepts

Host 1

Fabric 1

Front domain 1
(FC router 1)

Front domain 2
(FC router 2)

Xlate domain 1
(Fabric 2)

Xlate domain 2
(Fabric 3)

Target 1'

FIGURE 90

Target 2'

Target 3'

EX_Port phantom switch topology

All EX_Ports or VEX_Ports connected to an edge fabric use the same xlate domain ID for an
imported edge fabric; this value persists across switch reboots and fabric reconfigurations.
If you lose connectivity to the edge fabric because of link failures or the IFL being disabled, xlate
domains remain visible. This prevents unnecessary fabric disruptions caused by xlate domains
repeatedly going offline and online due to corresponding IFL failures. To remove the xlate domain
from the backbone, refer to “Identifying and deleting stale xlate domains.”
The combination of front domains and xlate domains allows routing around path failures, including
path failures through the routers. The multiple paths to an xlate domain provide additional
bandwidth and redundancy.
There are some differences in how the xlate domain is presented in the backbone fabric. The
backbone xlate domains are topologically connected to FC routers and participate in FC-FC routing
protocol in the backbone fabric. Front domains are not needed in the backbone fabric. As in the
case of an xlate domain in an edge fabric, backbone fabric xlate domains provide additional
bandwidth and redundancy by being able to present themselves as connected to single or multiple
FC routers with each FC router capable of connecting multiple IFLs to edge fabrics.
Use the fcrXlateConfig command to display or assign a preferred domain ID to a translate domain
or, in some scenarios, to prevent the creation of an unnecessary xlate domain.

Identifying and deleting stale xlate domains
If a remote edge fabric goes unreachable, the xlate domains created in other edge fabrics for this
remote edge fabric are retained and not removed unless there is any disruption in the local edge
fabric.
You can use the fcrXlateConfig command to identify and remove these stale xlate domains without
disrupting the fabric.

602

Fabric OS Administrator’s Guide
53-1002920-02

Setting up FC-FC routing

26

1. Connect to the FC router and log in using an account with admin permissions.
2. Enter the fcrXlateConfig --show command to identify any stale xlate domains.
3. Enter the fcrXlateConfig --del command to delete the stale xlate domains.
Example
sw0:root> fcrxlateconfig --show stalexd
Imported FID
Stale XD
Owner Domain
-------------------------------------------------012
002
007 ( this FCR )
sw0:root> fcrxlateconfig --del stalexd 12 2
Xlate domain 2 is deleted

FC router authentication
A Brocade FC router (FCR) is capable of forming a secure link across fabrics. The EX_Port-enabled
router exchanges DH-CHAP information with the edge fabric to enable authentication. Note that
while setting secret keys in the edge switch, the front phantom WWN should be used as the remote
switch WWN in the edge fabric. The front phantom domain's WWN is available through the
portCfgExport port command of the EX_Port connecting to the edge fabric. The FC router switch
should use the edge switch's WWN to configure the secret keys. Refer to “Secret key pairs for
DH-CHAP” on page 249 for more details.
FC-FC routing behaves passively to the authentication requests received from edge fabric switches.
An FC router never initiates authentication on an EX_Port and only responds to the edge fabric
requests.

NOTE
Changing the switch authentication policy mode does not affect online EX_Ports, so it is acceptable
to leave the default Passive policy configured on the FC router while the Active or On policy is
required on the edge switch.

Setting up FC-FC routing
To set up FC-FC routing, perform the following tasks in the order listed.
1. Verify that you have the proper setup for FC-FC routing. (Refer to “Verifying the setup for FC-FC
routing” on page 604.)
2. Assign backbone fabric IDs. (Refer to “Backbone fabric IDs” on page 605.)
3. Configure FCIP tunnels if you are connecting Fibre Channel SANs over IP-based networks.
(Refer to “FCIP tunnel configuration” on page 606.)
4. Configure IFLs for edge and backbone fabric connection. (Refer to “Inter-fabric link
configuration” on page 607.)
5. Modify port cost for EX_Ports, if you want to change from the default settings. (Refer to “FC
router port cost configuration” on page 613.)
6. Enable shortest IFL mode if you want to choose a lowest cost IFL path in the backbone fabric.
(Refer to “Shortest IFL cost configuration” on page 615.)
7.

Configure trunking on EX_Ports that are connected to the same edge fabric. (Refer to “EX_Port
frame trunking configuration” on page 619.)

Fabric OS Administrator’s Guide
53-1002920-02

603

26

Setting up FC-FC routing

8. Configure LSAN zones to enable communication between devices in different fabrics. (Refer to
“LSAN zone configuration” on page 620.)
Refer to Chapter 3, “Performing Advanced Configuration Tasks,” for more details about
configuration options for Brocade Backbones.

Verifying the setup for FC-FC routing
Before configuring a fabric to connect to another fabric, you must perform the following verification
checks on the FC router.
1. Log in to the switch or backbone as admin and enter the version command. Verify that
Fabric OS v7.2.0 or later is installed on the FC router, as shown in the following example.
switch:admin> version
Kernel:
2.6.14.2
Fabric OS: v7.2.0
Made on:
Fri Nov 18 01:15:34 2011
Flash:
Mon Nov 21 20:53:48 2011
BootProm:
1.0.9

2. If you are configuring a backbone, enter the slotShow command to verify that an FX8-24 blade
is present or an 8-Gbps or 16-Gbps port blade is present. The following example shows slots 1,
2, 3, 4, 9, 10, and 12 with 8-Gbps port blades enabled.
switch:admin> slotshow -m
Slot
Blade Type
ID
Model Name
Status
-------------------------------------------------1
SW BLADE
37
FC8-16
ENABLED
2
SW BLADE
37
FC8-16
ENABLED
3
SW BLADE
37
FC8-16
ENABLED
4
SW BLADE
39
FC8-16
ENABLED
5
CORE BLADE
52
CORE8
ENABLED
6
CP BLADE
50
CP8
ENABLED
7
CP BLADE
50
CP8
ENABLED
8
CORE BLADE
52
CORE8
ENABLED
9
SW BLADE
37
FC8-16
ENABLED
10
SW BLADE
55
FC8-32
ENABLED
11
UNKNOWN
VACANT
12
SW BLADE
51
FC8-48
ENABLED

Refer to Chapter 3, “Performing Advanced Configuration Tasks,” for a list of blades and their
corresponding IDs.
3. Enter the licenseShow command to verify that the Integrated Routing license is installed.
switch:admin> licenseshow
S9bddb9SQbTAceeC:
Fabric license
bzbzRcbcSc0c0SY:
Remote Fabric license
RyeSzRScycazfT0G:
Integrated Routing license

If you are connecting to a Fabric OS or M-EOS fabric and the Integrated Routing license is not
installed, you must install it, as described in Chapter 21, “Administering Licensing”. The
Integrated Routing license is not required if you are connecting to a Brocade Network OS fabric.
For configuring EX_Ports on an ICL, both the Integrated Routing license and the ICL POD license
are required.

604

Fabric OS Administrator’s Guide
53-1002920-02

Backbone fabric IDs

26

4. Verify that the Fabric-Wide Consistency Policy is not in “strict” mode by issuing the fddCfg
--showall command. When it is in strict mode, an ACL cannot support Fibre Channel routing in
the fabric.
switch:admin> fddcfg --showall
Local Switch Configuration for all Databases:DATABASE - Accept/Reject
--------------------------------------SCC - accept
DCC - accept
PWD - accept
Fabric-Wide Consistency Policy :- "SCC:S;DCC"

If the Fabric-Wide Consistency Policy has the letter “S” in it in the edge fabric or the backbone
fabric, do not connect the edge fabric to the FC router. The letter “S” (shown in the preceding
sample output) indicates the policy is strict. The fabric-wide policy must be tolerant before you
can connect fabrics to the FC router. Refer to Chapter 8, “Configuring Security Policies” for
information about configuring the Fabric-Wide Consistency Policy.
5. For 8-Gbps platforms, delete fabric mode Top Talker monitors, if they are configured. Refer to
“Deleting all fabric mode Top Talker monitors” on page 567 for instructions.
FC-FC routing and fabric mode Top Talker monitors are not concurrently supported on 8-Gbps
platforms.
FC-FC routing and fabric mode Top Talker monitors are concurrently supported only on the
Brocade 6510 and 6520 switches, and on the Brocade DCX Backbone family with only
16-Gbps-capable ports.

Backbone fabric IDs
If your configuration has only one backbone fabric, then you do not need to assign a backbone
fabric ID because the backbone fabric ID in this situation defaults to a value of 128. The default
backbone fabric ID is 1 if Virtual Fabrics is disabled.
All switches in a backbone fabric must have the same backbone fabric ID. You can configure the
backbone fabric ID using the fcrConfigure command. The backbone fabric ID must be unique from
the perspective of every attached edge fabric. Fabric ID changes made on a switch are not
propagated to other switches in the backbone fabric. Rather, the backbone fabric administrator is
responsible for making sure that all switches in the backbone have the same fabric ID. Because
fabric IDs are used heavily by the routing protocol between the Fibre Channel routers, using the
wrong fabric ID can affect both edge-to-edge and backbone-to-edge routing.
In addition to ensuring that the backbone fabric IDs are the same within the same backbone, you
must make sure that when two different backbones are connected to the same edge fabric, the
backbone fabric IDs are different, but the edge fabric ID should be the same. Configuration of two
backbones with the same backbone fabric ID that are connected to the same edge is invalid. In this
configuration, a RASLog message displays a warning about fabric ID overlap. When two backbone
fabrics are not connected to the same edge, they can have the same backbone fabric ID.

ATTENTION
In a multi-switch backbone fabric, modification of the FID within the backbone fabric will cause
disruption to local traffic.

Fabric OS Administrator’s Guide
53-1002920-02

605

26

FCIP tunnel configuration

Assigning backbone fabric IDs
1. Log in to the switch or backbone.
2. Enter the switchDisable command if EX_Ports are online.
3. Enter the fosConfig --disable fcr command to disable the FC-FC routing service.
The default state for the FC router is disabled.
4. Enter the fcrConfigure --bbfid command. At the prompt, enter the fabric ID, or press Enter to
keep the current fabric ID, which is displayed in brackets.
5. Verify the backbone fabric ID is different from that set for edge fabrics.
Multiple FC routers attached to the same backbone fabric must have the same backbone
fabric ID.
6. Enter the fosConfig --enable fcr command.
7.

Enter the switchEnable command.

Example
switch:admin> switchdisable
switch:admin> fosconfig --disable fcr
FC Router service is disabled
switch:admin> fcrconfigure --bbfid
FC Router parameter set.  to skip a parameter
Please make sure new Backbone Fabric ID does not conflict with any configured
EX-Port's Fabric ID
Backbone fabric ID: (1-128)[128]
switch:admin> fosconfig --enable fcr
FC Router service is enabled
switch:admin> switchenable

FCIP tunnel configuration
The optional Fibre Channel over IP (FCIP) Tunneling Service enables you to use “tunnels” to
connect instances of Fibre Channel SANs over IP-based networks to transport all Fibre Channel ISL
and IFL traffic. FCIP is a prerequisite for configuring VEX_Ports; if you are only using FC_Ports, then
there is no need to configure FCIP tunnels.
If using FCIP in your FC-FC routing configuration, you must first configure FCIP tunnels. Once a
tunnel is created, it defaults to a disabled state. Then configure the VE_Port or VEX_Port. After the
appropriate ports are configured, enable the tunnel.

NOTE
FCIP tunnel configuration is applicable only to Fabric OS fabrics and does not apply to Brocade
Network OS or M-EOS fabrics.
Refer to the Fibre Channel over IP Administrator’s Guide for instructions on how to configure FCIP
tunnels.

606

Fabric OS Administrator’s Guide
53-1002920-02

Inter-fabric link configuration

26

Inter-fabric link configuration
Configuring an inter-fabric link (IFL) involves disabling ports and cabling them to other fabrics,
configuring those ports for their intended uses, and then enabling the ports.
Before configuring an inter-fabric link, be aware that you cannot configure both IFLs (EX_Ports,
VEX_Ports) and ISLs (E_Ports) from a backbone fabric to the same edge fabric.

ATTENTION
To ensure that fabrics remain isolated, disable the port prior to inserting the cable. If you are
configuring an EX_Port, disable the port prior to making the connection.

Configuring an IFL for both edge and backbone connections
1. On the FC router, disable the port that you are configuring as an EX_Port (the one connected to
the Fabric OS switch) by issuing the portDisable command.
switch:admin> portdisable 7/10

You can verify that the port has been disabled by issuing the portShow command for the port.
2. Configure each port that connects to an edge fabric as an EX_Port or VEX_Port. Note the
following:

• portCfgVEXPort works only on VE_Ports.
• portCfgEXPort (only on the FC ports on the FC router) commands work only on ports that
are capable of FC-FC routing.
Use the portCfgEXPort or portCfgVEXPort command to:

• Enable or disable EX_Port or VEX_Port mode.
• Set the fabric ID (avoid using fabric IDs 1 and 128, which are the default IDs for backbone
connections).

• Configure an EX_Port to connect to a Brocade Network OS fabric (portCfgEXPort only).
The following example configures an EX_Port and assigns a Fabric ID of 30 to port 10.
switch:admin> portcfgexport 7/10 -a 1 -f 30
switch:admin> portcfgexport 7/10
Port
7/10
info
Admin:
enabled
State:
NOT OK
Pid format:
Not Applicable
Operate mode:
Brocade Native
Edge Fabric ID:
30
Preferred Domain ID:
160
Front WWN:
50:06:06:9e:20:38:6e:1e
Fabric Parameters:
Auto Negotiate
R_A_TOV:
Not Applicable
E_D_TOV:
Not Applicable
Authentication Type: None
DH Group: N/A
Hash Algorithm: N/A
Edge fabric's primary wwn: N/A
Edge fabric's version stamp: N/A

This port can now connect to another switch.

Fabric OS Administrator’s Guide
53-1002920-02

607

26

Inter-fabric link configuration

The following example configures an EX_Port for connecting to a Brocade Network OS fabric.
The -m 5 option indicates Network OS connectivity.
switch:admin> portcfgexport 1/5 -a 1 -f 15 -m 5
switch:admin> portcfgexport 1/5
Port
1/5
info
Admin:
enabled
State:
NOT OK
Pid format:
Not Applicable
Operate mode:
Brocade NOS
Edge Fabric ID:
15
(output truncated)

For related FC-FC routing commands, refer to fcrEdgeShow, fcrXlateConfig, fcrConfigure, and
fcrProxyConfig in the Fabric OS Command Reference.
A Fibre Channel router can interconnect multiple fabrics. EX_Ports or VEX_Ports attached to
more than one edge fabric must configure a different fabric ID for each edge fabric.
3. (Optional) Configure FC router port cost if you want to change the default values. For
information about using FC router port cost operations, refer to “FC router port cost
configuration” on page 613.
4. (Optional) Set up ISL or EX_Port trunking. For information on trunking setup, refer to
“Configuring EX_Port trunking” on page 578.
5. Enter the portEnable command to enable the ports that you disabled in step 1.
switch:admin> portenable 7/10

6. Physically attach ISLs from the Fibre Channel router to the edge fabric.
7.

Enter the portCfgShow command to view ports that are persistently disabled.
FC ports on the Brocade 7800 switches and FX8-24 blades are configured as persistently
disabled by default to avoid inadvertent fabric merges when installing a new FC router.
switch:admin> portcfgshow
Area Number:
Speed Level:
Trunk Port
Long Distance
VC Link Init
Locked L_Port
Locked G_Port
Disabled E_Port
ISL R_RDY Mode
RSCN Suppressed
Persistent Disable
NPIV capability
EX Port
Mirror Port
FC Fastwrite

7/10
74
AUTO
OFF
OFF
OFF
OFF
OFF
OFF
OFF
OFF
OFF
ON
ON
ON
ON

8. After identifying such ports, enter the portCfgPersistentEnable command to enable the port,
and then the portCfgShow command to verify the port is enabled.
switch:admin> portcfgpersistentenable 7/10
switch:admin> portcfgshow 7/10
Area Number:
74

608

Fabric OS Administrator’s Guide
53-1002920-02

Inter-fabric link configuration

Speed Level:
Trunk Port
Long Distance
VC Link Init
Locked L_Port
Locked G_Port
Disabled E_Port
ISL R_RDY Mode
RSCN Suppressed
Persistent Disable
NPIV capability
EX Port
Mirror Port
FC Fastwrite

26

AUTO
OFF
OFF
OFF
OFF
OFF
OFF
OFF
OFF
OFF
ON
ON
ON
ON

9. Enter either the portCfgEXPort or portShow command to verify that each port is configured
correctly.
switch:admin> portcfgexport 7/10
Port
7/10
info
Admin:
enabled
State:
NOT OK
Pid format:
Not Applicable
Operate mode:
Brocade Native
Edge Fabric ID:
30
Preferred Domain ID:
160
Front WWN:
50:06:06:9e:20:38:6e:1e
Fabric Parameters:
Auto Negotiate
R_A_TOV:
Not Applicable
E_D_TOV:
Not Applicable
Authentication Type: None
DH Group: N/A
Hash Algorithm: N/A
Edge fabric's primary wwn: N/A
Edge fabric's version stamp: N/A

switch:admin_06> portshow 7/10
portName:
portHealth: OFFLINE
Authentication: None
EX_Port Mode:
Fabric ID:
Front Phantom:
Fabric params:

Enabled
30
state = Not OK
R_A_TOV: 0

Pref Dom ID: 160
E_D_TOV: 0
PID fmt: auto

Authentication Type: None
Hash Algorithm: N/A
DH Group: N/A
Edge fabric's primary wwn: N/A
Edge fabric's version stamp: N/A
portDisableReason: None
portCFlags: 0x1
portFlags: 0x1
PRESENT U_PORT EX_PORT
portType: 10.0

Fabric OS Administrator’s Guide
53-1002920-02

609

26

Inter-fabric link configuration

portState: 2
Offline
portPhys: 2
No_Module
portScn:
0
port generation number:
0
portId:
014a00
portIfId:
4372080f
portWwn:
20:4a:00:60:69:e2:03:86
portWwn of device(s) connected:
Distance: normal
portSpeed: N4Gbps
LE domain: 0
FC Fastwrite: ON
Interrupts:
Unknown:
Lli:
Proc_rqrd:
Timed_out:
Rx_flushed:
Tx_unavail:
Free_buffer:
Overrun:
Suspended:
Parity_err:
2_parity_err:
CMI_bus_err:

0
0
0
0
0
0
0
0
0
0
0
0
0

Link_failure:
Loss_of_sync:
Loss_of_sig:
Protocol_err:
Invalid_word:
Invalid_crc:
Delim_err:
Address_err:
Lr_in:
Lr_out:
Ols_in:
Ols_out:

0
0
2
0
0
0
0
0
0
0
0
0

Frjt :
Fbsy :

0
0

Port part of other ADs: No

10. Enter the switchShow command to verify the EX_Port (or VEX_Port), edge fabric ID, and name
of the edge fabric switch (containing the E_Port or VE_Port) are correct.
11. Enter the fcrFabricShow command to view any edge fabric switch names and ensure links are
working as expected.

NOTE

The fcrFabricShow command displays the static IPv6 addresses for each FC router and each
edge fabric switch connected to the EX_Ports.
switch:admin> fcrfabricshow
FCR WWN: 10:00:00:05:1e:13:59:00, Dom ID: 2, Info: 10.32.156.52
1080::8:800:200C:1234/64,"Spirit-2"
"fcr_5300"
EX_Port FID Neighbor Switch Info (WWN, enet IP, name)
--------------------------------------------------------------7 10 10:00:00:05:1e:34:11:e5 10.32.156.33 "5300" 1080::8:8FF:FE0C:417A/64
4 116 10:00:00:05:1e:37:00:44 10.32.156.34 "5300"
FCR WWN: 10:00:00:05:1e:12:e0:00, Dom ID: 100, Info:10.32.156.50
1080::8:60F:FE0C:456A/64
"fcr_5300"
EX_Port FID Neighbor Switch Info (WWN, enet IP, name)
-----------------------------------------------------------------------4 95 10:00:00:05:1e:37:00:45 10.32.156.31 "5300"
FCR WWN: 10:00:00:05:1e:12:e0:00, Dom ID: 100, Info: 10.32.156.50,
"fcr_Brocade 5300"
EX_Port FID Neighbor Switch Info (WWN, enet IP, name)
-----------------------------------------------------------------------4 95 10:00:00:05:1e:37:00:45 10.32.156.31 "Brocade 5300"

610

Fabric OS Administrator’s Guide
53-1002920-02

Inter-fabric link configuration

26

5 95 10:00:00:05:1e:37:00:45 10.32.156.31 "Brocade 5300"
6 95 10:00:00:05:1e:37:00:45 10.32.156.31 "Brocade 5300"

12. Enter the iflshow command to display the FC router details and ensure the fabric is functioning
correctly.
switch:FID128:root> iflshow
E-Port
EX-Port
FCR-WWN
FCR-FID
FCR-Name
Speed
BW
-------------------------------------------------------------------------------1 : 350 --> 12
10:00:08:00:88:04:93:94
39
fcr_sw
4G
8G

TRUNK

Configuring EX_Ports on an ICL
The following restrictions apply when configuring EX_Ports on an ICL:

• Both the active and standby CP must be running Fabric OS 7.2.0 or later.
• EX_Ports on ICLs is supported only on the Brocade DCX 8510-4 and DCX 8510-8.
• EX-Port on ICL is supported only in DCX 8510-8 and DCX 8510-4 when all the port blades in the
chassis belong to one of these blade types: FC16-32, FC16-48, FC8-32E, FC8-48E).

NOTE

Device discovery will not occur properly with ICL EX-Port connected edge fabrics if the FC
Router has unsupported blades which are not mentioned in the above restriction.

• Switches must have Virtual Fabrics enabled.
• EX_Ports on ICLs are allowed only on the base switch. All user ports must be in the base
switch.

• You must be in Brocade native mode.
• You cannot configure a VEX_Port on an ICL
• The ICLs must be in a symmetric topology. For more information on topologies, refer to the
Inter-Chassis Links chapter.

• Bladeless configuration (where the chassis has core blades but no application blades) is not
supported.
To configure EX_Ports on an ICL, perform the following steps:
1. Configure an 8-bit zero-based Dynamic Area mode or a 10-bit Dynamic Area mode on the FC
router to bring up ICL EX_Ports.
Configure...
Enable a 256 Area Limit
(0 = No,
1 = Zero Based Area Assignment,
2 = Port Based Area Assignment): (0..2) [0]

The following options are available:

• 0: 10-bit Dynamic Area Mode. ICL EX_Ports, F_Ports, and regular EX_Ports have 8-bit
areas.

• 1: 8-bit Dynamic Area Mode. ICL_EX ports have 8-bit areas.
• 2: Not permitted if ICL ports are present in the logical switch.
Fabric OS Administrator’s Guide
53-1002920-02

611

26

Inter-fabric link configuration

2. On the FC router, disable all QSFP ports by issuing the portDisable command.
switch:admin> portdisable 6/20-23

You can verify that all ports have been disabled by issuing the portShow command for the
ports.
3. Configure EX_Ports on the ICL by issuing the portCfgEXPort command. If you configure EX_Port
on one of the QSFP ports, the configuration is automatically propagated to the other 3 QSFP
ports.
The following example configures EX_Port on one of the QSFP ports.
switch:admin> portcfgexport 6/20 -a 1 -f 45
2013/04/25-21:21:54, [FCR-1071], 29805, SLOT
is changed from non FCR port to FCR port.
2013/04/25-21:21:54, [FCR-1071], 29806, SLOT
is changed from non FCR port to FCR port.
2013/04/25-21:21:55, [FCR-1071], 29807, SLOT
is changed from non FCR port to FCR port.
2013/04/25-21:21:55, [FCR-1071], 29808, SLOT
is changed from non FCR port to FCR port.

4 | FID 2, INFO, Pluto, Port 6/20
4 | FID 2, INFO, Pluto, Port 6/21
4 | FID 2, INFO, Pluto, Port 6/22
4 | FID 2, INFO, Pluto, Port 6/23

4. (Optional) Configure FC router port cost if you want to change the default values. For
information about using FC router port cost operations, refer to “FC router port cost
configuration” on page 613.
5. Enter the portEnable command to enable the QSFP ports that you disabled in step 1.
switch:admin> portenable 6/20-23

6. Enter either the portCfgEXPort or portShow command to verify that each port is configured
correctly.
switch:admin> portcfgexport 6/20
Port
6/20 info
Admin:
enabled
State:
OK
Pid format:
Not Applicable
Operate mode:
Brocade Native
Edge Fabric ID:
50
Preferred Domain ID:
160
Front WWN:
50:00:51:e7:54:c0:0e:32
Fabric Parameters:
Auto Negotiate
R_A_TOV:
Not Applicable
E_D_TOV:
Not Applicable
Authentication Type: None
DH Group: N/A
Hash Algorithm: N/A
Edge fabric's primary wwn: N/A
Edge fabric's version stamp: N/A

7.

Enter the switchShow command to verify the EX_Port and edge fabric ID are correct.

8. Enter the fcrEdgeShow command to display the FIDs of all configured EX_Ports.
switch:admin> fcredgeshow
FID EX-port E-port Neighbor Switch (PWWN, SWWN )
-----------------------------------------------------------------------50
3/12
1180
50:00:51:e4:8f:8f:74:9c 10:00:00:05:1e:48:f8:00
50
3/13
1181
50:00:51:e4:8f:8f:74:9d 10:00:00:05:1e:48:f8:00

612

Fabric OS Administrator’s Guide
53-1002920-02

FC router port cost configuration

50
50

3/14
3/15

1182
1183

50:00:51:e4:8f:8f:74:9e
50:00:51:e4:8f:8f:74:9f

26

10:00:00:05:1e:48:f8:00
10:00:00:05:1e:48:f8:00

9. Enter the fcrIclPathBwMonitor command to monitor and report path bandwidth imbalances for
each edge fabric.
The following example enables the monitoring and reporting of bandwidth balances and
imbalances, and displays the ICL path bandwidth state for each fabric.
switch:admin> fcriclpathbwmonitor --show
ICL Path Bandwidth state :Disabled
switch:admin> fcriclpathbwmonitor --enable
ICL bandwidth balance functionality is enabled
Pluto:FID2:root> fcriclpathbwmonitor --show
ICL Path Bandwidth state :Enabled
FABRIC
SLOT-3 BW
SLOT-6 BW
STATE
===============================================
48
128
128
BALANCED
126
64
128
UNBALANCED

FC router port cost configuration
The FC router port cost is set automatically. This section provides information about the router port
cost and describes how you can modify the cost for a port if you want to change the default value.
FC routers optimize the usage of the router port links by directing traffic to the link with the
smallest router port cost. The FC router port cost is similar to the link cost setting available on
E_Ports, which allows you to customize traffic flow. The router port link cost values are either 0,
1,000, or 10,000. The router module chooses the router port path based on the lowest cost for
each FID connection. If multiple paths exist where one path costs less than the others, then the
lowest cost path is used. If exchange-based routing has not been disabled and multiple paths exist
with the same lowest cost, there will be load sharing over these paths.
The FC router port cost feature optimizes the usage of the router port links by directing the traffic to
a link with a smaller cost.
Every IFL has a default cost. The default router port cost values are:

• 1,000 for a legacy (v5.1 or XPath FCR) IFL
• 1,000 for an EX_Port IFL
• 10,000 for a VEX_Port IFL
The FC router port cost settings are 0, 1,000, or 10,000. If the cost is set to 0, the default cost will
be used for that IFL. The FC router port cost is persistent and is saved in the existing port
configuration file.
FC router port cost is passed to other routers in the same backbone. Link costs from the front
domain to the translate (xlate) domain remain at 10,000. You can use the lsDbShow command
from the edge fabric to display these link costs.

Fabric OS Administrator’s Guide
53-1002920-02

613

26

FC router port cost configuration

Port cost considerations
The router port cost has the following considerations:

• Router port sets are defined as follows:
- 0–7 and FCIP Tunnel 16–23
- 8–15 and FCIP Tunnel 24–31
• The router port cost does not help distinguish one IFL (or EX_ and VEX_Port link) from another,
if all the IFLs are connected to the same port set. Therefore, if you connect IFL1 and IFL2 to the
same edge fabric in port set 0–7 and then configure them to different router port costs, traffic
is still balanced across all the IFLs in the same port set.

• Use proper SAN design guidelines to connect the IFLs to different port sets for effective router
port cost use. For example, if both a low-speed IFL and a high-speed IFL are going to the same
edge fabric, connect the lower router cost IFL to a separate port group (for example, ports 0–7)
than the higher router cost IFL (for example, ports 8–15). For VEX_Ports, you would use ports
in the range of 16–23 or 24–31.
You can connect multiple EX_Ports or VEX_Ports to the same edge fabric. The EX_Ports can all be
on the same FC router, or they can be on multiple routers. Multiple EX_Ports create multiple paths
for frame routing. Multiple paths can be used in two different, but compatible, ways:

• Failing over from one path to another.
• Using multiple paths in parallel to increase effective data transmission rates.
EX_Ports and VEX_Ports, when connected, are assigned different router port costs and traffic will
flow only through the EX_Ports. Routing failover is automatic, but it can result in frames arriving out
of order when frames take different routes. The FC router can force in-order delivery, although
frame delivery is delayed immediately after the path failover.
Source EX_Ports can balance loads across multiple destination EX_Ports attached to the same
edge fabric using exchange IDs from the routed frames as keys to distribute the traffic.

Setting router port cost for an EX_Port
The router port cost value for an EX_Port is set automatically when the EX_Port is created. However,
you can modify the cost for that port. You can configure the EX_Port or VEX_Port with values of
either 1,000 or 10,000. If you want to differentiate between two EX_Port links with different
speeds, you can assign 1,000 to one link and 10,000 to the other link.
For details about the use of any of the following commands, refer to the Fabric OS Command
Reference.
1. Enter the portDisable command to disable any port on which you want to set the router port
cost.
switch:admin> portdisable 7/10

2. Enable EX_Port or VEX_Port mode with the portCfgEXPort or portCfgVEXPort command.
switch:admin> portcfgexport 7/10 -a 1

614

Fabric OS Administrator’s Guide
53-1002920-02

Shortest IFL cost configuration

26

3. Enter the fcrRouterPortCost command to display the router port cost for each EX_Port.
switch:admin> fcrrouterportcost
Port
Cost
-----------------------7/3
1000
7/4
1000
7/9
1000
7/10
1000
7/13
1000
10/0
1000

You can also use the fcrRouteShow command to display the router port cost.
To display the router port cost for a single EX_Port, enter the fcrRouterPortCost command with
a port and slot number.
switch:admin> fcrrouterportcost 7/10
Port
Cost
-----------------------7/10
1000

4. Enter the appropriate form of the fcrRouterPortCost command based on the task you want to
perform:

• To set the router port cost for a single EX_Port, enter the command with a port and slot
number and a specific cost:
switch:admin> fcrrouterportcost 7/10 10000

• To set the cost of the EX_Port back to the default, enter a cost value of 0:
switch:admin> fcrrouterportcost 7/10 0

5. Enter the portEnable command to enable the ports that you disabled in step 1.
switch:admin> portenable 7/10

Shortest IFL cost configuration
You can direct traffic flow to take the shortest path between host and target when multiple FC
routers in the backbone are connected to an edge fabric. Shortest inter-fabric link mode is disabled
by default. When you enable shortest IFL mode, an FC router can choose a lowest cost ISL path in
the backbone fabric. This feature is useful when an FC router has multiple connections to the
source edge fabric, and the backbone fabric has multiple FC routers connected through FCIP links
(VE_Ports) and FC links (E_Ports). The selection of a low cost path depends on individual ISL link
cost settings in the backbone fabric. Traffic originating from a domain in an edge fabric can choose
any equal cost path in order to reach the destination edge fabric. This traffic can be transmitted
through high cost paths, such as FCIP links, within the backbone fabric even though low cost paths,
such as FC links, are present.
When the shortest IFL mode is enabled, each FC router calculates the shortest path for each of its
locally-connected source edge fabrics to a remote destination edge fabric that is connected to
another FCR in the same backbone fabric. The FC router performs this calculation by identifying all
paths in the backbone fabric that can connect the source edge fabric to the destination edge
fabric. The cumulative ISL link cost for each path is then calculated:

Fabric OS Administrator’s Guide
53-1002920-02

615

26

Shortest IFL cost configuration

• For any path for which the cumulative ISL link cost of the path is greater than or equal to
10,000, the FC router sets the link cost from the front domain to translate the domain as
10,001.

• For any path for which the cumulative ISL link cost of the path is less than 10,000, the link cost
from front domain to translate domain will remain at 10,000, which is the shortest IFL path.

NOTE
The shortest IFL solution is applicable only when the edge fabric has multiple FC router connections
and the backbone fabric has at least one available low cost path. Shortest IFL mode should be
enabled on all FC routers in the backbone fabric for this feature to work correctly.
Figure 91 shows a shortest IFL solution. In order for Host 12 in Domain 1 in Edge fabric 1 to reach
Target 2 in Domain 3 in Edge fabric 3, traffic can choose one of two paths in the backbone fabric:

• Path 1: Edge fabric 1-FCR1 – FCR2 –FCR3 –Edge fabric 3.
• Path 2: Edge fabric 1-FCR4 – FCR3 –Edge fabric 3.
The second path is identified as the low cost path by setting the ISL link cost to 5,000. The first
path is considered the high cost path where the cumulative ISL cost is 10,000.

NOTE
Link cost should be set in the direction of traffic flow from the source edge fabric to the destination
edge fabric.

616

Fabric OS Administrator’s Guide
53-1002920-02

Shortest IFL cost configuration

FIGURE 91

26

Shortest IFL solution

Configuring shortest IFL cost
1. Enter the fcrFabricShow command to view the FC routers on the backbone fabric.
switch:admin>fcrfabricshow
FC Router WWN: 10:00:00:05:1e:58:bd:69, Dom ID:
10,
Info: 10.17.33.59, “DID_10"
EX_Port FIDNeighbor Switch Info (enet IP, WWN, name)
-----------------------------------------------------------------------34 110.17.33.68
10:00:00:05:1e:61:28:22
"DID_4_1"
switch:admin>fcrfabricshow
FC Router WWN: 10:00:00:05:1e:58:be:69, Dom ID:
20,
Info: 10.17.33.60, "DID_20"
EX_Port FIDNeighbor Switch Info (enet IP, WWN, name)
-----------------------------------------------------------------------2210.17.33.88
10:00:00:05:33:00:90:48
"DID_1_2"
switch:admin>fcrfabricshow
FC Router WWN: 10:00:00:05:1e:58:be:68, Dom ID:
30,
Info: 10.17.33.61, "DID_30"
EX_Port FIDNeighbor Switch Info (enet IP, WWN, name)
-----------------------------------------------------------------------31 310.17.33.153
10:00:00:05:33:7e:e8:7c
"DID_3_3"

Fabric OS Administrator’s Guide
53-1002920-02

617

26

Shortest IFL cost configuration

switch:admin>fcrfabricshow
FC Router WWN: 10:00:00:05:1e:58:be:67, Dom ID:
40,
Info: 10.17.33.62, "DID_40"
EX_Port FIDNeighbor Switch Info (enet IP, WWN, name)
-----------------------------------------------------------------------34 110.17.33.68
10:00:00:05:1e:61:28:22
"DID_4_1"

2. Enter the islshow command to identify the connections between the FC routers and the
destination edge fabric.
switch:admin>islshow
1: 10->1010:00:00:05:1e:58:be:69
QOS

20 DID_20

sp:

8.000G bw:

8.000G TRUNK

3. Identify all paths in the backbone fabric through which traffic can flow from the source edge
fabric to the destination edge fabric by examining the output generated in step 1 and step 2.
The following example step shows how to identify the number of paths from the source edge
fabric 1 to the destination edge fabric 2 in the backbone fabric based on the configuration
shown in step 1 and step 2:

• The path from FC router Domain ID 10 to FC router Domain ID 20.
• The path from FC router Domain 20: This FC router has no local connection to source edge
fabric 1, so cannot be considered for path identification.

• The path from FC router Domain 30: This FC router has no local connection to source edge
fabric 1, so cannot be considered for path identification.

• The path from FC router Domain ID 40 to FC router Domain ID 30 to FC router Domain ID
20.
Therefore, there are two available paths in the backbone fabric through which traffic can flow
from the source edge fabric 1 to the destination edge fabric 2.
4. Enter the linkcost command to obtain the cumulative ISL cost for each path identified by
adding all individual link costs, and select one single path to be the only low cost path.

• The following example shows the cumulative ISL cost for the first path identified in step 3:
from FC router ID Domain 10 to FC router ID Domain 20.
switch:admin>linkcost 10
Interface10 (E_PORT)Cost

500

The cumulative link cost for this path is 500. This path is now known as path 1.

• The following example shows the cumulative ISL cost for the second path identified in
step 3: first from FC router ID Domain 40 to FC router ID Domain 30, and then from FC
router Domain ID 30 to FC router ID Domain 20.
switch:admin>linkcost 10
Interface10 (E_PORT)Cost

500

switch:admin>linkcost 10
Interface10 (E_PORT)Cost

500

The cumulative link cost for this path is 1000. This path is now known as path 2.
Path 1 is selected as the low cost path.
5. Enter the linkcost command to set low cost values, ensuring that the cumulative ISL cost for
the selected path is lower than that of all other paths. A low cost path should have a
cumulative ISL cost of less than 10,000.

618

Fabric OS Administrator’s Guide
53-1002920-02

EX_Port frame trunking configuration

26

• In the following example, the ISL link cost of path 2 from FC router ID Domain 40 to FC
router Domain ID 30 is modified.
switch:admin>linkcost 10 5000
Interface10 (E_PORT)Cost
5000

• In the following example, the ISL link cost of path 2 from FC router Domain 30 to FC router
Domain 20 is modified.
switch:admin>linkcost 10 5000
Interface10 (E_PORT)Cost
5000

The modified cumulative link cost for path 2 is 10,000.
6. Enter the fcrConfigure --enable -shortestifl command to enable the shortest IFL mode in all the
FC routers in the backbone fabric.
switch:admin> fcrconfigure --enable -shortestifl
Shortest IFL path is enabled

7.

Enter the fcrConfigure -show command to verify that the shortest IFL mode is enabled in all the
FC routers in the backbone fabric.
switch:admin> fcrconfigure --show
Backbone fabric ID: 128
Shortest IFL feature is enabled

8. Verify that the link cost of both the front domain and translate domain in the source edge
fabric have been modified using the lsDbShow command.
linkCnt = 2, flags = 0x0
LinkId = 160, out port = 160 rem port =
= 1
LinkId = 161, out port = 161, rem port =
= 1

241, cost = 10001, costCnt = 0, type
242, cost = 10000, costCnt = 0, type

EX_Port frame trunking configuration
You can configure EX_Ports to use frame-based trunking just as you do regular E_Ports. EX_Port
frame trunking support is designed to provide the best utilization and balance of frames
transmitted on each link between the FC router and the edge fabric. You should trunk all ports
connected to the same edge fabrics.
The FC router front domain has a higher node WWN—derived from the FC router—than that of the
edge fabric. Therefore, the FC router front domain initiates the trunking protocol on the EX_Port.
After initiation, the first port from the trunk group that comes online is designated as the master
port. The other ports that come online on the trunk group are considered the slave ports. Adding or
removing a slave port does not cause frame drop; however, removing a slave port causes the loss
of frames in transit.
The restrictions for EX_Port frame trunking are the same as for E_Ports—all the ports must be
adjacent to each other using the clearly marked groups on the front of the product.

ATTENTION
This feature should be enabled only if the entire configuration is running Fabric OS v5.2.0 or later.

Fabric OS Administrator’s Guide
53-1002920-02

619

26

LSAN zone configuration

If router port cost is used with EX_Port trunking, the master port and slave ports share the router
port cost of the master port.
For information about setting up E_Port trunking on an edge fabric, refer to Chapter 24, “Managing
Trunking Connections”.

LSAN zone configuration
An LSAN consists of zones in two or more edge or backbone fabrics that contain the same devices.
LSANs provide selective device connectivity between fabrics without forcing you to merge those
fabrics. FC routers provide multiple mechanisms to manage inter-fabric device connectivity through
extensions to existing switch management interfaces. You can define and manage LSANs using
Brocade Advanced Zoning.

NOTE

For performance reasons, Brocade recommends that you do not configure LSANs for device sharing
between Fabric OS fabrics until after you activate the Integrated Routing license.

Use of Admin Domains with LSAN zones and FC-FC routing
You can create LSAN zones as a physical fabric administrator or as an individual Admin Domain
(AD) administrator. The LSAN zone can be part of the root zone database or the AD zone database.
FC-FC Routing harvests the LSAN zones from all administrative domains. If both edge fabrics have
the matching LSAN zones and both devices are online, FC-FC routing triggers a device import. To
support legacy applications, WWNs are reported based on the Admin Domain context. As a result,
you must not use the network address authority (NAA) field in the WWN to detect an FC router.
LSAN zone enforcement in the local fabric occurs only if the Admin Domain member list contains
both of the devices (local and imported device) specified in the LSAN zone.
For more information, refer to Chapter 20, “Managing Administrative Domains”.

Zone definition and naming
Zones are defined locally on a switch or backbone. Names and memberships, with the exception of
hosts and targets exported from one fabric to another, do not need to be coordinated with other
fabrics. For example, in Figure 87 on page 599, when the zones for Edge SAN 1 are defined, you do
not need to consider the zones in Edge SAN 2, and vice versa.
Zones that contain hosts and targets that are shared between the two fabrics must be explicitly
coordinated. To share devices between any two fabrics, you must create an LSAN zone in both
fabrics containing the port WWNs of the devices to be shared. Although an LSAN is managed using
the same tools as any other zone on the edge fabric, two behaviors distinguish an LSAN from a
conventional zone:

• A required naming convention. The name of an LSAN begins with the prefix “LSAN_”. The LSAN
name is not case-sensitive; for example, lsan_ is equivalent to LSAN_, Lsan_, and so on.

• Members must be identified by their port WWN because port IDs are not necessarily unique
across fabrics. The names of the zones need not be explicitly the same, and membership lists
of the zones need not be in the same order.

620

Fabric OS Administrator’s Guide
53-1002920-02

LSAN zone configuration

26

NOTE

The "LSAN_" prefix must appear at the beginning of the zone name. LSAN zones may not be
combined with QoS zones. Refer to “QoS zones” on page 419 for more information about the naming
convention for QoS zones.
To enable device sharing across multiple fabrics, you must create LSAN zones on the edge fabrics
(and optionally on the backbone fabric as well), using normal zoning operations to create zones
with names that begin with the special prefix “LSAN_”, and adding host and target port WWNs from
both local and remote fabrics to each local zone as desired. Zones on the backbone and on
multiple edge fabrics that share a common set of devices will be recognized as constituting a single
multi-fabric LSAN zone, and the devices that they have in common will be able to communicate
with each other across fabric boundaries.

LSAN zones and fabric-to-fabric communications
Zoning is enforced by all involved fabrics; any communication from one fabric to another must be
allowed by the zoning setup on both fabrics. If the LSANs are under separate administrative control,
then separate administrators maintain access control.

Controlling device communication with the LSAN
The following procedure illustrates how LSANs control which devices can communicate with each
other. The procedure shows the creation of two LSANs (called lsan_zone_fabric75 and
lsan_zone_fabric2), which involve the following devices and connections:

•
•
•
•
•
•
•

Switch1 and the host in fabric75.
Switch2, Target A, and Target B in fabric2.
Switch1 is connected to the FC router using an EX_Port or VEX_Port.
Switch2 is connected to the FC router using another EX_Port or VEX_Port.
Host has WWN 10:00:00:00:c9:2b:c9:0c (connected to switch1).
Target A has WWN 50:05:07:61:00:5b:62:ed (connected to switch2).
Target B has WWN 50:05:07:61:00:49:20:b4 (connected to switch2).

1. Log in as admin and connect to switch1.
2. Enter the nsShow command to list the WWN of the host (10:00:00:00:c9:2b:c9:0c).

NOTE

The nsShow output displays the LSAN zone status of a device, the port WWN, and the node
WWN; the port WWN must be used for LSANs.
switch:admin> nsshow
{
Type Pid
COS
PortName
NodeName
TTL(sec)
N
060f00;
2,3;
10:00:00:00:c9:2b:c9:0c;
20:00:00:00:c9:2b:c9:0c; na
FC4s: FCP
NodeSymb: [35] "Emulex LP9002 FV3.91A3 DV5-5.20A6 "
Fabric Port Name: 20:0f:00:05:1e:37:00:44
Permanent Port Name: 10:00:00:00:c9:2b:c9:0c
LSAN: Yes
The Local Name Server has 1 entry }

Fabric OS Administrator’s Guide
53-1002920-02

621

26

LSAN zone configuration

3. Enter the zoneCreate command to create the LSAN lsan_zone_fabric75, which includes the
host.
switch:admin> zonecreate "lsan_zone_fabric75", "10:00:00:00:c9:2b:c9:0c"

4. Enter the zoneAdd command to add Target A to the LSAN.
FID75Domain5:admin> zoneadd "lsan_zone_fabric75", "50:05:07:61:00:5b:62:ed"

5. Enter the cfgAdd or cfgCreate and cfgEnable commands to add and enable the LSAN
configuration.
switch:admin> cfgadd "zone_cfg", "lsan_zone_fabric75"
switch:admin> cfgenable "zone_cfg"
You are about to enable a new zoning configuration.
This action will replace the old zoning configuration with the
current configuration selected.
Do you want to enable 'zone_cfg' configuration (yes, y, no, n): [no] y
zone config "zone_cfg" is in effect
Updating flash …

6. Log in as admin to fabric2.
7.

Enter the nsShow command to list Target A (50:05:07:61:00:5b:62:ed) and Target B
(50:05:07:61:00:49:20:b4).
switch:admin> nsshow
{
Type Pid
COS
PortName
NodeName
TTL(sec)
NL
0508e8; 3;
50:05:07:61:00:5b:62:ed;
50:05:07:61:00:1b:62:ed; na
FC4s: FCP [IBM
DNEF-309170
F90F]
Fabric Port Name: 20:08:00:05:1e:34:11:e5
Permanent Port Name: 50:05:07:61:00:5b:62:ed
NL
0508ef; 3;
50:05:07:61:00:49:20:b4;
50:05:07:61:00:09:20:b4; na
FC4s: FCP [IBM
DNEF-309170
F90F]
Fabric Port Name: 20:08:00:05:1e:34:11:e5
Permanent Port Name: 50:05:07:61:00:49:20:b4
LSAN: Yes
The Local Name Server has 2 entries }

8. Enter the zoneCreate command to create the LSAN lsan_zone_fabric2, which includes the host
(10:00:00:00:c9:2b:6a:2c), Target A, and Target B.
switch:admin> zonecreate "lsan_zone_fabric2",
"10:00:00:00:c9:2b:c9:0c;50:05:07:61:00:5b:62:ed;50:05:07:61:00:49:20:b4"

9. Enter the cfgShow command to verify that the zones are correct.
switch:admin> cfgshow
Defined configuration:
zone: lsan_zone_fabric2
10:00:00:00:c9:2b:c9:0c; 50:05:07:61:00:5b:62:ed;
50:05:07:61:00:49:20:b4
Effective configuration:
no configuration in effect

10. Enter the cfgAdd and cfgEnable commands to create and enable the LSAN configuration.
switch:admin> cfgadd "zone_cfg", "lsan_zone_fabric2"
switch:admin> cfgenable "zone_cfg"
You are about to enable a new zoning configuration.

622

Fabric OS Administrator’s Guide
53-1002920-02

LSAN zone configuration

26

This action will replace the old zoning configuration with the
current configuration selected.
Do you want to enable 'zone_cfg' configuration (yes, y, no, n): [no] y
zone config "zone_cfg" is in effect
Updating flash ...

11. Log in as admin and connect to the FC router.
12. Enter the following commands to display information about the LSANs:

• lsanZoneShow -s shows the LSAN.
switch:admin> lsanzoneshow -s
Fabric ID: 2 Zone Name: lsan_zone_fabric2
10:00:00:00:c9:2b:c9:0c Imported
50:05:07:61:00:5b:62:ed EXIST
50:05:07:61:00:49:20:b4 EXIST
Fabric ID: 75 Zone Name: lsan_zone_fabric75
10:00:00:00:c9:2b:c9:0c EXIST
50:05:07:61:00:5b:62:ed Imported

• fcrPhyDevShow shows the physical devices in the LSAN.
switch:admin> fcrphydevshow
Device
WWN
Physical
Exists
PID
in Fabric
----------------------------------------75 10:00:00:00:c9:2b:c9:0c c70000
2 50:05:07:61:00:49:20:b4 0100ef
2 50:05:07:61:00:5b:62:ed 0100e8
Total devices displayed: 3

• fcrProxyDevShow shows the proxy devices in the LSAN.
switch:admin> fcrproxydevshow
Proxy
WWN
Proxy
Device
Physical
State
Created
PID
Exists
PID
in Fabric
in Fabric
---------------------------------------------------------------------------75
50:05:07:61:00:5b:62:ed 01f001
2
0100e8
Imported
2
10:00:00:00:c9:2b:c9:0c 02f000
75
c70000
Imported
Total devices displayed: 2

On the FC router, the host and Target A are imported, because both are defined by
lsan_zone_fabric2 and lsan_zone_fabric75. However, Target B is defined by lsan_zone_fabric2
and is not imported because lsan_zone_fabric75 does not allow it.
When a PLOGI, PDISC, or ADISC arrives at the FC router, the SID and DID of the frame are checked.
If they are LSAN-zoned at both SID and DID edge fabrics, the frame is forwarded to the DID. If they
are not zoned, only the PLOGI is dropped; for the remaining frames zoning enforcement takes place
in the edge fabrics.

Configuring backbone fabrics for interconnectivity
If you want devices in backbone fabrics to communicate with devices in edge fabrics, set up the
LSANs as described “Controlling device communication with the LSAN” on page 621. However,
instead of configuring the LSAN in the second edge fabric, configure the LSAN in the backbone
fabric.

Fabric OS Administrator’s Guide
53-1002920-02

623

26

LSAN zone configuration

Setting the maximum LSAN count
You can set the maximum number of LSAN zones, or LSAN count, that can be configured on the
edge fabrics. By default, the maximum LSAN count is set to 3,000. You can increase the maximum
LSAN count to 5,000 without disabling the switch.
The maximum number of LSAN devices supported is 10,000 (this includes both physical and proxy
devices). If you have 3,000 LSAN zones but have not exceeded the 10,000 device limit, you can
increase the LSAN count to 5,000.
All FC routers in the same backbone fabric should have the same maximum LSAN count defined, to
prevent the FC routers from running into indefinite state. Asymmetric LSAN configurations due to
different maximum LSAN counts could lead to different devices being imported on different FC
routers.
1. Enter the fcrlsancount command with no parameters to display the current LSAN limit.
switch:admin> fcrlsancount
LSAN Zone Limit 3000

2. Enter the fcrlsancount command and specify the new LSAN zone limit.
switch:admin> fcrlsancount 5000
LSAN Zone Limit 5000

For information on how to display the maximum allowed and currently used LSAN zones and
devices, refer to “Resource monitoring” on page 634.

NOTE

Because the maximum number of LSANs is configured for each switch, if there is a different
maximum LSAN count on the switches throughout the metaSAN, then the device import and export
will not be identical on the FC routers. You should enter the same maximum LSAN count for all the
FC routers in the same backbone that support this feature. Verify the configured maximum limit
against the LSANs configured using the fcrResourceShow command.

HA and downgrade considerations for LSAN zones
Be aware of how LSAN zones impact high availability and firmware downgrades:

• The LSAN zone matrix is synchronized to the standby CP.
• On a dual CP switch, both CPs must have Fabric OS v5.3.0 or later.
• If the feature is enabled on the active CP, introducing a CP with an earlier version of Fabric OS
as a standby will cause HA synchronization to fail.

• If the feature is enabled, before downgrading to an earlier Fabric OS version, you will be asked
to go back to the default mode.

• This feature does not have any impact on current HA functionality. LSANs will be synchronized
as usual after the limit is increased and new LSANs are created.

LSAN zone policies using LSAN tagging
You can create tags for LSAN zones to give them a special meaning.
LSAN zones are zones with names that start with the “lsan_” prefix. You can specify a tag to
append to this prefix that causes the LSAN zone to be treated differently.

624

Fabric OS Administrator’s Guide
53-1002920-02

LSAN zone configuration

26

You can specify two types of tags:

• Enforce tag – Specifies which LSANs are to be enforced in an FC router.
• Speed tag – Specifies which LSANs are to be imported or exported faster than other LSANs.
The LSAN tags are persistently saved and support configupload and configdownload.

Enforce tag
The Enforce tag reduces the resources used in an FC router by limiting the number of LSAN zones
that will be enforced in that FC router.
Use the Enforce tag to achieve better scalability in the FC router. This is useful when multiple FC
routers are connected to the same edge fabric. Without the Enforce tag, all FC routers import all
LSAN zones, even those that are not needed.
Normally, the FC router automatically accepts all zones with names that start with “lsan_”. You can
specify an Enforce tag to indicate that a particular FC router should only accept zones that start
with the prefix “lsan_tag”. For example, if you specify an Enforce tag of “abc”, the FC router accepts
only those LSAN zones that start with “lsan_abc” and does not import or export any other LSAN
zones.
The Enforce tag can be up to eight characters long and can contain only letters and numbers. The
Enforce tag is not case-sensitive; for example, the tag “abc” is equivalent to “ABC” and “Abc”.
If you specify “abc”, “xyz”, and “fab1” as Enforce tags, then the FC router accepts only those LSAN
zones with names that start with any of the following:
lsan_abc
lsan_xyz
lsan_fab1
In this example, the following LSAN zones would all be accepted:
lsan_abc
Lsan_xyz123456
LSAN_FAB1_abc
You can specify up to eight Enforce tags on an FC router.

Speed tag
During target discovery, the FC router process of presenting proxy devices and setting up paths to
the proxy devices may cause some sensitive hosts to time out or fail. The Speed tag allows you to
speed up the discovery process by importing the devices into the remote edge fabrics when the
devices come online, regardless of the state of the host. This helps sensitive hosts to quickly
discover the devices without timing out.
You set the Speed tag on the FC router, and then configure the LSANs in the target edge fabrics
with the tag.
For example, in Figure 92 on page 626, assume that the host, H1, needs fast access to target
devices D1 and D2. You could set up the Speed tag as follows:
1. In FC router 1 and FC router 2, configure the Speed tag as “super”.
2. In Edge fabric 2, configure two LSANs:

Fabric OS Administrator’s Guide
53-1002920-02

625

26

LSAN zone configuration

lsan_f2_f1 (H1, D1)
lsan_f2_f3 (H1, D2)
The LSAN in the host fabric does not need the tag.
3. In Edge fabric 1, configure the following LSAN:
lsan_super_f1_f2 (H1, D1)
4. In Edge fabric 3, configure the following LSAN:
lsan_super_f3_f2 (H1, D2)
5. Choose either the host or target to trigger the fast import process.
The “super” tag is needed only in the LSANs of the target fabrics.
The target proxies D1 and D2 are always present in the host fabric (Edge fabric 2), even if the host
is brought down. A target proxy is removed from the host fabric when the target device is offline.

D1

D2

H1

Edge fabric 1

Edge fabric 2

FC router 1

Edge fabric 3

FC router 2
= LSAN

FIGURE 92

Example of setting up Speed LSAN tag

Rules for LSAN tagging
Note the following rules for configuring LSAN tags:

• You configure the tags on the FC router, and not on the edge switches. If Virtual Fabrics is
enabled, you configure the tags on the base switch on which the EX_Ports and VEX_Ports are
located. You then must ensure that the LSAN zones in the edge fabrics incorporate the tags
correctly.

• The LSAN tags are configured per FC router, not per fabric. If the backbone fabric has multiple
FC routers, it is recommended that you configure the LSAN tags on all of the FC routers.

• The FC router must be disabled before you configure the Enforce tag. Configuring the Speed
tag does not require that the FC router be disabled; however, after configuring the Speed tag,
you must select the host or target port to trigger the fast import process.

626

Fabric OS Administrator’s Guide
53-1002920-02

LSAN zone configuration

26

• The tag is from 1 through 8 alphanumeric characters.
• You can configure only one Speed tag on an FC router, and up to eight Enforce tags on an FC
router. The maximum number of tags (Enforce and Speed) on an FC router is eight.

• Up to 500 Speed LSAN tags are supported.

Configuring an Enforce LSAN tag
1. Log in to the FC router as admin.
2. Enter the following command to disable the FC router:
switchdisable

3. Enter the following command to create an Enforce LSAN tag:
fcrlsan --add -enforce tagname

The tagname variable is the name of the LSAN tag you want to create.
4. Enter the following command to enable the FC router:
switchenable

5. Change the names of the LSAN zones in the edge fabrics to incorporate the tag in the names.
Example
sw0:admin> switchdisable
sw0:admin> fcrlsan --add -enforce enftag1
LSAN tag set successfully
sw0:admin> switchenable

Configuring a Speed LSAN tag
1. Log in to the FC router as admin.
2. Enter the following command to create a Speed LSAN tag:
fcrlsan --add -speed tagname

The tagname variable is the name of the LSAN tag you want to create.
3. Change the names of the LSAN zones in the edge fabrics to incorporate the tag in the names.
4. Choose the host or target port to trigger the fast import process.
Example
sw0:admin> fcrlsan --add -speed fasttag2
LSAN tag set successfully

Removing an LSAN tag
Use the following procedure to remove an LSAN tag. This procedure does not remove the LSAN
zone; it deactivates the tag so that LSAN zones with this tag in the name now behave as regular
LSAN zones.
You must disable the switch before removing an Enforce LSAN tag. You do not need to disable the
switch to remove a Speed LSAN tag.

Fabric OS Administrator’s Guide
53-1002920-02

627

26

LSAN zone configuration

1. Log in to the FC router as admin.
2. Enter the fcrlsan --remove command to remove an existing LSAN tag.
If you remove an Enforce LSAN tag, you must disable the switch first.
Example of removing an Enforce LSAN tag
sw0:admin> switchdisable
sw0:admin> fcrlsan --remove -enforce enftag1
LSAN tag removed successfully
sw0:admin> switchenable

Example of removing a Speed LSAN tag
sw0:admin> fcrlsan --remove -speed fasttag2
LSAN tag removed successfully

Displaying the LSAN tag configuration
1. Log in to the FC router as admin.
2. Enter the fcrlsan --show command.
Example
sw0:admin> fcrlsan --show -enforce
Total LSAN tags : 1
ENFORCE : enftag1
sw0:admin> fcrlsan --show -speed
Total SPEED tags : 1
SPEED : fasttag2
sw0:admin> fcrlsan --show -all
Total LSAN tags : 2
ENFORCE : enftag1
SPEED
: fasttag2

LSAN zone binding
LSAN zone binding is an optional, advanced feature that increases the scalability envelope for very
large metaSANs.

NOTE

LSAN zone binding is supported only on FC routers with Fabric OS v5.3.0 and later. The FC router
matrix feature is supported only on FC routers with Fabric OS v6.1.0 and later.
Without LSAN zone binding, every FC router in the backbone fabric maintains the entire LSAN zone
and device state database. The size of this database limits the number of FC routers and devices
you can have.

628

Fabric OS Administrator’s Guide
53-1002920-02

LSAN zone configuration

26

With LSAN zone binding, each FC router in the backbone fabric stores only the LSAN zone entries of
the remote edge fabrics that can access its local edge fabrics. The LSAN zone limit supported in
the backbone fabric is not limited by the capability of one FC router. In addition, due to the lower
LSAN count, the CPU consumption by the FC router is lower. If you configure the metaSAN such that
the backbone fabric has two groups of FC routers and there is no LSAN zone sharing and device
access between the two groups, the number of FC routers and devices supported in the backbone
fabric can be higher.
Figure 93 on page 629 shows a sample metaSAN with four FC routers in the backbone fabric.
Without LSAN zone binding, each FC router in the backbone fabric would store information about
LSAN zones 1, 2, 3, and 4.
LSAN zone 2

LSAN zone 1

Fabric 1

Fabric 2

FC
router 1

Fabric 3

Fabric 7
FC
router 2

Backbone fabric

FC
router 4

FC
router 3

Fabric 8

Fabric 9
Fabric 4

LSAN zone 3

FIGURE 93

Fabric 5

Fabric 6

LSAN zone 4

LSAN zone binding

After you set up LSAN zone binding, each FC router stores information about only those LSAN zones
that access its local edge fabrics. Table 94 shows what LSAN information is stored in each FC
router before and after LSAN zone binding is in effect.

Fabric OS Administrator’s Guide
53-1002920-02

629

26

LSAN zone configuration

TABLE 94

LSAN information stored in FC routers, with and without LSAN zone binding

Without LSAN zone binding

With LSAN zone binding

FC router 1

FC router 2

FC router 3

FC router 4

FC router 1

FC router 2

FC router 3

FC router 4

LSAN 1
LSAN 2
LSAN 3
LSAN 4

LSAN 1
LSAN 2
LSAN 3
LSAN 4

LSAN 1
LSAN 2
LSAN 3
LSAN 4

LSAN 1
LSAN 2
LSAN 3
LSAN 4

LSAN 1
LSAN 2

LSAN 2

LSAN 3
LSAN 4

LSAN 4

LSAN zone binding considerations
• Without LSAN zone binding, the maximum number of LSAN devices is 10,000.
• With LSAN zone binding, the metaSAN can import more than 10,000 devices and the
backbone fabric can support more FC routers.

• With LSAN zone binding, CPU consumption by an FC router is lower.

How LSAN zone binding works
LSAN zone binding uses an FC router matrix, which specifies pairs of FC routers in the backbone
fabric that can access each other, and an LSAN fabric matrix, which specifies pairs of edge fabrics
that can access each other.
You set up LSAN zone binding using the fcrLsanMatrix command. This command has two options:
-fcr and -lsan. The -fcr option is for creating and updating the FC router matrix, and the -lsan option
is used for creating and updating the LSAN fabric matrix.

NOTE

Best practice: Use this feature in a backbone fabric in which all FC routers are running Fabric OS
v6.1.0 or later.
When you set up LSAN zone binding on the local FC router (running Fabric OS v6.1.0 or later), the
resultant matrix database is automatically distributed to all of the Fabric OS v6.1.0 or later FC
routers in the backbone fabric. You do not need to set up LSAN zone binding on the other FC
routers unless those FC routers are running Fabric OS versions earlier than v6.1.0.
If a new FC router joins the backbone fabric, the matrix database is automatically distributed to
that FC router unless it has a different LSAN fabric matrix or FC router matrix or both defined
already.
Note the following for FC routers running a Fabric OS version earlier than 6.1.0:

• The matrix database is not automatically distributed from this FC router to other FC routers.
• You must manually configure the LSAN fabric matrix on these FC routers to match the other FC
routers in the backbone fabric.
If you have a dual backbone configuration, where two backbone fabrics share edge fabrics, the
LSAN fabric matrix and FC router matrix settings for the shared edge fabrics must be the same on
both backbone fabrics. The matrix databases are not automatically propagated from one backbone
fabric to another, so you must ensure that both backbone fabrics have the same matrix settings.

NOTE

You can use LSAN zone binding along with LSAN tagging to achieve better scalability and
performance. Refer to “LSAN zone policies using LSAN tagging” on page 624 for information about
using the Enforce LSAN tag.

630

Fabric OS Administrator’s Guide
53-1002920-02

LSAN zone configuration

26

FC router matrix definition
Depending on the structure of the backbone fabric, you can specify pairs of FC routers that can
access each other. For the metaSAN shown in Figure 93, the following FC routers can access each
other:

• FC router 1 and FC router 2
• FC router 3 and FC router 4
Because there is no device sharing between the two groups of FC routers, you can use the
fcrLsanMatrix command with the -fcr option to create the corresponding FC router matrix:
fcrlsanmatrix --add -fcr wwn1 wwn2
fcrlsanmatrix --add -fcr wwn3 wwn4

The variables wwn1, wwn2, wwn3, and wwn4 are the WWNs of the four FC routers.
Now edge fabrics 1, 2, 3, 7, and 8 can access each other, and edge fabrics 4, 5, 6, and 9 can
access each other; however, edge fabrics in one group cannot access edge fabrics in the other
group. The edge fabrics can still communicate with the backbone fabric.

LSAN fabric matrix definition
With LSAN zone binding, you can specify pairs of fabrics that can access each other. Using the
metaSAN shown in Figure 93 as an example, the following edge fabrics can access each other:

•
•
•
•

Fabric 1 and Fabric 2
Fabric 2 and Fabric 3
Fabric 4 and Fabric 5
Fabric 5 and Fabric 6

You can use the fcrLsanMatrix command with the -lsan option to create the corresponding LSAN
fabric matrix:
fcrlsanmatrix
fcrlsanmatrix
fcrlsanmatrix
fcrlsanmatrix

--add
--add
--add
--add

-lsan
-lsan
-lsan
-lsan

1
2
4
5

2
3
5
6

Fabrics that are not specified are part of the default binding and can access other edge fabrics that
are not specified. Thus, fabrics 7, 8, and 9 can access each other, but cannot access fabrics 1
through 6.

ATTENTION
The fcrLsanMatrix --add -lsan 0 0 command will erase the entire LSAN fabric matrix settings in the
cache.
The FC router matrix and the LSAN fabric matrix are used together to determine which fabrics can
access each other, with the LSAN fabric matrix providing more specific binding.

Fabric OS Administrator’s Guide
53-1002920-02

631

26

LSAN zone configuration

Setting up LSAN zone binding
1. Log in to the FC router as admin.
2. Enter the following command to add a pair of FC routers that can access each other:
FCR:Admin> fcrlsanmatrix --add -fcr wwn1 wwn2

The variables wwn1 and wwn2 are the WWNs of the FC routers.
3. Enter the following command to add a pair of edge fabrics that can access each other:
FCR:Admin> fcrlsanmatrix --add -lsan fid1 fid2

The variables fid1 and fid2 are the fabric IDs of the edge fabrics.
4. Enter the following command to apply the changes persistently:
FCR:Admin> fcrlsanmatrix --apply -all

Example
FCR:Admin> fcrlsanmatrix
10:00:00:60:69:c3:12:b3
FCR:Admin> fcrlsanmatrix
FCR:Admin> fcrlsanmatrix
FCR:Admin> fcrlsanmatrix
FCR:Admin> fcrlsanmatrix

--add -fcr 10:00:00:60:69:c3:12:b2
--add -lsan 4 5
--add -lsan 4 7
--add -lsan 10 19
--apply -all

Viewing the LSAN zone binding matrixes
1. Log in to the FC router as admin.
2. Enter the following command to view the FC router matrix:
fcrlsanmatrix --fabricview -fcr

3. Enter the following command to view the LSAN fabric matrix:
fcrlsanmatrix --fabricview -lsan

Example
FCR:Admin> fcrlsanmatrix --fabricview -fcr
SAVED FCR PAIRS
======================================================
FCR
FCR
-----------------------------------------------------10:00:00:60:69:c3:12:b2 (2)
10:00:00:60:69:c3:12:b3 (unknown)
FCR:Admin> fcrlsanmatrix --fabricview -lsan
LSAN MATRIX is activated
Fabric ID
Fabric ID
-------------------------------------4
5
4
7
10
19

632

Fabric OS Administrator’s Guide
53-1002920-02

Proxy PID configuration

26

Proxy PID configuration
When an FC router is first configured, the PIDs for the proxy devices are automatically assigned.
Proxy PIDs (as well as phantom domain IDs) persist across reboots.
The most common situation in which you would set a proxy PID is when you replace a switch. If you
replace the switch and want to continue using the old PID assignments, you can configure it to do
so; this value remains in the system even if the blade is replaced. To minimize disruption to the
edge fabrics, set the proxy PIDs to the same values used with the old hardware.
The fcrProxyConfig command displays or sets the persistent configuration of proxy devices. Used
with the -s slot option, it can also influence the assignment of the xlate domain port number (which
is used to determine the Area_ID field of the PID) and the Port_ID field. Like the PIDs in a fabric, a
proxy PID must be unique. If the slot argument results in a duplicate PID, it will be ignored. Proxy
PIDs are automatically assigned to devices imported into a fabric, starting at f001. For Proxy IDs
projected to an M-EOS edge fabric in McDATA fabric mode, use valid ALPAs (lower 8 bits).
Use the fcrXlateConfig command to display or assign a preferred domain ID to a translate domain.

Fabric parameter considerations
By default, EX_Ports and VEX_Ports detect, autonegotiate, and configure the fabric parameters
without user intervention.
You can optionally configure these parameters manually.

• To change the fabric parameters on a switch in the edge fabric, use the configure command.
Note that to access all of the fabric parameters controlled by this command, you must disable
the switch using the switchDisable command. If executed on an enabled switch, only a subset
of attributes is configurable.

• To change the fabric parameters of an EX_Port on the FC router, use the portCfgEXPort
command.

• To change the fabric parameters of a VEX_Port, use the portCfgVEXPort command.
The backbone fabric PID mode and the edge fabric PID mode do not need to match, but the PID
mode for the EX_Port or VEX_Port and the edge fabric to which it is attached must match. You can
statically set the PID mode for the fabric by using the -p option with the portCfgEXPort command.
Use the -t option to disable the negotiate fabric parameter feature; otherwise, the PID mode is
autonegotiated. The various edge fabrics may have different PID modes.
Fabric parameter settings, namely, E_D_TOV (error-detect timeout value), R_A_TOV
(resource-allocation timeout value), and PID format, must be the same on EX_Ports or VEX_Ports
and on the fabrics to which they are connected. You can set the PID format on an EX_Port when you
configure an inter-fabric link.
The default values for E_D_TOV and R_A_TOV for an EX_Port or VEX_Port must match those values
on other Fabric OS switches. You do not need to adjust these parameters for an EX_Port or
VEX_Port unless you have adjusted them for the edge fabric.
The default values for R_A_TOV and E_D_TOV are the recommended values for all but very large
fabrics (ones requiring four or more hops) or high-latency fabrics (such as ones using long-distance
FCIP links).

Fabric OS Administrator’s Guide
53-1002920-02

633

26

Inter-fabric broadcast frames

Inter-fabric broadcast frames
The FC router can receive and forward broadcast frames between edge fabrics and between the
backbone fabric and edge fabrics. Many target devices and HBAs cannot handle broadcast frames.
In this case, you can set up broadcast zones to control which devices receive broadcast frames.
(Refer to “Broadcast zones” on page 343 for information about setting up broadcast zones.)
By default, broadcast frames are not forwarded from the FC router to the edge fabrics.

Displaying the current broadcast configuration
1. Log in to the FC router as admin.
2. Enter the following command:
fcr:admin> fcrbcastconfig --show

This command displays only the FIDs that have the broadcast frame option enabled. The FIDs
that are not listed have the broadcast frame option disabled.

Enabling broadcast frame forwarding
1. Log in to the FC router as admin.
2. Enter the following command:
fcr:admin> fcrbcastconfig --enable -f fabricID

The fabricID variable is the FID of the edge or backbone fabric on which you want to enable
broadcast frame forwarding. Broadcast frame forwarding is enabled by default.

Disabling broadcast frame forwarding
1. Log in to the FC router as admin.
2. Enter the following command:
fcr:admin> fcrbcastconfig --disable -f fabricID

The fabricID variable is the FID of the edge or backbone fabric on which you want to disable
broadcast frame forwarding.

Resource monitoring
It is possible to exhaust resources, such as proxy PIDs. Whenever a resource is exhausted,
Fabric OS generates an error message. The messages are described in the Fabric OS Message
Reference.
You can monitor FC router resources using the fcrResourceShow command. The fcrResourceShow
command shows FC router resource limits and usage and includes the following:

• LSAN zones and LSAN devices — The information shows the maximum versus the currently
used zones and device database entries. Each proxy or physical device constitutes an entry. If
LSAN zones are defined in two edge fabrics, they are counted as two and not one. One device
imported into multiple edge fabrics counts multiple times.

634

Fabric OS Administrator’s Guide
53-1002920-02

Resource monitoring

26

The default maximum number of LSAN zones is 3,000. Refer to “Setting the maximum LSAN
count” on page 624 for information on changing this limit.

• Proxy Device Slots — The physical and proxy devices use the 10,000 device slots.
The information shows the maximum pool size for translate phantom node and port WWNs and
shows the number of translate node and port WWNs from this pool.

•
•
•
•

Phantom Node WWNs
Phantom Port WWNs
Max proxy devices
Max NR_Ports

The following example shows the use of the fcrResourceShow command to display physical port
(EX_Port) resources.
switch:admin> fcrresourceshow
Daemon Limits:
Max Allowed
Currently Used
------------------------------LSAN Zones:
3000
28
LSAN Devices:
10000
51
Proxy Device Slots:
10000
20

Phantom Node WWN:
Phantom Port WWN:

WWN Pool Size
Allocated
--------------------------------8192
5413
32768
16121

Port Limits:
Max proxy devices:
Max NR_Ports:

2000
1000

Currently
0 |
1 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
12 |
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |

Fabric OS Administrator’s Guide
53-1002920-02

Used(column 1: proxy, column 2: NR_Ports):
0
34
3
34
0
0
0
0
0
0
0
0
6
34
6
34
6
34
6
34
6
34
6
34
6
34
6
34
8
34
8
34
8
34
8
34
8
34
8
34
8
34
8
34

635

26

FC-FC routing and Virtual Fabrics

FC-FC routing and Virtual Fabrics
If Virtual Fabrics is not enabled, FC-FC routing behavior is unchanged. If Virtual Fabrics is enabled,
then in the FC-FC routing context, a base switch is like a backbone switch and a base fabric is like a
backbone fabric.
If Virtual Fabrics is enabled, the following rules apply:

• EX_Ports and VEX_Ports can be configured only on the base switch.
When you enable Virtual Fabrics, the chassis is automatically rebooted. When the switch
comes up, only one default logical switch is present, with the default fabric ID (FID) of 128. All
previously configured EX_Ports and VEX_Ports are persistently disabled with the reason
“ExPort in non base switch”. You must explicitly create a base switch, move the EX_Ports and
VEX_Ports to the base switch, and then enable the ports.
If you move existing EX_Ports or VEX_Ports to any logical switch other than the base switch,
these ports are automatically disabled.
If you want to change an EX_Port or VEX_Port on the logical switch to be a non-EX_Port or
VEX_Port, you must use the portCfgDefault command. You cannot use the portCfgExPort
command because that command is allowed only on the base switch.

• EX_Ports can connect to a logical switch that is in the same chassis or a different chassis.
However, the FID of the EX_Port must be set to a different value than the FID of the logical
switch to which it connects.

• EX_Ports and VEX_Ports — those in FC routers and those in a base switch — cannot connect to
any edge fabric with logical switches configured to use XISLs.
If you connect an EX_Port or VEX_Port to an edge fabric, you must ensure that there are no
logical switches with XISL use enabled in that edge fabric. If any logical switch in the edge
fabric allows XISL use, then the EX_Port or VEX_Port is disabled. Refer to “Configuring a logical
switch to use XISLs” on page 333 for instructions on disallowing XISL use.
Because XISL use is disallowed, dedicated links must be configured to route traffic across
switches in the same logical fabric, as shown in Figure 27 on page 316.

ATTENTION
If you connect an EX_Port or VEX_Port from an FC router running Fabric OS v6.1.x or earlier to a
logical switch that allows XISL use, the EX_Port or VEX_Port is not disabled; however, this
configuration is not supported.

• Backbone-to-edge routing is not supported in the base switch. Refer to “Backbone-to-edge
routing with Virtual Fabrics” on page 638 for information about how to configure legacy FC
routers to allow backbone-to-edge routing with Virtual Fabrics.

• All FC router commands can be executed only in the base switch context.
• The fcrConfigure command is not allowed when Virtual Fabrics is enabled. Instead, use the
lsCfg command to configure the FID.

• Although the Brocade 6510 and 6520 support up to four logical switches, if you are using
FC-FC routing, they can have a maximum of only three logical switches.

636

Fabric OS Administrator’s Guide
53-1002920-02

FC-FC routing and Virtual Fabrics

26

Logical switch configuration for FC routing
Figure 94 shows an example of two chassis partitioned into logical switches. This configuration
allows the device in Fabric 128 to communicate with the device in Fabric 15 without merging the
fabrics. Note the following:

• The base switch in Physical chassis 1 serves as an FC router and contains EX_Ports that
connect to logical switches in the two edge fabrics, Fabric 128 and Fabric 15.

• The other logical switches in Fabric 128 and Fabric 15 must be connected with physical ISLs,
and do not use the XISL connection in the base fabric.

• The logical switches in Fabric 1 are configured to allow XISL use. You cannot connect an
EX_Port to these logical switches, so the device in Fabric 1 cannot communicate with the other
two devices.
Physical chassis 2

Physical chassis 1
IFL

ISL

E Logical switch 1 E
(Default logical switch)
Fabric ID 128

Logical ISL

Logical switch 2
Fabric ID 1
Allows XISL use
F

Logical switch 3
Fabric ID 15

E Logical switch 5 F
(Default logical switch)
Fabric ID 128

ISL

E
E

Logical switch 6
Fabric ID 1
Allows XISL use
E

F

Logical switch 7
Fabric ID 15

IFL
EX Logical switch 4
(Base switch)
Fabric ID 8

FIGURE 94

EX
E

XISL

Logical switch 8
(Base switch)
E
Fabric ID 8

EX_Ports in a base switch

Figure 95 shows a logical representation of the physical chassis and devices in Figure 94. As
shown in Figure 95, Fabric 128 and Fabric 15 are edge fabrics connected to a backbone fabric.
Fabric 1 is not connected to the backbone, so the device in Fabric 1 cannot communicate with any
of the devices in the other fabrics.

Fabric OS Administrator’s Guide
53-1002920-02

637

26

FC-FC routing and Virtual Fabrics

Edge fabric
Fabric 128
Edge fabric
Fabric 15

SW3

SW5
E

SW1

SW7

E

EX

SW2

EX
Fabric 1

SW4
Backbone fabric
Fabric 8

FIGURE 95

SW6

SW8

Logical representation of EX_Ports in a base switch

Backbone-to-edge routing with Virtual Fabrics
Backbone-to-edge routing is not supported in the base switch, unless you use a legacy FC router.
A legacy FC router is an FC router configured on a Brocade 7500 switch.
Base switches can participate in a backbone fabric with legacy FC routers. You cannot connect
devices to the base switch because the base switch does not allow F_Ports. You can, however,
connect devices to the legacy FC router, thus enabling backbone-to-edge routing.
If you connect a legacy FC router to a base switch, you must set the backbone FID of the FC router
to be the same as that of the base switch.
In Figure 94, no devices can be connected to the backbone fabric (Fabric 8) because base
switches cannot have F_Ports. Figure 96 shows an FC router in legacy mode connected to a base
switch. This FC router can have devices connected to it, and so you can have backbone-to-edge
routing through this FC router. In this figure, Host A in the backbone fabric can communicate with
device B in the edge fabric with FID 20; Host A cannot communicate with device C, however,
because the base switches do not support backbone-to-edge routing.

638

Fabric OS Administrator’s Guide
53-1002920-02

26

Upgrade and downgrade considerations for FC-FC routing

Physical chassis 2

Physical chassis 1
IFL

E Logical switch 1 E
(Default logical switch)
Fabric ID 128

ISL

B

E Logical switch 5 F
(Default logical switch)
Fabric ID 128

Logical switch 2
Fabric ID 1
Allows XISL use

Edge fabric
FID 20

Logical switch 6
Fabric ID 1
Allows XISL use

C
F

Logical switch 3
Fabric ID 15

E

ISL

E
E

E

Logical switch 7
Fabric ID 15

IFL

IFL
EX Logical switch 4 EX
(Base switch)
E
Fabric ID 8

XISL

E

Logical switch 8
(Base switch)
Fabric ID 8

EX
E

A

E

ISL

Legacy FC router
Fabric ID 8

FIGURE 96

Backbone-to-edge routing across base switch using FC router in legacy mode

Upgrade and downgrade considerations for FC-FC routing
When you upgrade to Fabric OS v7.0.0 or later, EX_Ports remain functional and you can continue to
perform all FC router operations on the switch.
Brocade recommends that you save your FC-FC routing configuration (using the configUpload
command) before performing any downgrades.
For further instructions on downgrading, refer to Chapter 10, “Installing and Maintaining
Firmware”.

How replacing port blades affects EX_Port configuration
If you replace an FR4-18i blade with an 8-Gbps port blade or FX8-24 blade, the EX_Port
configuration remains the same for the first 16 ports on the 8-Gbps port blade (and for the first 12
FC ports on the FX8-24 blade). For all other ports on the blade, the EX_Port configuration is
cleared. No ports are persistently disabled.
If you replace an 8-Gbps port blade with an FX8-24 blade, the EX_Port configuration remains the
same for the first 12 FC ports on the FX8-24 blade.
If you replace an 8-Gbps port blade or FX8-24 blade with another 8-Gbps port blade, the EX_Port
configuration remains the same.

Displaying the range of output ports connected to xlate domains
The edge fabric detects only one front domain from an FC router connected through multiple output
ports. The output port of the front domain is not fixed to 0; the values can be in a range from 129
through 255. The range of the output ports connected to the xlate domain is from 1 through 128.
This range enables the front domain to connect to 127 remote xlate domains.

Fabric OS Administrator’s Guide
53-1002920-02

639

26

Displaying the range of output ports connected to xlate domains

1. Log in to a switch in the edge fabric.
2. Enter the lsDbShow command on the edge fabric.
In the lsDbShow output, ports in the range from 129 through 255 are the output ports on the
front domain.
The following example shows the range of output ports.
linkCnt = 2,
flags = 0x0
LinkId = 53, out port =
1, rem port = 35, cost = 500, costCnt = 0, type = 1
LinkId = 57, out port = 129, rem port = 18, cost = 500, costCnt = 0, type = 1

The following example also shows the use of the lsDbShow display on the edge fabric. The front
domain, domain 3, has two links representing two EX_Port connections with output ports 129
and 132.
Domain = 3, Link State Database Entry pointer
………
linkCnt = 4, flags = 0x0
LinkId = 199, out port = 129, rem port =
2,
LinkId = 199, out port = 132, rem port =
3,
LinkId =
2, out port =
1, rem port =
2,
LinkId =
1, out port = 32, rem port =
2,

640

= 0x100bbcc0

cost
cost
cost
cost

=
=
=
=

10000,
10000,
10000,
10000,

costCnt
costCnt
costCnt
costCnt

=
=
=
=

0,
0,
0,
0,

type
type
type
type

=
=
=
=

1
1
1
1

Fabric OS Administrator’s Guide
53-1002920-02

Appendix

Port Indexing

A

This appendix shows how to use the switchShow command to determine the mapping among the
port index, slot/port numbers, and the 24-bit port ID (PID) on any Brocade Backbone. Enter the
switchShow command without parameters to show the port index mapping for the entire platform.
Enter the switchShow -slot command for port mapping information for the ports on the blade in a
specific slot. Include the --qsfp option to list also the QSFP number, for slots that contain core
blades.
Example of port index mapping on a CR16-4 blade in a DCX 8510-4 Backbone

This example shows the output of the switchShow command for a CR16-4 core blade in slot 3 of a
Brocade DCX 8510-4 Backbone. The leftmost column shows the unique port index. The second
and third columns show the corresponding physical slot and port numbers, respectively. The
corresponding QSFP number for the port is also shown. For a core blade, no PID exists in the
Address column.
switch:FID128:admin> switchshow -slot 3 -qsfp
switchName: switch name
switchType: 121.3
switchState: Online
switchMode: Native
switchRole: Subordinate
switchDomain: 75
switchId: fffc4b
switchWwn: 10:00:00:05:1e:4f:eb:00
zoning: ON (zoning name)
switchBeacon: OFF
FC Router: OFF
Allow XISL Use: OFF
LS Attributes: [FID: 128, Base Switch: No, Default Switch: Yes, Address Mode 0]
Index
Slot
Port
QSFP
Address
Media
Speed
State
Proto
==========================================================================
256
3
0
0
-----id
16G
No_SigDet
FC
257
3
1
0
-----id
16G
No_SigDet
FC
258
3
2
0
-----id
16G
No_SigDet
FC
259
3
3
0
-----id
16G
No_SigDet
FC
260
3
4
1
------16G
No_Module
FC
261
3
5
1
------16G
No_Module
FC
262
3
6
1
------16G
No_Module
FC
263
3
7
1
------16G
No_Module
FC
264
3
8
2
------16G
No_Module
FC
265
3
9
2
------16G
No_Module
FC
266
3
10
2
------16G
No_Module
FC
267
3
11
2
------16G
No_Module
FC
268
3
12
3
------16G
No_Module
FC
269
3
13
3
------16G
No_Module
FC
270
3
14
3
------16G
No_Module
FC
271
3
15
3
------16G
No_Module
FC
736
3
16
4
------16G
No_Module
FC
737
3
17
4
------16G
No_Module
FC
738
3
18
4
------16G
No_Module
FC
739
3
19
4
------16G
No_Module
FC

Fabric OS Administrator’s Guide
53-1002920-02

641

A

Port Indexing

740
3
20
5
741
3
21
5
742
3
22
5
743
3
23
5
744
3
24
6
745
3
25
6
746
3
26
6
747
3
27
6
748
3
28
7
10:00:00:05:1e:39:e4:5a
749
3
29
7
10:00:00:05:1e:39:e4:5a
750
3
30
7
10:00:00:05:1e:39:e4:5a
751
3
31
7
10:00:00:05:1e:39:e4:5a

---------------------------------------------trunkmaster
-----trunkmaster
-----trunkmaster
-----trunkmaster

-16G
No_Module
-16G
No_Module
-16G
No_Module
-16G
No_Module
-16G
No_Module
-16G
No_Module
-16G
No_Module
-16G
No_Module
id
16G
Online
name (Trunk master)
id
16G
Online
name (Trunk master)
id
16G
Online
name (Trunk master)
id
16G
Online
name (Trunk master)

FC
FC
FC
FC
FC
FC
FC
FC
FC E-Port
FC E-Port
FC E-Port
FC E-Port

Example of port index mapping on an FC16-32 blade of a Brocade DCX 8510-8 Backbone

This example shows the truncated output of the switchShow command for an FC16-32 port blade
in slot 1 of a Brocade DCX 8510-8 Backbone. The Address column shows the PID.
switch:FID128:admin> switchshow -slot 1
switchName: DCX8510_8
(output truncated)
LS Attributes: [FID: 128, Base Switch: No, Default Switch: Yes, Address Mode 0]
Index
Slot
Port
Address
Media
Speed
State
Proto
=======================================================================
0
1
0
500000
-N16
No_Module
FC
1
1
1
500100
-N16
No_Module
FC
2
1
2
500200
-N16
No_Module
FC
(output truncated)

Example of port index mapping on an FC8-64 blade on a Brocade DCX Backbone.

This example shows the truncated switchShow output for an FC8-64 port blade on the Brocade
DCX Backbone. The assignment of port index numbers to PIDs will vary depending on blade type,
platform type, and slot number.
DCX:admin> switchshow
Index
Slot
Port
Address Media
Speed
State
=============================================================
0
1
0
0a0040
-N8
No_Module
1
1
1
0a0140
-N8
No_Module
2
1
2
0a0240
-N8
No_Module
(output truncated)
768
1
48
0a00c0
-N8
No_Module
769
1
49
0a01c0
-N8
No_Module
770
1
50
0a02c0
-N8
No_Module
(output truncated)
784
1
62
0a0ec0
-N8
No_Module
783
1
63
0a0fc0
-N8
No_Module
16
2
0
0a1040
-N8
No_Module
17
2
1
0a1140
-N8
No_Module
(output truncated)

642

Fabric OS Administrator’s Guide
53-1002920-02

Port Indexing

A

Example of port indexing on an FC8-64 blade on a Brocade DCX-4S Backbone.

The Brocade DCX-4S does not need a mapping of ports on port blades because it is a one-to-one
mapping. The order is sequential starting at slot 1 port 0 all the way through slot 8 port 255 for the
FC8-64 blade. For core blades, the port index mapping for the blade in slot 3 begins with port index
256, and port index mapping for the core blade in slot 6 begins with port index 736. There are no
shared areas on the Brocade DCX-4S.
The following example switchShow output is from a Brocade DCX-4S. It shows the index and PID
addressing. The output has been truncated.
DCX-4S:admin> switchshow
Index
Slot
Port
Address
Media
Speed
State
==========================================================
0
1
0
0a0000
-N8
No_Module
1
1
1
0a0100
-N8
No_Module
2
1
2
0a0200
-N8
No_Module
(output truncated)
48
1
48
0a3000
-N8
No_Module
49
1
49
0a3100
-N8
No_Module
50
1
50
0a3200
-N8
No_Module
(output truncated)
62
1
62
0a3e00
-N8
No_Module
63
1
63
0a3f00
-N8
No_Module
64
2
0
0a4000
-N8
No_Module
(output truncated)

Example of port indexing on an FX8-24 blade on a DCX 8510-8 Backbone

This example shows the truncated switchShow output for an FX8-24 application blade on the
Brocade DCX 8510-8 Backbone. The assignment of port index numbers to PIDs will vary depending
on blade type, platform type, and slot number.
switch:FID128:admin> switchshow
switchName: my8510-8
(output truncated)

-slot

10

Slot
Blade Type
ID
Model Name
Status
-------------------------------------------------10
AP BLADE
75
FX8-24
ENABLED
Index
Slot
Port
Address
Media
Speed
State
Proto
=============================================================
80
10
0
505000
id
4G
No_Light
FC
81
10
1
505100
-4G
No_Module FC
82
10
2
505200
id
4G
Mod_Inv
FC “Speed Mismatch /
Incompatible SFP”
83
10
3
505300
-4G
No_Module FC
84
10
4
505400
-4G
No_Module FC
(output truncated)
95
10
15
505f00 --Offline
VE
208
10
16
50d000 --Offline
VE
209
10
17
50d100 -4G
Offline
VE
(output truncated)

Fabric OS Administrator’s Guide
53-1002920-02

643

A

Port Indexing

Example of port indexing on an FS8-18 blade on a DCX 8510-8 Backbone

This example shows the truncated switchShow output for an FS8-18 encryption blade on the
Brocade DCX 8510-8 Backbone. The assignment of port index numbers to PIDs will vary depending
on blade type, platform type, and slot number.
switch:FID128:admin> switchshow -slot 2
switchName: myswitch
(output truncated)
Slot
Blade Type
ID
Model Name
Status
-------------------------------------------------2
AP BLADE
43
FS8-18
ENABLED
Index
Slot
Port
Address
Media
Speed
State
Proto
==================================================================
16
2
0
501000
-N8
No_Module
FC
17
2
1
501100
-N8
No_Module
FC
18
2
2
501200
-N8
No_Module
FC
19
2
3
501300
-N8
No_Module
FC
20
2
4
501400
-N8
No_Module
FC
(output truncated)
31
2
15
501f00
id
N4
No_Light
FC

644

Fabric OS Administrator’s Guide
53-1002920-02

Appendix

B

FIPS Support

In this appendix
• FIPS overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Zeroization functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• FIPS mode configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
• Preparing a switch for FIPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

645
645
647
651

FIPS overview
Federal information processing standards (FIPS) specify the security standards to be satisfied by a
cryptographic module utilized in Fabric OS v6.0.0 and later to protect sensitive information in the
switch.
As part of FIPS 140-2 level 2, compliance passwords, shared secrets, and the private keys used in
SSL, TLS, and system login need to be cleared out or zeroized. Before enabling FIPS compliance
mode, a power-on self-test (POST) is executed when the switch is powered on to check for the
consistency of the algorithms implemented in the switch. Known-answer tests (KATs) are used to
exercise various features of the algorithm and their results are displayed on the console for your
reference. Conditional tests are performed whenever an RSA key pair is generated. These tests
verify the randomness of the deterministic random number generator (DRNG) and the
non-deterministic random number generator (non-DRNG). They also verify the consistency of RSA
keys with regard to signing and verification and encryption and decryption.

ATTENTION
FIPS mode, when enabled, is a chassis-wide setting that affects all logical switches. Once enabled,
FIPS mode cannot be disabled.

Zeroization functions
Zeroization functions can be performed at the discretion of the security administrator. These
functions clear the passwords and the shared secrets. Core files and FFDC data are also removed
upon FIPS Zeroization. Table 95 lists the various keys used in the system that will be zeroized in a
FIPS-compliant Fabric OS module.

TABLE 95

Zeroization behavior

Keys

Zeroization CLI

Description

DH private keys

No command required

Keys will be zeroized within code before they are
released from memory.

FCAP private key

secCertUtil delete --fcapall
-nowarn

The secCertUtil delete --fcapall -nowarn command
removes all FCAP certificates and FCAP private keys.

Fabric OS Administrator’s Guide
53-1002920-02

645

B

Zeroization functions

TABLE 95

646

Zeroization behavior (Continued)

Keys

Zeroization CLI

Description

FCSP Challenge
Handshake
Authentication Protocol
(CHAP) Secret

secAuthSecret –-remove

The secAuthSecret -–create command is used to input
the keys, and the secAuthSecret -–remove command is
used to remove and zeroize the keys. All the DH-CHAP
and FCAP authenticated ports are disabled after
zeroization.

LDAP CA certificate

secCertUtil delete
–ldapcacert 

The given LDAP certificate file is zeroized and deleted
from the module.

Passwords

passwdDefault

The passwdDefault command removes user-defined
accounts in addition to default passwords for the root,
admin, and user default accounts. However, only the
root account has permissions for this command. Users
with securityadmin and admin permissions must use
fipsCfg –-zeroize, which, in addition to removing user
accounts and resetting passwords, also performs the
complete zeroization of the system.
Notes:
• In a dual CP system, executing passwdDefault
syncs with the standby. This means that when
passwdDefault is executed in the active CP, userdefined accounts are removed from both the
active and standby CPs and only the default
accounts (root, factory, admin, and user) will be
retained. These accounts will have the generic
default passwords set.
• To maintain FIPS 140-2 compliance, passwords
for the default accounts (admin and user) must be
changed after every zeroization operation.

RADIUS secret

aaaConfig –-remove

The aaaConfig --remove command zeroizes the secret
and deletes a configured server. The aaaConfig --add
command configures the RADIUS server.

RNG seed key

No command required

/dev/urandom is used as the initial source of seed for
RNG. The RNG seed key is zeroized on every random
number generation.

SFTP session keys

No command required

Automatically zeroized on session termination.

SSH RSA private key

sshUtil delprivkey

Key-based SSH authentication is not used for SSH
sessions.

SSH public keys

sshUtil delpubkeys

Zeroizes the SSH public.

SSH session key

No command required

This key is generated for each SSH session that is
established with the host. It automatically zeroizes on
session termination.

TLS authentication key

No command required

Automatically zeroized on session termination.

TLS pre-master secret

No command required

Automatically zeroized on session termination.

TLS private keys

secCertUtil delkey -all

The secCertUtil delkey -all command is used to zeroize
these keys. The secCertUtil genkey command creates
the keys. Only RSA keys of size 1024 or 2048 are
allowed.

TLS session key

No command required

Automatically zeroized on session termination.

Fabric OS Administrator’s Guide
53-1002920-02

FIPS mode configuration

B

Power-on self-tests
A power-on self-test (POST) is invoked by powering on the switch in FIPS mode and does not require
any operator intervention. If any KATs fail, the switch goes into a FIPS Error state, which reboots the
system to start the test again. If the switch continues to fail the FIPS POST, you will need to return
your switch to your switch service provider for repair. Refer to the Fabric OS Troubleshooting and
Diagnostics Guide for information about preparing a case for your service provider.

Conditional tests
These tests are for the random number generators and are executed to verify the randomness of
the random number generator. The conditional tests are executed each time prior to using the
random number provided by the random number generator.
The results of the POST and conditional tests are recorded in the system log or are output to the
local console. This action includes logging both passing and failing results. Refer to the Fabric OS
Troubleshooting and Diagnostics Guide for instructions on how to recover if your system cannot get
out of the conditional test mode.

FIPS mode configuration
By default, the switch comes up in non-FIPS mode. You can run the fipsCfg --enable fips command
to enable FIPS mode, but you must configure the switch first. Self-test mode must be enabled
before FIPS mode can be enabled. A set of prerequisites (as shown in Table 96) must be satisfied
for the system to enter FIPS mode. To be FIPS-compliant, the switch must be power-cycled. KATs are
run on the reboot. If the KATs are successful, the switch enters FIPS mode. If the KATs fail, then the
switch reboots until the KATs succeed. If the switch cannot enter FIPS mode and continues to
reboot, you must return the switch to your switch service provider. For information about how to
prepare a service provider case, refer to the Fabric OS Troubleshooting and Diagnostics Guide.
When the switch successfully reboots in FIPS mode, only FIPS-compliant algorithms are run.

NOTE
RPC is not supported in FIPS mode.
Table 96 lists Fabric OS features and their behaviors in FIPS and non-FIPS mode.

TABLE 96

FIPS mode restrictions

Features

FIPS mode

Non-FIPS mode

Authentication

All ports, including Access Gateway, FC
router, and F_Ports, adhere to FIPS
guidelines when authentication is enabled.

No restrictions.

Configupload/ download/
supportsave/ firmwaredownload

SCP only

FTP and SCP

DH-CHAP/FCAP
hashing algorithms

SHA-1

MD5 and SHA-1

DH-CHAP Shared Secret

Minimum length of 32 bytes for secret used
in in-flight encryption

Minimum length of 8 bytes for
secret

Fabric OS Administrator’s Guide
53-1002920-02

647

B

FIPS mode configuration

TABLE 96

FIPS mode restrictions (Continued)

Features

FIPS mode

Non-FIPS mode

FC-FC routing

If FIPS is enabled in the FC router and
disabled in the edge switch, the EX_Port is
disabled if the edge fabric switch has
Diffie-Hellman group 0 or hash group MD5.

No restrictions.

HTTP/HTTPS access

HTTPS only

HTTP and HTTPS

HTTPS algorithms

TLS/AES128 cipher suite

TLS AES 128 cipher suite
SSL is not supported.

In-flight encryption

Not supported

No restrictions

IPsec

Usage of AES-XCBC, MD5, and DH group 1
are blocked.

No restrictions

LDAP CA

CA certificate must be available.

CA certificate is optional.

Common certificate for FCAP and
HTTPS authentication

Not supported

Supported

RADIUS auth protocols

PEAP-MSCHAPv2

CHAP, PAP, PEAP-MSCHAPv2

Root account

Disabled

Enabled

Signed firmware download

Mandatory firmware signature validation
(SCP only)

Optional firmware signature
validation (FTP and SCP)

SNMP

Read-only operations

Read and write operations

SSH algorithms

HMAC-SHA1 (mac)
3DES-CBC, AES128-CBC, AES192-CBC,
AES256-CBC (cipher suites)

No restrictions

SSH public keys

RSA 1024 bit keys and RSA 2048 bit keys

RSA 1024 bit keys, RSA 2048
bit keys, and DSA 1024 bit keys

TACACS+ authentication

Not supported

Supported

Telnet/SSH access

Only SSH

Telnet and SSH

LDAP in FIPS mode
You can configure your Microsoft Active Directory server to use the Lightweight Directory Access
Protocol (LDAP) while in FIPS mode. There is no option provided on the switch to configure TLS
ciphers for LDAP in FIPS mode. However, the LDAP client checks if FIPS mode is set on the switch
and uses the FIPS-compliant TLS ciphers for LDAP. If the FIPS mode is not set and the Microsoft
Active Directory server is configured for FIPS ciphers, it uses FIPS-compliant ciphers.
Table 97 lists the differences between FIPS and non-FIPS modes of operation.

TABLE 97

648

FIPS and non-FIPS modes of operation

FIPS mode

non-FIPS mode

The certificate of the CA that issued the Microsoft Active
Directory server certificate must be installed on the switch.

There is no mandatory CA certificate installation on
the switch.

Configure FIPS-compliant TLS ciphers [TDES-168, SHA1,
and RSA-1024] on the Microsoft Active Directory server.
The host needs a reboot for the changes to take effect.

On the Microsoft Active Directory server, there is no
configuration of the FIPS-compliant TLS ciphers.

Fabric OS Administrator’s Guide
53-1002920-02

FIPS mode configuration

TABLE 97

B

FIPS and non-FIPS modes of operation (Continued)

FIPS mode

non-FIPS mode

The switch uses FIPS-compliant ciphers regardless of the
Microsoft Active Directory server configuration. If the
Microsoft Active Directory server is not configured for FIPS
ciphers, authentication will still succeed.

The Microsoft Active Directory server certificate is
validated if the CA certificate is found on the switch.

The Microsoft Active Directory server certificate is validated
by the LDAP client. If the CA certificate is not present on the
switch, then user authentication will fail.

If the Microsoft Active Directory server is configured
for FIPS ciphers and the switch is in non-FIPS mode,
then user authentication will succeed.

Setting up LDAP for FIPS mode
1. Log in to the switch using an account with admin or securityadmin permissions, or an account
with OM permissions for the RADIUS and switch configuration RBAC classes of commands.
2. Enter the dnsConfig command to configure the DNS on the switch.
Example of setting the DNS
switch:admin> dnsconfig
Enter option
1 Display Domain Name Service (DNS) configuration
2 Set DNS configuration
3 Remove DNS configuration
4 Quit
Select an item: (1..4) [4] 2
Enter Domain Name: [] domain.com
Enter Name Server IP address in dot notation: [] 123.123.123.123
Enter Name Server IP address in dot notation: [] 123.123.123.124
DNS parameters saved successfully
Enter option
1 Display Domain Name Service (DNS) configuration
2 Set DNS configuration
3 Remove DNS configuration
4 Quit
Select an item: (1..4) [4] 4

Specify the DNS IP address using either IPv4 or IPv6. This address is needed for the switch to
resolve the domain name to the IP address because LDAP initiates a TCP session to connect to
your Microsoft Active Directory server. A Fully Qualified Domain Name (FQDN) is needed to
validate the server identity as mentioned in the common name of the server certificate.
3. Set the switch authentication mode and add your LDAP server by using the commands shown
in the following example. Provide the Fully Qualified Domain Name (FQDN) of the Microsoft
Active Directory server for the host name parameter while configuring LDAP.
Example of setting up LDAP for FIPS mode
switch:admin> aaaconfig --add GEOFF5.ADLDAP.LOCAL -conf ldap -d
-p 389 -t 3
switch:admin> aaaconfig --authspec "ldap;local"
switch:admin> aaaconfig –show
RADIUS CONFIGURATIONS
=====================
RADIUS configuration does not exist.

Fabric OS Administrator’s Guide
53-1002920-02

adldap.local

649

B

FIPS mode configuration

LDAP CONFIGURATIONS
===================
Position
Server
Port
Domain
Timeout(s)

:
:
:
:
:

1
GEOFF5.ADLDAP.LOCAL
389
adldap.local
3

Primary AAA Service: LDAP
Secondary AAA Service: Switch database

4. Set up LDAP according to the instructions in “LDAP configuration and Microsoft Active
Directory” on page 181, and then configure the following additional Microsoft Active Directory
settings:
a.

To support FIPS-compliant TLS cipher suites on the Microsoft Active Directory server, allow
the SCHANNEL settings listed in Table 98.

TABLE 98

b.

Active Directory keys to modify

Key

Sub-key

Ciphers

3DES

Hashes

SHA1

Key exchange algorithm

PKCS

Protocols

TLSv1.0

Enable the FIPS algorithm policy on the Microsoft Active Directory.

LDAP certificates for FIPS mode
To utilize the LDAP services for FIPS between the switch and the host, you must generate a
certificate signing request (CSR) on the Active Directory server and import and export the CA
certificates. To support server certificate validation, it is essential to have the CA certificate
installed on the switch and Microsoft Active Directory server. Use the secCertUtil command to
import the CA certificate to the switch. This command will prompt for the remote IP and login
credentials to retrieve the CA certificate. The CA certificate should be in any of the standard
certificate formats, “.cer”, ”.crt,” or “.pem”.
LDAP CA certificate file names should not contain spaces when using the secCertUtil command to
import and export the certificate.

Importing an LDAP switch certificate
This procedure imports the LDAP CA certificate from the remote host to the switch.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the PKI RBAC class of commands.
2. Enter the secCertUtil import -ldapcacert command.
Example of importing an LDAP certificate
switch:admin> seccertutil import -ldapcacert
Select protocol [ftp or scp]: scp
Enter IP address: 192.168.38.206

650

Fabric OS Administrator’s Guide
53-1002920-02

Preparing a switch for FIPS

B

Enter remote directory: /users/aUser/certs
Enter certificate name (must have ".crt" or ".cer" ".pem" suffix):
LDAPTestCa.cer
Enter Login Name: aUser
Password: 
Success: imported certificate [LDAPTestCa.cer].

Exporting an LDAP switch certificate
This procedure exports the LDAP CA certificate from the switch to the remote host.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the PKI RBAC class of commands.
2. Enter the secCertUtil export -ldapcacert command.
Example of exporting an LDAP CA certificate
switch:admin> seccertutil export -ldapcacert
Select protocol [ftp or scp]: scp
Enter IP address: 192.168.38.206
Enter remote directory: /users/aUser/certs
Enter Login Name: aUser
Enter LDAP certificate name (must have ".pem" suffix): swLdapca.pem
Password: 
Success: exported LDAP certificate

Deleting an LDAP switch certificate
This procedure deletes the LDAP CA certificate from the switch.
1. Connect to the switch and log in using an account with admin permissions, or an account with
OM permissions for the PKI RBAC class of commands.
2. Enter the secCertUtil show -ldapcacert command to determine the name of the LDAP
certificate file.
3. Enter the secCertUtil delete -ldapcacert file_name command, where file_name is the name of
the LDAP certificate on the switch.
Example of deleting an LDAP CA certificate
switch:admin> seccertutil delete -ldapcacert swLdapca.pem
WARNING!!!
About to delete certificate: swLdapca.pem
ARE YOU SURE (yes, y, no, n): [no] y
Deleted LDAP certificate successfully

Preparing a switch for FIPS
It is important to prepare a switch for the following restrictions that exist in FIPS mode:

• The root account and all root-only functions are not available.
• HTTP, Telnet, and SNMP must be disabled. Once these ports are blocked, you cannot use them
to read or write data from and to the switch.

• The configDownload and firmwareDownload commands using an FTP server are blocked.

Fabric OS Administrator’s Guide
53-1002920-02

651

B

Preparing a switch for FIPS

Refer to Table 97 on page 648 for a complete list of restrictions between FIPS and non-FIPS
modes.

ATTENTION
You need both securityadmin and admin permissions to enable FIPS mode.

Overview of steps
1. Remove legacy OpenSSH DSA keys.
2. Optional: Configure the RADIUS server or the LDAP server.
3. Optional: Configure any authentication protocols.
4. For LDAP only: Install an SSL certificate on the Microsoft Active Directory server and a CA
certificate on the switch for using LDAP authentication.
5. Create separate IP filter policies for IPv4 and IPv6 and block access to Telnet (TCP port 23) or
HTTP (TCP port 80).
6. Set the SNMP security level to off.
7.

Configure the switch for signed firmware.

8. Disable in-flight encryption.
9. Disable IPsec for Ethernet and IPsec for FCIP.
10. Disable in-band management.
11. Disable authspec modes if TACACS + authentication or non-PEAP RADIUS are configured.
12. Disable root access.
13. Enable the KATs and the conditional tests.
14. Disable the boot PROM access.
15. Enable FIPS.
16. Perform zeroization as described in “Zeroizing for FIPS” on page 655.

Enabling FIPS mode
1. Log in to the switch using an account with securityadmin permissions.
2. Enter the sshutil delpubkeys and sshutil delprivkey commands to remove legacy OpenSSH DSA
keys.
These keys, which previously were the default keys, migrate to Fabric OS v7.0.0 but are no
longer supported in FIPS mode. You must remove these keys to remain FIPS-compliant.

NOTE

Support for RSA keys is retained. You can implement RSA keys using the sshutil command.
3. Optional: Select the appropriate authentication method based on your needs:

• If the switch is set for RADIUS, enter the aaaConfig --change or aaaConfig --remove
command to modify each server to use only PEAP-MSCHAPv2 as the authentication
protocol.

652

Fabric OS Administrator’s Guide
53-1002920-02

Preparing a switch for FIPS

B

The RADIUS server must also be configured to use only PEAP-MSCHAPv2. Note that among
the Windows RADIUS servers supported, only Windows 2000-, Windows 2003-, and
Windows 2008-based RADIUS servers may be used in a FIPS-compliant configuration.

• If the switch is set for LDAP, refer to the instructions in “Setting up LDAP for FIPS mode” on
page 649.
4. Optional: Set the authentication protocols.
a.

Enter the authUtil --set -h sha1 command to set the hash type for MD5, which is used in
the DH-CHAP and FCAP authentication protocols.

b.

Enter the authUtil --set -g n command (where n represents the DH group) to set the DH
group to 1, 2, 3, or 4.

5. Install the LDAP CA certificate on the switch and Microsoft Active Directory server. Refer to
“LDAP certificates for FIPS mode” on page 650.
6. Enter the ipFilter --show command and verify that no active IP filter policy permits access to
Telnet or HTTP ports, even if a higher priority policy explicitly denies such access. If an active IP
policy does permit any of these ports, you must modify or deactivate the policy. Create
separate policies for IPv4 and IPv6, and block access on Telnet and HTTP ports.
a.

Enter the ipFilter command to create IP filter policies for IPv4 and IPv6. Refer to “Creating
an IP Filter policy” on page 254.

b.

Add rules to each IP filter policy. Refer to “Adding a rule to an IP Filter policy” on page 259.
You can use the following modifications to the rule to block access to Telnet and HTTP
ports:
ipfilter --addrule policyname -rule rule_number -sip source_IP -dp
dest_port -proto protocol -act deny

• The -sip option can be given as any.
• The -dp options for the port numbers for Telnet and HTTP are 23 and 80, respectively.
• The -proto option should be set to TCP.
c.

Activate each IP filter policy. Refer to “Activating an IP Filter policy” on page 255.

d.

Save each IP filter policy. Refer to “Saving an IP Filter policy” on page 255.

Example
ipfilter --create http_block_v4 -type ipv4
ipfilter --addrule http_block_v4 -rule 1 -sip any -dp 80 -proto tcp -act deny
ipfilter --activate http_block_v4

7.

Use the snmpConfig --set seclevel command to turn on SNMP security. When prompted to
select the SNMP SET Security Level, enter 3, for no access.
Example
switch:FID128:admin> snmpconfig --set seclevel
Select SNMP GET Security Level
(0 = No security, 1 = Authentication only, 2 = Authentication and Privacy, 3 =
No Access): (0..3) [0]
Select SNMP SET Security Level
(0 = No security, 1 = Authentication only, 2 = Authentication and Privacy, 3 =
No Access): (0..3) [0] 3

8. Enter the fipsCfg --disable bootprom command to block access to the boot PROM.

Fabric OS Administrator’s Guide
53-1002920-02

653

B

Preparing a switch for FIPS

NOTE

This command can be entered only from the root account. It must be entered before disabling
the root account.
9. Enter the configure command and respond to the following prompts to enable signed firmware:

•
•
•
•

System services: No
cfgload attributes: Yes
Enforce secure config Upload/Download: Press Enter to accept the default.
Enforce firmware signature validation: Yes

Example
switch:admin> configure
Not all options will be available on an enabled switch.
To disable the switch, use the "switchDisable" command.
Configure...
System services (yes, y, no, n): [no]
…
cfgload attributes (yes, y, no, n): [no] yes
Enforce secure config Upload/Download (yes, y, no, n): [no]
Enforce firmware signature validation (yes, y, no, n): [no] yes

10. Enter the portCfgEncrypt --disable command to disable in-flight encryption. You must first
disable the port.
Example
myswitch:root> portdisable 0
myswitch:root> portcfgencrypt --disable 0
myswitch:root> portenable 0

11. Enter the ipSecConfig --disable command to disable Ethernet IPsec.
12. Disable IPsec for FCIP connections. The procedure depends on the type of extension blade
used.
For FX8-24 extension blades, enter the portCfg fciptunnel [slot/]port modify -ipsec 0
command.
13. Enter the portCfg --mgmtif delete command to disable in-band management.
14. Enter the aaaconfig --authspec local command to disable to authspec mode if TACACS +
authentication, PAP, or CHAP are configured.
15. Enter the fipsCfg --enable selftests command to enable KAT and conditional tests on the
switch.
16. Enter the fipsCfg --verify fips command to verify the switch is FIPS-ready.
17. Enter the userConfig --change root -e no command to block access to the root account.
By disabling the root account, RADIUS and LDAP users with root permissions are also blocked
in FIPS mode.
18. Enter the fipsCfg --enable fips command.
19. Reboot the switch. For a director, reboot both CPs.

654

Fabric OS Administrator’s Guide
53-1002920-02

Preparing a switch for FIPS

B

Zeroizing for FIPS
1. Log in to the switch using an account with admin or securityadmin permissions, or a user
account with OM permissions for the FIPSCfg RBAC class of commands.
2. Enter the fipsCfg --zeroize command.

NOTE

Passwords of the default accounts (admin and user) should be changed after every zeroization
operation to maintain FIPS 140-2 compliance.
3. Power-cycle the switch.

Displaying FIPS configuration
1. Log in to the switch using an account with admin or securityadmin permissions, or a user
account with OM permissions for the FIPSCfg RBAC class of commands.
2. Enter the fipsCfg --showall command.

Fabric OS Administrator’s Guide
53-1002920-02

655

B

656

Preparing a switch for FIPS

Fabric OS Administrator’s Guide
53-1002920-02

Appendix

Hexadecimal Conversion

C

Hexadecimal overview
Hexadecimal, also known as hex, is a numeral system with a base of 16, usually written by means
of symbols 0–9 and A–F (or a–f). Its primary purpose is to represent the binary code that
computers interpret in a format easier for humans to remember. It acts as a form of shorthand, in
which one hexadecimal digit takes the place of four binary bits. For example, the decimal numeral
79, with the binary representation of 01001111, is 4F (or 4f) in hexadecimal where 4 = 0100, and
F = 1111.
Hexadecimal numbers can have either an 0x prefix or an h suffix. The address 0xFFFFFA is the
same address as FFFFFAh.This type of address with 6 digits representing 3 bytes, is called a hex
triplet. Fibre Channel uses hexadecimal notation in hex triplets to specify well-known addresses
and port IDs.

Example conversion of the hexadecimal triplet Ox616000
Notice the PID (610600 - bolded) in the nsShow output is in hexadecimal.
switch:admin> nsshow
{
Type Pid
COS
PortName
NodeName
TTL(sec)
N
610600;
2,3;10:00:00:00:c9:29:b3:84;20:00:00:00:c9:29:b3:84; na
FC4s: FCP
NodeSymb: [36] "Emulex LP9002 FV3.90A7 DV5-5.10A10 "
Fabric Port Name: 20:08:00:05:1e:01:23:e0
Permanent Port Name: 10:00:00:00:c9:29:b3:84
Port Index: 6
Share Area: No
Device Shared in Other AD: No
Redirect: No
LSAN: Yes
The Local Name Server has 1 entry }

1. Separate the 6 digits into triplets by inserting a space after every 2 digits: 61 06 00
2. Convert each hexadecimal value to a decimal representation:
61 = Domain ID = 97
06 = Area (port number) = 06
00 = Port (ALPA) = 0 (not used in this instance, but is used in loop, shared areas in PID
assignments on blades, NPIV, and Access Gateway devices)
Result: hexadecimal triplet 610600 = decimal triplet 97,06,00

Fabric OS Administrator’s Guide
53-1002920-02

657

C

Hexadecimal Conversion

Decimal-to-hexadecimal conversion table
TABLE 99

658

Decimal-to-hexadecimal conversion table

Decimal

01

02

03

04

05

06

07

08

09

10

Hex

01

02

03

04

05

06

07

08

09

0a

Decimal

11

12

13

14

15

16

17

18

19

20

Hex

0b

0c

0d

0e

0f

10

11

12

13

14

Decimal

21

22

23

24

25

26

27

28

29

30

Hex

15

16

17

18

19

1a

1b

1c

1d

1e

Decimal

31

32

33

34

35

36

37

38

39

40

Hex

1f

20

21

22

23

24

25

26

27

28

Decimal

41

42

43

44

45

46

47

48

49

50

Hex

29

2a

2b

2c

2d

2e

2f

30

31

32

Decimal

51

52

53

54

55

56

57

58

59

60

Hex

33

34

35

36

37

38

39

3a

3b

3c

Decimal

61

62

63

64

65

66

67

68

69

70

Hex

3d

3e

3f

40

41

42

43

44

45

46

Decimal

71

72

73

74

75

76

77

78

79

80

Hex

47

48

49

4a

4b

4c

4d

4e

4f

50

Decimal

81

82

83

84

85

86

87

88

89

90

Hex

51

52

53

54

55

56

57

58

59

5a

Decimal

91

92

93

94

95

96

97

98

99

100

Hex

5b

5c

5d

5e

5f

60

61

62

63

64

Decimal

101

102

103

104

105

106

107

108

109

110

Hex

65

66

67

68

69

6a

6b

6c

6d

6e

Decimal

111

112

113

114

115

116

117

118

119

120

Hex

6f

70

71

72

73

74

75

76

77

78

Decimal

121

122

123

124

125

126

127

128

129

130

Hex

79

7a

7b

7c

7d

7e

7f

80

81

82

Decimal

131

132

133

134

135

136

137

138

139

140

Hex

83

84

85

86

87

88

89

8a

8b

8c

Decimal

141

142

143

144

145

146

147

148

149

150

Hex

8d

8e

8f

90

91

92

93

94

95

96

Decimal

151

152

153

154

155

156

157

158

159

160

Hex

97

98

99

9a

9b

9c

9d

9e

9f

a0

Decimal

161

162

163

164

165

166

167

168

169

170

Hex

a1

a2

a3

a4

a5

a6

a7

a8

a9

aa

Fabric OS Administrator’s Guide
53-1002920-02

C

Hexadecimal Conversion

TABLE 99

Decimal-to-hexadecimal conversion table (Continued)

Decimal

171

172

173

174

175

176

177

178

179

180

Hex

ab

ac

ad

ae

af

b0

b1

b2

b3

b4

Decimal

181

182

183

184

185

186

187

188

189

190

Hex

b5

b6

b7

b8

b9

ba

bb

bc

bd

be

Decimal

191

192

193

194

195

196

197

198

199

200

Hex

bf

c0

c1

c2

c3

c4

c5

c6

c7

c8

Decimal

201

202

203

204

205

206

207

208

209

210

Hex

c9

ca

cb

cc

cd

ce

cf

d0

d1

d2

Decimal

211

212

213

214

215

216

217

218

219

220

Hex

d3

d4

d5

d6

d7

d8

d9

da

db

dc

Decimal

221

222

223

224

225

226

227

228

229

230

Hex

dd

de

df

e0

e1

e2

e3

e4

e5

e6

Decimal

231

232

233

234

235

236

237

238

239

240

Hex

e7

e8

e9

ea

eb

ec

ed

ef

ee

f0

Decimal

241

242

243

244

245

246

247

248

249

250

Hex

f1

f2

f3

f4

f5

f6

f7

f8

f9

fa

Decimal

251

252

253

254

255

Hex

fb

fc

fd

fe

ff

Fabric OS Administrator’s Guide
53-1002920-02

659

C

660

Hexadecimal Conversion

Fabric OS Administrator’s Guide
53-1002920-02

Index

Numerics
10 Gbps operation on an FC port, enabling, 528
10-bit addressing mode, 84
10G license, 527–530
128-bit encryption, in browser, 200
16-link ICL license, 524
1st POD ICL license, 523
256-area addressing mode, 85
2nd POD ICL license, 524
8G license, 525
8-link ICL license, 524

A
AAA service requests, 168
aaaConfig command, 169, 171, 189, 193, 194, 652
accepting distributed user databases locally, 158
access
API, 229
browser security support, 200
changing account parameters, 157
creating accounts, 156
deleting accounts, 157
HTTP, 229
IP address changes, 59
log in fails, 59
management server, 229
NTP, 74
password, changing, 64
remote access policies, 178
secure using SSL, 200
serial, 229
SNMP, 229
switch defaults, 229
telnet, 229
blocking, 227
unblocking, 228
using SSL, HTTPS, 200
Access Control List. See: ACL.
Access Gateway
authentication, 248
configuring F_Port trunking on, 580
Fabric OS Administrator’s Guide
53-1002920-02

considerations for Advanced Performance Monitoring,
553
F_Port trunking for, 579
F_Port trunking requirements on, 580
N_Port failover with FA-PWWN, 484
shared secrets, 250
accessing
devices, 229
hosts, 229
switches and fabrics, 229
zones, 229
account ID, 60
account management for Virtual Fabrics, 319
accounts, 151–194
changing parameters, 157
creating, 156
deleting, 157
displaying information, 156
local database of users, 155–158
lockout policy, 161
lockout policy, duration, 162
lockout policy, threshold, 162
lockouts and denial of service implications, 163
managing passwords, 158
password policies, 159–163
unlocking, 162
user-defined, 155
See also: user accounts.
ACL
activating policy changes, 233
adding member, 48
adding policy member, 234
deleting member, 49
deleting policy, 233
distributing local policies, 263
how policies are stored, 231
manually distributing policy database, 260
policies, 231–235
Admin Domain considerations, 232
storage, 231
viewing, 233
Virtual Fabric considerations, 232
policy database distribution, 260
policy database management, 260
policy distribution to other switches, 262

661

policy management, 232–235
policy members, 232
removing policy member, 234
resolving conflicting ACL policies, 264
activating
ACL policy changes, 233
Admin Domains, 498
IP Filter policy, 255
licenses, 533
ports on demand, 535
TI zones, 406
ad command, 494, 498, 499, 500, 501, 502, 503, 506,
507, 508
AD0, ACL management, 232
AD0, and Admin Domains, 488
AD255, ACL management, 232
AD255, and Admin Domains, 489
Adaptive Networking
bottleneck detection, 413
Ingress Rate Limiting, 414
overview, 413–414
Quality of Service, 414
Top Talkers, 413
Traffic Isolation Zoning, 413
adding
a Top Talker monitor to a port (port mode), 565
Admin Domain members, 499
alias members, 347
end-to-end monitors, 553
frame monitors to a port, 560
licensed features, 533
members to a zone configuration, 363
ports to logical switches, 329
public key to switch, 198
rules to an IP Filter policy, 259
switch or fabric to a zone, 371
switches to a zone, 371
Top Talker monitors on all switches in fabric (fabric
mode), 565
zone members, 351
address
IPv4 filter policy, 256
IPv6 filter policy, 256
addressing mode
10-bit, 84
256-area, 85
core PID, 84
fixed, 84, 474
Admin Domain number and domain ID, 486
Admin Domains
about, 485

662

access levels, 487
ACL policy considerations, 232
activating, 498
AD list
Microsoft Active Directory, 183
OpenLDAP, 188
RADIUS, 174
TACACS+, 191
AD0, 488
AD255, 488, 489
adding members, 499
assigning to an existing user account, 497
assigning users to, 496
compatibility, availability, and merging, 494
configDownload command, 512
configUpload command, 512
configuration, defined, 494
configuration, displaying, 508
creating, 495
deactivating, 499
defined AD configuration, 494
deleting, 501, 502
device members, 491
effective AD configuration, 494
features, 487
home AD, 490
Microsoft Active Directory, 183
OpenLDAP, 188
RADIUS, 174
TACACS+, 191
implementing, 495
interaction with Fabric OS features, 509
LDAP server, 183, 187
logging in to, 490
LSAN zones, 511
managing, 485–512
member types, 491–492
Microsoft Active Directory service, 183
numbering, 485
OpenLDAP server, 187
physical fabric administrator, 487
RADIUS configuration, 173
RADIUS server configuration, 173
recommended maximum number, 485
removing from user accounts, 498
removing members, 500
renaming, 500
requirements, 487
role considerations, 153
switch members, 492
switch port members, 491
switch WWN, 492
switching context, 508
Fabric OS Administrator’s Guide
53-1002920-02

system-defined, 488
TACACS+ service, 191
TI zone considerations, 398
transaction model, 494
trunk area, 576
user-defined, 488
using, 506
validating members, 506
VF mode and, 325
Virtual Fabrics permissions, 151
zone database, 510
admin lockout policy, disabling, 162
admin lockout policy, enabling, 162
Administrative Domains. See: Admin Domains.
Advanced Performance Monitoring, 551–553
Access Gateway considerations, 553
monitor types, 551
restrictions, 552
Virtual Fabrics considerations, 552
advanced zoning, 337–376
advanced zoning commands, 338
alerts
bottleneck, 431, 433–440
congestion or latency, 433
using bottleneckMon command, 433
alias
adding members, 347
creating, 347
deleting, 349
removing members, 348
Alias server, described, 46
aliCreate command, 347
aliDelete command, 349
aliRemove command, 348
aliShow command, 349
AP Dedicated Link policy, 124
AP policies and routing policies, 124
AP route
policies, 124
setting route policy, 125
AP Shared Link policy, 124
applications
blade compatibility, 98
listener applications blocked, 228
used by switches, 229
aptPolicy command, 122, 125
assigning user-defined roles, 155
assigning users to Admin Domains, 496
audit log
configuration, 107

Fabric OS Administrator’s Guide
53-1002920-02

configuring for specific event classes, 109
auditCfg command, 109
auditDump command, 109
AUTH module, Virtual Fabric considerations, 244
AUTH policy, 243, 244
distributing fabric-wide, 253
authenticating devices, 246
Virtual Fabrics considerations, 247
authenticating E_Ports, 244
authentication
algorithms used, 269
authentication policy restrictions, 247
configuring and enabling, 453
configuring for devices, 247
configuring for E_Ports, 245
configuring incoming SSH, 198
configuring outgoing SSH, 199
configuring policy, 243–253
enabling, 171
fabric license, 243
FCAP, starting, 253
joining FC routers and edge fabric switchws, 453
key generation, 448
protocols, 248
re-authenticating an E_Port, 246
setting protocols, 248
TACACS+ service, 189
using with Access Gateway, 248
viewing parameter settings, 248
authentication server, 193
adding, 193
data, 168
deleting, 193
reordering, 193
authentication service
configuring, 167–170
disabling, 193, 194
enabling, 193
local, 194
Microsoft Active Directory and LDAP, 181
modifying, 193
OpenLDAP, 184–189
RADIUS, 172
remote, 167–194
authorization, fabric-wide distribution of policy, 253
authUtil command, 245, 246, 247, 248, 253, 453, 653
auto-assigned FA-PWWN behavior, 480
auto-leveling, FR4-18i blade, 298, 304
automatic PID assignment, enabling, 86

663

B
Backbone
assigning fabric IDs, 606, 617
blade compatibility, 98
fabric ID, 605–606
fabric, described, 596
port blades, described, 88
port configurations supported, 321
port restrictions, 321
shutdown, 81
upgrading firmware, 297
Backbone fabric, and TI zones, 389
Backbone firmware, 296–299
download, 296
download process overview, 296
version testing, 304
Backbone-to-edge routing, 600, 605
backing up a configuration, 279
base fabric, 319
base switch
about, 316
creating, 326
defined, 317
extended ISLs and, 316
blade
application compatibility, 98
Backbone compatibility, 98
compatibility, 95
control processor (CP), 97
core, 97
disabling, 99
disabling port blades, 98
enabling, 99
FX8-24 compatibility, 98
limits, 98
port area ID, 90
port compatibility, 98
port identification, 89
port indexing, 90
port numbering schemes, 89
powering off, 103
powering on, 103
swapping, 99–102
terminology and compatibility, 95
types, 89
upgrading firmware, 297
bladeCfgGeMode command, 529
bladeDisable command, 99
bladeEnable command, 99
bladeSwap command, 99

664

blocked listener applications, list, 228
blocking telnet access, 227
bond0 logical network interface, 65
boot PROM password, 163–167
Backbone with recovery string, 164
Backbone without recovery string, 166
switch with recovery string, 163
switch without recovery string, 165
bottleneck, 427–443
access gateway considerations, 430
alert-related parameters, 435
alerts, 433–440
congestion or latency alerts, 433
congestion type, 428
defined, 427
disabling detection, 442
displaying statistics, 442
enabling detection, 431
excluding a port from detection, 440
high availability considerations, 430
latency type, 428
reporting, 428
settings retention, 435
trunking considerations, 430
types, 428
upgrade and downgrade considerations, 430
virtual fabrics considerations, 430
bottleneck detection, 427–443
advanced settings, 439
alert status, 431
configuration retention, 430
configurations, 429
considerations
access gateway, 430
high availability, 430
trunking, 430
upgrades and downgrades, 430
virtual fabrics, 430
disabling, 442
displaying status, 431
enabling, 431
history, 428
history retention time, 428
licensing, 427
limitations, 429
parameters, 435
slave ports, 440
bottleneckMon command, 428, 431, 433, 435, 436,
440, 442
Broadcast server, described, 46
broadcast zones, 337, 343
name restriction, 350

Fabric OS Administrator’s Guide
53-1002920-02

Brocade 7800, upgrade license, 516, 523
Brocade 7800, XISL restriction, 320
Brocade adapters, configuring F_Port trunking for, 581
Brocade adapters, F_Port trunking for, 581
Brocade configuration setup form, 287
Brocade DCX, 519, 543, 547
auto-leveling, 290
ICLs, 546
Brocade DCX 8510, 518, 543
auto-leveling, 290
ICLs, 544
Brocade DCX 8510-4, 519
Brocade DCX 8510-8, 519
Brocade DCX-4S, 547
Brocade FC16-48 port blade enabling exceptions, 99
Brocade FC8-48 port blade enabling exceptions, 99
Brocade FC8-48E port blade enabling exceptions, 99
Brocade FC8-64 port blade enabling exceptions, 99
Brocade fixed-port switches, upgrading firmware, 295
Brocade FX8-24
compatibility, 98
enabling 10-GbE ports, 529
XISL use and VE_Ports, 321
Brocade MIB files, 213
Brocade Network Advisor, 57
Brocade Vendor-Specific Attribute. See: VSA.
browser
128-bit encryption, 200
displaying encryption support, 200
root certificates in Firefox, 205
root certificates in Internet Explorer, 205
security certificate configuration, 204
browser support, 200
buffer credit
management, 135
buffer credit recovery, 146
buffer credits, 119
allocating, 137–142
for average-size frames, 140
for F_Ports, 142
for full-size frames, 137
by switch model, 143
buffer-to-buffer credits, 119, 135

C
capitalization in commands, 58
certificate signing request. See: CSR.
certificates

Fabric OS Administrator’s Guide
53-1002920-02

browser, configuring, 204
certificate authorities (CA), 201
FCAP, 244
importing for FCAP, 252
installing on switch, 203
installing root certificate for Java plugin, 205
IPsec and, 271
IPSECCA.pem certificate name, 271
obtaining, 203
private key generation, 201
public key generation, 201
root, 203
root, configuring, 205
security, 196
signing requests, generating and storing, 202
SSL, 196, 200, 251
SSL certificate files, 203
switch, 203, 251
See also: root certificates or security certificates.
cfgActvShow command, 361, 367
cfgAdd command, 363, 425
cfgClear command, 367
cfgCreate command, 363
cfgDelete command, 365
cfgDisable command, 365
cfgEnable command, 364, 382, 406
cfgRemove command, 364, 426
cfgSave command, 349
cfgShow command, 133, 350, 356, 366, 367
cfgSize command, 362
cfgTransAbort command, 366
cfgTransShow command, 377
Challenge Handshake Authentication Protocol. See: CHAP.
changing
authentication server configuration, 193
authentication server contact order, 193
chassis name, 77
logical switch to base switch, 331
passwords, 63
CHAP, 646
alternatives, 177
password encryption requirement, 177
See also: DH-CHAP.
chargen listener application, 228
chassis management IP interface, setting, 68
chassis names, 77
chassis, changing name of, 77
chassisDistribute command, 260, 261
chassisName command, 77
ChassisRole

665

Microsoft Active Directory, 183
OpenLDAP, 188
RADIUS, 173
TACACS+, 188
chassisShow command, 104
CIDR block notation, 67
class 2 and 3 traffic support, 115
classConfig command, 153
classless inter-domain routing. See: CIDR.
clearing performance monitor counters, 557
clearing zone configurations, 367
CLI
capitalization in, 58
command history, 61
commands to display switch configuration, 281
commands to modify switch configuration, 281
Fabric OS, 58–61
cliHistory command, 61
command
getting help on, 60
aaaConfig, 169, 171, 189, 193, 194, 652
ad, 494, 498, 499, 500, 501, 502, 503, 506, 507,
508
aliCreate, 347
aliDelete, 349
aliRemove, 348
aliShow, 349
aptPolicy, 122, 125
auditCfg, 109
auditDump, 109
authUtil, 245, 246, 247, 248, 253, 453, 653
bladeCfgGeMode, 529
bladeDisable, 99
bladeEnable, 99
bladeSwap, 99
bottleneckMon, 428, 431, 433, 435, 436, 440, 442
capitalization in, 58
cfgActvShow, 361, 367
cfgAdd, 363, 425
cfgClear, 367
cfgCreate, 363
cfgDelete, 365
cfgDisable, 365
cfgEnable, 364, 382, 406
cfgRemove, 364, 426
cfgSave, 349
cfgShow, 133, 350, 356, 366, 367
cfgSize, 362
cfgTransAbort, 366
cfgTransShow, 377
chassisDistribute, 260, 261

666

chassisName, 77
chassisShow, 104
classConfig, 153
cliHistory, 61
configDefault, 284, 512
configDownload, 280, 283, 285, 370, 531
restrictions, 281
Virtual Fabrics mode restrictions, 286
configShow, 277
configUpload, 277, 279, 284, 285, 292, 370, 531
Virtual Fabrics mode restrictions, 286
configure, 111, 197, 301, 323, 334, 654
configureChassis, 326, 418
date, 72
defZone, 361
distribute, 152, 238, 262, 263, 264
dlsReset, 126
dlsSet, 126, 131
dlsShow, 126
dnsConfig, 649
errClear, 293
fabricName, 78
fabricShow, 75, 105
described, 307
fanShow, 104
fcrConfigure, 605, 606, 619
fcrXlateConfig, 602
fddCfg, 238, 260, 262, 263, 264, 605
fipsCfg, 647, 653, 655
firmwareCommit, 296, 303, 304, 305
firmwareDownload, 290, 291, 293, 294, 295, 296,
297, 299, 300, 301, 302
firmwareDownloadStatus, 294, 307
firmwareKeyUpdate, 300, 301
firmwareRestore, 303, 304
firmwareShow, 292, 295, 297, 302, 303
described, 307
fmMonitor, 558, 559, 560, 561, 562
for advanced zoning, 338
fosConfig, 324, 325, 606
fosExec, 328
frameLog, 128
haDisable, 164
haFailover, 165, 306
haShow, 104, 296, 297, 305
haSyncStart, 297
help, 60
ifModeSet, 93
iodReset, 127
iodSet, 127
iodShow, 127
ipAddrSet, 68, 69, 70, 259, 333

Fabric OS Administrator’s Guide
53-1002920-02

ipAddrShow, 66, 69, 70
ipFilter, 227, 228, 254, 255, 259, 653
ipSecConfig, 267, 269, 271, 272, 274, 654
islShow, 452, 573
keyTool, 206
killTelnet, 59
ldapAdd, 189
ldapCfg, 171, 181, 182, 183, 184, 186
licenseAdd, 528, 529, 533
licenseIdShow, 40
licensePort, 538, 539, 540, 541
licenseRemove, 534
licenseShow, 528, 529, 531, 532, 533, 534, 604
licenseSlotCfg, 528, 529, 531
lsCfg, 325, 326, 329, 330, 331, 332
msCapabilityShow, 47
msConfigure, 48, 49
msPlatShow, 47, 50
msPlClearDb, 51
msplMgmtActivate, 46, 47
msplMgmtDeactivate, 46, 47
mstdDisable, 52
mstdEnable, 51
mstdReadConfig, 51
nsAllShow, 54, 105
described, 307
nsShow, 54, 105
described, 307
passwd, 63
passwdCfg, 159, 161
passwdDefault, 646
perfAddEeMonitor, 554
perfCfgClear, 568
perfCfgRestore, 568
perfCfgSave, 568
perfMonitorClear, 557
perfMonitorShow, 556, 557
perfSetPortEEMask, 555
perfTTmon, 565, 566, 567
portBufferCalc, 450
portBufferShow, 142, 450
portCfg, 654
portCfgCompress, 456, 457
portCfgEncrypt, 455, 457, 654
portCfgExport, 603
portCfgFillWord, 90, 589
portCfgISLMode, 119, 121
portCfgLongDistance, 113, 589
portCfgNpivPort, 475, 476
portCfgOctetSpeedCombo, 95, 527, 528
portCfgPersistentEnable, 92
portCfgQos, 415, 425, 426

Fabric OS Administrator’s Guide
53-1002920-02

portCfgShow, 476
portCfgSpeed, 94, 448, 527, 528
portCfgTrunkPort, 574, 581
portDecom, 93
portDisable, 92, 574
portEnable, 92, 537
portEncCompShow, 448, 450, 452
portLoginShow, 478
portName, 89
portShow, 477, 537
portStatsShow, 450
portSwap, 84, 91
portSwapEnable, 91
portTrunkArea, 574, 576, 581
portZoneShow, 342
powerOffListSet, 103
powerOffListShow, 103
psShow, 104
roleConfig, 154
secAuthSecret, 249, 250, 453
secCertUtil, 201, 202, 203, 252, 253, 271, 650
secModeEnable, 229
secPolicyAbort, 234
secPolicyActivate, 233, 236, 237, 264
secPolicyAdd, 234
secPolicyCreate, 236, 240, 243
secPolicyDelete, 233, 239, 240
secPolicyFCSMove, 237
secPolicyRemove, 234
secPolicySave, 240
secPolicyShow, 233
setContext, 125, 333, 334
slotPowerOff, 103
slotPowerOn, 103
slotShow, 104, 604
snmpConfig, 653
snmpWalk, 216, 217
ssh-keygen, 198
sshUtil, 198, 200, 652
sshutil, 291
supportSave, 40
switchCfgPersistentDisable, 102
switchCfgSpeed, 94
switchCfgTrunk, 574
switchDisable, 79, 111, 125, 541
switchEnable, 79, 80, 111, 336
switchName, 77
switchShow, 90, 104, 105, 334, 336, 452, 473, 477,
538, 541
switchShow, 641
switchStatusPolicySet, 106
switchStatusPolicyShow, 106

667

switchStatusShow, 104
syslogDIpAdd, 109
sysShutdown, 80, 81
tac_plus, 190
topologyShow, 402
trunkShow, 574
tsClockServer, 74
tsTimeZone, 72, 73
usbStorage, 299
userConfig, 155, 497, 498, 654
version, 604
wwn, 40
wwnAddress, 87
zone, 132, 133, 369, 395, 402, 405, 406, 407, 409
zoneAdd, 351
zoneCreate, 350, 424
zoneDelete, 355
zoneHelp, 338
zoneObjectRename, 370
zoneObjectReplace, 353
zoneRemove, 352
zoneShow, 356
command line interface. See: CLI.
commands
errShow, 210
switchStatusPolicySet, 210
compression
configuring, 456
disabling, 457
enabling, 448, 456
enabling on a port, 450, 451
in-flight, 445
license, 445
long distance ports, 450
ratios for enc/comp enabled ports, 450
restrictions, 446
viewing configuration, 452
concurrent zone transactions, 376
conditional tests for FIPS, 647
configDefault command, 284, 512
configDownload command, 280, 283, 285, 370, 531
restrictions, 281
Virtual Fabrics mode restrictions, 286
configShow command, 277
configUpload command, 277, 279, 284, 285, 292, 370,
531
in Admin Domain context, 512
Virtual Fabrics mode restrictions, 286
configuration
configDownload command, 280
configuration management for Virtual Fabrics, 285
FA-PWWN upload and download considerations, 483
668

format of configuration file, 278
in fabrics, 284
modifying for switches, 281
restoring, 282
saving for frame monitors, 560
security considerations, 284
setup form, 287
supported for FA-PWWN, 483
without disabling a switch, 282
zones, 370
configuration file
backing up, 279
backup, 279
chassis section, 278
configDownload command, in Admin Domain context,
512
display settings, 277
downloading, 512
format, 278
information not saved, 279
restoring, 280
save to a host, 277
sharing between identical switch models, 284
switch section, 278
uploading, 512
in interactive mode, 279
with Virtual Fabrics enabled, 285
configuration settings, 277–287
configure command, 111, 197, 301, 323, 334, 654
configureChassis command, 326, 418
configuring
a switch for signed firmware, 301
access methods, Web Tools, 57
audit log, 107
authentication, 453
authentication policy, 243–253
browser security certificates, 204
compression, 456
date and time, 72
device authentication, 247
device-switch connection, 90
DHCP, 69
Enforce LSAN tag, 627
extended ISLs, 589
F_Port trunking on an Access Gateway, 580
FA-PWWNs, 481–482
FCAP, 251
FIPS mode, 647–651
FLOGI-time handling of duplicate PWWNs, 110
HTTPS access, 200
incoming SSH authentication, 198
in-flight encryption, 455

Fabric OS Administrator’s Guide
53-1002920-02

interfabric link, 607
IPv6 automatically, 71
links through a gateway, 121
lossless DLS, 131
NTP, 74
outgoing SSH authentication, 199
remote authentication, 167–170
remote authentication on switch, 192
root certificates, 205
security certificates, 200
Speed LSAN tag, 627
SSL, 200, 201–205
TACACS+ service, 189
zone, rules for, 342
conflicting ACL policies, resolving, 264
congestion bottleneck type, 428
congestions versus over-subscription, 119
connected devices and logical switches, 313
connecting
device to a switch, 90
multiple EX_Ports to an edge fabric, 603
switches running different firmware versions, 81
to devices, 82
to switch, 82
connection
restrictions, 154
ssh, 59
telnet, 59
consistency policies, matching fabric-wide, 265
consistency policies, non-matching fabric-wide, 265
console session on serial port, 58
control processor. See: CP.
converting hexadecimal numbers, 657–659
core blades, 97
core/edge topology and ISL trunking, 573
core-edge topology, 549
CP blades, 97
accessing, 175
licensed features and, 533
standby, 175
swapping, 533
CP8 blade
devices supporting dual port, 66
creating
Admin Domains, 495
alias, 347
base switches, 326
DCC policies, 239
FCS policies, 236
frame monitors, 559
frame redirect zones, 132

Fabric OS Administrator’s Guide
53-1002920-02

frame types to be monitored, 559
logical switches, 326
SCC policies, 243
TI zones, 402
user-defined roles, 154
zone configurations, 362, 363
zones, 350
CS_CTL auto mode, using at the chassis level, 418
CS_CTL-based frame prioritization, 416–418
considerations, 418
default and auto modes, 417
disabling, 418
enabling, 417
high availability considerations, 417
supported configurations, 417
CSR
defined, 650
exporting for FCAP, 252
generating and storing, 202
generating for FCAP, 251
obtaining certificates, 203
customizing the switch name, 75
cut-through routing, 117

D
D_Port (diagnostic port), 459
saving port mappings, 464
topologies, 463
viewing information, 465
with HBAs, 467
without HBAs, 465
D_Port, described, 88
daemon processes and High Availability, 55
daemon, tac_plus, 190
daemons automatically restarted, 55
date and time, 72
date change license restriction, 531
date command, 72
date settings, 72
daytime listener application, 228
DCC
creating policy, 239
deleting policy, 240
policies, 232, 238–242
policy member, 232
policy restrictions, 239
policy, maximum name length, 239
DCC policies

669

for NPIV ports, 241
policy behavior with fabric-assigned PWWNs, 241
Virtual Fabric considerations, 239
deactivating
Admin Domains, 499
TI zones, 406
decimal to hexadecimal conversion table, 658
decommissioning ports, 92
default
account passwords, 63
accounts, listed, 63
Fabric OS roles, 152
IP Filter policy names, 254
IP Policy Rules, 258
logical switch, 310
zone access mode, viewing current, 361
zone mode, 360, 495
zoning mode, setting, 361
default logical switch
base switch restriction, 321
XISL restriction, 321
defZone command, 361
deleting
a Top Talker monitor on a port, 567
ACL policy, 233
Admin Domains, 501, 502
alias, 349
all fabric mode Top Talker monitors, 567
DCC policy, 240
end-to-end monitors, 556
frame monitors, 560
frame redirect zones, 133
IP Filter policy, 255
LDAP certificates, 651
logical switches, 329
private key from switch, 200
public key from switch, 200
rule from an IP Filter policy, 259
TI zones, 407
zone configurations, 365
zones, 355
delivery order, forcing for frames, 127
deploying secure protocols, 196
device
accessing, 229
configuring authentication, 247
connecting, 82
CP8 blade dual port support, 66
limiting traffic from, 415
login, 53–55
proxy devices, 599

670

recovery, 55
verifying connectivity, 105
device authentication policy, 246
and Virtual Fabrics considerations, 247
Device Connection Control. See: DCC.
device-based routing, 122, 123, 126
FICON environments only., 123
DH-CHAP, 248, 449
and F_Ports, 247
and FIPS, 647
authentication on E_Ports, 244
configuring authentication, 243
secret key pairs, 249
specifying as authentication protocol, 248
See also: CHAP.
DHCP
activation, 69
configuration, 69
disabling, 70
enabling, 69
stateful IPv6 addresses, 69
diagnostic port. See: D_Port.
dictionary.brocade, 173
Diffie Hellman-Challenge Handshake Authentication
Protocol. See: DH-CHAP.
Directory server, described, 45
disabled zone configuration, defined, 341
disabling
admin lockout policy, 162
bottleneck detection, 442
compression, 457
CS_CTL-based frame prioritization, 418
DHCP, 70
F_Port trunking, 585
failover in TI zones, considerations, 381
in-flight encryption, 456
ingress rate limiting, 415
ISL trunking, 574
local switch protection, 262
NPIV, 476
port, 92
QoS zone-based traffic prioritization, 426
remote authentication, 193
switches, 79, 102
topology discovery, 52
Virtual Fabrics, 325
zone configurations, 365
disabling MIBs, 210
discard listener application, 228
displaying
Admin Domain configuration, 508

Fabric OS Administrator’s Guide
53-1002920-02

configuration settings, 277
current routing policy, 122
domain IDs, 75
encryption support in browser, 200
existing zones, 350
F_Port trunking information, 585
frame monitors, 561
logical switch configuration, 330
LSAN tags, 628
monitor counters, 557
network interface settings, 66
port license assignments, 538
switch name server contents, 54
TI zones, 407
trunking information, 574
distance vector, in routing, 115
distribute command, 152, 238, 262, 263, 264
Distributed Management Server
FCS policy, 47
management server database, 47
topology discovery, 51
well-known address, 46, 47
distributing
authorization policy fabric-wide, 253
FCS policies, 238
IP Filter policy, 260
local ACL policies, 263
local user account database, 158
distribution policy states, 238
DLS
computation trigger, 125
effect on other logical switches, 131
overview, 125
rebalancing triggers, 129
See also: Dynamic Load Sharing.
dlsReset command, 126
dlsSet command, 126, 131
dlsShow command, 126
dnsConfig command, 649
domain ID
and Admin Domain number, 486
domain IDs, 75–76
conflicts, 326
displaying, 75
displaying top talking flows for, 566
domain ID 0, 75
setting, 76
downgrading firmware, 291
download configuration file, 512
DPS
described, 123
device-based routing, 123
Fabric OS Administrator’s Guide
53-1002920-02

support on Virtual Fabrics, 124
dropped frames, discovering why, 127
DSA key pair generation, 198
duplicate
F_Port login, 111
NPIV port login, 111
Port World Wide Name (PWWN), 55
duplicate PWWNs, handling, 110
Dynamic Load Sharing. See: DLS.
Dynamic Path Selection. See: DPS.
dynamic PID binding, 83

E
E_Port
authentication, 244
authentication using DH-CHAP or FCAP, 244
configuring authentication, 245
described, 88
inter-switch link, 118
login process, 53
re-authenticating, 246
Virtual Fabric considerations, 245
echo listener application, 228
edge fabric
described, 596
TI zones and, 388
edge-to-edge routing, 605
EE monitors
about, 553
adding, 554
clearing statistic counters, 557
defined, 551
deleting, 556
displaying counters, 557
maximum number, 553
setting a mask for, 555
supported port configurations for, 554
effective AD configuration, 494
effective zone configuration, defined, 341
ELP mode, 121
enabling
10 Gbps operation on an FC port, 528
10-GbE ports on an FX8-24 blade, 529
admin lockout policy, 162
authentication, 453
bottleneck detection, 431
compression, 456
CS_CTL-based frame prioritization, 417
DCC policy on a trunk area, 586
DHCP, 69
671

FIPS mode, 652
FIPS mode, permissions needed, 652
ISL trunking, 574
local switch protection, 262
NPIV, 476
port, 92
remote authentication, 193
switches, 79, 80
Virtual Fabrics mode, 324
zone configurations, 364
Encapsulating Security Payload. See: ESP.
encryption
128-bit in browser, 200
algorithms used, 269
configuring in-flight, 455
disabling in-flight, 456
enabling, 448
enabling on a port, 450, 451
in-flight, 445
key generation, 448
license, 445
restrictions, 446
using SSL, 200
viewing configuration, 452
end-to-end (EE) monitoring, 553
end-to-end monitors
deleting, 556
restoring configuration, 567
saving configuration, 567
setting a mask, 555
end-to-end performance monitoring, 553
end-to-end transport tunnel mode, example, 274
enforce LSAN tag, 625
enforcement of zones, 342
Enhanced TI zones. See: ETIZ.
ensuring fabric domains share policies, 236
enterprise ICL license, 524
equipment status, 104
errClear command, 293
errShow command, 210
ESP, described, 269
eth0 port on CP8 blade, 65
eth3 port on CP8 blade, 65
ethernet address, static, 67
ethernet interface
on switch, 64
Virtual Fabrics, 65
ethernet IP address, setting static, 68
ETIZ
configuration rules, 396

672

defined, 384
platform restrictions and FC router domains, 397
See also: Enhanced TI zones.
events, logging date and time, 72
EX_Port, 635
configuring trunking, 578
described, 88, 596
displaying trunking information, 578
FCR authentication, 603
masterless trunking, 577
supported trunking configurations and platforms, 578
EX_Port trunking, 577–579
configuring, 578
displaying information, 578
masterless, 577
supported configurations and platforms, 578
Exchange Link Parameters mode. See: ELP mode.
exchange-based routing, 122, 123, 126
expired licenses, 531
removing, 532
exporting
CSR for FCAP, 252
LDAP certificates, 651
public key from switch, 199
extended fabrics
about, 588
buffer credit management, 135
buffer credit recovery, 146
buffer requirement calculation, 137
buffer-to-buffer credits, 135
extended ISLs, 589
F_Port buffer credits, 142
ISL, 136
long-distance mode, 138
port buffer credit, 137
QoS buffer credit requirements, 145
time-division multiplexing, 590
extended ISL
about, 317
and base switches, 316
and fmsmode, 323
logical fabric creation, 334
restrictions, 323
See also: XISL.
extending a universal temporary license, 532

F
F_Port
configuring trunking for Brocade adapters, 581

Fabric OS Administrator’s Guide
53-1002920-02

configuring trunking on an Access Gateway, 580
described, 88
DH-CHAP protocol failure, 247
disabling trunking, 585
displaying trunking information, 585
duplicate logins, 111
trunking considerations, 582
trunking for access gateways, 579
trunking for Brocade adapters, 581
trunking requirements on an Access Gateway, 580
F_Port trunking, 579–586
and Virtual Fabrics, 584
configuring for Brocade adapters, 581
considerations, 582
for access gateways, 579
for Brocade adapters, 581
fabric
access, 229
adding Top Talker monitors, 565
addresses. See: PID.
authentication availability, 243
authentication license, 243
authentication policies, 243–253
changing name, 78
configurations in, 284
connectivity, 105
deleting all Top Talker monitors, 567
domain policy sharing, 236
element authentication, 243
fabric login. See: FLOGI.
fabric-assigned PWWN, 479
fabric-assigned PWWNs and DCC policy behavior, 241
fabric-wide consistency policy
distribution, 260
setting, 264
viewing, 263
fabric-wide distribution of authorization policy, 253
fabric-wide policy enforcement, 263
joining switches to, 264
login command, 53
login process, 54
matching fabric-wide consistency policies, 265
name, 77
high avaliability issues, 78
upgrade and downgrade considerations, 78
non-matching fabric-wide consistency policies, 265
optimizing behavior, 413–426
parameters for ISLs, 118
segmentation and zoning, 373
fabric administrator user account, 497
Fabric configuration server. See: FCS.
Fabric controller, described, 45

Fabric OS Administrator’s Guide
53-1002920-02

fabric ID
assigning for Backbones, 606, 617
conflicts, 326
described, 598
logical switches and, 311
See also: FID.
Fabric Login server, described, 45
Fabric mode Top Talker monitor, described, 563
Fabric OS
authentication services, external, 167
command line interface, 58, 58–61
default roles, 152
feature interaction with Virtual Fabrics, 322
interaction with Virtual Fabrics, 322
policies, 232
protocols supported, 196
security protocols supported, 195
user accounts, 171–172
on RADIUS servers, 172–181
user accounts through LDAP, 171
web server, 204
Fabric Shortest Path First. See: FSPF.
fabricName command, 78
fabricShow command, 75, 105
described, 307
fabric-wide consistency policy, 232, 605
failover, for traffic isolation, 380–382
fanShow command, 104
FA-PWWN, 479
auto-assigned behavior, 480
configuration upload and download considerations,
483
configuring, 481–482
DCC policy behavior, 241
N_Port Access Gateway failover, 484
priority, 480
restrictions, 484
security considerations, 483
supported switches and configurations, 483
user-assigned behavior, 480
FC routers
setting QoS zone-based traffic prioritization over, 426
FCAP
and FIPS, 647
authentication on E_Ports, 244
certificates, 244
configuration overview, 251
CSR export, 252
fabric element authentication, 243
generating key and CSR, 251
importing security certificate, 252

673

importing switch certificate, 252
PKI certificates required, 243
specifying as authentication protocol, 248
starting authentication, 253
FC-FC routing
and FCIP, 606
and Virtual Fabrics, 636
backbone-to-edge, 600
configurations supported, 595
edge-to-edge, 600
fabric mode Top Talker monitors, 605
license requirements, 594
platforms supported, 594
routing service, 593
setup, 603–605
setup verification, 604
Top Talker monitors and, 563
topologies, 600
See also: FCR and Fibre Channel routing
FCIP
and FC-FC routing, 606
tunnel configuration, 606
tunnel hop support, 321
FC-NAT, defined, 117
FCoE, NPIV required, 476
FCR
and traffic isolation, 386
authentication, 603
Brocade 7800 logical switches, 594
fcrConfigure command, 605, 606, 619
fcrXlateConfig command, 602
FCS
creating policy, 236
modifying policy, 235
modifying switch order, 237
policies, 232, 235, 235–238
policy distribution, 238
policy member, 232
policy restrictions, 235
policy states, 235
fddCfg command, 238, 260, 262, 263, 264, 605
feature licenses, 515
listed, 516
FEC
configuration requirements, 112
disabling, 113
disabling for long distance ports, 113
enabling, 111, 112
enabling for long distance ports, 113
limitations, 112
viewing settings, 113

674

Federal Information Processing Standards. See: FIPS.
Fibre Channel
NAT, 117
See also: FC.
Fibre Channel Authentication Protocol. See: FCAP.
Fibre Channel Common Transport (FC-CT) protocol service,
described, 46
Fibre Channel fabrics, and port ID, 117
Fibre Channel Over IP service. See: FCIP.
Fibre Channel port, 88
Fibre Channel port, enabling 10 Gbps operation, 528
Fibre Channel protocol auto discovery process, 54
Fibre Channel router, described, 596
Fibre Channel routers
domains and ETIZ platform restrictions, 397
SCC policies, 242
TI zone limitations, 390
Fibre Channel routing, 596
concepts, 596–603
See also: FCR and FC-FC routing.
Fibre Channel services, 45–56
overview, 45
FICON CUP environment considerations, 291
FID. See: fabric ID.
FIPS
conditional tests, 647
described, 300
DH-CHAP, 647
displaying configuration, 655
error state, 647
FCAP, 647
firmware considerations, 300
firmwareDownload command, 301
LDAP certificates
deleting, 651
displaying and deleting, 650
exporting, 651
installing, 650
overview, 645
permissions needed to enable, 652
power-on self tests, 647
preparing a switch, 651
RADIUS server configuration, 653
RPC unsupported, 647
support, 645–655
zeroization functions, 645
zeroizing for, 655
FIPS mode
configuration, 647–651
enabling, 652
LDAP, 648

Fabric OS Administrator’s Guide
53-1002920-02

LDAP certificates, 650
restrictions, 647
fipsCfg command, 647, 653, 655
Firefox
root certificate installation and verification, 205
SSL support, 200
firmware, 289–307
Backbone, 296–299
Backbone download process overview, 296
Backbone version testing, 304
downgrading, 291
download process, 289
downloading without a password, 291
finding version, 293
for switches, 294–295
obtaining and decompressing, 293
power-on checksum test for FIPS, 302
signed, 301
switch version testing, 302
upgrading, 291
upgrading for Brocade fixed-port switches, 295
upgrading on Backbones, 297
upgrading on blades, 297
firmware download, 290
auto-leveling, 304
Backbones, 296
connected switches, 293
FICON CUP considerations, 291
FIPS, 300
high availability synchronization, 291
preparing for download, 292
process overview, 294
protocol, FTP and SCP, 290
switches, 294
test and restore on Backbones, 304
test and restore on switches, 302
testing different firmware versions, 304
USB device, 299–300
validating, 306
verify progress, 290
firmwareCommit command, 296, 303, 304, 305
firmwareDownload command, 290, 291, 293, 294, 295,
296, 297, 299, 300, 301, 302
firmwareDownloadStatus command, 294, 307
firmwareKeyUpdate command, 300, 301
firmwareRestore command, 303, 304
firmwareShow command, 292, 295, 297, 302, 303
described, 307
fixed addressing mode, 84
fixed-port switches
port configurations supported, 320

Fabric OS Administrator’s Guide
53-1002920-02

port restrictions, 320
FL_Port, described, 88
FLOGI
defined, 53
FC-SP bit setting, 247
process, 54
rejected, 247
request frame header value, 54
fmMonitor command, 558, 559, 560, 561, 562
fmsmode, and XISL, 323
forcing frame delivery order, 127
forward error correction. See: FEC.
fosConfig command, 324, 325, 606
fosExec command, 328
frame delivery and topology changes, 127
frame monitoring, 558
frame monitors
adding to a port, 560
clearing statistic counters, 562
creating, 559
defined, 552
deleting, 560
displaying, 561
removing from a port, 560
restoring configuration, 567
saving configuration, 560, 567
frame order, specifying delivery, 126
frame prioritization, CS_CTL-based, 416–418
frame redirection, 132
frame redirection zones, 337
creating, 132
deleting, 133
viewing, 133
frame types
creating to be monitored, 559
deleting, 560
Frame Viewer
filtering results, 128
using, 127
frameLog command, 128
frames
creating frame redirect zones, 132
deleting frame redirect zones, 133
discovering why dropped, 127
forcing delivery order, 127
restoring unordered delivery order, 127
viewing frame redirect zones, 133
FreeRADIUS
clients, enabling, 176
configuring, 175
Fabric OS user setup, 173

675

user, adding, 176
vendor attributes, 175
See also: RADIUS and Linux.
FSPF
described, 116
number of routes supported, 116
path calculation, 117
traffic isolation routing rules, 383
FSPF-1009 RASLOG message, 398
ftp listener application, 228

G
G_Port, described, 88
gateway links, 120
buffer credits, 588
gateway, configuring a link through, 121
generating
DSA or RSA key pairs, 198
key and CSR for FCAP, 251
PKI key pairs, 199

H
HA. See: High Availability.
haDisable command, 164
haFailover command, 165, 306
haShow command, 104, 296, 297, 305
haSyncStart command, 297
help command, 60
help for commands, 60
hexadecimal conversion, 657–659
hexadecimal to decimal conversion table, 658
High Availability
daemon processes, 55
failover and passwords, 159
failover on RADIUS server, 175
QoS zone-based traffic prioritization considerations,
422
support for trunking, 572
synchronization, 291
verifying features, 104
history of CLI commands, 61
home Admin Domain, 490
Microsoft Access Directory, 183
OpenLDAP, 188
RADIUS, 174
TACACS+, 191
home LF
676

Microsoft Active Directory, 183
OpenLDAP, 188
RADIUS, 174
TACACS+, 191
host syslog, verifying, 109
hosts, accessing, 229
HTTPS protocol, 200
described, 195
secure protocol, 196

I
IAS
configuring, 178
remote access policies, 178
ICL
16-link license, 524
1st POD license, 523
2nd POD license, 524
8-link license, 524
about Inter-Chassis Links, 543
core-edge topology, 549
enterprise license, 524
for DCX 8510 family, 544
for DCX family, 546
license, 544
licensing, 523–524
limitations for lossless core, 130
maximum numbers of, 545
mesh topology, 547
prohibited connections, 546
topologies supported, 547–550
triangular topology, 547
trunking on DCX and DCX-4S, 547
Virtual Fabric considerations, 547
ICL ports and XISL, 321
identifying ports, 89–90
IE, root certificate installation and verification, 205
IFL
about, 596
configuration, 607
configuring, 607
described, 596
ifModeSet command, 93
IKE
policies and IPsec, 271
policies, null encryption support, 275
implementing Admin Domains, 495
indexing ports, 641–644
in-flight compression and port decommissioning, 448

Fabric OS Administrator’s Guide
53-1002920-02

in-flight encryption
configuring, 455
disabling, 456
license, 445
port decommissioning, 448
restrictions, 446
in-flight encryption and compression, 445
overview, 445
ingress rate limiting, 414–415
disabling, 415
Virtual Fabrics considerations, 414
in-order frame delivery, forcing, 127
installing
certificates on switch, 203
LDAP certificates, 650
root certificate to Java plugin, 205
Integrated Routing license, 594
Inter-Chassis Links. See: ICL.
inter-fabric link See: IFL.
Internet Explorer and SSL support, 200
Internet Explorer. See: IE.
inter-switch link. See: ISL.
iodReset command, 127
iodSet command, 127
iodShow command, 127
IP addresses
configuring for a Virtual Fabric, 333
removing from a Virtual Fabric, 333
IP Filter
aborting transaction, 259
activating policy, 255
adding rule to policy, 259
cloning policy, 254
creating policy, 254
default policy names, 254
default policy rules, 258
deleting policy, 255
deleting rule from policy, 259
displaying policy, 254
implicit rules, 258
IP Filter policy destination port, 256
IP Filter policy source address, 256
IP Filter policy traffic type and destination IP, 258
policy, 253
policy and Virtual Fabrics considerations, 254
policy distribution, 260
policy enforcement, 258
policy rules, 255
policy rules using service names, 256
saving policy, 255
supported actions, 257

Fabric OS Administrator’s Guide
53-1002920-02

supported protocols, 257
supported services and port numbers, 256
IP interface for chassis management, 68
ipAddrSet command, 68, 69, 70, 259, 333
ipAddrShow command, 66, 69, 70
ipFilter command, 227, 228, 254, 255, 259, 653
IP-NAT, 117
IPsec
algorithms, 269
Authentication Header protocol, 269
configuration on the management interface, 266
Encapsulating Security Payload protocol, 269
flushing security associations, 275
IKE policies, 271
key management, 271
manual key entry, 272
null encryption support for IKE policies, 275
policies, 270–275
policy described, 270
pre-shared key, 271
protocol, described, 195
protocols, 269
sa-proposal, 269
security association, 269
security certificate, 271
traffic selector, 270
transform set, 270
tunnel configurations, 267–268
IPSECCA.pem, certificate name, 271
ipSecConfig command, 267, 269, 271, 272, 274, 654
IPv4 filter policy address, 256
IPv6
autoconfiguration, 71
DHCP and stateful IPv6 addresses, 69
filter policy address, 256
IPv6 policies tunneling IMCP traffic, 275
ISL, 82
best practices, 118
configuring extended, 589
fabric parameters, 118
logical fabrics and, 315
maximum distances in LO mode, 82
ISL R_RDY mode, 121
ISL trunking
disabling, 574
enabling, 574
over long distance fabrics, 576
islShow command, 452, 573

677

J
Java
installing root certificate in plugin, 205
installing root certificate to plugin, 205
support for SSL, 200
supported version, 201
Java plugin, installing root certificate for, 205
joining a switch to a fabric, 264

K
key
adding public key to switch, 198
deleting private from switch, 200
deleting public from switch, 200
generating for FCAP, 251
generation, 201
key management and IPsec, 271
key pair generation for RSA or DSA, 198
manual key entry and IPsec, 272
PKI key pair generation on switch, 199
pre-shared, and IPsec, 271
private key generation, 201
secret pairs for DH-CHAP, 249
setting a secret key pair, 250
viewing list of secret key pairs, 249
key generation, 448
keyTool command, 206
killTelnet command, 59

L
L_Port, described, 88
latency bottleneck type, 428
Layer 2 routing. See also: FSPF.
LDAP
Active Directory LDAP versions supported, 181
authentication, non-FIPS mode, 184
certificates for FIPS mode, 650
configuring for Microsoft Active Directory, 182
creating Fabric OS user accounts, 171
deleting certificates, 651
exporting certificates, 651
in FIPS mode, 648
installing certificates, 650
IPv4 and IPv6 support, 181
non-FIPS mode restrictions, 181
role mapping and OpenLDAP, 186
678

role mapping, and Microsoft Active Directory, 181
secure service, 168
LDAP server
adding, 193
deleting, 193
reordering, 193
LDAP service
configuration, displaying, 194
configuring, 181
configuring for OpenLDAP, 184–189
disabling, 193, 194
enabling, 193
group assignment, 186
groups, creating, 182
modifying, 193
overview, 152
role, assigning, 183
users, adding, 182, 185
vendor attributes, 184
ldapAdd command, 189
ldapCfg command, 171, 181, 182, 183, 184, 186
LDAPS protocol, described, 195
Ldifde.exe utility, using, 183
LF role
Microsoft Active Directory, 183
OpenLDAP, 188
RADIUS, 174
TACACS+, 191
licence. See: license.
license
10G, 527–530
8G, 525
activating, 533
adding features, 533
Brocade 7800 upgrade, 523
compression, 445
date change restriction, 531
encryption, 445
enterprise ICL, 524
expired, 531
Extended Fabrics, 587
fabric authentication, 243
ICL, 523–524, 544
ICL 16-link, 524
ICL 1st POD, 523
ICL 2nd POD, 524
ICL 8-link, 524
in-flight encryption, 445
installation requirements and location, 519
Integrated Routing, 594
preserving, 515
purchasing keys, 536

Fabric OS Administrator’s Guide
53-1002920-02

removing expired, 532
removing features, 534
requirements for SID/DID prioritization, 416
requirements for trunking, 571
reserving for POD, 540
slot-based, 526–527
temporary, 530–532
time-based, 530–532
universal temporary
extending, 532
shelf life, 532
universal temporary, described, 532
viewing installed, 532
licenseAdd command, 528, 529, 533
licensed features, 515
listed, 516
licenseIdShow command, 40
licensePort command, 538, 539, 540, 541
licenseRemove command, 534
licenseShow command, 528, 529, 531, 532, 533, 534,
604
licenseSlotCfg command, 528, 529, 531
licensing, 515–541
overview, 515–523
limiting traffic from a device, 415
link operating mode, 93
link state database, 116
link state, in routing, 115
link, configuring through a gateway, 121
Linux
FreeRADIUS and Fabric OS user setup, 173
LDAP authentication, 184–189
RADIUS server support, 175
TACACS+ authentication on, 190
LISL. See: Logical ISL.
list of port types, 88
listener application
blocked
chargen, 228
daytime, 228
discard, 228
echo, 228
ftp, 228
rexec, 228
rlogin, 228
rsh, 228
rstats, 228
rusers, 228
time, 228
blocked list, 228
chargen, 228

Fabric OS Administrator’s Guide
53-1002920-02

daytime, 228
discard, 228
echo, 228
ftp, 228
rexec, 228
rlogin, 228
rsh, 228
rstats, 228
rusers, 228
time, 228
loading brocade MIBs, 212
local ACL policies, distributing, 263
local authentication
backup, as, 194
overview, 152
local clock, 74
local database user accounts, 155–158
local user account database
distributing, 158
local user account database distribution, 158
local user account passwords, 157
LOCL, 74
logging timestamp, 72
logical fabrics, 315–319
about, 315
changing context, 334
creating using XISLs, 334
defined, 310
formation, 319
ISLs and, 315
logical ISLs, 317
logical network interface, bond0, 65
logical ports
zoning, 350
logical ports in ISL, 319
Logical SANs, described, 597
logical switches, 310–314
about, 310
allowing XISL use, 333
basic configuration values, 326
changing to a base switch, 331
commanding in a different context, 328
connected devices and, 313
creating, 326
deleting, 329
displaying configuration, 330
DLS effect on, 131
fabric IDs and, 311
management model, 314
moving ports, 329
multiple FIDs, 316

679

number, 311
number per chassis, 323
port assignment, 312
restoring configuration, 285
Top Talkers and, 329
unique names for, 77
login
changing password, 157
command for fabric, 53
fails, 59
process for fabric, 54
with Admin Domains, 490
login sessions, maximum allowed, 154
long distance fabrics, and ISL trunking, 576
long distance ports, compression, 450
lossless core, 130
ICL limitations, 130
traffic flow limitations, 130
lossless DLS, 129–131
configuring, 131
in Virtual Fabrics, 131
lossless dynamic load sharing. See: lossless DLS.
LSAN, 620
LSAN tags, 624
LSAN zone binding, 628
LSAN zone name length consideration, 512
LSAN zones, 337
in Admin Domains, 511
lsCfg command, 325, 326, 329, 330, 331, 332

M
M_Port, described, 88
making basic connections, 81
management channel, 154
management interface
IPsec configuration, 266
security, 266–275
management server
displaying ACL, 48
viewing database, 50
management server database, 47–51
Management server, described, 46
managing
Admin Domains, 485–512
trunking connections, 569
user accounts, 151–194
user-defined roles, 154–155
zoning configurations in a fabric, 368
manually distributing ACL policy database, 260
680

mask for end-to-end monitors, setting, 555
masterless EX_Port trunking, 577
masterless trunking, 570
matching fabric parameters, 603
maximum ISL distances in LO mode, 82
maximum zone database size, 362
members
policy, 232
merging zones, 362
mesh topology, 547
MetaSAN, described, 598
MIB loading order, 214
Microsoft Active Directory and LDAP role mapping, 181
Microsoft Active Directory LDAP service
configuring, 182
Microsoft Active Directory service
configuring for LDAP, 181
groups, creating, 182
role, assigning, 183
users, adding, 182
vendor attributes, adding to schema, 184
mirror port. See also: M_Port.
modifying
FCS policy, 235
FCS switch order, 237
TI zones, 405
zoning configurations, 362
monitor configuration, restoring, 567
monitoring
end-to-end performance, 553
frames, 558
trunks, 567
Mozilla Firefox. See: Firefox.
msCapabilityShow command, 47
msConfigure command, 48, 49
msPlatShow command, 47, 50
msPlClearDb command, 51
msplMgmtActivate command, 46, 47
msplMgmtDeactivate command, 46, 47
mstdDisable command, 52
mstdEnable command, 51
mstdReadConfig command, 51

N
N_Port ID Virtualization. See: NPIV.
N_Port, Access Gateway failover with FA-PWWN, 484
name
chassis, 77
Fabric OS Administrator’s Guide
53-1002920-02

fabric, 77
security certificate name, 271
switch, 76
name server contents, displaying, 54
naming ports, 89
NAT, 117
network address translation, see NAT
network interface
displaying settings, 66
logical (bond0), 65
Network OS connectivity, 593
Network OS connectivity, unsupported configurations,
595
network prefix length. See: CIDR.
network time protocol, 74
NPIV, 473–478
10-bit addressing mode, 474
configuring, 475–476
disabling, 476
enabling, 476
F_Ports, 476
FCoE requirement, 476
fixed addressing mode, 474
overview, 473–474
PIDs, 86
upgrade considerations, 474
viewing PID login information, 478
viewing port configuration information, 476
NPIV ports
DCC policy behavior, 241
duplicate login, 111
nsAllShow command, 54, 105
described, 307
nsShow command, 54, 105
described, 307
NTP, 74
NTP access, 74
null encryption support for IKE policies, 275

O
on-demand ports, 535–541
activating, 537
available ports, 535
disabling dynamic, 539
displaying installed licenses, 536
dynamic, 537
enabling dynamic, 538
supported devices, 535

Fabric OS Administrator’s Guide
53-1002920-02

Open LDAP
See also: LDAP.
OpenLDAP
configuring, 184–189
group assignment, 186
group membership, 185
group membership, enabling, 185
LDAP role mapping and, 186
memberOf overlay, 185
restrictions, 184
server configuration, 184
users, adding, 185
users, modifying, 186
over-subscription versus congestions, 119

P
passwd command, 63
passwdCfg command, 159, 161
passwdDefault command, 646
password, 60
changing, 63, 158
changing defaults, 64
CHAP encryption requirement, 177
default for accounts, 63
limits, 63
recovery string, 165
recovery string, boot PROM password, 163
Password Authentication Protocol (PAP), 177
password database, distribution restrictions, 158
password expiration policy, 161
password history policy, 160
password policy
account lockout, 161
strength, 159
password strength policy, 159
passwordless firmware download, 291
passwords
boot PROM, 163–167
Backbone with recovery string, 164
Backbone without recovery string, 166
switch with recovery string, 163
switch without recovery string, 165
local user accounts, 157
policies for, 159–163
rules, 157
path calculation using FSPF, 117
path selection for routing, 116
paths, defined, 116

681

PEAP-MSCHAPv2, 177, 648, 652
perfAddEeMonitor command, 554
perfCfgClear command, 568
perfCfgRestore command, 568
perfCfgSave command, 568
perfMonitorClear command, 557
perfMonitorShow command, 556, 557
performance data collection, 568
perfSetPortEEMask command, 555
perfTTmon command, 565, 566, 567
permissions assigned to roles, 153
phantom domains, 601–603
described, 599
physical fabric administrator, 487
physical fabric administrator user account, creating, 497
PID, 83–87
10-bit addressing mode, 84
assigning static, 87
automatic assignment, 86
binding overview, 83
clearing binding, 87
core addressing mode, 84
maximum number of assignments, 86
showing assignments, 87
static and NPIV, 86
swapping port area IDs, 91
WWN-based assignment, 86
WWN-based Virtual Fabrics assignment, 86
PKI
generating DSA or RSA key pairs, 198
generating key pairs, 199
private key generation, 201
public key generation, 201
used with SSL, 200
platform database, 47
platform services, 46–47
disabling, 47
enabling, 47
Virtual Fabrics, 47
platforms, FC-FC routing supported, 594
PLOGI, 54
defined, 53
POD
enabling ports, 92
releasing a port from a set, 540
reserving a port license, 540
See also: ports on demand.
policies
account lockout, 161
account lockout duration, 162

682

account lockout threshold, 162
ACL, 231–235
ACL and Admin Domain considerations, 232
ACL and Virtual Fabric considerations, 232
ACL changes, activating, 233
ACL, storage, 231
ACL, viewing, 233
adding member to ACL, 234
admin lockout, 162
AUTH, 244
AUTH, distributing fabric-wide, 253
authentication for fabric, 243–253
DCC, 232, 238–242
DCC and Virtual Fabric considerations, 239
distributing local ACL, 263
ensuring fabric domains share policies, 236
fabric-wide consistency, 232, 605
fabric-wide consistency policy distribution, 260
fabric-wide policy enforcement, 263
FCS, 232, 235–238
FCS policy distribution, 238
IPsec, 270–275
IPv6 tunneling IMCP traffic, 275
manually distributing ACL policy database, 260
matching fabric-wide consistency policies, 265
non-matching fabric-wide consistency policies, 265
password, 159–163
password expiration, 161
password history, 160
removing member from ACL, 234
resolving conflicting ACL policies, 264
routing, 122–125
saving changes without activating, 233
SCC, 232, 242–243
SCC and Virtual Fabric considerations, 242
security, 231–275
switch database distribution setting, 260
policy
ACL deleting, 233
ACL distribution, 262
activating IP Filter, 255
adding rule to an IP Filter policy, 259
authentication restrictions, 247
cloning an IP Filter, 254
creating DCC, 239
creating FCS, 236
creating for IP Filter, 254
creating SCC, 243
DCC deleting, 240
DCC restrictions, 239
default IP Filter policy rules, 258
deleting IP Filter, 255

Fabric OS Administrator’s Guide
53-1002920-02

deleting rule from an IP Filter policy, 259
device authentication, 246
device authentication and Virtual Fabrics
considerations, 247
displaying IP Filter, 254
enforcing IP Filter, 258
FCS restrictions, 235
IP Filter, 253
IP Filter policy distribution, 260
management of ACL, 232–235
members, identifying, 232
modifying FCS, 235
password strength, 159
rules for IP Filter, 255
saving IP Filter, 255
using service names in IP Filter rules, 256
policy database distribution, 260
settings, 261
viewing settings, 262
Virtual Fabric considerations, 260
policy set, defined, 232
port, 88–95
activating POD, 537
activation, 92
adding frame monitors to, 560
and blade compatibility, 98
assignment in logical switches, 312
blade enabling exceptions, 99
compression for long distance, 450
compression ratios
for compression-enabled ports, 450
for encryption-enabled ports, 450
configuration of ports, 229
configurations supported for Backbones, 321
configurations supported for fixed-port switches, 320
configuring E_Port authentication, 245
deactivation, 92
decommissioning, 92
deleting Top Talker monitor on, 567
disabling, 92
disabling dynamic POD, 539
disabling on blades, 98
displaying license assignments, 538
displaying the top n bandwidth-using flows, 566
dynamic POD, 537
enabling, 92
enabling compression, 450, 451
enabling dynamic POD, 538
enabling encryption, 450, 451
excluding from bottleneck detection, 440
FEC disabling for long distance, 113
FEC enabling for long distance, 113

Fabric OS Administrator’s Guide
53-1002920-02

ICL, 546
ID and Fibre Channel fabrics, 117
identification
by index, 90
by port area ID, 90
by slot and port number, 89
logical and zoning, 350
logical in ISL, 319
lossless dynamic load sharing, 129–131
moving, 313
naming, 89
port login command, 53
port login process, 54
port types, 88
ports and applications used by switches, 229
re-authenticating an E_Port, 246
releasing from a POD set, 540
removing frame monitors from, 560
reserving a POD license, 540
restrictions on Backbones, 321
restrictions on fixed-port switches, 320
restrictions on moving, 324
serial connection, 58
setting mode, 93
setting speed for a port octet, 95
slave port bottleneck detection, 440
SNMP filtering, 217
speed and number of encryption/compression ports,
447
Top Talker monitor, adding, 565
port area ID, 90
port area IDs, swapping, 91
port decommissioning
on port with in-flight encryption/compression, 448
port groups for trunking, 571
port identifier. See also: PID.
port index, 641
port indexing, 90, 641–644
port information, viewing, 465
port mirroring, 88
Port mode Top Talker monitor, described, 562
port numbering schemes, 89
port speeds, setting, 94
port type
D_Port, 88
diagnostic, 88
E_Port, 88
EX_Port, 88
F_Port, 88
FL_Port, 88
G_Port, 88

683

L_Port, 88
M_Port, 88
mirror, 88
U_Port, 88
VE_Port, 88
VEX_Port, 88
Port World Wide Name. See also: PWWN.
port-based routing, 122, 123, 126
portBufferCalc command, 450
portBufferShow command, 142, 450
portBufferShow command, 450
portCfg command, 654
portCfgCompress command, 456, 457
portCfgEncrypt command, 455, 457, 654
portCfgExport command, 603
portCfgFillWord command, 90, 589
portCfgISLMode command, 119, 121
portCfgLongDistance command, 113, 589
portCfgNpivPort command, 475, 476
portCfgOctetSpeedCombo command, 95, 527, 528
portCfgPersistentEnable command, 92
portCfgQos command, 415, 425, 426
portCfgShow command, 476
portCfgSpeed command, 94, 448, 527, 528
portCfgTrunkPort command, 574, 581
portDecom command, 93
portDisable command, 92, 574
portEnable command, 92, 537
portEncCompShow command, 448, 450, 452
portLoginShow command, 478
portName command, 89
ports on demand, 535–541
activating, 537
available ports, 535
disabling dynamic, 539
displaying installed licenses, 536
dynamic, 537
enabling dynamic, 538
licence restrictions, 535
supported devices, 535
See also: POD.
portShow command, 477, 537
portStatsShow command, 450
portSwap command, 84, 91
portSwapEnable command, 91
portTrunkArea command, 574, 576, 581
portZoneShow command, 342
power management, 103
powering down, 80

684

powerOffListSet command, 103
powerOffListShow command, 103
power-on self tests for FIPS, 647
preparing a switch for FIPS, 651
preserving licenses, 515
pre-shared key, and IPsec, 271
primary FCS, 47
primary FCS modifying, 237
Principal ISLs, 116
principal switch
defined, 53
principal switch, capabilities, 53
priority groups, for virtual channels, 119
private key
deleting from switch, 200
generation, 201
PRLI, 54
protocol
Fibre Channel Common Transport (FC-CT), described,
46
HTTPS, described, 195
IPsec, described, 195
LDAPS, described, 195
SCP, described, 195
secure
HTTPS, 196
SCP, 196
SNMPv1, 196
SNMPv2, 196
SNMPv3, 196
SSHv2, 196
SNMP, described, 195
SSH, described, 196
SSL, 200
SSL, described, 196
telnet, 226
protocols
authentication, 248
IPsec, 269
items needed to deploy secure, 196
secure, 196
security, 195
setting for authentication, 248
supported for IP Filter, 257
proxy device, 599–600
described, 597
proxy PID, 633
proxy PID, described, 597
psShow command, 104
public key

Fabric OS Administrator’s Guide
53-1002920-02

adding to switch, 198
authentication, 198
deleting from switch, 200
generation, 201
public key infrastructure and encryption, 200
public key infrastructure. See also: PKI.
PWWN
assigned by fabric, 479
configuring FLOGI-time handling of duplicates, 110
duplicates, 55
handling duplicates, 111
See also: Port World Wide Name.

Q
QoS, 414
buffer credit requirement, 145
described, 119
enabled by default, 425
on E_Ports, 421
over FC routers, 421
SID/DID traffic prioritization, 415
traffic prioritization, 421
QoS CS_CTL-based frame prioritization, 416–418
QoS port buffer configuration, 141
QoS zone-based traffic prioritization, 419
disabling, 426
High Availability considerations, 422
limitations and restrictions, 424
setting, 424
ssetting over FC routers, 426
supported configurations, 423
trunking considerations, 424
Virtual Fabrics considerations, 422
QoS zones, 119, 338
defined, 419
name prefix specified in an LSAN zone, 420
QSFP ports in DCX 8510 chassis, 544
Quality of Service. See: QoS.

R
RADIUS client
configuration, 176
enabling, 176
RADIUS server
adding, 193
configuration for FIPS, 653
configuration with Admin Domains or Virtual Fabrics,

Fabric OS Administrator’s Guide
53-1002920-02

173
configuring support with Linux, 175
configuring support with Windows 2000, 177
deleting, 193
High Availability failover on, 175
reordering, 193
RSA setup, 179
setup, 175–181
RADIUS service
ADList, 174
configuration, displaying, 194
configuring, 175
ContextRoleList, 174, 191
disabling, 193, 194
enabling, 193
homeAD, 174
Linux server-based, 175
modifying, 193
overview, 152, 172
user, adding, 176
Virtual Fabrics HomeContext, 174
Windows 2008 support, 178
Windows server-based, 177
RASLOG message
FSPF-1009, 398
ZONE-1062, 359
RBAC, 152
Admin Domain considerations, 153
and Fabric OS, 58
role permissions, 153
recommendations for trunk groups, 572
recovering a device, 55
redirecting frames, 132
Registered State Change Notification, 54
rejecting distributed user databases locally, 159
releasing a port from a POD set, 540
remote access policies, 178
remote authentication, 167–194
adding server to the switch configuration, 193
changing authentication server configuration, 193
changing authentication server contact order, 193
deleting server from the switch configuration, 193
displaying current configuration, 194
enabling and disabling, 193
switch configuration, 192
removing
Admin Domain members, 500
Admin Domains from user accounts, 498
alias members, 348
frame monitors, 560
frame monitors from a port, 560
licensed features, 534
685

LSAN tags, 627
members from a zone configuration, 364
ports from logical switches, 329
zone configuration members, 364
zone members, 352
renaming Admin Domains, 500
requirements
Admin Domains, 487
for F_Port trunking on an Access Gateway, 580
for trunk groups, 572
restoring
configuration file, 280
logical switch configuration, 285
monitor configuration, 567
unordered frame delivery, 127
restrictions
authentication policies, 247
Backbone ports, 321
compression, 446
encryption, 446
fixed-port switch ports, 320
in-flight encryption, 446
on FA-PWWN, 484
on moving ports, 324
QoS zone-based traffic prioritization, 424
upgrading temporary slot-based licenses, 531
Virtual Fabrics, 322
XISLs, 323
rexec listener application, 228
rlogin listener application, 228
Role-Based Access Control. See: RBAC.
roleConfig command, 154
roles
Admin Domain considerations, 153
assigning user-defined, 155
creating user-defined, 154
default, 152
managing user-defined, 154–155
role permissions, 153
root certificates
in Firefox, 205
in Internet Explorer, 205
installing in Java plugin, 205
installing to the Java plugin, 205
See also: security certificates.
route selection for routing, 116
route selection, defined, 116
routes, number supported using FSPF, 116
routing
AP policies, 124
AP route policies, 124

686

Backbone-to-edge, 600, 605
cut-through, 117
device-based, 122, 123, 126
displaying current policy, 122
distance vector, 115
dynamic load sharing, 125
exchange-based, 122, 123, 126
FC-FC setup, 603–605
frame order delivery, 126
frame redirection, 132
link state, 115
lossless dynamic load sharing, 129
out-of-order exchanges, 127
overview, 115–117
path and route selection, 116
performance, 122
port-based, 122, 123, 126
route selection, 125
setting AP route policy, 125
setting policy, 125
Virtual Fabrics, 124
routing policies, 122–125
VE_Ports, 123
RPC, unsupported in FIPS mode, 647
RSA key pair generation, 198
RSA RADIUS server, 179
RSA RADIUS server, setup, 179
RSA SecurID, 179
RSCN, 77
RSCN. See: Registered State Change Notification.
rsh listener application, 228
rstats listener application, 228
rule
adding to an IP Filter policy, 259
configuring zones, 342
deleting from an IP Filter policy, 259
passwords, 157
rusers listener application, 228

S
sa-proposal, 269
saved zone configuration, defined, 341
saving monitor configuration, 567
SCC
creating policy, 243
policies, 232, 242–243
Virtual Fabric considerations, 242
policy member, 232
SCC policies, 242

Fabric OS Administrator’s Guide
53-1002920-02

SCP
configuration for uploads and downloads, 197
described, 196
for certificates, 202
protocol, described, 195
secure protocol, 196
SCR, defined, 53
secAuthSecret command, 249, 250, 453
secCertUtil command, 201, 202, 203, 252, 253, 271,
650
secModeEnable command, 229
secPolicyAbort command, 234
secPolicyActivate command, 233, 236, 237, 264
secPolicyAdd command, 234
secPolicyCreate command, 236, 240, 243
secPolicyDelete command, 233, 239, 240
secPolicyFCSMove command, 237
secPolicyRemove command, 234
secPolicySave command, 240
secPolicyShow command, 233
secret key pair
communicating, 249
for DH-CHAP, 249
length, 249
setting, 250
viewing list of, 249
secure copy protocol. See: SCP.
Secure Fabric OS policies, 232
secure LDAP, 168
secure protocol
HTTPS, 196
items needed to deploy, 196
SCP, 196
SNMPv1, 196
SNMPv2, 196
SNMPv3, 196
SSHv2, 196
Secure Shell protocol. See: SSH.
Secure Sockets Layer protocol. See: SSL.
security
AUTH policy, 243
browser support, 200
certificates, 196
encryption and SSL, 200
IAS remote access policies, 178
IP policy rules, 258
key generation, 201
management interface, 266–275
obtaining certificates, 203
policies, ACL, 231

Fabric OS Administrator’s Guide
53-1002920-02

protocols supported, 195
public key authentication, 198
root certificates in Firefox, 205
root certificates in Internet Explorer, 205
root certificates in Java plugin, 205
secure protocols, supported, 196
SSL certificates, 196
tunnel creation, 272–275
zoning and, 371
Security Association Database (SADB), described, 269
security associations, 269
flushing, 275
IPsec and, 269
security certificates. See: certificates or root certificates.
security considerations for FA-PWWN, 483
security policies, 231–275
security protocols, 195–196
security scenarios, 196
serial number, location on switch, 40
serial port connection, 58
serial port, console session, 58
Server Application Optimization. See: SAO.
sessions, maximum allowed, 154
setContext command, 125, 333, 334
setting
changing passwords, 64
chassis configurations, 95
chassis management IP interface, 68
date, 72
default zone mode, 495
fabric-wide consistency policy, 264
mask for end-to-end monitors, 555
port speeds, 94
QoS zone-based traffic prioritization, 424
QoS zone-based traffic prioritization over FC routers,
426
static ethernet IP address, 68
switch date and time, 72
time, 72
time zone, 73
time zone interactively, 73
settings, configuration, 277–287
shared ISL. See: extended ISL.
shared secrets on Access Gateway, 250
shelf life of a universal temporary license, 532
shutdown
Backbone, 81
switch, 80
SID/DID traffic prioritization, 415
signed firmware, 301

687

Simple Network Management Protocol. See: SNMP.
slapd.conf file, 185
slave port bottleneck detection, 440
slot-based licensing, 526–527
slotPowerOff command, 103
slotPowerOn command, 103
slotShow command, 104, 604
SNMP
configuring, 206
described, 195
filtering ports, 217
host access, 229
password change, 157
SNMPv1
secure protocol, 196
SNMPv2
secure protocol, 196
SNMPv3
secure protocol, 196
switch and chassis context enforcement, 217
snmpConfig command, 653
snmpWalk command, 216, 217
special zones, 337
specification of ACL policy members, 232
Speed LSAN tag, 625
speed, setting for ports, 94
SSH
allowed-user, 198
configuring incoming authentication, 198
configuring outgoing authentication, 199
connection, 59
encrypted sessions, 197
protocol, described, 196
public key authentication, 198
SSHv2 protocol, 196
support, 197
supported OpenSSH protocol version, 197
sshd daemon, 197
ssh-keygen command, 198
sshUtil command, 198, 200, 652
sshutil command, 291
SSL, 251
browser support, 200
certificate files, 203
configuring, 201–205
encryption strength, 200
protocol, 200
protocol, described, 196
security certificates, 200
SSL security certificates, 196
standby CP blade, 175

688

State Change Registration. See: SCR.
static ethernet address, 67
static ethernet IP address, setting, 68
static PID, assigning, 87
static PIDs, NPIV, 86
statistics, bottleneck, 442
status of equipment, 104
status policy threshold values, setting, 106
status policy threshold values, viewing, 105
supported browsers, 200
supportSave command, 40
swapping blades, 99–102
switch
access, 229
access methods, Web Tools, 57
ACL policy distribution, 262
activation and deactivation, 78
adding public key, 198
applications used, 229
buffer credits by model, 143
certificates, installing, 203
changing name, 77
configuring for signed firmware, 301
configuring without disabling, 282
connecting, 82
connecting to a device, 90
connecting with different firmware, 81
default access, 229
deleting private key, 200
deleting public key, 200
disabling, 79, 102
disabling local switch protection, 262
disabling port, 92
displaying name server contents, 54
enabling, 79, 80
enabling local switch protection, 262
ethernet interface, 64
exporting public key, 199
firmware download, 294
firmware version testing, 302
firmware version, finding, 293
host access, 229
joining to fabric, 264
LDAP certificates
deleting, 651
exporting, 651
installing, 650
modifying FCS order, 237
modifying switch configuration, 281
name limitations, 77
names, 76

Fabric OS Administrator’s Guide
53-1002920-02

naming, 75
PKI key pair generation, 199
ports used, 229
restoring a configuration, 282
serial number location, 40
setting date and time, 72
setting port speed, 94
setting status policy threshold values, 106
shutdown, 80
switch database distribution setting, 260
unique names for logical, 77
user-defined accounts, 155
viewing status policy threshold values, 105
switch authentication mode, setting, 171
switch authentication policy, 244
See also: AUTH.
Switch Connection Control. See: SCC.
switch firmware, 294–295
switch WWN in Admin Domains, 492
switchCfgPersistentDisable command, 102
switchCfgSpeed command, 94
switchCfgTrunk command, 574
switchDisable command, 79, 111, 125, 541
switchEnable command, 79, 80, 111, 336
switches supported for FA-PWWN, 483
switchName command, 77
switchShow command, 641
switchShow command, 90, 104, 105, 334, 336, 452,
473, 477, 538, 541
switchStatusPolicySet command, 210
switchStatusPolicySet command, 106
switchStatusPolicyShow command, 106
switchStatusShow command, 104
syslogDIpAdd command, 109
sysShutdown command, 80, 81
system-defined Admin Domains, 488

T
tac_plus command, 190
tac_plus daemon, 190
tac_plus.cfg file, 190
TACACS+ server
adding, 193
deleting, 193
failover, 189
installing, 190
reordering, 193
retry, 190

Fabric OS Administrator’s Guide
53-1002920-02

supported protocols, 189
timeout, 189
TACACS+ service
ADList, 191
Admin Domains, configuring, 191
authentication service, 189
configuration, 189
configuration, displaying, 194
disabling, 193, 194
enabling, 193
home Virtual Fabric, 191
homeAD, 191
LINUX based, 190
modifying, 193
overview, 152
password expiration, configuring, 192
user, adding, 190
vendor attributes, 190
Virtual Fabrics, configuring, 191
Windows server-based, 192
tags for LSAN zones, 624
telnet
blocking access, 227
connection, 59
protocol, 226
unblocking access, 228
temporary licenses, 530–532
defined, 530
temporary slot-based licenses, upgrade restrictions, 531
Terminal Access Controller Access-Control System Plus
protocol. See: TACACS+.
TI zones, 338
activating, 406
admin domain considerations, 398
changing state, 406
creating, 402
creating in a base fabric, 404
deactivating, 406
defined, 380
deleting, 407
displaying, 407
failover, 380–382
FC router limitations, 390
FC routers, 386
limitations and restrictions, 398
modifying, 405
supported configurations, 396
trunk port violation handling, 395
trunking, 397
Virtual Fabric considerations, 402
Virtual Fabrics considerations, 399
with Virtual Fabrics, 401

689

within a Backbone fabric, 389
within an edge fabric, 388
time and date, 72
time listener application, 228
Time server, described, 45
time settings, 72
time zone
setting, 73
setting interactively, 73
time zone settings, 72–74
time, synchronizing local and external, 74
time-based licenses, 530–532
Top Talker monitors
adding on all switches in fabric, 565
adding to aport (port mode), 565
and FC-FC routing, 563
defined, 552
deleting all in fabric, 567
deleting on a port, 567
fabric mode, described, 563
limitations, 565
port mode, described, 562
Top Talkers, 562
logical switches and, 329
topologies
core-edge, 549
mesh, 547
supported for ICL connections, 547–550
topology database, 116
topology discovery, 51–52
disabling, 52
displaying status, 51
enabling, 51
topologyShow command, 402
traffic flow limitations for lossless core, 130
traffic isolation, 379–412
enhanced zones, 384–386
failover, 380–382
FSPF routing rules, 383
over FCR, 386
over FCR with Virtual Fabrics, 401
overview, 379
traffic isolation zones, 338
traffic isolation zoning, 379–412
Traffic Isolation. See also TI.
traffic patterns, planning for, 573
traffic prioritization, 421
QoS zone-based, 419
SID/DID, 415
traffic selector, and IPsec, 270

690

traffic support, 115
traffic, limiting from a device, 415
transaction model for managing Admin Domains, 494
transform set, and IPsec, 270
transform set, defined, 270
traps, 209
trunk area and admin domains, 576
trunk area, enabling DCC policy on, 586
trunk groups
configuring, 573
recommendations, 572
requirements, 572
trunk monitoring, 567
trunk port violation handling, for TI zones, 395
trunking
configuring F_Port for Brocade adapters, 581
disabling, 574
disabling F_Port trunking, 585
disabling ISL, 574
displaying F_Port information, 585
displaying information, 574
enabling, 574
enabling ISL, 574
EX_Port, 577–579
F_Port, 579–586
F_Port considerations, 582
F_Port for access gateways, 579
F_Port for Brocade adapters, 581
F_Ports and Virtual Fabrics, 584
High Availability support, 572
ICL on DCX and DCX-4S, 547
ISL over long distance fabrics, 576
license requirements, 571
managing, 569
masterless, 570
overview, 569
port groups, 571
supported configurations, 571
supported platforms, 571
types, 570
with TI zones, 397
trunkShow command, 574
tsClockServer command, 74
tsTimeZone command, 72, 73
tunnel
configurations using IPsec, 267–268
creation, 272–275

Fabric OS Administrator’s Guide
53-1002920-02

U
U_Port, described, 88
unblocking telnet access, 228
understanding MIBs, 208
understanding SNMP bASICs, 207
universal temporary license
defined, 530
described, 532
extending, 532
shelf life, 532
unlocking an account, 162
unordered frame delivery, restoring, 127
upgrading firmware, 291
upgrading temporary slot-based licenses, restrictions, 531
uploading AD configuration file, 512
USB device, 299, 299–300
usbStorage command, 299
user account
assigning Admin Domains to, 497
creating a physical fabric administrator, 497
for managing Admin Domains, 497
user accounts, 151–194
distributing local database, 158
Fabric OS and LDAP, 171
Fabric OS, 171–172
Fabric OS and RADIUS servers, 172–181
local database distribution, 158
local database of, 155–158
managing, 151–194
overview, 151
password policies, 159–163
password rules, 157
removing Admin Domains, 498
user authentication, 171
user databases, 158
accepting distributed locally, 158
rejecting distributed locally, 159
user-assigned FA-PWWN behavior, 480
userConfig command, 155, 497, 498, 654
user-defined accounts, 155
user-defined Admin Domains, 488
user-defined role
assigning, 155
creating, 154
managing, 154–155
User-Principal-Name, 181
users
assigning to Admin Domains, 496
authenticating, 152

Fabric OS Administrator’s Guide
53-1002920-02

using security certificates, 200

V
validating a zone, 358
validating Admin Domain members, 506
VE_Ports
described, 88
routing policy, 123
XISL and FX8-24, 321
verification check, 604
verifying
device connectivity, 82, 105
High Availability features, 104
host syslog, 109
version command, 604
VEX_Port
described, 88
FCIP required, 606
in Fibre Channel, 596
VF mode
Admin Domains and, 325
definition, 324
See also: Virtual Fabrics
viewing
ACL policies, 233
alias, 349
authentication parameter settings, 248
compression configuration, 452
current default zone access mode, 361
encryption configuration, 452
fabric-wide consistency policy, 263
frame redirect zones, 133
installed licenses, 532
list of secret key pairs, 249
NPIV port configuration information, 476
policy database distribution settings, 262
port information, 465
virtual PID login information, 478
zone configuration information
all, 366
in effective zone database, 367
selected zones, 367
zones, 356
virtual channels, 119
priority groups, 119
Virtual Fabrics
account management, 319
ACL policy considerations, 232
AUTH module considerations, 244

691

base switch
about, 316
creating, 326
changing logical switch to base switch, 331
configDownload restrictions, 286
configUpload restrictions, 286
configuration management, 285
considerations
for Adv. Perf. Monitoring, 552
for WWN-based PID assignment, 86
considerations for ICLs, 547
ContextRoleList, 174, 191
date settings, 72
DCC policy considerations, 239
disabling, 325
DPS support, 124
E_Port considerations, 245
enabling, 324
ethernet interface, 65
extended ISL (XISL), 317
F_Port trunking, 584
FC-FC routing, 636
home Virtual Fabric, TACCS+, 191
HomeContext, 174
ingress rate limiting, 414
interaction with Fabric OS features, 322
interaction with other Fabric OS features, 322
IP address removal, 333
IP address setup, 333
IP Filter policy considerations, 254
LDAP server, 183, 187
logical fabrics
about, 315
context change, 334
logical ISL (LISL), 317
logical switch
creating, 326
default, 310
deleting, 329
displaying configuration, 330
overview, 310
lossless dynamic load sharing, 131
Microsoft Active Directory service, 183
OpenLDAP server, 187
overview, 309
password database distribution restrictions, 158
permissions and Admin Domains, 151
platform services, 47
policy database distribution considerations, 260
ports, moving, 329
QoS zone-based traffic prioritization considerations,
422
RADIUS configuration, 173
692

RADIUS server configuration, 173
restrictions, 322
SCC policy considerations, 242
supported platforms, 320
TACACS+ service, 191
TI zone considerations, 399, 402
with traffic isolation over FCR, 401
XISL, allowing on logical switches, 333
zone alias considerations, 347
zone database size considerations, 362
virtual PID login information
viewing, 478
VSA
permissions assignment, 171
VSA-based account role syntax, 171

W
web server used by Fabric OS, 204
Web Tools access methods, configuration, 57
well-known address, described, 45
Windows 2000
IAS and Fabric OS user setup, 172
RADIUS authentication on, 177
Windows 2008 RADIUS (NPS) support, 178
Windows IAS, Fabric OS user setup, 172
Windows server
LDAP authentication on, 182
RADIUS authentication on, 177
TACACS+ authentication on, 192
World Wide Name command. See: wwn command.
WWN, 533
format for logical ports, 319
switch WWNs in Admin Domains, 492
wwn command, 40
wwnAddress command, 87
WWN-based PID assignment, 86
considerations for Virtual Fabrics, 86

X
XISL
Brocade 7800 restriction, 320
default logical switch restriction, 321
ICL port restriction, 321
on FX8-24, 321
See also: extended ISL.
xlate domain ID, 602
xlate domains, 601
Fabric OS Administrator’s Guide
53-1002920-02

Z
zeroization functions for FIPS, 645
zeroizing for FIPS, 655
zone
access mode, viewing current, 361
accessing, 229
adding a new switch or fabric, 371
adding members, 351
administering security, 371
alias
adding members, 347
deleting, 349
removing members, 348
viewing, 349
Virtual Fabrics considerations, 347
wildcard usage, 351, 352, 354
all access, 360
broadcast, 337, 343
broadcast (reserved name), 350
concepts, 338
concurrent transactions, 376
configuration management, 370
configurations, 341
adding members, 363
creating and maintaining, 362
managing, 368
configuring rules, 342
creating, 350
creating a configuration, 363
creating frame redirect, 132
creation and maintenance, 350–360
database changes, examining, 356
database configurations, viewing, 367
database size, 362
database size and Virtual Fabric considerations, 362
default zone mode, 360, 495
deleting, 355
configurations, 365
deleting frame redirect, 133
disabled zone configuration, defined, 341
disabling a configuration, 365
effective zone configuration, defined, 341
enabling a configuration, 364
existing, 350
frame redirection, 337
LSAN, 337
maximum database size, 362
merging, 362, 371–376
merging scenarios, 373
no access, 360
objects, 340

Fabric OS Administrator’s Guide
53-1002920-02

optimizing resources, 338
QoS, 338
QoS zones, defined, 419
removing members, 352
from a configuration, 364
replacing member, 353
saved zone configuration, defined, 341
schemes, 341
setting default zoning mode, 361
special, 337
splitting a fabric, 373
terminology, 338
TI, 338
TI zones, defined, 380
traffic isolation, 338
types, 339
viewing
database transactions, 377
frame redirect, 133
zones, 356
viewing configuration information
for all, 366
for selected zones, 367
in effective zone database, 367
zone configuration, defined, 341
zone object, 340
zone types, 337
zone command, 132, 133, 369, 395, 402, 405, 406,
407, 409
zone configuration database, maximum items, 362
zone configurations
clearing, 367
creating, 363
deleting, 365
disabling, 365
enabling, 364
removing members, 364
zone database and Admin Domains, 510
zone database, maximum size, 362
zone object
copying, 368
deleting, 369
maintenance, 368–370
renaming, 370
zoneAdd command, 351
zoneCreate command, 350, 424
zoneDelete command, 355
zoneHelp command, 338
zoneObjectRename command, 370
zoneObjectReplace command, 353
zoneRemove command, 352

693

zoneShow command, 356
zoning
advanced, 337–376
advanced commands, 338
defined, 338
enforcement, 342
on logical ports, 350
overview, 338

694

Fabric OS Administrator’s Guide
53-1002920-02



Source Exif Data:
File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.6
Linearized                      : No
Format                          : application/pdf
Creator                         : Dell Inc.
Title                           : Fabric OS Administrator’s Guide v7.2.0
Subject                         : Administrator Guide6
Description                     : Administrator Guide6
Producer                        : Acrobat Distiller 9.5.5 (Windows); modified using iTextSharp 5.1.3 (c) 1T3XT BVBA
Keywords                        : Servers, Storage and Networking#Networking#powerconnect b dcx4s#powerconnect-b-dcx4s#Administrator Guide6#Fabric OS Administrator’s Guide v7.2.0
Create Date                     : 2013:09:05 16:40:05Z
Creator Tool                    : FrameMaker 9.0
Modify Date                     : 2013:12:06 06:05:29-06:00
Page Mode                       : UseOutlines
Page Count                      : 694
Author                          : Dell Inc.
Copyright                       : 2007-2010
Productcode                     : powerconnect-b-dcx4s
Typecode                        : ag6
Typedescription                 : Administrator Guide6
Languagecodes                   : en-us
Publishdate                     : 2013-12-06 00:00:00
Expirydate                      : 9999-09-09 00:00:00
Manualurl                       : http://ftp.dell.com/manuals/all-products/esuprt_ser_stor_net/esuprt_networking/powerconnect-b-dcx4s_Administrator Guide6_en-us.pdf
Readytocopy                     : false
Isdeleted                       : False
Businesskeywords                : Fabric OS Administrator’s Guide v7.2.0
Futureproductindication         : No
Categorypathforfutureproducts   : 
Filesize                        : 9892
Creationdate                    : D:20130905164005Z
Moddate                         : D:20131206030726-06'00'
EXIF Metadata provided by EXIF.tools

Navigation menu