NOS_2_Analysis_Handbook_Ref_60459300M_Dec88_Part_1 NOS 2 Analysis Handbook Ref 60459300M Dec88 Part 1

NOS_2_Analysis_Handbook_Ref_60459300M_Dec88_Part_1 NOS_2_Analysis_Handbook_Ref_60459300M_Dec88_Part_1

User Manual: Pdf NOS_2_Analysis_Handbook_Ref_60459300M_Dec88_Part_1

Open the PDF directly: View PDF .
Page Count: 1000 [warning: Documents this large are best viewed by clicking the View PDF Link!]

NOS Version 2

Analysis Handbook

Reference

yams

This product is intended for use only as

described in this document Control Data

cannot be responsible for the proper

functioning of undescribed features and

parameters.

Publication Number 60459300

Manual History

Revision System

Version

PSR

Level Date

A2.0 562. April 1982

B2.1 580. January 1983

C2.2 596. October 1983

D2.3 617. October 1984

E2.4.1 630. March 1985

F2.4.2 642. October 1985

G2.5.1 664. September 1986

H2.5.1 670. December 1986

J2.5.2 678. April 1987

K2.5.3 688. September 1987

L2.6.1 700. April 1988

M2.7.1 716. December 1988

Revision M of this manual, printed December 1988, reﬂects NOS 2.7.1 at PSR level

716. This edition obsoletes all previous editions. It documents the following:

• Addition of Job Scheduler enhancements containing pseudo-control points; revisions

to IPRDECK entries and DSD commands (SERVICE and DELAY); and new features

including IPRDECK entries (TRACE and FLEXIBLE PARTITIONS), DSD commands

(ENPR and FLEXIBLE PARTITIONS), DIS command (,), and CMR dump directives

(MCT, PCP, and TBDUMP).

• Support of the 9853 Disk Storage Subsystem, the CC598 console, and the 960 and

994 Computer Systems.

• Changes to the TRACER and PROBE utilities.

• Addition of new DSD debugging commands (TRACE and TRAP).

• Addition of EQPDECK entries (UP and DOWN) and a IPRDECK entry

(HARDWARE FAULT INJECTION).

Miscellaneous editorial and technical corrections are made. Technical changes in this

manual are indicated by bars in the margins.

Printed in the United States of America. /^^S

2 NOS Version 2 Analysis Handbook Revision M

Contents

About This Manual 11

Audience 11

Organization 11

Conventions 12

Related Publications 13

Submitting Comments 20

CYBER Software Support

Hotline 20

Disclaimer 20

Controlware Utilities 1-1

Loading Controlware 1-1

Dumping Controlware 1-4

Deadstart '.. 2-1

Modifying the CMRDECK 2-3

Modifying the EQPDECK 2-4

Modifying the APRDECKs 2-22

Modifying the IPRDECK 2-23

L o a d i n g t h e S y s t e m 2 - 2 4

Initiating Job Processing 2-26

Preparing for Recovery Deadstart 2-28

Le ve l 0 In it ia l De ad st ar t 2- 3 6

Error Processing 2-36

Deadstart Decks 3-1

CMRDECK 3-2

EQPDECK 3-13

APRDECK 3-103

IPRDECK 3-107

LIBDECK 3-159

DIS Operations 4-1

DIS Job Dayﬁle Display (A) 4-3

DIS Job Status Display (B) 4-4

DIS Memory Displays (C, D, F,

G ) 4 - 6

DIS Exchange Package Display

(X) 4-8

DIS Directory Display (Z) 4-9

Console Operation 4-10

Display Selection Commands 4-13

DIS Keyboard Entries 4-14

Memory E n try Comma n ds 4-18

PP Call Commands 4-20

DSD Commands 5-1

Display Selection Commands 5-2

Dayﬁle Commands 5-3

Queued File Utility Commands ... 5-4

Job Processing Control

Commands 5-4

Peripheral Equipment Control

Commands 5-24

Subsystem Control Commands 5-50

System Control Commands 5-56

Secured System Control

Commands 5-70

Memory E n try Comma n ds 5-72

Channel Control Commands 5-74

Extended Memory Flag Register

Commands 5-75

Breakpoint Package Commands .. 5-75

Debugging Commands 5-84

Express Deadstart Dump

Interpreter (DSDI) 6-1

Calling the Express Deadstart

Dump Interpreter. 6-3

Input Directives 6-6

Interactive Use of DSDI 6-55

Printer Output Listing Examples . 6-65

Install Command 7-1

K-Display Utilities 8-1

FLAW K Display 8-3

I N I T I A L I Z E K D i s p l a y 8 - 8

Machine Recovery (MREC)

Utility K Display 8-16

MREC Procedures 8-17

MREC Unit and Controller

Reservations 8-23

Mass Storage Extended

Subsystem (MSE) K Display 8-25

Network Access Method (NAM)

K Display. 8-27

Queue File Transfer Facility

(QTF) K Display 8-41

Revision M Contents 3

REDEFINE K Display 8-61

Remote Batch Facility (RBF) K

Display 8-72

Remote Host Facility (RHF) K

Display 8-75

SCOPE 2 Station Facility (SSF)

K Displays 8-85

File Transfer Limit Commands ... 8-87

Transaction Facility (TAF) K

Displays 8-89

L-Display Utilities 9-1

FOTD L Display 9-2

LIDOU L Display 9-3

QDSPLAY L Display 9-6

SCTD L Display 9-10

S D S P L A Y L D i s p l a y 9 - 1 2

SUBSYST L Display 9-20

LID/RHF Conﬁguration Files 10-1

LID Conﬁguration File 10-1

RHF Conﬁguration Files 10-9

QTF Conﬁguration Requirements 10-23

Mass Storage Extended

Subsystem (MSE). 11-1

Introduction 11-1

MSE Utilities 11-13

MSE Operational Procedures 11-78

Tape Alternate Storage 12-1

Introduction 12-1

MAGNET Command 12-8

GENPFD Utility 12-9

Multimainframe Operations 13-1

Introduction 13-1

Linked SDM Operation 13-2

Independent SDM Operation 13-10

NAD Maintenance Utilities 14-1

Dump NAD Memory (DMPNAD) . 14-1

Maintenance Host Facility

(MHF) 14-3

Listing NAD Dumps 14-3

Network Operations 15-1

Network Organization 15-1

NAM Startup 15-5

NAM Shutdown 15-8

Network Control by HOP 15-9

Network Control by NOP 15-21

Network Control by DOP 15-43

Recent HISTORY Command 15-52

Report Unsolicited Status

Command 15-53

Send Message Command 15-54

Status Network Element

Command 15-55

Summary of Network Operation

Commands 15-59

Network Utilities 16-1

NAM Initialization (NAMI) 16-2

Network Dump Collector

(COLLECT) 16-6

Load File Generator (LFG) 16-9

NPU Dump Analyzer (NDA) .... 16-12

PIP Dump Analyzer (LISTPPM) . 16-31

A I P T r a c e W i t h I A F 1 6 - 3 2

Network Description Files 16-34

IAF Stimulator 16-36

Permanent File Utilities 17-1

Introduction 17-1

Calling the Utilities 17-37

Parameters for the Permanent

File Utilities 17-46

Permanent File Utility Routines 17-61

Queued/DayfUe Utilities 18-1

Queued File Utilities 18-1

Dayﬁle Dumping Utilities 18-56

SYSEDIT 19-1

Directives 19-3

Record Types 19-9

System File Initialization 20-1

ISF 20-1

Deadstart Sequencing 20-4

,(<S^u

' '**^S\

4 NOS Version 2 Analysis Handbook Revision M

T r a c e r / P r o b e U t i l i t i e s 2 1 - 1

Tracer Utility 21-1

PROBE Utility 21-37

Character Sets A-l

Character Set Anomolies A-2

C h a r a c t e r S e t T a b l e s A - 3

Glossary B-l

Scope 2 Station Facility C-l

Equipment Conﬁguration C-l

SSF File Transfers C-2

Error Logout C-4

Status/Control Register

Simulator D-l

Using the Simulator D-2

Simulator Commands D-5

580 Programmable Format

Control E-l

Building Programmable Format

Control Arrays E-2

Adding Programmable Format

Control Arrays E-3

533/536, 537, and 585 Printer

Electronic Vertical Format Unit

(EVFU) F-l

Figures

2-1. Deadstart Process 2-1

2-2. Equipment Status Display 2-9

2-3. Mass Storage Status Display ... 2-13

2-4. Mass Storage Initialization

Status Display 2-16

2-5. Controlware Status Display 2-19

2-6. Disk Thresholds Display 2-20

2-7. Level 3 Deadstart Left Screen

Display. 2-31

2-8. Level 3 Deadstart Right Screen

Display. 2-32

3-1. Expander Addressing Map 3-49

3-2. 844 Expander Conﬁguration

With 20 Drives 3-50

EVFU Directives for 533/536

Printers. F-l

881/883 Pack Reformatting

Utility G-l

FORMAT Command G-2

Input Formats G-4

Output Formats G-6

Accessing Disk Devices G-ll

N O S / V E A d d r e s s F o r m a t s H - l

Management Of Storage Media

Defects 1-1

Detecting Failing Devices 1-1

Dealing With Known Media

Defects 1-3

Display Disk File (DDF) Utility ... J-l

Bringing Up DDF J-l

Keyboard Input J-4

Console Messages J-5

DDF Commands J-6

DDF Examples J-12

Pack Recovery J-16

PACKER Utility K-l

Operational Overview K-l

Command Format K-4

PACKER Examples K-6

3-3. Vertically Adjacent Three-Unit

Device 3-51

3-4. Horizontally Adjacent Two-Unit

Device 3-52

3-5. Hardware Conﬁgured into

Seven Devices 3-53

3-6. Example of Ranges of Service

Class Priorities 3-117

4-1. DIS Job Dayﬁle Display (A) .... 4-3

4-2. DIS Job Status Display (B) 4-4

4-3. DIS Data Storage Display (F) ... 4-6

4-4. DIS Program Storage Display

(G) 4-7

4-5. DIS Exchange Package Display

(X) 4-8

Revision M Contents 5

4-6. DIS Directory Display (Z) 4-9

5-1. Record of Original Values in

W,M Display (Delay Values) 5-12

5-2. Record of Original Values in L

Display SDSPLAY 5-16

5-3. Record of Original Values in L

Display SDSPLAY 5-23

5-4. PP Breakpoint V Display 5-78

5-5. CPUMTR Breakpoint Display .. 5-82

6-1. C Directive Output 6-57

6-2. CP Directive Output 6-58

6-3. D Directive Output 6-59

6-4. PP Directive Output 6-60

6-5. Q Directive Output 6-61

6-6. Sample Alternate Directives

File 6-64

6-7. Sample PP Directive Output ... 6-66

6-8. Sample AP Directive Output ... 6-67

6-9. Sample CP Directive Output ... 6-68

6-10. Sample MST Directive Output 6-73

6-11. Sample C Directive Output ... 6-74

6-12. Sample FNT Directive Output 6-75

6-13. Sample PRORF Directive

Output 6-76

8-1. FLAW K Display (Left Screen) .. 8-3

8-2. FLAW K Display (Right

Screen) 8-4

8-3. FLAW Utility K Display (Left

Screen) 8-7

8-4. INITIALIZE Command K

Display. 8-9

8-5. Machine Conﬁgurations 8-16

8-6. MREC K Display (Left Screen) . 8-18

8-7. MREC K Display (Right

Screen) (Page 1 of 2) 8-21

8-8. MREC K Display (Right

Screen) (Page 2 of 2) 8-21

8-9. MSE K Display (Message

Format) 8-26

8-10. MSE K Display (Storage

Module Format) 8-26

8-11. NAM K-Display Format 8-27

8-12. NAM K Display 8-28

8-13. NAM Status Display 8-37

8-14. Connection Slot Array and

PID/Selection Class Matrix 8-42

8-15. QTF HELP K Display (Right

Screen) 8-50

8-16. General Format of QTF Status

K Display 8-53

8-17. QTF STATUS.PID K Display .. 8-55

8-18. QTF STATUS,SC K Display ... 8-57

8-19. QTF STATUS/TRANSFER K

Display. 8-59

8-20. REDEFINE K Display (Left

Screen) 8-63

8-21. REDEFINE K Display (Right

Screen) 8-63

8-22. Reconﬁguration Run Output .. 8-67

8-23. RBF K Display 8-72

8-24. Sample LCN 8-77

8-25. Application Table Display 8-79

8-26. Network Path Table Display .. 8-81

8-27. TAF Initialization K Display

(Left Screen) 8-89

8-28. TAF Status K Display (Left

Screen) 8-94

8-29. CRMTASK Commands K

Display (Left Screen) 8-100

8-30. CRM Status K Display (Left

Screen) 8-102

8-31. CRM Data Base Status K

Display (Left Screen) 8-102

8-32. CRM File Status K Display

(Left Screen) 8-103

9-1. FOTD L Display 9-2

9- 2 . LI DO U L Di spl ay ( LI D ) 9- 4

9-3. LIDOU L Display (HELP)

( P a g e 1 o f 2 ) 9 - 4

9-4. LIDOU L Display (HELP)

( P a g e 2 o f 2 ) 9 - 5

9-5. Initial QDSPLAY L Display

( P a g e 1 o f 2 ) 9 - 7

9-6. Initial QDSPLAY L Display

( P a g e 2 o f 2 ) 9 - 7

9-7. SCTD L Display (Page 1 of 2) . 9-11

9-8. SCTD L Display (Page 2 of 2) . 9-11

9-9. SDSPLAY-HELP Display (Page

1 o f 3 ) 9 - 1 2

9-10. SDSPLAY-HELP Display

( P a g e 2 o f 3 ) 9 - 1 3

9-11. SDSPLAY-HELP Display

( P a g e 3 o f 3 ) 9 - 1 3

9-12. SDSPLAY-CLASS Display

( P a g e 1 o f 3 ) 9 - 1 4

9-13. SDSPLAY-CLASS Display

( P a g e 2 o f 3 ) 9 - 1 5

9-14. SDSPLAY-CLASS Display

( P a g e 3 o f 3 ) 9 - 1 5

10-1. Equipment Conﬁguration 10-4

10-2. LIDCMid Files for Example

Equipment Conﬁguration. 10-5

10-3. Sample LCN Network 10-20

6 NOS Version 2 Analysis Handbook Revision M

/0*S.

10-4. Network Conﬁguration

Statements for the RHF on

Mainframe MFA 10-21

10-5. Network Conﬁguration

Statements for the RHF on

Mainframe MFD 10-22

10-6. Procedure File Example 10-26

11-1. SSLABEL Report File 11-23

11-2. SSMOVE Report: List of

Directives 11-37

11-3. SSMOVE Report: Run-time

Parameter Values 11-37

11-4. SSMOVE Report: Optional List

of Files 11-38

11-5. SSMOVE Report: Device

Status Report 11-39

11-6. SSMOVE Report: Destage

Abandonment 11-39

11-7. SSVAL Validation Report .... 11-48

11-8. SSUSE Basic Report 11-54

11-9. SSUSE Optional Report A ... 11-56

11-10. SSUSE Optional Report B .. 11-57

11-11. SSUSE Optional Report C .. 11-58

11-12. SSUSE Optional Report D .. 11-59

11-13. SSDEBUG: Executing SSVAL

in Problem-Fixing Mode 11-65

11-14. Full Stripe Disk File Format 11-67

11-15. Partial Stripe Disk File

Format 11-68

1 1 - 1 6 . S M K D i s p l a y 1 1 - 7 6

11-17. CU K Display 11-77

13-1. Deadstarting a Machine in a

Multimainframe Complex 13-5

15-1. Sample 255x NPU Network ... 15-3

16-1. Sample NDA Output File .... 16-19

16-2. Sample Session File With

Tasks for Use by STIMULA 16-38

16-3. Sample Session for Use by

STIMULA 16-39

16-4. Sample Session File for

Internal Stimulation 16-41

16-5. STIMULA K Display (Left

Screen) 16-41

16-6. STIMULA Parameters (Left

Screen) 16-43

16-7. STIMULA K-Display

Commands (Right Screen) 16-46

16-8. Parameter Matrix (Left

Screen) 16-49

16-9. Mixed Mode Parameters

(Right Screen) 16-49

16-10. Output Statistics 16-51

16-11. Postprocessing Example 16-52

17-1. Functional Overview of the

Permanent File Utility Overlays... 17-1

17-2. Example of Multiﬁle Archive

File Structure 17-3

17-3. Track Reservation Table

Word 17-22

17-4. Bit Settings for Track Link

Bytes 17-23

17-5. Linkage Format With

Example 17-24

17-6. Example of Families in Two

Conﬁgurations 17-26

17-7. Example of Transferring

Family Access 17-28

17-8. Initial Permanent File

Utilities K Display (Left Screen).. 17-39

17-9. Permanent File Utilities K

Display (Right Screen) 17-39

17-10. PFDUMP K Display (Left

Screen) (Page 1 of 5) 17-41

17-11. PFDUMP K Display (Left

Screen) (Page 2 of 5) 17-41

17-12. PFDUMP K Display (Left

Screen) (Page 3 of 5) 17-42

17-13. PFDUMP K Display (Left

Screen) (Page 4 of 5) 17-42

17-14. PFDUMP K Display (Left

Screen) (Page 5 of 5) 17-43

17-15. Sample Directory Produced

by PFATC,LO=T 17-65

17-16. Sample Directory Produced

PFCAT,LO = T,DN=40,UI = 4054... 17-67

17-17. Mass Storage Table Report . 17-68

17-18. Cumulative Statistics by

User Index PFCAT,LO = S,DN = 40. 17-69

17-19. Cumulative Statistics for

Entire Device

P F C AT, L O = S , D N = 40 1 7 - 7 0

17-20. Catalog of Catalog Image

Record 17-78

17-21. PFDUMP/PFLOAD Example

(Sheet 1 of 3) 17-85

17-22. PFDUMP/PFLOAD Example

(Sheet 2 of 3) 17-86

17-23. PFDUMP/PFLOAD Example

(Sheet 3 of 3) 17-87

18-1. QFSP K Display (Left Screen) 18-3

18-2. QFSP K Display (Right

Screen) (Page 1 of 4) 18-5

18-3. QFSP K Display (Right

Screen) (Page 2 of 4) 18-5

Revision M Contents 7

18-4. QFSP K Display (Right

Screen) (Page 3 of 4) 18-6

18-5. QFSP K Display (Right

Screen) (Page 4 of 4) 18-6

18-6. DFTERM K Display (Left

Screen) 18-29

18-7. DFTERM Dayﬁle List (Right

Screen) 18-30

18-8. QALTER/QFTLIST K Display

(Left Screen) (Page 1 of 3). 18-33

18-9. QALTER/QFTLIST K Display

(Left Screen) (Page 2 of 3) 18-33

18-10. QALTER/QFTLIST K Display

(Left Screen) (Page 3 of 3) 18-34

18-11. QALTER/QFTLIST LIST

Command Display (Right Screen).. 18-35

18-12. QALTER/QFTLIST LIST=qft

Command Display (Right Screen).. 18-35

18-13. QALTER Printer

Listing/Output File 18-36

18-14. Initial QDUMP K Display

(Left Screen) (Page 1 of 3) 18-38

18-15. Initial QDUMP K Display

(Left Screen) (Page 2 of 3) 18-38

18-16. Initial QDUMP K Display

(Left Screen) (Page 3 of 3) 18-39

18-17. Secondary QDUMP K

Display (Left Screen) 18-39

18-18. QDUMP Output File 18-40

18-19. QLOAD K Display (Left

Screen) (Page 1 of 3) 18-43

18-20. QLOAD K Display (Left

Screen) (Page 2 of 3) 18-43

18-21. QLOAD K Display (Left

Screen) (Page 3 of 3) 18-44

18-22. QLOAD Output File 18-44

Tables

2 - 1 . E Q P D E C K E n t r i e s 2 - 5

2-2. Levels of System Deadstart:

I n f o r m a t i o n R e c o v e r e d 2 - 2 9

2-3. Mass Storage Device Recovery . 2-38

3-1. Table of Central Memory Sizes. 3-12

3-2. Required and Optional

Keywords 3-19

3-3. Mass Storage Functions for

Various Device Types 3-42

3-4. Mass Storage Conﬁguration for

Two Systems at One Site 3-44

3-5. PF Entry Parameter Settings

(type, dm, name, device, and sm)... 3-82

18-23. Initial QMOVE K Display

(Left Screen) (Page 1 of 3) 18-47

18-24. Initial QMOVE K Display

(Left Screen) (Page 2 of 3) 18-47

18-25. Initial QMOVE K Display

(Left Screen) (Page 3 of 3) 18-48

18-26. Secondary QMOVE K Display

(Left Screen) 18-48

18-27. QMOVE Output File 18-49

18-28. Initial QREC K Display (Left

Screen) (Page 1 of 3) 18-52

18-29. Initial QREC K Display (Left

Screen) (Page 2 of 3) 18-53

18-30. Initial QREC K Display (Left

Screen) (Page 3 of 3) 18-53

18-31. Inactive Queues K Display

(Right Screen) 18-54

18-32. Secondary QREC K Display

(Left Screen) 18-54

18 - 33 . QR EC O u tp ut Fi le 1 8 -5 5

21-1. Example of TRACER Output .. 21-9

21-2. Example of PROBE Output .. 21-39

A-l. Conversion Differences A-17

D-l. Simulator K Display (Left

Screen) D-3

D-2. Simulator Commands K

Display (Right Screen) D-4

F-l. EVFU Load File F-4

G-l. FORMAT Output,

MODE=RESTORR G-7

G-2. FORMAT Output,

MODE=ALTER G-9

G-3. FORMAT Output,

MODE = FETCH G-10

3-6. Information for Setting or

C l e a r i n g F l a w A r e a s . 3 - 1 0 6

3-7. Example Set of Job Control

Parameters, Part L 3-115

3-8. Example Set of Job Control

Parameters, Part II 3-116

3-9. CPM Default Values for ni .... 3-119

4-1. PP Call Formats 4-20

8-1. Flawing Information for Disks

a n d E x t e n d e d M e m o r y 8 - 7

8-2. Device Initialization Options ... 8-10

8-3. Track Flawing Options 8-11

8-4. MREC Parameters 8-20

A^S

8 NOS Version 2 Analysis Handbook Revision M

00$S

8-5. MREC Commands 8-20

8-6. NA M K -Disp l ay Fiel d s 8-27

8-7. NAM Mode Commands 8-30

8-8. Equipment Requirements for

Reconﬁguration 8-61

8-9. Reconﬁguration Parameters 8-65

8-10. Reconﬁguration Commands ... 8-66

10-1. Suggested Code Conversion

Parameters 10-17

15-1. Parameter Record Feature

Descriptions 15-7

15-2. NS Activity 15-14

15-3. Load/Dump Status Messages . 15-14

17-1. Permanent File Utility

Parameters 17-46

18-1. Queued File Utility

Parameters (DFTERM through

QFTLIST) 18-12

18-2. Queued File Utility

Parameters (QLIST through

QREC) 18-13

20-1. Initialize and Release

Parameters of Fast-Attach Files 20-3

21-1. Data Items 21-19

A-l. Character Sets for Interactive

Jobs. A-5

A-2. Character Sets for Batch Jobs .. A-9

A-3. ASCII to 6/12-Bit Display Code

Conversion A-13

A-4. Nine-Track ASCII Coded Tape

Conversion A-18

A-5. Nine-Track EBCDIC Coded

Tape Conversion A-20

A-6. Seven-Track Coded Tape

Conversions A-23

E-l. Released Programmable

Format Control Arrays E-4

F-l. Released EVFU Load Image:

Print Lines and Channel Numbers.. F-5

0m*<

Revision M Contents 9

0m>K

0m\

About This Manual

This manual describes the CONTROL DATA® Network Operating System (NOS)

Version 2. NOS 2 operates on the following computer systems:

• CDC® CYBER 180 Computer Systems Models 810, 830, 835, 840, 845, 850, 855,

860, 870, 960, 990, 994, and 995

• CDC CYBER 170 Computer Systems Models 171, 172, 173, 174, 175, 176, 720, 730,

740, 750, 760, 815, 825, 835, 845, 855, 865, and 875

• CDC CYBER 70 Computer Systems Models 71, 72, 73, and 74

• CDC 6000 Computer Systems

Audience

This manual assumes you are a site analyst. It assumes you are familiar with the

hardware of your computer system(s) and that you understand the functions of the

various components of NOS.

Organization

This manual includes information required for the day-to-day maintenance of the

operating system and for troubleshooting. Topics discussed include the mass storage

subsystems, network operations, the K and L utilities, backing up and reloading ﬁles,

deadstart, and DIS operations.

Since the sections of the manual are self-contained in that they do not build on each

other, the sections are ordered alphabetically by title. The appendixes include character

set tables; a glossary; and descriptions of the SCOPE 2 Station Facility, the

status/control register simulator, programmable format control for 580 printers, disk

pack reformatting for 881/883 units, address formats for NOS/VE, management of

storage media defects, and the display disk ﬁle utility.

Revision M About This Manual 11

Conventions

The following conventions are used in this manual:

examples Examples of user entries and computer responses are shown in a

font that resembles computer output.

lowercase In a format, lowercase letters represent values you choose.

Numbers All numbers are decimal unless otherwise noted.

UPPERCASE In a format, uppercase letters represent reserved words deﬁned by

the system for speciﬁc purposes. You must use these words exactly

as shown.

Vertical bar A vertical bar in the margin indicates a technical change.

The CDC 18002-2 console is available as an option for CYBER 180 Models 810 and 830

Systems using NOS 2.3, PSR level 617 or later operating systems. This product -^

includes a CDC 634B display terminal (also known as the 721-21 display terminal) and )

an AV117A cable. This console is referred to throughout the manual as the CC634B.

The CDC 19003 console is available as an option for certain CYBER 180-class

machines. This product includes a video monitor; keyboard; 40-Mbyte hard disk

(Winchester) drive; 1.2-Mbyte, 5-1/2-in ﬂoppy disk drive; 640-Kbyte RAM memory; one

parallel printer port; and nine RS-232-C serial ports. This console is referred to

throughout this manual as the CC598B console.

Models 815, 825, 835, 845, and 855 of the CYBER 170 Computer Systems share many **\

of the functional and architectural attributes of the CYBER 180 Computer Systems.

This manual uses the term CYBER 180-class machines when describing these similar

models collectively.

Extended memory for models 865 and 875 and CYBER 180-class machines is uniﬁed

extended memory (UEM) and may also include either extended core storage (ECS),

extended semiconductor memory (ESM), or STORNET. Extended memory for model 176

is large central memory extended (LCME) and may also include ECS, ESM, or

STORNET. Extended memory for all other NOS computer systems is either ECS, ESM,

o r S T O R N E T . 1

In this manual, ECS refers to both ECS, ESM, and STORNET; and extended memory

refers to all forms of extended memory unless otherwise noted. However, when

referencing extended memory in the context of a linked shared device multimainframe

complex or distributive data path (DDP) access, UEM and LCME are excluded. ECS,

ESM, and STORNET are the only forms of extended memory that can be shared in a

linked shared device multimainframe complex and can be accessed by a DDP.

(Manuals dealing with the various form of extended memory are listed, under Related

Publications.)

12 NOS Version 2 Analysis Handbook Revision M

0$ms

0^S.

Related Publications

AU of the following manuals are available through Control Data sales ofﬁces or

through:

Control Data

Technology and Distribution Services

308 North Dale Street

St. Paul, MN 55103

The reader should be thoroughly familiar with the material in the following NOS

publications.

Publication

Manual Title Number

NOS Version 2 Operations Handbook 60459310

NOS Version 2 Reference Set, Volume 2, Guide to System Usage 60459670

NOS Version 2 Reference Set, Volume 3, System Commands 60459680

The following lists contain manuals that provide additional information about NOS and

its product set. For the reader's convenience, these are grouped according to topic:

CDCNET manuals, hardware manuals, NOS 2 manuals, and optional product manuals.

In addition, the NOS System Information manual contains brief descriptions of all NOS

operating system and NOS product set manuals. It is accessed by logging into NOS and

entering the EXPLAIN command.

Revision M About This Manual 13

CDCNET Manuals

The following list contains manuals that describe the Control Data Distributed

Communications Network (CDCNET).

Manual Title

CDCNET Batch Device User Guide

CDCNET Conceptual Overview

CDCNET Conﬁguration and Site Administration

CDCNET DI Installation and Checkout

CDCNET Network Analysis

CDCNET Network Operations

CDCNET Network Performance Analyzer

CDCNET Systems Programmer's Reference Manual, Volume 1

Base System Software

CDCNET Systems Programmer's Reference Manual, Volume 2

Network MEs and Layer Interfaces

CDCNET Systems Programmer's Reference Manual, Volume 3

Network Protocols

CDCNET Terminal Interface Usage

Publication

Number

60463863

60461540

60461550

60460580

60451590

60461520

60461510

60462410

60462420

60462430

60463850

y^^SK

14 NOS Version 2 Analysis Handbook Revision M

Extended Memory Manuals

Programming information for the various forms of extended memory can be found in

the COMPASS Version 3 Reference Manual (publication number 60492600) and in the

appropriate computer system hardware reference manual. Hardware descriptions of the

various forms of extended memory can be found in the following manuals.

Publication

Manual Title Number

CYBER 5380-100 STORNET Subsystem (SNSS) 60000188

Hardware Reference

Extended Core Storage Reference Manual 60347100

Extended Core Storage II and Distributive Data Path 60430000

Reference Manual

Extended Semiconductor Memory Hardware Reference Manual 60455990

0ims

0$ms.

Revision M About This Manual 15

/^ ■S<§\

Hardware Manuals

The following list contains manuals that describe Control Data computer systems and

related equipment.

Manual Title

Publication

Number

CYBER 70 Model 71 Computer System Hardware Reference Manual 60453300

CYBER 70 Model 72 Computer System Hardware Reference Manual 60347000

CYBER 170 Computer Systems Models 171 through 175 60420000

(Levels A, B, C) Model 176 (Level A, B, C)

Hardware Reference Manual

CYBER 170 Computer Systems Models 720, 730, 740, 750, and 760 60456100

Model 176 (Level B/C)

Hardware Reference Manual

CYBER 170 Computer Systems Models 815 and 825 60469350

Hardware Reference Manual

CYBER 170 Computer Systems Models 835, 845, and 855

CYBER 180 Computer Systems

Models 835, 840, 845, 850, 855, 860, and 990

CYBER 990E, 994, and 995E Computer Systems CYBER 170 State

Hardware Reference Manual 60469290

CYBER 170 Computer Systems Models 835, 845, and 855 60458390

CYBER 180 Computer Systems Models 835, 845, and 855

Hardware Operator's Guide

CYBER 170 Computer Systems Models 865 and 875 60458920

Hardware Reference Manual

CYBER 180 Models 810 and 830 Computer Systems 60469440

Hardware Operator's Guide

CYBER 180 Models 810 and 830 Computer Systems 60469420

Hardware Reference Manual

CYBER 840A, 850A, 860A, and 870A Computer Systems 60463560

Hardware Reference Manual

5870 Printer User's Reference Manual 60462720

CYBER 960 Computer Systems 60000127

CYBER 170 State

Hardware Reference Manual

19003 System Console CC598-A/B Operations and 60463610

Maintenance Guide

380-170 Network Access Device Hardware Reference Manual 60458500

>S3SN

16 NOS Version 2 Analysis Handbook Revision M

NOS 2 Manuals

The following list contains NOS 2 manuals.

Manual Title

COMPASS Version 3 Reference Manual

CYBER Initialization Package (CIP) Reference Manual

CYBER Loader Version 1 Reference Manual

CYBER Record Manager Advanced Access Methods Version 2

Reference Manual

CYBER Record Manager Basic Access Methods Version 1.5

Reference Manual

Modify Version 1 Reference Manual

Modify Instant

NOS Online Maintenance Software Reference Manual

NOS Version 2 Administration Handbook

NOS Version 2 Applications Programmer's Instant

NOS Version 2 Diagnostic Index

NOS Version 2 Installation Handbook

NOS Version 2 Reference Set, Volume 1

Introduction to Interactive Usage

NOS Version 2 Reference Set, Volume 4

Program Interface

NOS Version 2 Security Administrator's Handbook

NOS Version 2 System Overview

NOS Version 2 Systems Programmer's Instant

Publication

Number

60492600

60457180

60429800

60499300

60495700

60450100

60450200

60454200

60459840

60459360

60459390

60459320

60459660

60459690

60460410

60459270

60459370

Revision M About This Manual 17

>ﬁ5^

Optional Product Manuals

The following list contains manuals that describe optional products.

Publication

Manual Title Number

Binary Maintenance Log (BML) Message Formats 60459940

COBOL Version 5 Reference Manual 60497100

Communications Control Program Version 3 Diagnostic Handbook 60471500

CYBER Cross System Version 1 Build Utilities Reference Manual 60471200

CYBER Supermini Operations User's Guide 60459850

FORTRAN Extended Version 4 Reference Manual 60497800

FORTRAN Version 5 Reference Manual 60481300

Message Control System Version 1 Reference Manual 60480300

MSSI Version 3 Reference Manual 60458820

Network Access Method Version 1/Communications Control Program 60480600

Version 3 Terminal Interfaces Reference Manual

Network Access Method Version 1 Host Application Programming 60499500

Reference Manual

Network Access Method Version 1 Network Deﬁnition Language 60480000

Reference Manual

NOS Version 2 Full Screen Editor User's Guide 60460420

N O S Ve r s i o n 2 Sc r e e n F o r m a t t i n g Re f e r e n c e M a n u a l 60 4 6 0 4 3 0

NOS Version 2 Tape Management System (TMS) 60463350

Site Operations Manual

N O S V e r s i o n 2 Ta p e M a n a g e m e n t S y s t e m ( T M S ) 6 0 4 6 3 1 1 0

User Reference Manual

NOS/VE System Analyst Reference Set 60463915

System Performance and Maintenance Manual

Remote Batch Facility Version 1 Reference Manual 60499600

Remote Host Facility Access Method Reference Manual 60459990

Remote Host Facility Usage 60460620

><^^y

18 NOS Version 2 Analysis Handbook Revision M

yams

Publication

Manual Title Number

SCOPE Version 2.1 Operator's Guide 60455090

TAF Version 1 Reference Manual 60459500

TAF Version 1 User's Guide 60459520

TA F / C R M D a t a Ma n a g e r Ve r s i o n 1 R e f e r e n c e Ma n u a l 60 4 5 9 5 1 0

Update Version 1 Reference Manual 60449900

j0m\.

Revision M About This Manual 19

Submitting Comments

The last page of this manual is a comment sheet. Please use it to give your opinion on

the manual's usability, to suggest speciﬁc improvements, and to report any errors. If

the comment sheet has already been used, you can mail your comments to:

Control Data

Technology and Publications Division, ARH219

4201 Lexington Avenue North

St. Paul, MN 55126-6198

Additionally, if you have access to SOLVER, an online facility for reporting problems,

you can use it to submit comments about the manual. Use NS2 as the product

identiﬁer. Include the name and publication number of the manual.

Address questions about the physical packaging and/or distribution of printed manuals

to Literature and Distribution Services at the following address:

Control Data ^

Technology and Publications Services

308 North Dale Street

St. Paul, Minnesota 55103

or you can call (612) 292-2101. If you are a Control Data employee, call (612)

292-2100.

CYBER Software Support Hotline

Control Data's CYBER Software Support maintains a hotline to assist you if you have

trouble using our products. If you need help beyond that provided in the documentation

or ﬁnd that the product does not perform as described, call one of the following

numbers and a support analyst will work with you.

From the USA and Canada: (800) 345-9903

From other countries: (612) 851-4131

Disclaimer

NOS and its product set are intended to be used only as described in this document.

Control Data cannot be responsible for the proper functioning of undescribed features or

parameters.

20 NOS Version 2 Analysis Handbook Revision M

Controlware Utilities

Loading Controlware • 1"!

Control Module Controlware ■■ • • • • • 1 - 1

Disk Controlware ..... ......... ■. ■ 1 - 2

Network Access Device (NAD) Controlware 1-2

Dumping Controlware ....... 1-4

(P*

v_,

0$m*t.

Controlware Utilities

This section describes the utilities used for loading and dumping controlware.

Loading Controlware

By using the LOADBC utility, you can download control module controlware, disk

controlware, or network access device (NAD) controlware to the associated controller.

Control Module Controlware

You can use the LOADBC utility to load controlware into a control module for the 834

or 836 Disk Storage Subsystem. The calling job must be of system origin or you must

be validated for system origin privileges, and the system must be in engineering mode

(refer to the DSD ENABLE command in section 5). LOADBC will issue appropriate

messages to indicate the success or failure of the attempt to load controlware.

The format of the command is:

LOADBC,EQ=est,F=loadﬁle.

Parameter Description

EQ=est est is the EST ordinal of the control module in which to load the

controlware.

F = loadﬁle Name of the local ﬁle from which control module controlware is to

be loaded. If F = loadﬁle is speciﬁed, local ﬁle loadﬁle must contain

the control module controlware in binary format and an appropriate

header (refer to the NOS Version 2 Installation Handbook). If

F = loadﬁle is omitted, controlware is read from the system library

SYSTEM.

Revision M Controlware Utilities 1-1

Disk Controlware

You can initiate downloading of disk controlware only from the system console. Also,

you can load the disk controlware to a channel only if it is either active or down and

unassigned. The calling job must be of system origin or you must be validated for

system origin privileges. The system must be in engineering mode (refer to the DSD

ENABLE command in section 5). LOADBC will issue appropriate messages to indicate

the success or failure of the disk controlware load attempt.

The format of the command is:

LOADBC,C=ch,F=loadﬁle,D=dumpﬂle.

Parameter Description

C=ch ch is a 2-digit octal number of the channel to which the disk

controlware is to be loaded. The controlware can be loaded only if

the channel status is UP or if the channel status is DOWN and not

assigned to a maintenance user.

F = loadﬁle Name of the local ﬁle from which disk controlware is to be loaded.

If F=loadﬁle is speciﬁed, local ﬁle loadﬁle must contain the disk

controlware in binary format and an appropriate header (refer to

the NOS Version 2 Installation Handbook). If F = loadﬁle is omitted,

controlware is read from the system library SYSTEM.

D=dumpﬁle Name of the local ﬁle to which 7155/7165/7255 disk controlware is

to be dumped before reloading. This parameter is ignored for other

types of controllers and an informative message is issued. LOADBC

performs a binary comparison between the old and new controlware

and writes this data to a ﬁle that can be processed by DSDI by

using the DMB parameter (refer to section 6 for information on

DSDI).

Network Access Device (NAD) Controlware

The LOADBC utility can be used to load NAD controlware into local NADs (380-170)

and remote NADs (380-170, 380-200, 380-370, and 380-110). Since the NAD controlware

is not automatically loaded at deadstart, LOADBC must be used before a local NAD

can be used by the operating system. NAD controlware may be automatically loaded by

the Remote Host Facility (RHF) when RHF is initiated. Refer to the RHF K display in

section 8.

LOADBC can be called from the console or a batch job. When loading 380-170

controlware into a local NAD, the EST entry associated with the NAD's channel

number must be OFF or the controlware-not-loaded ﬂag must be set.

When loading a remote NAD, a local NAD that is not reserved for maintenance must

be deﬁned in the EST. The EST entry must be ON. Controlware must be loaded and

running in the local NAD before loading the remote NAD.

Remote NAD loading operations can occur concurrently with RHF use of the local

NAD. However, extreme care should be exercised when performing a remote NAD load

to ensure that the correct remote NAD is being loaded and that the remote NAD is

not being used by the mainframe to which it is connected. LOADBC will issue

appropriate messages to indicate the success or failure of the NAD controlware load

attempt.

1-2 NOS Version 2 Analysis Handbook Revision M

^*®^Ss\

Network Access Device (NAD) Controlware

/0!m^.t

The format of the command is:

LOADBC,Pi,P2, •••,Pn-

Pi Description

C=ch

F=ﬁlename

The 2-digit octal number of the channel to which the NAD

controlware is to be loaded. This parameter is required.

Name of the local ﬁle from which NAD controlware is to be loaded.

If F=ﬁlename is speciﬁed, local ﬁle ﬁlename must contain the

NAD controlware in binary format and an appropriate header (refer

to the NOS Version 2 Installation Handbook). If F=ﬁlename is not

speciﬁed, the NAD controlware type speciﬁed by the TY parameter

is read from the system library SYSTEM.

The following parameters apply only when loading remote NAD controlware.

Parameter Description

AC=aaaa

LT = totlt2t3

ND = nn

TY = value

The 4-digit hexadecimal number specifying the remote NAD's access

code (refer to the RHF K display in section 8). The default is

AC = 0000.

The 4-digit binary bit pattern specifying the local trunk control

units (TCUs) that are enabled. At least one TCU enable must be

speciﬁed for remote NAD controlware loading. tn = l enables TCUn.

For example, LT=1010 indicates that the local TCUs 0 and 2 are

enabled.

The 2-digit hexadecimal number specifying the remote NAD's

address (refer to the RHF K display in section 8). This parameter

is required for remote NAD loads.

Type of controlware to be loaded.

value Description

170

IBM

CYBER 170 controlware (380-170)

IBM controlware (380-370)

MIN Minicomputer controlware (380-110)

The default value is TY=170.

Under certain conditions, a remote NAD loading operation will fail on the ﬁrst attempt

but a second loading attempt will succeed. This loading problem can be prevented by

always preceding a remote NAD loading operation with a remote NAD dumping

operation to ensure the remote NAD controlware is halted before loading is attempted.

For example, to load a remote NAD with NAD address 7F and access code F0F0

connected to TCU 0 of the local NAD on channel 5, enter the following commands.

X.DMPNAD(CH=05,ND=7F,AC=FOFO,LT=1000)

X.LOADBC(C=05,ND=7F,AC=F0F0,LT=1000)

Revision M Controlware Utilities 1-3

Dumping Controlware

The DMPCCC utility provides the capability to dynamically dump the CYBER channel J

coupler (CCC) memory in an online environment. The calling job must be of system

origin or you must be validated for system origin privileges, and the system must be

in engineering mode (refer to the DSD ENABLE command in section 5).

The format of the command is:

DMPCCC,C=ch,L=out ﬁle.

Parameter Description

C=ch ch is a 1- or 2-digit octal number of the channel from which the

CCC memory is to be dumped. The speciﬁed channel number must

be in the range from 0 to 13s or from 208 to 33s. The default is

C = 0.

L=outﬁle The 1- to 7-character name of the ﬁle to which the dump is to be ^v

written. The default is L=OUTPUT. 1

DMPCCC lists the CCC memory in line format. Each line consists of 16 CCC memory

words in hexadecimal format. Repetitive lines are suppressed.

1-4 NOS Version 2 Analysis Handbook Revision M

f"'

{*'

xf^'

Deadstart 2

Modifying the CMRDECK 2-3

Modifying the EQPDECK .......... 2-4

Deadstart Displays _ 2-9

Equipment Status Display # 2-9

Mass Storage Status Display . 2-13

Mass Storage Initialization Status Display 2-16

Controlware Status Display . . 2-19

Disk Thresholds Display 2-20

Modifying the APRDECKs 2-22

Modifying the IPRDECK 2-23

Loading the System 2-24

Messages .. 2-24

Entering the Date and Time 2-25

Initiating Job Processing . 2-26

Preparing for Recovery Deadstart , 2-28

Level 3 Recovery Deadstart 2-31

Level 1 Recovery Deadstart 2-34

Level 2 Recovery Deadstart .. , 2-35

Level 0 Initial Deadstart 2-36

Error Processing 2-36

Recovery Functions ............... 2-37

Error Idle Recovery 2-44

Removable Device 2-44

Nonremovable Device 2-44

Deadstart

Deadstart is the process that makes the system operational and ready to process jobs.

After performing the appropriate deadstart procedures described in the CIP User's

Handbook, you can continue the deadstart process as shown in ﬁgure 2-1 and described

in detail following the ﬁgure.

This section also describes what you can do to recover if you experience system

problems during the deadstart process.

PERFORM APPROPRIATE

DEADSTART PROCEDURES

OESCRIBED IN CIP USER'S

HANDBOOK

MODIFY

CMRDECK

ENTER

NEXT.

MODIFY

EQPDECK

ENTER NEXT.

UNTIL YOU

LOCATE THE

DESIRED

APRDECK

MODIFY

APRDECK

ENTER

IPR.

MODIFY

IPRDECK

ENTER TIME

AND DATE

(IF NECESSARY)

WAIT FOR

DEADSTART

TO COMPLETE

INITIATE

JOB

PROCESSING

Figure 2-1. Deadstart Process

Revision M Deadstart 2-1

Deadstart

NOTE

Modifying the deadstart decks (CMRDECK, EQPDECK, APRDECKs, and IPRDECK) ""'***)

does not apply to a level 3 recovery deadstart. Modiﬁcations made during the last

level 0, 1, or 2 deadstart are recovered during a level 3 recovery deadstart.

2-2 NOS Version 2 Analysis Handbook Revision M

0$m\.

Modifying the CMRDECK

If bit 6 of word 13 (word 12 for warmstart on a CYBER 70 or 6000 Computer System

with an active PP) is set (ppp = 001), or if you select the D = Y option on the *P*

display, an instruction display entitled CMRINST appears on the console screen(s) after

the CTI displays on a level 0, 1, or 2 deadstart. .All valid CMRDECK entries are

deﬁned in this display. Several of the entries listed are assigned system default values.

These values are assumed if the entries do not appear in the CMRDECK being used.

To view the contents of the CMRDECK being used, toggle from the CMRINST display

to the CMRDECK display. If either the CMRDECK or CMRINST overﬂows two

screens, the display can be paged.

Modify the CMRDECK by entering the appropriate changes or additions from the

console keyboard. These entries can be made while either CMRDECK or CMRINST is

being displayed. Generally, each console entry supersedes the value currently speciﬁed

in the CMRDECK (or default value in CMRINST).

Refer to section 3, Deadstart Decks, for complete information on all CMRDECK

entries.

NOTE

The modiﬁed CMRDECK remains in effect only until the next level 0 deadstart is

performed. Changes to the CMRDECK are not recovered for the next deadstart unless

a new deadstart ﬁle is created. If you want these changes to take place on the next

level 0 deadstart, make the appropriate changes to the CMRDECK after NOS is up

and running and use LIBEDIT to replace the record on the deadstart ﬁle.

After all CMRDECK modiﬁcations have been made and you want to modify an

EQPDECK, APRDECK, or IPRDECK, enter:

NEXT.

Refer to Modifying the EQPDECK, Modifying the APRDECKs, or Modifying the

IPRDECK in this section. Otherwise, to indicate that all modiﬁcations to the

CMRDECK are complete and you want to begin loading the system, enter:

GO.

Revision M Deadstart 2-3

Modifying the EQPDECK

After completing all CMRDECK modiﬁcations, you can also modify the default

EQPDECK, an APRDECK, or the IPRDECK being used. If no changes need to be

made to any EQPDECK, but you do need to modify an APRDECK or the IPRDECK,

refer to Modifying the APRDECKs or Modifying the IPRDECK later in this section.

To modify an EQPDECK, enter

NEXT.

while the CMRDECK or CMRINST is being displayed.

You can make changes when the EQPDECK, EQPINST, or any one of the deadstart

displays is displayed at the console screen (refer to Deadstart Displays described later

in this section).

Table 2-1 describes the entries that can be made only at the console keyboard at

deadstart time and cannot be stored in the EQPDECK on the deadstart ﬁle. Refer to s*%i

section 3, Deadstart Decks, for complete information concerning all EQPDECK

entries.

After making the changes to EQPDECK you can toggle through the deadstart

displays to ensure all the changes are made.

NOTE

The modiﬁed EQPDECK remains in effect only until the next level 0 deadstart is

performed. Changes to the EQPDECK are not recovered for the next deadstart unless

a new deadstart ﬁle is created. If you want these changes to take place on the next

level 0 deadstart, make the appropriate changes to the EQPDECK after NOS is up

and running and use LIBEDIT to replace the record on the deadstart ﬁle.

After all EQPDECK modiﬁcations have been made and you want to modify an

APRDECK or IPRDECK, enter:

NEXT.

Refer to Modifying the APRDECKs or Modifying the IPRDECK later in this section.

Otherwise, to indicate that all modiﬁcations to the EQPDECK are complete and you

want to begin loading the system, enter:

GO.

2-4 NOS Version 2 Analysis Handbook Revision M

Modifying the EQPDECK

Table 2-1. EQPDECK Entries

Entry

AUTOLOAD.

GRENADE.

0P*\

INITIALIZE ,op,esti ,est2,... ,estn.

Function

Toggles the selection of buffer controller

autoloading for all 7054/7154/7152/7155/7165/7255

control module controllers. This entry is valid only

when entered from the console keyboard. That is,

the AUTOLOAD entry cannot be included as part

of the EQPDECK on the deadstart ﬁle. This entry,

which affects all 7054/7154/7152/7155/7165/7255

control module controllers deﬁned on the

mainframe being deadstarted, is not necessary for

normal system operation but is provided as an aid

to hardware checkout.

Toggles the selection of the grenade function. This

entry is valid only when entered from the console

keyboard. That is, the GRENADE entry cannot be

included as part of the EQPDECK on the

deadstart ﬁle. The grenade function is issued to

all 7054/7154/7152/7155/7255 control module

controllers, once the controlware is loaded. This

function causes unit reservations to be cleared on

all 834, 836, and 844 units physically connected to

each controller. This entry is normally used when

a unit reservation from a downed mainframe

exists on a device. Use this entry with caution

since it can interrupt the operation of another

machine that could be accessing affected units

through another controller.

Creates new labels for the speciﬁed mass storage

devices during a level 0 deadstart. This entry is

valid only when entered from the console

keyboard. That is, the INITIALIZE entry cannot

be included as part of the EQPDECK on the

deadstart ﬁle. Before any mass storage device

deﬁned in the EQPDECK (by an EQ entry) can be

used, it must have a label that can be recognized

by the system. Existing labels are normally

recovered automatically during all levels of system

deadstart. However, if the existing label is

destroyed (for example, during maintenance

operations on the device) or if a new mass storage

device is added to the system, you enter the

INITIALIZE command to create a new label.

(Continued)

Revision M Deadstart 2-5

Modifying the EQPDECK

Table 2-1. EQPDECK Entries (Continued)

Entry

INITIALIZE,op,esti,est2,...,estn.

Function

■"^5\

(Continued)

Parameter Description

op Level of initialization:

AL Total initialization.

PF Permanent ﬁles.

QF Queued ﬁles.

DF System dayﬁle.

AF Account dayﬁle.

EF Error log.

FP Format pack (844 or 895).

MF Binary maintenance log.

esti The 1- to 3-digit number speciﬁed in

the EQ entry for the device (for

example, EQ005...). This is also the

EST ordinal for the device. Any

number of devices can be initialized

with the same level of initialization,

provided the command is no more

than 72 characters.

Depending upon the levels of initialization

selected, all or part of the previously existing

information on the device is lost when the new

label is created. Total initialization of 844 or 895

format pack (AL or FP options) destroys all

information on a device. The other options

selectively purge information. A separate

INITIALIZE entry is required for each option

selected. Selecting FP results in an automatic

system selection of AL. The system deletes all

existing ﬁles, including a system deadstart ﬁle,

from a device initialized with the AL option. CTI,

MSL, and HIVS information is not deleted when a

device is initialized with the AL option. You

cannot initialize the device from which you are

deadstarting. You should initialize a device if you

just loaded CTI, HIVS, or MSL on it.

(Continued)

2-6 NOS Version 2 Analysis Handbook Revision M

Modifying the EQPDECK

0m*\

Table 2-1. EQPDECK Entries (Continued)

Entry Function

INITIALIZE,op,esti ,est2,... ,estn. (Continued)

No options (except AL and FP) are processed until

deadstart is completed. At that time, the K

display is requested and you must enter the

family name (FM) and device number (DN) of the

device to be initialized. This is a ﬁnal check to

ensure that the correct device is being initialized;

the selected options are then processed.

If permanent ﬁles are to reside on the device

being initialized, the EQPDECK should contain a

PF entry for that device. The PF entry

corresponds to the EST ordinal speciﬁed in the EQ

entry and indicates that permanent ﬁles can reside

on the device. If the EQPDECK displayed contains

a PF entry for the device being initialized, a new

PF entry is not required unless the existing entry

is to be altered, or the associated EQ entry is

altered. Redeﬁning the existing EQ entry clears

all associated attributes. In addition, if PF entries

do not exist in the EQPDECK, initializing the

device causes a default family name and device

number to be assigned. Thus, it is necessary to

reestablish the PF entry via the console keyboard

if the device is to remain a permanent ﬁle device.

For this reason, it is recommended that the PF

entry for all mass storage devices used for

permanent ﬁles reside in the EQPDECK. Although

this is recommended, it is not required.

If the EQ entry in the EQPDECK displayed

indicates that the status of a particular mass

storage device is off when the INITIALIZE entry

is made, initialize status is maintained and occurs

automatically when the DSD command ON is

entered for that device during normal system

operation.

Note that initialization of mass storage devices

can also be accomplished during normal system

operation via the DSD command INITIALIZE.

(Continued)

Revision M Deadstart 2-7

Modifying the EQPDECK

Table 2-1. EQPDECK Entries (Continued)

Entry Function

PRESET,n.

PRESET=esti ,est2,...,estn.

Initializes tables (MST, TRT, MRT, and DAT) on

the link device that are required for the

management of shared multimainframe mass

storage devices. The entry is valid only for level 0

deadstarts by the ﬁrst mainframe in the

multimainframe complex to deadstart. This entry

is valid only when entered from the console

keyboard. That is, the PRESET entry cannot be

included as part of the EQPDECK on the

deadstart ﬁle.

nDescription

n Number of shared devices.

If you do not specify n, the link device is preset,

and the amount of table space reserved for the

shared devices is determined by the number of

shared device entries in the EQPDECK.

Presets independent shared devices in a

multimainframe complex. It is valid only on a

level 0 deadstart by the ﬁrst mainframe in the

multimainframe complex. This entry is valid only

when entered from the console keyboard. That is,

the PRESET entry cannot be included as part of

the EQPDECK on the deadstart ﬁle.

esti Description

esti EST ordinal of independent shared

device.

RESET=esti ,est2,... ,estn. Rescinds all device-related attributes except those

speciﬁed on the EQ entry. It restores the values

speciﬁed with the last encountered EQest entry. If

the DOWN command was speciﬁed, EQestn will be

up, but off. Ranges of ordinals are not allowed;

each ordinal must be entered individually. This

entry is valid only when entered from the console

keyboard. That is, the RESET entry cannot be

included as part of the EQPDECK on the

deadstart ﬁle.

y=S^\

2-8 NOS Version 2 Analysis Handbook Revision M

Deadstart Displays

When processing the EQPDECK, the following deadstart displays are available in

addition to the EQPDECK and EQPINST displays. You can page through the deadstart

displays in a circular manner.

Display Description

Equipment Status

Mass Storage Status

Mass Storage

Initialization Status

Controlware Status

Disk Thresholds

Shows the hardware conﬁgurations (refer to ﬁgure 2-2).

Shows how the system allocates ﬁles on a mass storage

device (refer to ﬁgure 2-3).

Shows the initialization status of the devices (refer to

ﬁgure 2-4).

Shows the status of the controlware loaded (refer to

ﬁgure 2-5).

Shows the disk threshold values at deadstart time (refer to

ﬁgure 2-6).

Equipment Status Display

The equipment status display lists the status of all the devices.

Figure 2-2 illustrates the equipment status display.

EQUIPMENT STATUS

EST TYPE ST EQ UN CHANNELS

0. RD ON 0. 00. 00.

1. DS ON 7. 00. 10.

2. NE ON 0. 00. 00.

3. TE ON 0. 00. 00.

Figure 2-2. Equipment Status Display

Revision M Deadstart 2-9

Equipment Status Display

Each entry in the display appears in the following format:

est type st eq un channels

Header Description

| est EST ordinal.

| type Device type. The following device types can appear in the

I equipment status display.

| type Description

CC Satellite Coupler.

CM Control module for an 834 Disk Storage Subsystem.

CP 415 Card Punch.

CR 405 Card Reader.

DB-i 885-42 Disk Storage Subsystem (1 ^ i ^ 3; full

track).

DC-i 895 Disk Storage Subsystem (1 ^ i ^ 2; full track).

DD-i 834 Disk Storage Subsystem. (1 ^ i ^ 8; full track).

DE Extended memory.

DF-i 887 Disk Storage Subsystem (4K sector; 1 ^ i ^ 3;

full track).

DG-i 836 Disk Storage Subsystem (1 ^ i ^ 3; full track).

DH-i 887 Disk Storage Subsystem (16K sector; 1 ^ i ^ 2;

full track).

Dl-i 844-21 Disk Storage Subsystem (1 ^ i ^ 8; half

track).

DJ-i 844-41/44 Disk Storage Subsystem (1 ^ i ^ 8; half

track).

DK-i 844-21 Disk Storage Subsystem (1 ^ i ^ "8; full

track).

DL-i 844-41/44 Disk Storage Subsystem (1 ^ i ^ 8; full

track).

DM-i 885-11/12 Disk Storage Subsystem (1 ^ i ^ 3; half

track).

DN 9853 Disk Storage Subsystem (2K sector; full track).

DP Distributive data path to extended memory.

DQ-i 885-11/12 Disk Storage Subsystem (1 ^ i ^ 3; full

track).

2-10 NOS Version 2 Analysis Handbook Revision M

Equipment Status Display

0SS

Header Description

type (Continued)

type Description

DV 819 Disk Storage Subsystem (single density).

DW 819 Disk Storage Subsystem (double density).

LQ Any line printer.

LR 580-12 Line Printer.

LS 580-16 Line Printer.

LT 580-20 Line Printer.

LX 5870 Printer.

MT Magnetic Tape Drive (7 track).

NC 380-170 Network Access Device.

ND CDCNET Device Interface (MDI or MTI).

NP 255x Network Processing Unit.

NT Magnetic Tape Drive (9 track).

RM Two-port multiplexer (models 865, 875, and CYBER

180-class machines).

SS MSE Controller.

T T I n t e r n a l s t i m u l a t i o n d e v i c e .

The system creates the following device types at deadstart for

internal use. Except for the display console, DS, physical hardware

does not exist for these device types. The device types appear in

the second column of the equipment status display along with the

real device types.

type Description

DS Display console (EST ordinal 1).

NE Null equipment (EST ordinal 2).

Revision M Deadstart 2-11

Equipment Status Display

Header

type

channels

Description

(Continued)

type Description

Used for online reconﬁguration of mass storage (EST

ordinal 0).

Tape equipment (EST ordinal 3).

TT Used for assignment of terminal ﬁles (EST ordinal 4).

Equipment status (ON, OFF, IDLE, or DOWN).

Equipment number that corresponds to switch settings on the

controller that connects the equipment to the channel(s). The site

engineer is responsible for setting these switches. Ask your

customer engineer for the correct switch settings if you are

unsure.

Unit number (serves as ID code for unit record devices). The ID

code provides a method of grouping unit record devices when a

site has several units. Output from a job read in through a card

reader with identiﬁer un can only be directed to a device with

the same identiﬁer. Changing the identiﬁer code via the ROUTE

command can direct program output to a special printer.

Most equipment has the unit number labeled on the equipment.

Some equipment has the unit number as part of a switch or

button. Ask your customer engineer for the correct unit numbers

if you are unsure.

Channel(s) on which equipment is available. An asterisk (*)

instead of a period (.) following the channel number entry

indicates that the channel is down.

2-12 NOS Version 2 Analysis Handbook Revision M

Mass Storage Status Display

The mass storage status display provides detailed information about all mass storage

devices.

Figure 2-3 illustrates the mass storage status display.

0m\.

MASS STORAGE STATUS

EST TYPE

5. DE

6. DB

7. DB

10. DB

11. DB

14. DQ

15. DQ

16. DQ

17. DQ

20. DQ

21. DQ

22. DL

23. DL

24. DL

25. DL

STATUS

B—

S 3

S—

F-

~R—

—R—

FILES

R--T

R—T

Figure 2-3. Mass Storage Status Display

Revision M Deadstart 2-13

Mass Storage Status Display

Each entry in the display appears in the following format:

est type status ﬁles

Header Description

est EST ordinal,

type Device type:

type Description

DB-i 885-42 Disk Storage Subsystem (1 ^ i ^ 3; full

track).

DC-i 895 Disk Storage Subsystem (1 ^ i ^ 2; full track).

DD-i 834 Disk Storage Subsystem (1 ^ i ^ 8; full track).

DE Extended memory.

DF-i 887 Disk Storage Subsystem (4K sector; 1 ^ i ^ 3;

full track).

DG-i 836 Disk Storage Subsystem (1 ^ i ^ 3; full track).

DH-i 887 Disk Storage Subsystem (16K sector; 1 ^ i ^ 2;

full track).

Dl-i 844-21 Disk Storage Subsystem (1 ^ i ^ 8; half

track).

DJ-i 844-41/44 Disk Storage Subsystem (1 ^ i ^ 8; half

track).

DK-i 844-21 Disk Storage Subsystem (1 ^ i ^ 8; full

track).

DL-i 844-41/44 Disk Storage Subsystem (1 ^ i ^ 8; full

track).

DM-i 885-11/12 Disk Storage Subsystem (1 ^ i ^ 3; half

track).

DN 9853 Disk Storage Subsystem (2K sector; full track).

DP Distributive data path to extended memory.

DQ-i 885-11/12 Disk Storage Subsystem (1 ^ i ^ 3; full

track).

DV 819 Disk Storage Subsystem (single density).

DW 819 Disk Storage Subsystem (double density).

,tf*3^v

2-14 NOS Version 2 Analysis Handbook Revision M

Mass Storage Status Display

Header

status

ﬁles

Description

Status conditions. Any combination of conditions can exist. The

following codes are listed in the order in which they appear on the

display.

status Description

A copy of the system resides on this device.

Device is shared by more than one mainframe.

Device is removable.

Independent shared device.

Alternate system device.

Initialization requested or format is pending.

Preset of the device requested.

System dayﬁle on this device.

Account dayﬁle on this device.

Error log on this device.

Binary maintenance log on this device.

System "default family on this device.

Checkpoint device. A copy of the checkpoint ﬁle resides

on this device.

Types of ﬁles that are allowed on this device. Any combination of

types can exist. The following codes are listed in the order in

which they appear on the display. Refer to the MSAL EQPDECK

entry in section 3, Deadstart Decks.

status Description

SSecondary rollout.

BLGO.

LLocal.

PPrimary.

D Job dayﬁle.

RRollout.

0Output.

IInput.

TTemporary.

Revision M Deadstart. 2-15

Mass Storage Initialization Status Display

The mass storage initialization status display shows the initialization status of mass

storage devices on the system.

Figure 2-4 illustrates the mass storage initialization status display.

MASS STORAGE INITIALIZATION STATUS

EST TY OPTIONS I AM DAM FM/PN DN NC

5. DE QSA--

6. DB E-

7. DB —PQSAED

10. DB FT

11. DB

14. DQ 377. 377. FEATURE 40. 000.

15. DQ 000. 377. R4IAE 000.

16. DQ

17. DQ

20. DQ

21. DQ

Figure 2-4. Mass Storage Initialization Status Display

Each line in the display appears in this format:

est type options t iam dam fm/pn dn nc

Header Description

est EST ordinal.

type Device type.

type Description

DB-i 885-42 Disk Storage Subsystem (1 ^ i ^ 3; full ]

track).

DC-i 895 Disk Storage Subsystem (1 ^ i ^ 2; full track).

DD-i 834 Disk Storage Subsystem (1 ^ i ^ 8; full track).

DE Extended memory.

DF-i 887 Disk Storage Subsystem (4K sector; 1 ^ i -£ 3;

full track).

DG-i 836 Disk Storage Subsystem (1 ^ i ^ 3; full track).

DH-i 887 Disk Storage Subsystem (16K sector; 1 ^ i ^ 2;

full track).

Dl-i 844-21 Disk Storage Subsystem (1 *s i ^ 8; half ^^

track).

2-16 NOS Version 2 Analysis Handbook Revision M

Mass Storage Initialization Status Display

rHeader

type

options

Description

(Continued)

type Description

DJ-i

DK-i

DL-i

DM-i

DQ-i

844-41/44 Disk Storage Subsystem (1 ^ i ^ 8; half

track).

844-21 Disk Storage Subsystem (1 «s i ^ 8; full track).

844-41/44 Disk Storage Subsystem (1 ^ i ^ 8; full

track).

885-11/12 Disk Storage Subsystem (1 ^ i as 3; half

track).

9853 Disk Storage Subsystem (2K sector; full track).

Distributive data path to extended memory.

885-11/12 Disk Storage Subsystem (1 ^ i ^ 3; full

track).

DV 819 Disk Storage Subsystem (single density).

DW 819 Disk Storage Subsystem (double density).

Initialize options. Maximum of eight options are displayed.

options Description

FFormat pack.

TTotal system.

PPermanent ﬁles.

Q Queue ﬁles.

sSystem dayﬁle.

AAccount dayﬁle.

EError log.

BBinary maintenance log.

Type of unit device.

t Description

Family device.

Auxiliary device.

0ms

Revision M Deadstart 2-17

Mass Storage Initialization Status Display

Header Description

iam Indirect access mask,

dam Direct access mask,

fm/pn Family name/pack name.

fm/pn Description

F Family name.

X Pack name,

dn Device number if the device unit is the family device,

nc Number of permanent ﬁle catalog tracks.

2-18 NOS Version 2 Analysis Handbook Revision M

/$foft\

jﬁfﬁQiﬁrSy

Controlware Status Display

The controlware status display shows what type of controlware is loaded to which

channels.

Figure 2-5 illustrates the controlware status display.

CONTROLWARE STATUS

CHANNEL CONTROLWARE

00.

01.

02.

03. NN

04.

05.

07. FM

10.

11. NN

12.

13.

Figure 2-5. Controlware Status Display

Each line in the display appears in this format:

channel controlware

Header Description

channel

controlware

Channel on which equipment is available.

Name of the controlware loaded. Refer to the LBC EQPDECK

entry in section 3, Deadstart Decks, for a list of controlware

types.

Revision M Deadstart 2-19

Disk Thresholds Display

The disk thresholds display shows the disk threshold values at deadstart time. The

system uses these threshold values as limits when monitoring disk veriﬁcation failures,

available disk space, and disk error processing. If a threshold value is exceeded, the

system performs a corrective action, such as notifying the operator or restricting

activity on the affected disk.

Figure 2-6 illustrates the disk thresholds display.

EST TYPE SIZE VF RA LS RE UE

6DL-3 3150 100 315 146 100

Figure 2-6. Disk Thresholds Display

Each line in the display appears in the following format:

e s t t y p e s i z e v f r a i s r e u e

Header Description

est EST ordinal,

type Device type.

type Description

DB-i 885-42 Disk Storage Subsystem (1 ^ i ^ 3; full

track).

DC-i 895 Disk Storage Subsystem (1 ^ i ^ 2; full track).

DD-i 834 Disk Storage Subsystem (1 ^ i ^ 8; full track).

DE Extended memory.

DF-i 887 Disk Storage Subsystem (4K sector; 1 ^ i ^ 3;

full track).

DG-i 836 Disk Storage Subsystem (1 ^ i ^ 3; full track).

DH-i 887 Disk Storage Subsystem (16K sector; 1 ^ i ^ 2;

full track).

Dl-i 844-21 Disk Storage Subsystem (1 ^ i ^ 8; half

track).

DJ-i 844-41/44 Disk Storage Subsystem (1 ^ i ^ 8; half

track).

DK-i 844-21 Disk Storage Subsystem (1 ^ i ^ 8; full

track).

/<^^\

2-20 NOS Version 2 Analysis Handbook Revision M

Disk Thresholds Display

Header Description

type (Continued)

type Description

DL-i 844-41/44 Disk Storage Subsystem (1 ^ i ^ 8; full

track).

DM-i 885-11/12 Disk Storage Subsystem (1 -* i ss 3; half

track).

DN 9853 Disk Storage Subsystem (2K sector; full track).

DP Distributive data path to extended memory.

DQ-i 885-11/12 Disk Storage Subsystem (1 ^ i ^ 3; full

track).

DV 819 Disk Storage Subsystem (single density).

DW 819 Disk Storage Subsystem (double density),

size Total number of logical tracks on the corresponding device type,

vf Veriﬁcation failure threshold,

r a R e s t r i c t e d a c t i v i t y t h r e s h o l d .

Is Low space threshold,

re Recovered error threshold,

ue Unrecovered error threshold.

Revision M Deadstart 2-21

Modifying the APRDECKs

After completing all EQPDECK modiﬁcations, you can also modify the default

APRDECK, the APRDECK for a speciﬁc equipment, or the IPRDECK being used. You

can modify an APRDECK only when you are initializing the corresponding equipment.

If no changes need to be made to any APRDECK, but you do need to modify the

IPRDECK, refer to Modifying the IPRDECK later in this section.

The APRDECK contains entries identifying areas of mass storage that are not usable

(ﬂaws). The APRDECK used can vary from equipment to equipment. One of the

parameters speciﬁed when an equipment is deﬁned in the EQPDECK is the

APRDECK number that applies to that equipment. The default (APRDOO) is selected

if this parameter is not speciﬁed.

To modify an APRDECK, enter

NEXT.

while the EQPDECK or EQPINST is being displayed. The APRINST display describes ^

the valid entries. You can toggle between the APRDECK and APRINST. Enter the

changes or additions to the APRDECK from the console keyboard (refer to section 3,

Deadstart Decks, for a description of the entries).

If there are no changes to the APRDECK displayed, enter

NEXT.

to go to the next APRDECK. Repeat this process until the appropriate APRDECK is

displayed or until you have changed all APRDECKs needing changes. '"^

After all APRDECK modiﬁcations are complete, you can skip to the IPRDECK by

entering:

IPR.

Refer to Modifying the IPRDECK, next, for more information. Otherwise, to indicate

that all modiﬁcations to the APRDECKs are complete and you want to begin loading

the system, enter:

GO.

2-22 NOS Version 2 Analysis Handbook Revision M

00®S

(

Modifying the IPRDECK

The IPRDECK contains installation parameters that describe the mode of system

operation. IPRDECK modiﬁcation is seldom required during deadstart since nearly all

IPRDECK commands are also valid DSD commands that make the same changes

during normal system operation. Generally, installation parameters changed during

normal operations (with DSD commands or by modifying the IPRDECK) are retained

only across a level 3 recovery deadstart.

After entering

IPR.

when all the CMRDECK, EQPDECK, or APRDECK modiﬁcations are complete or after

repeatedly entering

NEXT.

to step through all the APRDECKs, the instruction display entitled IPRINST appears

on the console screen(s). This display deﬁnes all valid IPRDECK entries. Most of these

entries are also valid DSD commands.

To view the contents of the IPRDECK being used, toggle from the IPRINST display

to the IPRDECK display. If either the IPRDECK or IPRINST overﬂows two screens,

you can page the display.

Enter the appropriate changes or additions from the console keyboard. These entries

can be made while either IPRINST or IPRDECK is being displayed. A console entry

supersedes the value currently speciﬁed in the IPRDECK.

NOTE

The modiﬁed IPRDECK remains in effect only until the next level 0, 1, or 2

deadstart is performed. Changes to the IPRDECK are retained if a level 3 recovery

deadstart is performed. If you want these changes to take place on the next level 0

deadstart, make the appropriate changes to the IPRDECK after NOS is up and

running and use LIBEDIT to replace the record on the deadstart ﬁle.

For complete information concerning IPRDECK entries, refer to section 3, Deadstart

Decks, and to section 5, DSD Commands.

To indicate that changes to the CMRDECK, EQPDECK, APRDECK, and/or IPRDECK

are completed and you want to begin loading the system, enter:

GO.

Revision M Deadstart 2-23

Loading the System

If you are performing a level 0 or level 2 deadstart, the system library is

automatically loaded from the deadstart ﬁle to each mass storage device speciﬁed in

the EQPDECK as a system device. If no system device is speciﬁed, the system is

loaded on the ﬁrst nonremovable mass storage device in the equipment status table.

Mass storage labels are validated for all levels of deadstart. This ensures that the

conﬁguration matches the one speciﬁed in the EQPDECK. Normally, the device label

validation display appears for only a few seconds. However, this display remains

longer (from 30 seconds to 3 minutes) if 834, 836, 887, or 9853 disk units must ﬁrst

be automatically spun up in order to validate their labels.

If you specify a level 1 or level 3 recovery deadstart, the system library is not

reloaded. In this case, the deadstart ﬁle is rewound and is not accessed again until

another deadstart operation is performed. The system library is recovered from

checkpoint information on mass storage. Central memory tables such as the system

ﬁle name table (FNT), executing job table (EJT), queued ﬁle table (QFT), equipment

status table (EST), and track reservation table (TRT) are either recovered from

checkpoint information for level 1 or from central memory (and the link device,

extended memory, if in multimainframe mode) for level 3.

Messages

If a deadstart error occurs, a message appears on the right console screen and,

depending upon the nature of the error, deadstart processing may halt. Refer to Error

Processing later in this section for complete information and corrective action. If the ^_

system is being loaded (level 0 or 2 only), the name of each system library program is j

also displayed on the right console screen as it is being loaded. This allows you to

monitor deadstart progress.

The left console screen may display the message ENTER DATE YY/MM/DD. You can

enter the date while the system is being loaded.

2-24 NOS Version 2 Analysis Handbook Revision M

Entering the Date and Time

Each time a system deadstart function is performed, it is necessary to enter the

current date and time (except for CYBER 180-class machines; they get the date and

time from CTI during deadstart). The system uses the date and time (updated every

second) for dayﬁle messages and for permanent ﬁle catalogs and directories for ﬁles

being accessed. This includes the creation, last modiﬁcation, and last access date and

time for each permanent ﬁle in the system. It is important to enter the correct date

and time in order to accurately maintain these system records. If you are performing a

level 3 recovery deadstart, it is possible to recover the date and time from the previous

system deadstart. However, this is not recommended since the new date and time

recorded for system records would no longer be accurate.

When the system loading (or recovery) phase of deadstart is about to begin, the

system checks for the correct date and time. If CTI is unable to determine the correct

date and time or the mainframe is other than a CYBER 180-class, the following

message appears on the left console screen requesting entry of the current date.

ENTER DATE YY/MM/DD.

Enter the current date, followed by CR, in this format:

yy/mm/dd.

Entry Description

yy Year; 00 through 99.

mm Month; 01 through 12.

dd Day; 01 through nn (nn is the number of days in the month).

For deadstart levels 0, 1, and 2, pressing CR without ﬁrst entering the date causes the

system to assume the date that the deadstart ﬁle was created. For a level 3 recovery

deadstart, pressing CR alone recovers both the previous date and time (time entered

during the last deadstart plus time accumulated until this deadstart).

When the system accepts the date entry, it displays the following request for entry of

the current time.

ENTER TIME HH.MM.SS.

Enter the current time in this format:

hh.mm.ss.

Entry Description

hh Hour; 00 through 23.

mm Minute; 00 through 59.

ss Second; 00 through 59.

For deadstart levels 0, 1, and 2, pressing CR without ﬁrst entering the time causes the

system to set the time to 00.00.00. If you enter CR alone following the date prompt on

ra level 3 recovery, this prompt to enter time does not appear (the previous time is

Revision M Deadstart 2-25

Initiating Job Processing

DSD commands speciﬁed in the IPRDECK automatically initiate normal job processing

except on a level 3 recovery where you must enter AUTO. If a level 1 or level 3

recovery deadstart is being performed, the system recovers all jobs and active ﬁles and

resumes normal operation immediately. However, if an initial deadstart (level 0) or

level 2 recovery deadstart is being performed, job processing may not be initiated

immediately. The time of initiation depends upon the time it takes to load the system

from the deadstart ﬁle (you can monitor progress on the right console screen). If ﬁle

loading is not completed when the time entry is made, the DSD commands speciﬁed in

the IPRDECK are displayed on the left screen. Until ﬁle loading completes, you can

clear one or more of the DSD commands.

Clearing a command prevents it from being executed when ﬁle loading completes. In

this case, you must manually enter the commands necessary to initiate job processing

from the console keyboard.

To initiate job processing, enter either:

AUTO.

MAINTENANCE

Following entry of the AUTO or MAINTENANCE command during an initial (level 0)

deadstart, the deadstart sequencing process begins. Deadstart sequencing causes job

processing to be suspended until all system ﬁles in the default family are initiated. To

initiate a family other than the default, enter the command:

X.ISF(FM=family)

Parameter Description

family Alternate family of devices.

y ^ = 5 ^ >

For additional information concerning the ISF command and deadstart sequencing, refer

to section 20, System File Initialization.

2-26 NOS Version 2 Analysis Handbook Revision M

Initiating Job Processing

Normal job processing begins after the deadstart sequencing job completes. If the

J^v AUTO command is entered, the subsystems enabled in the IPRDECK are automatically

V assigned to speciﬁed control points. Assuming that all standard subsystems are set to

be enabled, the system calls them to speciﬁc control points as shown in the following

example:

Control Point Job Sequence

Number Name Activity

1 IAF Interactive Facility.

2 NAM Network Access Method.

3 RHF Remote Host Facility.

4 SMF Screen Management Facility.

Magnetic tape subsystem executive routine.

n-2x MAG

n_1 BIO Central site batch input/output.

nRBF Remote Batch Facility.

n + 1 S Y S S y s t e m .

The MAINTENANCE command performs the same function as the AUTO command.

Additionally, it assigns several maintenance routines, according to mainframe type, to

available control points and runs them as normal jobs with minimum queue and CPU

priorities. These are CPU or central memory test routines designed to detect hardware

errors. The routines display error messages either at the control point on the B,0

display or in the system error log. To display the error log, enter:

A,ERROR LOG.

You should monitor these routines from time to time. If a maintenance routine displays

an error message indicating a hardware malfunction occurred, call a customer engineer.

It is recommended that these programs be run at all times. The maintenance programs

use little memory, are run at minimum CPU and queue priority, and are automatically

^^ rolled out if necessary; thus, system performance is not severely affected. Descriptions

^^ of the maintenance routines are in the NOS Online Maintenance Software Reference

Manual.

You can use the SCHEDULE command to schedule jobs, but it does not initiate

subsystems or maintenance jobs. Refer to the NOS Version 2 Operations Handbook

for a description of the SCHEDULE command.

1. n is the number of control points deﬁned by the NCP entry in the CMRDECK.

Revision M Deadstart 2-27

Preparing for Recovery Deadstart

Sometimes during system operation an uncorrectable error occurs that prevents further

system activity. Often the situation can be corrected by deadstarting the system and

recovering prior activity. The success of such a recovery depends upon the severity of

the problem and the extent to which system information is destroyed. There are three

levels of recovery deadstart available (levels 1, 2, and 3). Table 2-2 lists each deadstart

level, including level 0 (initial deadstart), and describes the extent of recovery possible.

If the MS VALIDATION installation option is enabled during a level 1 or 2 recovery,

or if both the MS VALIDATION and PF VALIDATION options are enabled during a

level 3 recovery, the system:

• Verifies selected mass storage files.

• Checks files identified in the queued file table (QFT) to ensure that all tracks in

the chain are reserved and that no circular linkage exists.

• Depending on file type, checks the track reservation table (TRT) to ensure that

the ﬁle is preserved.

• Checks all preserved files for proper length.

If the system encounters a veriﬁcation failure, it clears the queued ﬁle table entry but

does not release disk space assigned to the ﬁle. If a length error is detected, the

system sets error idle status and terminates recovery operations on the device. To

recover from an error idle condition, refer to Error Idle Recovery under Error

Processing later in this section.

During level 0 deadstart, the system veriﬁes the length of preserved ﬁles regardless

of the setting of the MS VALIDATION option. If a length error is detected, the

system reads the disk chain to determine the correct length of the ﬁle, issues a

message to the B,0 display, and stops recovery of the device. To alter the EOI for

the ﬁle and proceed with recovery, enter:

GO,SYS. or GO,.

To terminate recovery of the device, enter:

PAUSE,SYS.

2-28 NOS Version 2 Analysis Handbook Revision M

Preparing for Recovery Deadstart

y * ^ \ Table 2-2. Levels of System Deadstart: Information Recovered

Dead-

start

Level Jobs Queued Files

Perma

nent

Files

Yes.

System

No.

System

Dayﬁles

0Batch jobs are

rerun. Recovered as

inactive. Yes.2

1Recovered from

last checkpoint.

Active queues

recovered from

last checkpoint.

Yes. Recovered

from last

checkpoint.

Yes.

2Recovered from

last checkpoint.

Active queues

recovered from

last checkpoint

Yes. No. Yes.

3Recovered from

CM copy of EJT.3

Active queues

recovered from

CM copy of QFT.

Yes. Yes. Yes.

1. The input and output queues are recovered. Rolled out and executing batch jobs

are also rerun. The input ﬁles associated with these jobs are returned to the input

queues, unless the device they reside on is initialized. Permanent ﬁles are recovered

unless the device is initialized.

2. Dayﬁles are recovered unless initialized by an INITIALIZE entry in the

EQPDECK.

3. Jobs that are rolled out continue. Jobs that are in CM are aborted with EXIT

processing and then rerun if possible.

Revision M Deadstart 2-29

Preparing for Recovery Deadstart

CAUTION

Before attempting any level of recovery deadstart, whether it is a level 0, 1, 2, or 3 )

deadstart, enter a CHECKPOINT SYSTEM command, and wait for the CHECKPOINT

COMPLETE message to appear on the system console. The CHECKPOINT SYSTEM

performs the following functions:

1. All currently executing jobs are rolled out. (Ensures that these jobs can roll back

and continue executing after a level 3 deadstart.)

2. All mass storage tables and track reservation tables are copied to disk, and all

pending disk I/O operations are completed. (Ensures the integrity of the disks

after a level 0, 1, or 2 deadstart.)

3. A system checkpoint ﬁle (which includes a copy of all the system tables in central

memory) is written to disk. (Provides a system checkpoint ﬁle for a level 1 or 2

deadstart.)

If , du e t o a s y st em h an g or o th er p ro b l e m, i t i s n o t p o s si bl e to e nt er a . <* * % ^

CHECKPOINT SYSTEM, or if the CHECKPOINT COMPLETE message does not /

appear, the disk updates may not have been performed. If a level 0 deadstart is

performed at this time, permanent ﬁle information may be lost. To prevent such a

loss, always perform a level 3 deadstart after a system hang. Select the ABORT

option on the level 3 deadstart display to abandon the deadstart; the ABORT option

performs the disk updates and then abandons the deadstart.

The following topics provide general information concerning each level of system

deadstart and recommended steps of preparation.

2-30 NOS Version 2 Analysis Handbook Revision M

Level 3 Recovery Deadstart

Usually you perform a level 3 recovery deadstart following an equipment or system

malfunction (for example, channel or PP hung), providing the system remains intact

Basically, the system FNT, QFT, EJT, TRT, EST, and control-point areas of central

memory must be intact in order to successfully perform a level 3 recovery deadstart.

However, unless you can determine that central memory is no longer intact, attempt a

level 3 recovery deadstart before a level 0 deadstart. This is recommended because

system activity, as it existed at the time of the malfunction, can best be recovered by

performing a level 3 recovery deadstart. Only PP memory conﬁdence testing occurs

during a level 3 recovery deadstart; central memory is not affected.

Requests for device checkpoint are retained over a level 3 recovery. Therefore, if a

system malfunction prevents a device checkpoint from being done, the checkpoint is

processed after level 3 recovery is successfully completed. If a level 3 recovery fails,

always do another level 3 recovery with the ABORT option selected before doing a '

level 0 deadstart. This ensures that all checkpoint processing is done correctly.

f^' On a level 3 deadstart the CMRDECK, the EQPDECK, the APRDECKs, and the

IPRDECK cannot be viewed or changed. The CMRDECK, the EQPDECK, the

APRDECKs, and the IPRDECK speciﬁed during the last level 0 deadstart remain in

effect. If you set the CMRDECK switch (bit 6 in word 13 of the deadstart panel) or

select the D = Y option on the CTI *P* display; the system halts and displays level 3

deadstart options. The options and their default values are displayed on the left

screen (ﬁgure 2-7) and instructions on the right screen (ﬁgure 2-8).

LEVEL 3 OPTIONS

A B O R T . N O

ABORT,B. NO

AUTOLOAD. YES

GRENADE. NO

AUTO. YES

Figure 2-7. Level 3 Deadstart Left Screen Display

Revision M Deadstart 2-31

Level 3 Recovery Deadstart

INSTRUCTIONS FOR SELECTING LEVEL 3 RECOVERY OPTIONS.

ENTER COMMAND TO TOGGLE SELECTION.

ENTER GO. TO CONTINUE RECOVERY.

ABORT.

CHECKPOINT ALL DEVICES AND ABORT LEVEL 3 RECOVERY.

SELECTING -ABORT.* DESELECTS -ABORT,B.*

ABORT,B.

CHECKPOINT ALL NONBUFFERED DEVICES AND ABORT LEVEL

3 RECOVERY. SELECTING *ABORT,B.* DESELECTS -ABORT.*.

AUTOLOAD.

TOGGLE THE SELECTION OF BUFFER CONTROLLER AUTOLOADING.

GRENADE.

TOGGLE THE SELECTION OF THE GRENADE FUNCTION. THE

GRENADE FUNCTION IS ISSUED ONCE THE CONTROLWARE IS

LOADED, CAUSING UNIT RESERVATIONS TO BE CLEARED ON

ALL 844 UNITS PHYSICALLY CONNECTED TO EACH CONTROLLER.

AUTO.

TOGGLE THE SELECTION OF THE DSD AUTO COMMAND.

Figure 2-8. Level 3 Deadstart Right Screen Display

Entering a command will toggle the level 3 deadstart selections. The ABORT command

checkpoints all the devices and aborts level 3 recovery. Selecting the ABORT command

automatically deselects the ABORT,B command. The ABORT,B command checkpoints

all the nonbuffered devices and aborts level 3 recovery. Selecting the ABORT,B

command automatically deselects the ABORT command. The AUTOLOAD command

toggles the selection of buffer controller autoloading. The GRENADE command toggles

the selection of the grenade function. The AUTO command toggles the selection of the

DSD AUTO command. Refer to table 2-1 for more information about the AUTOLOAD

and GRENADE commands. When you are done making changes, enter ^

GO.

to continue the deadstart recovery. If you choose not to display the level 3 options, the

default values are used.

A level 3 recovery deadstart is impossible after:

• An attempted checkpoint recovery (level 1).

• An aborted level 0 (initial) deadstart.

• The MREC utility (refer section 8, K-Display Utilities) has been run for the

machine to be deadstarted while in multimainframe mode.

2-32 NOS Version 2 Analysis Handbook Revision M

Level 3 Recovery Deadstart

It is recommended that you stop system activity prior to beginning the system

deadstart procedure. To accomplish this, enter the following DSD commands:

Command Description

E,M.

UNLOCK.

CHECK POINT

SYSTEM.

STEP.

NOTE

Displays the disk status display (E,M).

Unlocks the system console. This command is necessary only if the

system console is locked.

Provides for termination of job processing and for writing the

contents of central memory tables to mass storage. For a complete

description of this process, refer to the CHECK POINT SYSTEM

command in section 5, DSD Commands.

Prevents the system from processing PP requests. This stops all

central memory input/output (I/O) operations. You should enter the

STEP command after all device checkpoints are completed.

Determine checkpoint status from the disk status display (E,M)

(refer to the NOS Version 2 Operations Handbook).

To recover interactive users after a level 3 recovery, sense switch 1 must be set in

the IAF procedure. Sense switch 1 is set by default in the released IAF procedure.

Revision M Deadstart 2-33

Level 1 Recovery Deadstart

Usually you perform a level 1 recovery deadstart to resume normal processing

following maintenance procedures. The system, all jobs, and all active ﬁles are

recovered from checkpoint information on mass storage.

NOTE

A level 1 recovery deadstart is not intended to be a recovery process after a system/

equipment malfunction. You should never attempt it after a level 3 recovery

deadstart fails.

Level 1 recovery is also useful in system test situations. If two systems are being

alternated, separate mass storage devices and tapes must be available for both

systems. Tapes are not repositioned after a level 1 deadstart. Thus, if a job was

previously assigned to the tape unit that has been used for deadstarting, the job cannot

be recovered. The tape unit should be left unloaded after recovery until it is no longer

assigned to the job (job aborted).

The following rules apply when performing a level 1 recovery deadstart.

• The DSD command CHECK POINT SYSTEM must have been successfully

completed immediately before the end of the last NOS operating period.

« The contents of the extended memory must not be destroyed from the time of the

CHECK POINT SYSTEM command.

® Memory dumps must be completed before level 1 recovery deadstart begins since

memory conﬁdence testing destroys the contents of both central memory (except '^

on CYBER 180-class machines) and PPs.

• The mass storage equipment configuration must be the same as specified during

the most recent level 0 deadstart; that is, the same EQPDECK must be used.

must be the same as or fewer than those speciﬁed during the most recent level 0

deadstart.

It is recommended that you stop system activity before beginning the system deadstart /es%,

procedure. To accomplish this, enter the following DSD commands:

Command Description _____

UNLOCK. Necessary only if console is currently locked.

CHECK POINT Provides for termination of job processing and for writing the

SYSTEM. contents of central memory tables to mass storage.

STEP. Prevents the system from processing PP requests. This stops all

central memory I/O operations.

2. Separate tapes are necessary only if tape jobs are being checkpointed.

2-34 NOS Version 2 Analysis Handbook Revision M

Level 2 Recovery Deadstart

0ms

V Usually you perform level 2 recovery deadstart in system test situations; it is not

recommended for the normal production environment. If you select level 2 recovery, all

jobs and active ﬁles are recovered from checkpoint information on mass storage as in

level 1 recovery. However, no attempt is made to recover the system. Instead, the

system is loaded from the deadstart ﬁle as in level 0 deadstart. In all other respects,

level 2 recovery is identical to that described for level 1, and all level 1 rules apply'

Revision M Deadstart 2-35

Level 0 Initial Deadstart

Use level 0 or initial deadstart in cases where a recovery deadstart is not possible.

This is a complete or initial load from the deadstart ﬁle. Only preserved ﬁles, which

include permanent ﬁles, queued ﬁles, and system dayﬁles, are recovered (preserved

ﬁles are recovered on all levels of system deadstart). Because memory conﬁdence

testing destroys the contents of central memory (except on CYBER 180-class machines)

and PPs, all memory dumps must be completed before deadstart begins.

NOTE

If the machine is the ﬁrst machine being deadstarted in a shared device

multimainframe environment, you must enter a PRESET command (refer to section

13, Multimainframe Operations).

Error Processing

If no CTI display appears when you initiate a deadstart, perform the following steps as

needed. After each step, reinitiate the deadstart to see if the problem has been

eliminated.

Deadstart from tape:3

1. If the unit select switch on the deadstart tape unit is not on (tape does not move),

check the channel, controller, and unit selections on the deadstart panel or display

to ensure they are set correctly.

2. If the unit select switch is on, the correct unit was selected; however, check word

11 of the deadstart panel or display to ensure it is set correctly.

3. Ensure that a 7-track tape is not mounted on a 9-track drive or vice versa. Also,

ensure that a deadstart tape is not mounted on a tape unit that does not support

the density of the deadstart tape.

4. Ensure that the deadstart tape is an l-format unlabeled tape.

5. Ensure that the card reader and tape unit (667 or 669 only) are not on the same ^\

channel and that the card reader is not on a channel with a PP. Also, ensure

that two or more units do not have the same physical unit number.

6. If still no display appears after initiating the deadstart, there might be a parity

error on one of the ﬁrst records of the deadstart tape or the magnetic tape

controller might have detected a channel parity error on a CYBER 170 Computer

System.

3. When deadstarting a CYBER 180-class machine, the disk containing the CIP module must be used.

2-36 NOS Version 2 Analysis Handbook Revision M

Recovery Functions

[ Deadstart from disk:

1. Ensure that disk power is on and the disk is ready for operation.

2. Ensure that the disk has the CIP module loaded.

3. Ensure that the deadstart panel or display is set correctly.

4. Select an alternate channel.

5. If still no display appears after initiating the deadstart, there might be a parity

error on one of the ﬁrst records of the deadstart ﬁle or the disk controller might

have detected a channel parity error on a CYBER 170 Computer System.

For a proper understanding of the problems that can occur during deadstart, you

should be familiar with several basic concepts. For example, because most errors that

occur involve mass storage devices, you should be familiar with their use in the

system. Each mass storage device has a label that contains descriptive information

about its contents. For certain levels of recovery deadstart, this information must be

consistent with corresponding information either contained in central memory or

provided through deadstart procedures. Conﬂicts can result in the system issuing

deadstart error messages. An attempt is made to recover all mass storage devices

deﬁned in the EST during all levels of system deadstart.

Recovery Functions

The speciﬁc recovery function performed depends upon the level of deadstart selected.

Table 2-3 describes the recovery function performed for each deadstart level and the

types of errors you can encounter. The system response to errors and the recommended

action are also listed.

Refer to the NOS Version 2 Operations Handbook for information concerning all

deadstart messages.

Revision M Deadstart 2-37

Recovery Functions

Table 2-3. Mass Storage Device Recovery

■'<:S%v

i J o

8 «

M 0 1 > n

t l W T J

oi c <s

X O 91

tO E U

*J X i O >H

k . J -

O f H E

»H Ck

TJ «k« rH

0> 0» 91

u i a i x

T l 9 1 r H

u . •

h U T J T )

4 1 0 > 0 )

X S i t t X

H - H o

3 H b

6 0 ) X O

U 01

D < H C

< H > r t O

X C l 9 1 - t

k l t i J P k l

r a o .

IM Q> rH Ti

U i X k . it

U T J U

t o n rH

• « J P u

c o t j t j j p

u g i < t

0 3 T J 0 1

H t l U f kl

» 0 1 <H U

u u kl

i i K a o kl

CL 01 tO 4J 91

«l

una

01 h h t n

0 1 6 0 < D J C • r H O

O U C 0 1 k l O t a k i c

k > - H W H r H U •M «H O "H

> C W r l k l r H k l k l

r A ) x i s a < * i 3 > ■H M* U

I O T J C O - H k l O 0 1 C - H k l K

t u a s i t n • 3 K « J O J = T J T < O l r H

9 Q ( i w e u

< 0 £ D 0 1 » H O U ^ « l <

O . k i 0 ) E B ^ 9 1 • » « J P » k l M

i m n - h t o * » j : k l * - r t - >

hH »-H Gtokiki

O T J r H t O d O I -

k l O M

« t j t j a s

o i t o u t u r H t a e

• • k H « a k i o 9 1 0 01 rH

Bumk l C

01 « «H Q T J ( I I J U ^ I I 1 0

O « u h « j p t j 6

o i e c 6 - H « o i

O l c

-4 SO U O k i t t ) ID NH

•h oiuuHsaiHtiu

6 . o u x - i t a t o

■H 01 JP B

> k i B

kl 91

s n i i

>%kftntDft)fg-Ha>>% 01 tx

PJ TJ TJ to as W Q . - H - t C U r t 9 k i C O tr.

Cl D O S k l k i 01

U H

ﬁg

^4 N-l 00 ki cc 0 1 X

k l > > k l 01 00-H CO t k , CO to

OVH T J 6 k i T J

- J N - l t o » - H

CO CO 3 01 0) to to

O * t l 01 Ol O JB TJ >. M t n >

ki rH kl o u 9 t o

BT , T J ^s k l B

B r t d t n O 6 B 1 9 3

o i > e 01 <k l O 18 r CO HI

H I I O CO - 4 C O I d O O) kl

■M TJ kl kl <M 01 CO >«kl IM

v - 3 01 IU X kl CO rH tv, TJ 01

kl kl 10 rH <0 U O B r H

kl j C k l U O . X

B - 5 . t a -» CO

"' IS pa

O l 0 1 to 4 1 0 1 P. H 0 0 isrH 3"BENk. B O CO TJ B

not • r l N H in 01

eu 6 > V S I C O

i t 0 1 H H 0 1 £8a. 0 1 c i

TJ -H

Vt}t

Ck-H kl g"S 0 1 TJ to 0 01

o . t o O O 83 o >

U 01

- h j p 01 co B O

«M 01 B n t o i m X e » O k i

i t u e s i a .KKn9 r - 8h

k i s n

1ii d r

B - H k .

CO kl

n c j t o

x * - M X

t l k l

01 to

O 0 C O 4J

> k l r H CO

0 1 3 t

a t u n

US*

U O k , o

kl 00 k) TJ

01 01 0)

O rH TJ

■H JO N*

> 10 CO

01 > 01

T J O k l

e o i a

01 ki 0)

k l B k l

C O O C O

> . B > ,

t n * t r a

kl <r» Ux

k l • •H 1 O

3 T J ■Mtl

JO 4) k i O c o

CO H w t l to

B B

to -H 01 ?«801 kl JP • 01

kl 01 ti 01 CD oi ■H

>01 O.—1 kl

01 k i 0 1

Si. Ol J B V

TJ UHH

TJ -H oc

r H k l a. CO CO B CS

01 O 0) IM

X c k> <M Ck kl kl

set to O U l O

Tl TJ Cl T B ci 01

■Hk l i t

0 ) 3 ki TJ T i - * U l Ol 0)

■S O B 3 3 j:ki

H O t u Tt 3 S X ki a.

oCl

IM Ol ti

cL.

BrH

rH 01

01 tn J9

Xi tO to

aarH

ki Ou

44 ty

Xjp ti to

ti TJ

0) nH

M•H

co IM •rt 01

B 6 u

-rt Cl nH

■001 •M

oo a 0> Ol

C O C l k l - r t

at -( co eg a tj tj

(Continued)

ytSS^K

2-38 NOS Version 2 Analysis Handbook Revision M

Recovery Functions

Table 2-3. Mass Storage Device Recovery (Continued)

m k i e

0 ) O k , o

k i k i A h e SK

( M J J W 0 1 03

C O 0 ) 0 ) C B k l r 4 9 1

t a o i t j m c o e u c

C B - r t

S s 0 > > « S 3 > >

C O r H - 1 0 ) 0 0 1 0 • r t » ) B •

H b l l T J - r t T J r t M 6

C k B t l I M I M •

Bc o o o i o e o t j Z k ) k l H

er t O k , O 9 1 - k l 0 1 k l

T J k l r H O k l k l C d C O T J C l

k> C D O U t o 0 ) k l C O

Cl s r - < c i > o i x > ( U C A O

xamlne E, eadstart eterntne ctlon. erform le r redeﬁn o match t elng reco Examine thi after dead: complete ti corrective

H T J T J C 0 O . O k l j O

0 1 O S

4 1 r H 3

B01 C l T J k l

O k l 9> k . - h c W

I T _ o TJ nH N H > N - l k l

> M H I B T J C O N - l k i g l kl 0 1 0 1 C k 0 )

o t n k i b o i o O k l t> £ T J C O r

kl X * 1 0 C B r - 4 t X to CO k l - r t z j 6 4 1 O T J

o0 1 0 1 © C O * < i T J k l N - i O M k l n H

kl I B C r i C k > , k l I M O < M k l - 4 0 1 O C r H 9 1 r t

M3 H X O . B J - ' 0 ) -t

H > kJfBrJOlOlStTJtQ

to CO CO COrHIQX CB.ﬁ U B t o C O • Q r t r i 9 1 >

u u h C k j r w c o o o - t X c n T J H J P o i k l J <

oo i T , c i t a u , m k l C J oicoeotojptscioi

uc o c s t o o o - M e o 3 . u B — t n a t o - t C U C H O J | l b

o i 7 t o t j c CJ 01 j r o i > H k . H H H U £ I t

Ol k l = > t o t o - t I C a X rH k l O < M 3 B f * . C O ( J

01 C O E C O 0 1 o > t i 01 kl k l H U H > 0 1 0 ! > s

BC I C 9 3 0 1 H 0 ) M l . kl > k . T J > O S 0 1 H C O X 0 ) B C U

MttuePtiox

0N - i 3 e o o i > s n O TJ OJ e) 01 41 CO

Ck T J k i k , c n C l M b ] 01 01 r H k l E T J 0 ) 3 0 1 3 r t X k l

CD E C O l M 0 0 E O O r t k i X T T 0 ) 3 C l C U k l C k > k l

01 n - i M S J E O k , l - l n J O TJ t t 0 1 o U t O 0 1 k l • 4 1 t M

OS 0 > > H H U B , Z h T J C O C CB i h g I S k l I M « » ■. T J 4 > n H

4 1 » O k l 01 01 CO JP C O r H 0 1 H u

eTJTJCO Oklkl tt] O kl Ot TJ CO M s r . t i h c o o o t o o i

0) O S r t r t X k i C £ > > « CO ra > c i t i r t X X 6 C

kl O t O r t 0 0 C O O h - r t k l B • r l O 9 1 N H r t I C B l N r t T J H

CR C l O r t k l D C ( - k i 0 1 01 E k . B B H H B E c j B

. >. I h t i C l w i < n K U ( J M t a x k l 0 1 T i 3 H n H r t e t i

CR 0 O - - • T J 1 0 n £ < 0 )

k i U > 0 1 0 ) 1 0

U b | | X X «

C O 0 1 T J H k l T J

■H » i m k . T J t a x a t c s h o i

—'NHuUkiEnki

h b H O I I l l £ I I

< o 3 Q . t n k . u t j

• 0 1

0 1 « - « > •

> C O - r t 0 1

N H t t l k l 0 )

01 k l r H C J < H

Cl C J H C B H

C O I M k l I M

>O X

01 O C O 4 - 1 C O

a6 C O I I r t I D

s ^ 0 1 - H S U

MH O J P O

00 1 O 1 9 0 1 O

r H C B > H I I 01

01 x o x JP

CL t t i t B C O k i to

>s > C l 0 1 > C l

Remo dtre Nonr remo dire 01

• M » f l

k , C O - " • k . 0 1

o L i m k . O H

ki k > t a r ^ | 0 k i T J

OU 3 T J < m S ^ - n . • k , E

u0 1 J P 0 1 k i o T J t i 0 1 - H

uN H O H W C C B.

to C < B t u H H C f O O )

o t o n h o i 3 a i t n

c e c c n

O E r t 0 1

IM nHOIIiJPOII ci <""• oi r t H u j p

ck l m 0 1 C O C l 0 1 C O t i

t o > h n i i i o . k a

k l E T t

t e h 3 k i

BC l « ! > M o c i t e b o

OT l « I I • | l £ t l k l r t E c

HI M X I X C J T J k l —i C M o IM 01 IM

UH - i T J C r t t i O t )

Ou liHUCtntOCOT k . H r

>M o i a i o e c j n h u o i O i 6 t n 3 T J

M> < C ( U U O r 1 b > O 0 ) O 0 >

Cl c o x c o o t n c r a h c i h

01 H H T J « U — 1 3 H H T J n W

£Q J i M H U i l f t 1 1 O C J l r l

J l I I 3 H j h h u J P J P r t C l M

COXOOBESSuh

j h u t i i N H 3 e j j o

tr B C l. CO 41

_ 3 _ i i d a >

I M < - l

0 k ,

to fvl

- 4 k l

OJ 01

5> TJ CM

01 <0

J 0 1

o-*

H 0 1 * t e •

Bt e h t j . - » OJ

0C l H T J • O J T J — M t l

r t A " ) H C I I l l « 1 t n o r t

kl CC OiSOHrtHEBOl U r t C J >

Cl 41 >» klO-rt9 0.0.3rtrtt5 o H 0 1

c— . j r t n e o < » t o o t . ( - > M TJ

3X CL ^bCCIrtOeniCU '3 CO 0)

u. r e x t r . - H o < r r t u x t o - t - t t i w

> M 4 1 u c ^ X — ' 4 ) 4 1 « > — i

U J k . 0 1

9*. O - t > 9 1 O t — 1 - 3 X C O C l B . C E O " C

ue 3 r t X 0 < ? £ m 4 ) t n - c B C T , i J C

o> 4 i M E m > — ' 1 o c u n i o - h

>t . e n 3 s e > ^ ■ i t i u u h o i x m o i t j a

oOICIErt-H^-^-. «rt >I-HHS)H 41 01 K

oE O 0 1 h C M 0 0 J 3 < — 0 1 t O T J C C l 0 0 r t

01 rtrttnual QOJ^rt-BWBtirt rt C ki

oc t O > C H C v < T T J r t C J 1 r t O u -

£noc«4cio-ig^eeo.u,ii

> t e r t

0 1 X 3

UTIUSwCOigOltaODHtvOt j c i e

(Continued)

Revision M Deadstart 2-39

Recovery Functions

Table 2-3. Mass Storage Device Recovery (Continued)

.-^mu.

0 1 0 0 C O

o o y _ B 0 )

t n H 0 0 r t r H

9 . > 6 k l r t

< Q 4 ) r t O I M

Om O TJ 03 rt

3 K k l

k i I M 0 1 4 1 E

O H f O 4 )

ij «e e •

• N r t Q k l

C O • » H B C O

> i T J T J rr t S d t, O

CO Ol SI CO ti 01 H

« 0 1 t i r t 0 1 a

O O k l r t C J — X

u 6 e ﬂ rt

0 1 O . 0 1 H a 4 ) H

5 . t i x h

t s o c o a e g < o h

HUHHB.H 3

IBM

H O O

kl S» • O l 0 0 C O - n .

09 01 t H C l B 0 1 O h

TJ TJ C O r t 0 0 r t r t 0 u • k l

CO S • > C k l r t C O T J r t

<Q0lrt0!IM 41 B

C m C J T J t H r t I M r t O

01 41

T J X

X t i p •

• ki C B

tl 3 X k l r t U i

ok i I M C J 0 ) C r t T J

SI O r - i X T J 0 1 O B *

bl CO 01 O H M k l 4 1 C C • U C O > t

r N r t C O k l a k i

C O • - t , B 0 9 C O k l k l

> • T J T J r t C O E O U C

u k l r t 01 rt

Ol co ol k l > H

X T J T J C l oi a

H r t lAOISICOlHOIHacOOl

kl » O J k l r t J ) C i . « k l

01 TJ kl 01 Eb O u t i n u o i 3 r a t v

B > ( 9 O k l r t U — X t 3 r 3

k l C B S H 3 r , C 0 t Or t t i 0 1 H

B 4 1 k l k l

O r t

IM 0 1 C l 4 ) r t a l 4 1 r t l i o i - n n

a ) k i s i t l H T J C L t i X r t k l T J U .

X >M rt 41 0 1 B rococo te a Tt e si o.

W 1 0 O . t i Oh 03 ( r t k l r t H O . H 3 C J I h t n

Cd 01 03 qj kl TJ 4)

H 6 B 0 0

I ^ I M C O

CO Ih im

_ CO O O <-*

r t 0 ) U H C O 0 1 H O

_ . 6 w 0 - a r t

k i o i C k a k , o i u h

t k i o >

<c o u s

0 1 ! » . 0 0 T J

X t o E 6

k l H r t C O k l

4 1 O H

B H t O

r t T J U J

co se oi

X - X

W W k l

ta cs rt eo a o w

C k k l C O E T J 0 1

B 0 3 > » h o ) t o t l

0 1 T J H H 0 1 4 > r t

k l C O « 3 r t T J r t >

k l 0 ) B C B 0 1 H 0 1

< TJ IB UH ti UH TJ

k> B

r t t j »

a C B

0 1 C O ca » r i

O.SU

Ck CB rt f- CO >

CO rt CB od 3 a

ax k l B

01 CO Id CB 3

00 rt 00 in kl

co tj e ﬁt o k i

C O r t UH

a oi co Id TJ 01

4] rt CO

B 0 01 e rt

MCB

a a U. CO

00 6 o Ul r t H

B 0 U Cu (H

t\ t i e u S" 00 01

3OS l-l o l c i

O k l k l SC CO T J r t

r t X I h ﬁCO O 8 >

rt CO <e U V

O H k l TJ

UJ ti CO

TJ as ki

o —

01 01 03 k , f - l

X X 0 1 bISH k l T J

. 2

c o p a

, ki

"to 2 *

• X f H

e o C O

5 8

0 O C J

c o c i

0 0 B O

O kl kl

r t X t i

r t 0 0 t o

01 01 CB

X X 0 1

fH kl T J

6. SK kl

a t o r t

e g h t o

a > a

0 0 r t o o

CB TJ 6

C B r t

CO 01 CO

8 H C OO 0 1

CB Cl

0 0 B O

B O t i

r t c i a

O k l k l

r t X k l

H 00 C B

OrtM

Uh ti CO

II 01 CB

X X 01

H k l T J

• > to

sss r t 0 0 t o t o ,

O r t M <

O X

kl CO

I I r t 0 9

H 3 0 1

rt SHrt

3x i n

Cft V O t M

01 CO CO o

U 3 _ 1 0

ucb 3 m

3 Cl 41 kl

0 1 T J

kl X 10 01

B TJ 00 CB

r i B H

co 3 r t U J

S >»Nrta)

>Cl UH

Tt SN tkl

k l W C l k , p

r t r t 0 1 k l C S

tj x a b Ou

W H « 41 <

tn X O T3 >

-n» 01 01

k l « J r H t i TJ

e g b O l k i 0 )

CO TJ

T J r t

01 5

2 -*

01 Ui

•H 01

II H

(Continued)

2-40 NOS Version 2 Analysis Handbook Revision M

Recovery Functions

Table 2-3. Mass Storage Device Recovery (Continued)

v i C Q

01 CB

T J 0 1 t i « C 9 0 1 C l 01 01

B 6 k i C B H 0 1 0 1 B t . k . r 4J

A r t 0 ) k l O C O X k l o > • Sk kl 3 — 0 9 9 K CB

B r M C l O O r t M B

. « C < k C B X T J m > T 3 4 I

00 41 CO HklCO<!oiUuBEOO

0 1 ' K CB k l C O r t « r t Cl

U H T J S k 0 1 0 1 0 1 H k > r t t -

c i a U H k l B X S k U 0 1 C B 0 1 r C T J k l e

• k l w 0 > T J U H B J C O B r X 0 1 r O T J 1 4 CO ^rt H T t C l C l k l C O r t B O k ,

tdvaicokikiTjti o Brtki O l k l C J k l T J 0 1 E r t 0 3 r - i > ^ ^ 4 U i 0 1 r t I I M

TJUCOOk, O 41 s*xkirta X r t t i E B r t 0 1 T J r t l i b l C O I ) > k l e g u t j l i o i

. - r t 4 1 U j C S k X B O O I H O k i u h > e g t o U U i 01 ci tj oi g

r t T J 8 0 1 0 1 E r t

I I r t X B T J E

B t ﬂ O > k l U U h U U V D , Q H

• u « a o . o o i c s o ) u h O o i k i 3 r t ^ X k i kl

OU h T J C O k l k l C l C J . k . I I • 0 0 O O r t a r t 3 r t

Ti < TJKBSJSrQklknCI II

c b i i 3 o : o n e n h • a , k ,

O O l T J t i 0 1 b l W e g T J l i O X 0 0 k l — i B t i - 4 U H

U h 5 C l O 0 C O

14 B T J t O C B 9 K a o h k i E 3

U o i s k o o t j t i < T t > k i x t e 0 1 O 0 1 0 1 k l r t t o a . o i a 0 I . r t C B S k c j c a c j 4 '

«t X C B B B 0 1 O • 4 1 U u 6 • • r t T J C B

eg oi w

X T J k l l i k . N H C e g w k , r t

k i r t r t t o m • i n k , a k i t j a « ~ O 0 0 I I U i 4 ) T J k i e g OJ O O r t T J T J U T J 0 0 O J r t

a r t t i O o i j ^ k , k i c c c _ B > r t l i C O C l X 01 01 e g 0 1

• 0 1 I k ,

k . 6 X u h

I ) t O H M O r t M M C l O C O I ) 0 < r t O r t O C l O l r t m e g 0 1 T J M U h 0 ) t o 0 1 r t 0

B r t a C U u < £ 6 0 1 X 6 T J 0 1 k l 3 B 0 1 T J T J 0 0 r t 0 1 o 0 1 B 4 1 0 1 B k l O O T J 3 0 9

r t T J U H B O S S k h C X C O r t C O C O C J

0 4 1 T J r t 4 1 C J r t 4 1 C J r t

O O 4 1 a E 0 1 O U i > B S S > O k , > r t t n e g c j b e o

8 T J 0 1 t l 4 1

u h r t o i o e g o s H O S r t a oi

b a o o

tSSCIMOC3>TJ TJnHT3HS> t i - i C B M B U h 4 1 C l e g t o c n o i o i c i

X -X MSIBOIBOBCBOIrtCI 91 O K u . O 9 K 0 1 U h X 0 1 X 0 1 0 1 k i X C l

C d I d k l < T J O H ( I I O U H M u h T J a u h i i b l < 0 C l M k . O M U b l T J X a k l O B

c E 01 09

o • 01

ca oo 41

C O T O b l C O TJ C l 0 1

t i B h k , 6 • • •

Lo < j T J U H 4) U 0 1 X 09

ti 0 3 C B • r H t o e g • c m s b J E 5 =

0 1 0 ) X i O O O u

B O e g e o k i Ui 01

0 0 1 C O < I d U to to

ti a S k k l b . a s k k i a to u t o E

ba t o h c u a r t t i 0 1 H Z M C l 0 1 T t O

bl 1 0 H 0 3 C J C B r t O J O X U H 0 . 0 0 1 4 1 I d

a x z a x b l M B eg M B 0 1 >

O 0 1 C O r t 0 1 0 9 k, r t r t O S 0 w

M 0 0 r t 0 0 C J 0 0 r t 0 0 0 1 C l o . . T J I I r t b l TJ

C S T J 6 S E t O T J 6 k> X U h C O O J t i I n

41 C O r t b l

a o i o s p C O r t 1 0 4 1 - t o U h ( - 9 K t d

O 1 0 r S

O 4 1 k i

- t X V

CO C Q 0 1 C O §*

Bc j h c b t y

ﬂ O 4 1 b l

4 1 H D r t

B O 0 1 k l

(Mm T J k l O u f- > B

aM B Uh k l • C B • eo 0 1 3 B

c o c j t o a o e o c j CO U4 01 B t i 0 1 R C O C O

3 0 X C B J O C

blT J S C r t

09 0 0 B O 0 0 E O 0 1 Z 0 1 O

01 B O I h C h b o t . e y f a k i O k . k l O - H O

a r t c j a e e - c j a W g t j t e • 01 41 O k . U H - t o t

3 < 6 r t « 1 4 1 t o Ih « J 6 U h O C

B O k l k l f _ O k l k l r t t e e g 0 1 TJ eg k l r t 0 0 I d

oH X l i C O r t X k l - 3 B O O r t B E 6 " 09

1 4 • r t 0 0 « C r t e e a b . Cl CO 0 1 r t 0 9 - r t C B b l

a k , 3 o j r t OJ

aO r t k l H S O r t M X t l T 3 > 01 II rt

SK u h k , C O b l u h k . c o O B w 4 1 k , 3 CJ U c o ( 0 0 C 7 > - • tn nH

cn T J O T J C J 0 1 0 T J O k l M 4 1 0 1 H t o b .

O i o i C B 4 i o i a M 1 0 C J - 4 a r t

x x o i X X 0 1 I h k l 01 6 h a o

H k l T J l - l k l T J b l 0 9 TJ r t U H C B U h

X •

M . i - *

Cl T l V 41

-i 3 c l Cl

>Ert

41 4 1 0 J 0 1

a41

C J r t T J

r t X H

X > e g c o

0«3

O l » 0 1

t j o u 0)

CJ kl

a 6 B 4 1 B CO

9k Ol 0 1 k , 0 1 9k

Hti

k l E k l

BOO)

S K 6 9 k

C O < r t C O

-4

SK c o t s

-4 t c t s

Xt , O 4 1 3

O 0 1 B a • o i • H c l

k, -TTJ ti 0 3 k l J t C l C J

U H r t C O r t B

g C C l k i

O 0 1 0 1 o H H e s O

u0 0 k l 3 C O 0 1 X 0 3 > o e g u i •

k, B C B I S 0 1 O 8 O l B 0 0 4 1 k l C l I h

Cd r t r t C O B

O k l U

B r t C J T J J *

t a > B o i c i

0 1 X 0 9 C l I I O

— 1 ( r t r t 0 1 C O k l

UH e t i o t o co oi a u Sk eg k . C J

o0 1 6 k i •

3 r t t i S O l

- a o r t u a

B s » e g •

T J J S r t M 0 1

1 0 • C l T J C O 0 9

B C O l O C l r t 0 1 0 1 0 1 H 0 1 O i M e g u

O 0 1 0 1 H B X -u rt B I n O t i I M r t B

rt t n x o t j > C O C O k l r t 1 0 u u > O u t 4 1

kl v . O l 0 1 a i e C J X e o O O k l

aO l < X 3 k l 4 ) 0 9 0 0 k l C O

U O 0 1 r t o i O k i e g o i C S 0 1 E 3 S K

k> « 3 B 0 ) k l O l > X X 0 1 U n H X 0 9

Cl k l C O B 0 0 I I k l k i E 0 k l t i B e g E

03 C B T J 3 3 C O T J 0 1 r t 3 3 0 1 U C B 1 0

T J i H C B O t i 0 1 X k l 0 1 n « M r t - i n H

« J 3 C l C l o 0 > B M > r t C O T J 0 0 C B M

O l O 0 1 B k l 3 1 8 3 0 1 k . C O I B S k 6 O J X O

b i t n n H X C l B

a c j x ai to (-■ X 6 BOX

U h k l

0 t i

- 4 k l

01 C O

> T J

0) a

- 1 0 1

ert

rt T J 0 9

01 09

CJ n a i n a i o j

sC O 0 ) 3 I I t l C l

3a c i t o r t

B r t B r t

O C l

tk r t C B

> 0 1 U H ti

Sk r t O • 0 1 M

ti r t T J O M S k M C l

a> e g k i 6 k i E 0 1

>0 1 0 1 r t r t k ,

o9 k 0 0 B 6 l i r t r

Cl u h t o o i t o e o b T J 0 1

01 r t k l k l ﬂ 0 1

t i O 0 9 t , M

I B 0 1

Of 0 1 k l r t

0 1 k l S k 4 1 B

> o o e g a r t

H O H

C J U H U H

/fF*> (Continued)

Revision M Deadstart 2-41

Recovery Functions

Table 2-3. Mass Storage Device Recovery (Continued)

.-■h-Shu, a

B to

0 1 t i a -

u c o O b l

s a o i tl r O

SkH

09 rt kl Uh

U h I h O

OJ 5k «

B e g k i k i 0 0

TJ Ut

tl 10

3 I d T J H

C O O S r H UJ

CO rH 3

T t C l O

U h 3 r t r t 1 1

• O 0 1 U u U H r H

H 6 T J T J T t

3 01 X OJ DO

C O k i u B B

rt 9k 00 E k.

£ g

I d 0 1 ki

S X C O

rt X TJ O

t - i I n k l O

rt TJ ti k,

9k O TJ B ki 01

u k i e e g s i a

0 0 H r t 3

r t T J D 0 1

X O 0 1 3

O U K

0 9 B O

c o s a t

E09 Ui I H r t 0) co

0eg r t 0 kl

41 • 0

T J U H CO

8iu U h 1 ) CO to

k l r . rt O J X O X

CO UH ti I H 0 1 01

0 1 H iu 0 1 T J ki

0 0 r t Uh 01 W B kl

CO UH B r t 3 •

X 3 k, O S >

Uh 0 t i 01 C J k .

S"" ﬁT* C l 0 1

Oi to B 0 9 M t i ca

BOZ301 S> B k .

PC jj w eg B 0*

C k l DC 3 O

0 OC

3 9 k t i T l Cl PC

0 e o eg O H to bl

B01 6 U i

H T J TJ rt 41 0) co bl

0 B rH CO 00

&,H

bl rt - CO eg

rJ UH O 01 09 UH I d b .

a c j 09 9k

X O H C J a)

r t r t 0 1

r t u . t ;

r t 0 3 0 0

BOH bE

6 r t 3 SP

N H U H 0 O

3 9 k Cl CJ

o a

r t T J 9k k i e c

HIh in o

O OJ OJ O J o c

u h X O-oc

.***^k

41 kl

Uh TJ 0)

>H 01 CO

tl rH

^H rt

aid

0 9 f -

«J

14 uh CB

TJ ki kl

.X kl

Cl 03 Cl

0os OJ 01

-4 in eg

01 ki

kl rH

BNH co

MUh rH

T O

u u bi

CO 41

(Continued)

2-42 NOS Version 2 Analysis Handbook Revision M

Recovery Functions

0ms

JtrW^K.

Table 2-3. Mass Storage Device Recovery (Continued)

k l T J

O 01 t d 0 0 1 41

0 1 T l N 6 > II

t i U h E •

t i T t J l u t l k i e s H 0 1

O C l 41 TJ S"s B 01 CB 0 0 h

Cl 01 > 0 1 0 1 X 3 E X

ar t - t H T J t i M 01 *rt

4 1 C O r k l U h r H I I I I k l o i O

X U O H k l k l O Z k i t < b U H U H r t b c n

k l c s u B J O O k l m o e g CB Uh rt b i • 01 O

ot , t d

k i q 41 E eg

B S . k l

H o i

0 1 X H c s

k l r t 0 1

CO TJ r. c £S —t

01 8 §■

• O K O C O U h C O > M 01 01 T J M r t k l r t

4J r t T J c i r t a E 0 3 I B O X E X 01 09

ok l C J O k i e g r t 3 H 0 1 C l 3 C l rt 3 0 9

<t 01 b l O O C J H X t j o o 0 e g b t o C l r t

9 k C 4 1 Oj UH H TJ eg ki bl cj 4 1 E r t 6 t O M

U h r t X b CO es t i r t CO SK

r t X k l 41 0 1 H essB 9 k I I U T J b B <S

I h C J B B C e g U U h B t l 01 01 b M B 0 1

41 eg B

> B r t

3 r t r t — O SB Ou k, 0 1 n H r t 0 1 U X O D > b >

O t X X U l ■M TJ 4i X C l X M 01 M U H T J O O

C l ( J r t t i e g t j X M 0 1 C l t i

rt a e g o k l o b t o U h C l

• g es E

<N g B rt I I I I C u B C 4 J t t l B 4 1

O u t j e g o bl co B u si a P u T J > t a t

Ee B Bc

a t j c j

C S T J bl CO TJ t n t j i n t j Id

t i B M B < .

e g a r c c

b B b C O b B ku

ue g e g • c b e g • e g t o • t d t o e g r

0 0 ) C B

O. Sk M « , c s h

a s k m 4 1 c s

a 9 k k i 2.sk2 § O i a

6.5km

•w a a - H a o h _ ) a t s r t a t s h i d a eg rt

U3 eg rt eg a r t e g i d < 0 rt eg e g h t o H eg rt as

ax CO a x c o ax a x x ax

OBJ S3 0 1 0 9 <

0 0 r t o o J o i t o O J 0 9 C d oi cs

Ai 00 rt CO 6 0 r t 0 0 f- 0 C - r t 0 0 6 0 r t 0 0

e g t j E

eg tj c es TJ B C B T J 6 0 3 T J C Z

tu a r t 0 9 r t 9 - C S r t ■<0 ) r t r t C S r t 9-

CO 03 01 CO 06 c o o i c o o c 0 9 I I C S c s b j t n CS OJ CS o s •

cJ l H 0 3

a o o i bl C J r t o o

ﬂ O 4 1 X

b j h m

B o o i

9" o j r t i n o OJ rt CS o t o

aCB i- B O O ) b l B O 0 1 X ot

eo ci I D C J W C O CJ i n o z to Cl Si

co PS0 0 0 6 O X 0 0 6 O 0 0 E O r n 0 0 B O

£B O M B O M O 6 O l i bl E O b U . E O b OC

r t c i a r t c i a a z

3 U 3

r t c i a oc r t c i a i d r t c i a a t d

3oc 3 O td

Ol O M k l

H X t i O k l k l P O

r t X k . Z U .

O M M

H X b

CJ O k i k i Z

r t X b 5 S-

O M 1 4

h j e b §£

uH 0 0 C S r t 0 0 C S b l h 0 0 e g r t 0 0 1 0 h o o e g b i r t

aO r t M O r t M U . U t

U H M c s X O

O H M O r t M © O r t M f P C

9k U H t i C S U H U C O U H b C O H U h b 6 S X <

CO TJ T J b l Z TJ T J X TJ b l 0 .

01 oi eg 01 01 CO 01 Ol eg oi oi eg O i oi eg

X X o i X X 0 1 X X 0 1 X X O J X X b j

( - . M T J H k l T J ( - k l T J h " T J H k l T J

Tt 9k Sk 9k 9s 9k

>ki In

B01

TJ TJ TJ TJ TJ

2. 01 41 Ol 01 01

TJ TJ TJ TJ TJ

H01 41 01

uM M kl 14

XX X X X

bl bl bl bl Id

9k OJ

M 0 1 C S B H

6 6 O J

C J r t 8

t j X e g

0 1 e g o >

o,-. 0 ) T J 4 1 iiE'Srt 0 1 0 0

ti r t c l b B e s e g u « h O X E k l B

kl 81 ts Uh M S k n h m TJ 3 T J X k l r a C c c i 0 1 3 r t

Ul 4 1 B B C l k . X r t 0 1 . E Q t S I S X O 0 3 JC Cl 03

b t C B Q k l B

M B r t r t

k l 0 1 g ] r t 1 ) M X X C O r t B CO 01 09

00 eg

9 k E 1

Cl to o j X k i u kl M X 0 1

ok l 0 1 3 r H 0 1 3 a x m 9 k b 3 b 01 Cl 9s

t i r t c a o i b c o b I I C l O J 0 9 U

BO E I I - 4 T J T J 0 1 E H E 0 1 0 1 O X O l X

8 u l X k l

e g o t o

o B rt > 41 TJ E 01 CO rt r t 01 T J 1 4 3 k i

NH Ol eg rt

B M M

X 4 1 3 6 I I X H C S 0 1 r t 41 O

kl I B T J O r t O n E T J M 0 9 C l B 0 1 B k l 0 0 o

a E u H r . r t O J O J r t 01 CB r t 6 r t

T J O e g . X E E b T J T J - X X r t U TJ

b 0 1 c i - i T J M k l 4 1 0 1 * C l r t r t B I I T J k l E 01 M Cl 41

Cl T J U H r t r t * O M B CJ e g t o 9 s r t T J 0 1 O a 3 TJ

cc E o i O 3 r t o i B 0 1 r t t d a m t j b 6 0 0 9 k b ﬂ b

0 1 r t i m e g t j x a C C o t o

o i r t s a r t 0 1 C X t o O 0 1 k i CJ

k i X e s > t i C O I I Cl oc k i e g I d b M C S

X e s d m o e g I . £5 X t o O 0 1 X X X b £ b M 6

C d M r t r t < u i 3 a I n B U h a H 0 1 C l O X I d S 3 r t 01

rt Cl

U H t i

O M

CB CM r\ CM

H k l

oi a

> T J rt

o t o

^ c j

Tt rt

ukJ

a2 at

3B B

u h r t

b01

01 9 k B

>k . e g •

o01 Ih 1)

tl > U H o

01 o e. rt

BJ! C l r t >

01 S3 1)

O f ﬂ T J

Revision M Deadstart 2-43

Error Idle Recovery

Setting the device error idle status helps to prevent error propagation by inhibiting

new ﬁle assignment on the device and by inhibiting any new PFM, PFLOAD, or

QLOAD activity from occurring on the device. The error idle status is indicated by the

EI error code on the E,M display. Error idle status is set when:

• Mass storage read or write errors occur in the catalog or permit chains.

® Direct or indirect access files are too long or too short.

Correcting the error idle condition requires total initialization of the affected device.

This can be accomplished online for a removable device, but must be performed during

a level 0 deadstart for a nonremovable device. Suggested recovery procedures follow.

Removable Device

1. Dump inactive queued ﬁles, if present on the device (QDUMP).

2. Dump permanent ﬁles (PFDUMP).

3. Perform a total initialization of the device (INITIALIZE,AL,est.). If the error idle

status was caused by a mass storage error, the appropriate track should be

ﬂawed.

4. Reload inactive queued ﬁles, if applicable (QLOAD).

5. Reload permanent ﬁles (PFLOAD).

Nonremovable Device

1. Dump queued ﬁles (QDUMP).

2. Terminate any dayﬁles that must be saved (DFTERM).

3. Dump permanent ﬁles (PFDUMP).

4. Idle the system and perform a level 0 deadstart.

5. Perform a total initialization of the device affected by the error idle condition

(INITIALIZE,AL,est.). If the error idle status was caused by a mass storage error,

the appropriate track should be ﬂawed.

6. Reload queued ﬁles (QLOAD).

7. Reload permanent ﬁles (PFLOAD).

2-44 NOS Version 2 Analysis Handbook Revision M

Deadstart Decks

jp^

CMRDECK .. .... 3-2

Central Memory Descriptions .. 3.3

EQPDECK 3.13

Dayﬁle Descriptions 3_jg

General Description of Equipment Assignment Entries . 3-18

Clear EST Assignment Entry ........ 3-18

Equipment Assignments: Nonmass Storage 3-21

Nonstandard Equipment EST Entry '.[] 3-21

Pseudoequipments 3_22

Unit Record Equipment EST Entry 3-23

Magnetic Tape Equipment EST Entry ..... 3-27

M a s s S t o r a g e E x t e n d e d S u b s y s t e m E S T E n t r y . . . 3 - 3 0

Stimulator Equipment EST Entry ... .... .' 3.31

Network Processing Unit EST Entry 3-32

CDCNET Device Interface EST Entry .'!."".!!.!!!!!...'..'.'!.'.'!!!! 3-34

Network Access Device EST Entry 3.35

Two-Port Multiplexer EST Entry 3-36

MAP III or IV Equipment EST Entry 3-37

6683 Satellite Coupler EST Entry .. 3-38

CYBERPLUS Ring Port EST Entry ...'. 3-39

Equipment -Assignments: Mass Storage 3-40

NOS Mass Storage Concepts 3-41

Alternate System Device 3-41

Alternate Permanent File Family 3-44

Auxiliary Device , 3.45

Family Device 3.46

Link Device 3-46

Master Device .. 3.46

Multispindle Device 3-48

844 Expander .... 3.48

Nonremovable Device 3.54

Removable Device .. 3-54

Shared Device 3.54

System Device 3-54

Temporary File Device .... 3-54

Buffered Disks .,-. 3-55

Extended Memory Overview ........... 3-56

Physical Conﬁguration 3-56

Logical Conﬁguration 3.57

CM and UEM Size Speciﬁcation and Determination 3-59

Simultaneous Use of Two Types of Extended Memory ... 3-59

Specifying Ranges of EST Ordinals 3-6O

-ESQ — Disk Equipment EST Entry 3-60

EQ — Control Module EST Entry 3-66

EQ — Extended Memory EST Entry 3-68

Examples of EQPDECK Entries for Extended Memory 3-72

Single Mainframe Examples 3.72

Multimainframe Examples 3.75

MSAL — Mass Storage Allocation Control Entry 3-78

DOWN — Down Channel Entry 3-80

UP — Up Channel Entry 3-80

PF — Permanent Files Device Assignment Entry 3-81

SYSTEM — System Library Device Assignment Entry .-...■ 3-85

A S R — A l t e r n a t e S y s t e m L i b r a r y D e v i c e A s s i g n m e n t E n t r y 3 - 8 6 ^ v

SGKP — System Checkpoint File Device Assignment Entry 3-86 f

FAMILY — Default Family Name Assignment Entry . 3-87

REMOVE — Removable Device Assignment Entry 3-87

SHARE — Shared Deyice Entry ......... .. .. 3-88

ISHARE — Independent Shared Device Entry 3-89

LBC — Load Buffer Controllers Entry 3-90

XM — Declare Extended Memory Space for Buffers or User Access 3-92

UEMIN — UEM Equipment Initialization 3-93

ACCESS — Set Access Level Limits , 3-94

THRESHOLD — Set Disk Thresholds 3-95

EQPDECK Entries Made Only During Deadstart 3-96

AUTOLOAD — Toggle Autoloading 3-96

GRENADE — Clear Unit Reservations ....... 3-96

INITIALIZE — Initialization Entry ... 3-96

PRESET — Preset the Link Device Entry . 3-101

PRESET — Preset the Independent Shared Device Entry ... 3-102 ^^

RESET — Reset Device Attributes 3-102 ^

APRDECK 3-103

APRDECK Format ,. 3-103

CAF — Clear All Flaw Reservations .. 3-104

SLF — Set Logical Flaws on Any Mass Storage Device , 3-105

CLF — Clear Logical Flaws on .Any Mass Storage Device 3-105

SPF .— Set Physical Extended Memory Track or Disk Area Flaws 3-105

CPF — Clear Physical Extended Memory Track or Disk Area Flaws 3-106

IPRDECK . , 3-107 1

Job Control Information ....... 3-108

Job Scheduling .... ■.....■ 3-108

Control Points ... • • • 3-110

Pseudo-control Points ...... .. 3-111

Number of Control Points and Pseudo-control Points Available ............ 3-111

M e m o r y C o n t r o l . • • • • • • 3 - 1 1 2

Example of Job Control Parameters 3-112

IPRDECK Entries Made Only During Deadstart .3-118

COMLIB 3-118

C P M . ■•■ ■• 3 - 1 1 8

CSM • 3-121

DISK VALIDATION ., 3-121

D S D . . . • • • 3 - 1 2 2

EI ............... .- 3-123

EOTJSNDED STACK PURGING 3-123

HARDWARE FAULT INJECTION 3-123

KEYPM 3-124

MEMORY CLEARING 3-124

MICRO . ....... 3-125

NAMIAF -.. 3-125

PROBE 3-125

SCP • 3-126

SCRSIM 3-126

SECCATS 3-127

SPC - 3-127

SPD 3-128

SPL 3-128

SPW 3-128

^s'

SUBCP 3_12o,

TCVM 3_129

TDEN 3_130

T D T R • • 3 _ 1 3 0

™S- ..3-131

TMSTO. 3-131

TRACE 3-132

IPRDECK Entries '.[ .. .][[......... .....'.. \..' 3.133

Subsystems 3 233

AUTORESTART '.'.'.'.'.'.'"" " * " 3.136

CARTRIDGE PF STAGING ............'.'.......... 3-136

GLASS 3^37

CPTT ,. 3-138

DEBUG 3.138

DELAY . 3_139

DDP ROLLOUT PATH "*."". " 3.140

dfpt. ..;;;;;:::;:; 3-uo

ENGR 3_U1

FLEXIBLE PARTITIONS 3-141

LOCK .'....'..'.'.'.'.'.'.'.'.'. 3-142

LOGGING 3.142

MASTERMSE ' '" '" 3.143

M S V A L I D A T I O N . " * ' 3 . 1 4 3

OQSH 3.144

PCLASS 3.144

PF VALIDATION 3.145

PRIVILEGED ANALYST MODE 3-145

PRIVILEGED RDF 3-146

QUEUE .....' 3.146

REMOVABLE PACKS 3-148

RESIDENT RDF 3.449

SECONDARY USER COMMANDS 3-149

SECURES 3-150

SERVICE 3-151

SPINDOWN 3-156

SRST 3-156

SYSTEM DEBUG 3-157

TAPE PF STAGING 3-158

UNLOCK 3-158

USER EXTENDED MEMORY ,..,.. 3-158

LIBDECK 3-159

0ms

Deadstart Decks

Deadstart decks are text records that reside on the deadstart ﬁle. This section

describes the following deadstart decks.

Deck Description

CMRDECK Central memory resident deck.

EQPDECK Equipment deck.

APRDECK Auxiliary mass storage parameter deck.

IPRDECK System installation parameter deck.

LIBDECK System library deck.

This section contains detailed information about how you can modify the released

versions of the deadstart decks.

jg^S

Revision M Deadstart Decks 3-1

CMRDECK

The central memory resident deck (CMRDECK) resides on the deadstart ﬁle as a text

record that is processed during system initialization. It contains entries deﬁning the

following types of information.

• Central memory.

• Table sizes.

• Configuration information not oriented to equipment.

The deadstart ﬁle can contain up to 1008 CMRDECKs. Having several CMRDECKs on

the same deadstart ﬁle is advantageous because one ﬁle can deadstart several

conﬁgurations. You can obtain a listing of all CMRDECKs by accessing the system ﬁle

SYSTEM with an ASSIGN or COMMON command, then using the T parameter on the

CATALOG command (refer to the NOS Version 2 Reference Set, Volume 3 for more

information concerning these commands). CMRDECKs are named CMRDnn, where nn

is from 00 to 778.

You can modify the released settings of the CMRDECK in two ways: type a new

entry when the CMRDECK is displayed during deadstart, or create a new deadstart

ﬁle. The usual method of modifying a CMRDECK follows.

1. Deadstart, using the released deadstart ﬁle, and select the CMRINST and the

CMRDECK to be displayed (refer to section 2, Deadstart).

CMRINST lists all valid CMRDECK entries. Default values, described in this

section, are assumed if the entries do not appear in the CMRDECK being used. If

either CMRDECK or CMRINST overﬂows two screens, you can page the display.

2. Modify the released version of CMRDECK by entering the changes or additions

from the system console while either the CMRDECK or CMRINST is displayed.

Each console entry supersedes the value currently speciﬁed in the CMRDECK (or

the default value).

NOTE

The modiﬁed CMRDECK remains in effect only until the next deadstart is

performed, except for a level 3 deadstart. That is, changes to the CMRDECK are

not recovered across level 0, 1, and 2 deadstarts unless a new deadstart ﬁle is

created to reﬂect them.

3. To expedite subsequent deadstarts, modify the CMRDECK on the deadstart ﬁle

using SYSGEN (refer to the NOS Version 2 Installation Handbook).

When constructing or modifying a CMRDECK, separate parameters with commas and

terminate each entry with a period. When an error exists in an entry in a CMRDECK

on the deadstart ﬁle, the CMRDECK is displayed and the error is indicated. This

occurs even if you do not select the display CMRDECK option.

/*s%

'^ H,

3-2 NOS Version 2 Analysis Handbook Revision M

Central Memory Descriptions

The general function of central memory description entries is to assign the amount of

central memory to be used for central memory resident (CMR) and the amount to be

used for job processing. The simpliﬁed relationship is: the more central memory that is

assigned to tables in CMR, the less is available for job ﬁeld lengths.

If you will run only a few batch jobs, fewer control points may be required. Thus,

you could decrease the control point area in CMR, which requires 2008 words per

control point.

The following entries are speciﬁed in the SET program with the released default

values listed.

Entry Format

Released

Default Value Signiﬁcance

/ms. BSP=option. Speciﬁes whether NOS should build the SCI

parameter table. If the table is built by NOS,

then one copy of SCI services MDD mode and the

NOS/VE console in a dual-state environment.

NOTE

For compatibility, if NOS 2.6.1 level 700 is run in

a dual-state environment with a NOS/VE system

which is pre-1.3.1 level 700, then you must enter

the BSP entry in order to use MDD.

option

Description

If Y is speciﬁed, the SCI parameter

table is built in the NOS CMR.

If N is speciﬁed, the SCI parameter

table is not built in the NOS CMR

unless the NOS console is a CC598B or

CC634B. In this case, the table is

always built.

Revision M Deadstart Decks 3-3

Central Memory Descriptions

Entry Format

Released

Default Value

C LT = n u m b e r 0

DFT=mode

EJT=number. 6208

EQP = number. Number of

CMRDECK

Signiﬁcance

Speciﬁes the octal number of common library table

(CLT) entries allowed. The number can range from

0 to 1008.

If the table exists, the ﬁrst entry is reserved for

the system; so you should equate CLT with the

number of entries you want, plus 1. For example, if

you want three entries, enter CLT=4.

Common library table entries are described in the

IPRDECK entry COMLIB.

The dedicated fault tolerance (DFT) entry speciﬁes

whether to run DFT in dedicated mode or in

nondedicated mode. In dedicated mode, the DFT PP

is loaded during the NOS deadstart process and

DFT resides in the PP for the life of the system.

Dedicated mode allows the halt-on-error hardware

feature of the model 990 to be used. It also results

in capturing the relevant data about hardware

errors closer to the actual point of failure. In

nondedicated mode, DFT is called to process

mainframe errors when necessary.

mode Description

If D is speciﬁed, DFT runs in dedicated

mode. DFT also runs in dedicated mode if

the VE CMRDECK entry is speciﬁed (other

than VE = *).

If N is speciﬁed, DFT runs in nondedicated

mode (unless the VE CMRDECK entry is

speciﬁed).

Speciﬁes an octal number of entries for the

executing job table (EJT). The system uses the EJT

entries to keep track of executing jobs.

The maximum value for number is 7777s; the

minimum value is 3.

Speciﬁes the number of the EQPDECK to use at

deadstart. The EQPDECK contains equipment

assignments. Up to 1008 EQPDECKs can exist on a

deadstart ﬁle; the number can range from 00 to 77s.

If an EQP entry is not included in the CMRDECK,

the system uses the EQPDECK with the same

number as the CMRDECK being used.

3-4 NOS Version 2 Analysis Handbook Revision M

Central Memory Descriptions

Released

Entry Format Default Value Signiﬁcance

FNT=number. 23s

FOT=number. 108

Sets the octal number of entries allowed in the

system ﬁle name table (FNT). The system FNT

contains the system ﬁle and all fast-attach ﬁles.

Determine the necessary number of FNT entries by

using the formula:

number = SY+RS+(VL+PR)*FM+SFA

Variable Description

SEA

Number of system ﬁles.

Number of resource ﬁles.

Number of VALIDUs ﬁles per family.

Number of PROFILa ﬁles per family.

Number of families that can be active at

any one time.

Site deﬁned fast-attach ﬁles.

For a system installed with the released defaults

and with no additional ﬁles added to the system

FNT by local code, the maximum number of FNT

entries necessary is 2018.

The maximum value for number is 7777s; the

minimum value is 3.

Sets the number of entries in the family ordinal

table (FOT). Each family is allowed one entry in

the table. The ﬁrst entry in the FOT is reserved for

system use. The system uses family name and user

index for job ownership and ﬁle routing. The family

ordinal is a 6-bit number that corresponds to a

particular family name. The FOT maintains the

family name to family ordinal relationship. If a

family is unloaded and later reloaded, it continues

to use the same family ordinal.

The maximum value for number is 100s; the

minimum value is 3. The size of the FOT need not

be the same for each mainframe in a

multimainframe environment.

Revision M Deadstart Decks 3-5

Central Memory Descriptions

Released

Entry Format Default Value Signiﬁcance

INB=number 0

IPD=number. First

IPRDECK on

deadstart ﬁle

LDT = number. 0

LIB = number. 0

MID=id. AA

Speciﬁes a block of memory for site use. The size of

the installation block can range from 0 to 77778

central memory words.

Speciﬁes the number of the IPRDECK to use at

deadstart. The IPRDECK contains installation

parameters. Up to 100s IPRDECKs can exist on a

deadstart ﬁle; the number can range from 00 to 77s.

If an IPD entry is not included in the CMRDECK,

the ﬁrst IPRDECK on the deadstart tape is

processed and is not displayed.

Speciﬁes the number of central memory words

allocated to the LID table. The minimum value is

0; the maximum value is 16610s. The value to be

speciﬁed can be determined by using this formula:

(3+lid)*pid

Parameter Description

lid

pid

Maximum number of LIDs allowed

per PID.

Total number of PIDs in all networks

deﬁned on this mainframe.

One central memory word is always allocated for

the LID table header.

As an example, assume lid = 5 and pid=6. By using

the formula, you come up with (3 + 5)*6 = 48. Adding

one word for the LID table header makes 49, or

618, central memory words that should be allocated

to the LID table.

Speciﬁes the number of the LIBDECK to use at

deadstart; the number can range from 00 to 77s.

LIBDECK is a directive record used by SYSEDIT.

Speciﬁes the 2-character machine identiﬁcation (id)

that is associated with the mainframe. The id

characters must be alphanumeric.

jr«£-fflroSv

3-6 NOS Version 2 Analysis Handbook Revision M

Central Memory Descriptions

Entry Format

Released

Default Value Signiﬁcance

MINCM = size. 14008 Reserves an amount of central memory to be used

by NOS when running in dual state with NOS/VE.

The VEMEM utility can help you determine how

NOS and NOS/VE will share central memory (refer

to the DUAL installation procedure in the Special

Product Installation Information chapter in the NOS

Version 2 Installation Handbook for more

information about VEMEM). (UEM is deﬁned in the

extended memory equipment EST entry). The

remaining central memory is available for use by

NOS/VE. Use the VE CMRDECK entry to reserve

central memory for NOS/VE if UEM is deﬁned. The

minimum and default value for size is 1400s.

Parameter Description

size The size of central memory to

reserve in words/1008. Size is a 1- to

4-digit octal value (1- to 5-digit octal

value for a model 865, 875, or

CYBER 180-class machine) that

restricts the actual central memory

size. The value cannot be 0 and

cannot exceed the total number of

words of memory present in the

machine. Refer to table 3-1.

Size CM Words CM Words

(Octal) (Octal) (Decimal)

1400 140000 49152

2000 200000 65536

3000 300000 98304

4000 400000 131072

6000 600000 196608

10000 1000000 262144

14000 1400000 393216

20000 2000000 524288

24000 2400000 655360

30000 3000000 786432

34000 3400000 917504

40000 4000000 1048576

44000 4400000 1179648

50000 5000000 1310720

54000 5400000 1441792

60000 6000000 1572864

64000 6400000 1703936

70000 7000000 1835008

74000 7400000 1966080

100000 10000000 2097152

Revision M Deadstart Decks 3-7

Central Memory Descriptions

Entry Format

NAME=date

line.

Released

Default Value Signiﬁcance

NETWORK

OPERATING

SYSTEM

NCP = number. 12s

Speciﬁes the system date line, which is displayed on

the system console display and on the terminal

when an interactive user logs in to the system.

Parameter Description

date line Alphanumeric string; must be fewer

than 39 characters. If the date line is

not terminated with a period, one

will be appended. If the date line is

38 characters, the last character

must be a period.

Sets the number of control points available for job

processing. Refer to Job Control Information later in

this section for a discussion of the proper number of

control points to select.

Parameter Description

number Number of control points available in

central memory; number can be from

2 to 348.

3-8 NOS Version 2 Analysis Handbook Revision M

Central Memory Descriptions

Released

Entry Format Default Value Signiﬁcance

PCP=number. 0 Sets the number of pseudo-control points available

for job processing. Refer to Job Control Information

later in this section for a discussion of the proper

number of pseudo-control points to select.

Parameter Description

number Number of pseudo-control points

available in central memory; number

can be from 2 to 34s.

OPSECM = n. 0 Speciﬁes the operating system security mode. The

site security administrator supplies the value for

this entry. Refer to the NOS Version 2 Security

Administrator's Handbook for information on this

entry.

n Description

0 or Sets the system to unsecured mode,

omitted

1 Enables multilevel security. The values oft

the system access level limits can be set |

by the SECURES IPRDECK entry or by |

the console command. The SECURES

console command can be used to raise or |

lower system access level limits.

2 Enables multilevel security. The values oft

the system access level limits are set by I

the SECURES IPRDECK entry. The

SECURES console command can be used |

to raise but not to lower system access

level limits.

3 Enables multilevel security. The values of|

the system access level limits are set by |

the SECURES IPRDECK entry only; the j

SECURES console command is not

allowed.

Revision M Deadstart Decks 3-9

Central Memory Descriptions

Entry Format

Released

Default Value Signiﬁcance

PPU=ppi,

PP2,...,PPn.

All available

PPs are active

QFT = number. 620s

Toggles the active status of the physically available

PPs except for PPs 1, 2, and 10, which are always

active. Active means the PP is available for system

use; inactive means it is not available for system

use. The PPU entry is not in the released

CMRDECK; therefore, all available PPs are active.

PPU is a toggle entry: each entry changes the

active status of the PPs.

/«^!\

Parameter Description

ppi Number of the PP; from 0 to lis and

208 to 318. Specifying an asterisk (*)

on the entry instead of ppi toggles

the status of PPs 20s to 31s.

This entry may be useful if PP memory is failing

or if a channel is causing problems on its associated

PP. For example, the following entry deactivates PP

3 and PP 4 (assuming no other entries have been

made for PP 3 and PP 4).

PPU=3,4.

Sets the number of entries allowed for the queued

ﬁle table (QFT). The system uses the QFT to

manage all ﬁles in the input and output queues,

number is a 1- to 4-digit number from 3 to 77778.

^ftsj.

1. APP that has been turned off by CTI is physically unavailable and cannot be turned on by the PPU

entry.

3-10 NOS Version 2 Analysis Handbook Revision M

Central Memory Descriptions

rEntry Format

Released

Default Value

VE = integer.

Signiﬁcance

Reserves an amount of central memory to be used

by NOS/VE when running in dual state with NOS.

The VEMEM utility can help you determine how

NOS/VE and NOS will share central memory (refer

to the DUAL installation procedure in the Special

Product Installation Information chapter in the

NOS 2 Installation Handbook for more information

about VEMEM).

If you want to run NOS/VE in dual-state mode and

you do not have UEM (speciﬁed by either the

extended memory equipment EST entry or XM

EQPDECK entry), enter VE = 0 and use the MINCM

entry to reserve an amount of central memory for

NOS.

If you want to run NOS/VE in dual-state mode and

you have UEM, you must enter VE = integer to

reserve an amount of central memory for NOS/VE.

The memory is reserved even when NOS/VE is not

running.

Use VE = * if you do not want to run NOS/VE in

dual-state mode.

On a dual-CPU 830 machine, NOS automatically

disables processor 1 at deadstart time if there is no

VE = * entry. This is done because NOS/VE does not

support dual state in both processors. For all other

CYBER 180 machines, NOS uses only processor 0.

Parameter Description

integer Octal number divided by 1000s of

central memory words to reserve for

NOS/VE when you have UEM. If you

do not have UEM and want to run

NOS/VE in dual-state mode, use the

MINCM entry and VE = 0 entry.

Refer to table 3-1 for a list of

memory sizes. NOS/VE requires at

least 6.5 megabytes of central

memory.

Revision M Deadstart Decks 3-11

Central Memory Descriptions

Entry Format

Released

Default Value Signiﬁcance

VERSION=name. NOS 2.5.3

688/688

Speciﬁes the system version that is displayed on

the system console display.

Parameter Description

name Alphanumeric string; must be

fewer than 18 characters. If name

is not terminated with a period,

one will be appended. If name is

17 characters, the last character

must be a period.

The operating system version

number should be in character

positions 5 through 7 of the name

parameter, since the value in

character positions 5 through 7 is

available to users through the

symbolic name VER.

Table 3-1. Table of Central Memory Sizes

CM Words CM Bytes

CM Words CM Words (Hexa CM Bytes (Hexa

CM Size1 (Octal) (Decimal) decimal) (Decimal) decimal)

lMbyte/128K 400000 131072 20000 1048576 100000

1.5Mbyte/192K 600000 196608 30000 1572864 180000

2Mbyte/256K 1000000 262144 40000 2097152 200000

3Mbyte 1400000 393216 60000 3145728 300000

4Mbyte 2000000 524288 80000 4194304 400000

5Mbyte 2400000 655360 OAOO00 5242880 500000

6Mbyte 3000000 786432 ocoooo 6291456 600000

7Mbyte 3400000 917504 OE0O00 7340032 700000

8Mbyte 4000000 1048576 100000 8388608 800000

9Mbyte 4400000 1179648 120000 9437184 900000

10Mbyte 5000000 1310720 140000 10485760 0A00000

11Mbyte 5400000 1441792 160000 11534336 0B00000

12Mbyte 6000000 1572864 180000 12582912 0C00000

13Mbyte 6400000 1703936 1A0000 13631488 0D00O00

14Mbyte 7000000 1835008 1C0000 14680064 0E00000

15Mbyte 7400000 1966080 1E0000 15728640 0F00000

16Mbyte 10000000 2097152 200000 16777216 1000000

■'*:^%\

A*^*S

1. K is 1,024 60-bit words. Mbyte represents megabytes. One megabyte is 1,048,576

8-bit bytes, which is equivalent to 100,000 hexadecimal 8-bit bytes or 131,072 60-bit

words.

3-12 NOS Version 2 Analysis Handbook Revision M

EQPDECK

i^S

EQPDECK

The equipment deck (EQPDECK) resides on the deadstart ﬁle as a text record that is

processed during system initialization. It contains entries deﬁning the following types of

information.

• Dayfile buffer size.

• Hardware configuration.

• File residence on mass storage devices.

• Permanent file family residence.

• Multimainframe configurations.

• Software attributes associated with hardware (such as forms code).

An EQPDECK entry is described under the type of information it deﬁnes.

The deadstart ﬁle can contain up to 100s EQPDECKs. Having several EQPDECKs on

the same deadstart ﬁle is advantageous because one ﬁle can deadstart several

conﬁgurations. You can obtain a listing of all EQPDECKs by accessing the system

ﬁle SYSTEM with a COMMON command, then using the T parameter on the

CATALOG command (refer to the NOS Version 2 Reference Set, Volume 3 for more

information concerning these commands). EQPDECKs are named EQPDnn, where nn

is from 00 to 77s.

rYou can modify the released settings of the EQPDECK in two ways: type a newentry when the EQPDECK is displayed during deadstart, or create a new deadstart

ﬁle. The usual method of modifying an EQPDECK follows.

1. Deadstart, using the released deadstart ﬁle, and select the EQPINST and the

EQPDECK to be displayed (refer to section 2, Deadstart).

EQPINST lists all valid EQPDECK entries. Default values, described in this

section, are assumed if the entries do not appear in the EQPDECK being used. If

either EQPDECK or EQPINST overﬂows two screens, you can page the display.

^^ 2. Modify the released version of EQPDECK by entering the changes or additions

from the system console while either the EQPDECK or EQPINST is displayed.

Each console entry supersedes the value currently speciﬁed in the EQPDECK (or

the default value).

NOTE

The modiﬁed EQPDECK remains in effect only until the next level 0, 1, or 2

deadstart is performed. That is, changes to the EQPDECK are retained across a

level 3 deadstart, but not across a level 0, 1, or 2 deadstart (unless a new

deadstart ﬁle is created to reﬂect them).

3. To expedite subsequent deadstarts, modify the EQPDECK on the deadstart ﬁle

using SYSGEN (refer to the NOS Version 2 Installation Handbook).

Several equipment status displays are available during deadstart (refer to section 2,

Deadstart).

jp*.

Revision M Deadstart Decks 3-13

EQPDECK

When constructing or modifying an EQPDECK, the following restrictions apply:

device (such as assigning it for permanent ﬁle, system, or temporary ﬁle use). If

you redeﬁne the EQ entry (by clearing the entry and retyping a new entry or by

changing the device type in the entry), reenter all other assignments for that

equipment. If you need to modify an EQ entry, you need only enter the changes

rather than reentering the entire entry. You do not need to reenter other

assignments for that equipment. This is especially useful for changing channels or

units and for reconﬁguring around hardware malfunctions.

Example:

The following mass storage assignments are in the EQPDECK.

EQ7=DI,ST=ON,UN=60,CH=1.

PF=7,F,125,125,SYSTEM,3,200.

If you want to specify unit 70 rather than 60, you need only enter:

EQ7=DI,UN=70.

You.do not need to modify the PF entry.

• EST ordinals range from 1 to 777s. This range depends on the value of ESMX

(refer to PPCOM parameters in the NOS Version 2 Installation Handbook).

Ordinals 1 through 4 are reserved for pseudoequipment EST entries that are

automatically deﬁned by the system and cannot be changed. Ordinal 5 is reserved

for extended memory if it is used as a link device in a multimainframe

environment (refer to EQ - Extended Memory EST Entry later in this section).

Otherwise, EST ordinal 5 can be used for any equipment type.

• The device from which you are deadstarting must be defined.

• Commas must separate parameters.

• A period must terminate each entry.

• Except where explicitly specified that controllers and/or equipment can be shared

between mainframes, assume they cannot be shared. For example, NOS does not

support sharing a two-channel tape controller between mainframes; nor does it

support sharing mass storage controllers, except as speciﬁed in section 13,

Multimainframe Operations.

When an error exists in an entry in an EQPDECK on the deadstart ﬁle, the

EQPDECK is displayed and the error is indicated. This occurs even if you do not select

the display EQPDECK option or if you enter GO after displaying the EQPDECK.

.■^%j.

3-14 NOS Version 2 Analysis Handbook Revision M

J^S.

J^S

Dayﬁle Descriptions

One of the functions of dayﬁle description entries is to assign the amount of central

memory to be used for buffers. The more central memory assigned to dayﬁle buffers in

CMR, the less central memory available for job ﬁeld lengths.

If, for example, you need a large portion of central memory to run a job, it might be

advisable to decrease the size of the dayﬁle buffers area in CMR to accommodate that

job. However, when dayﬁle buffers are smaller, the information stored in them is

written to mass storage more often, which requires more system overhead.

The following entries are speciﬁed in the SET program with the released default

values listed.

Entry Format

Released

Default Value Signiﬁcance

ACCOUNT=est,length. 400s Sets the residence of the account dayﬁle

and the length of the account dayﬁle

buffer.

The account dayﬁle is an accounting

record containing information such as

type and quantity of resources used, and

job and execution times. This account

information is written to the central

memory account dayﬁle buffer during job

processing; the central memory buffer is

written to mass storage when it is full.

The account dayﬁle buffer resides in

CMR in the dayﬁle buffer area.

Parameter Description

est

length

The 1- to 3-digit octal

equipment status table

(EST) ordinal of the

equipment on which the

account dayﬁle is to

reside. A null value means

the system uses the ﬁrst

system device it can ﬁnd.

If the existing account

dayﬁle is recovered, the

account dayﬁle buffer

resides on that equipment

and the est parameter is

ignored.

The 3- or 4-digit octal

length of the account

dayﬁle buffer in CMR;

must be a multiple of 100s

and less than or equal to

10008. If 0 (zero) is

speciﬁed, messages issued

to the account dayﬁle are

discarded.

Revision M Deadstart Decks 3-15

Dayﬁle Descriptions

Entry Format

Released

Default Value

DAYFILE=est,length. 400s

ERRLOG=est,length. 1008

Signiﬁcance

Sets the residence of the system dayﬁle

and the length of the system dayﬁle

buffer.

The dayﬁle buffer contains the dayﬁle

information, which is maintained in the

same way as the account dayﬁle buffer. It

resides in CMR in the dayﬁle buffer area.

Parameter Description

est The 1- to 3-digit octal EST

ordinal of equipment on

which the dayﬁle is to

reside. A null value means

the system uses the ﬁrst

system device it can ﬁnd.

The residence of this dayﬁle

is normally determined by

the recovery of the existing

dayﬁle. Use this parameter

if no system dayﬁles are

recovered.

length The 3- or 4-digit octal

length of the system dayﬁle

buffer in CMR; must be a

multiple of 100s and less

than or equal to 10008. If 0

is speciﬁed, messages issued

to the system dayﬁle are

discarded.

Sets the residence of the error log and the

length of the error log buffer.

The error log is a record of hardware

error and status messages. This

information is maintained in the same way

as the account dayﬁle buffer.

Parameter Description

est The 1- to 3-digit octal EST

ordinal of equipment on

which the error log is to

reside. A null value means

the system uses the ﬁrst

system device it can ﬁnd

the existing error log is

recovered, the error log

buffer resides on that

equipment and the est

parameter is ignored.

3-16 NOS Version 2 Analysis Handbook Revision M

Dayﬁle Descriptions

0$ms Entry Format

Released

Default Value Signiﬁcance

ERRLOG=est,length.

(Continued) Parameter Description

MAINLOG=est,length. 1008

yfP^N.

length The 3- or 4-digit octal

length of the error log

buffer in CMR; must be a

multiple of 100s and less

than or equal to 10008. If 0

is speciﬁed, messages issued

to the error log are

discarded.

Sets the residence of the binary

maintenance log and the length of the

binary maintenance log buffer. The binary

maintenance log is a record of hardware

diagnostic information.

Parameter

est

length

Description

The 1- to 3-digit octal EST

ordinal of equipment on

which the binary

maintenance log is to reside.

A null value means the

system uses the ﬁrst system

device it can ﬁnd. If the

existing binary maintenance

log is recovered, the binary

maintenance log buffer

resides on that equipment

and the est parameter is

ignored.

The 3- or 4-digit octal

length of the binary

maintenance log buffer in

CMR; must be a multiple of

1008 and less than or equal

to 10008. If 0 (zero) is

speciﬁed, messages issued to

the binary maintenance log

are discarded.

j0m\

Revision M Deadstart Decks 3-17

S ^ ^ S

General Description of Equipment Assignment Entries

The general format of the EQ entry is:

EQest =type,keywordi=valuei,... ,keywordn=valuen.

Keyword=value entries are order independent.

Depending on the type of equipment being deﬁned, certain keywords are required,

while others are optional, as shown in table 3-2.

You can redeﬁne an EST entry by entering an EQ entry for the EST ordinal with a

different equipment type. This action clears the original entry before creating the new

entry. You can also modify an EST entry by entering an EQ entry with the same

equipment type and only the keyword(s) you want changed.

Clear EST Assignment Entry

Use the following entry to clear an assignment that currently exists for an EST -*a%.

ordinal. Clearing the assignment does not clear ﬂaw entries for that equipment. The

EQest=0 entry is not required when you are assigning the EST entry to a different

type of equipment.

EQest=0.

EQest=.

Parameter Description

est EST ordinal of the equipment; est can be from 5 to 777s.

3-18 NOS Version 2 Analysis Handbook Revision M

Clear EST Assignment Entry

rTable 3-2. Required and Optional Keywords

Equipment

Type

Required

Keywords

CC EQ,CH

CM EQ,CH

CP EQ,CH

CR EQ,CH

DB UN,CH

DC UN,CH

0m*s,

DF UN,CH

DG UN

DH UN,CH

DI UN,CH

DJ UN,CH

DK UN,CH

DL UN,CH

DM UN,CH

DN EQ,UN,CH

00&S DP SZ,CH

DQ UN,CH

Optional

Keywords

ST,IB

CW,IB

ST,ID,FC,IB

ST,ID,IB

ST,AP,IB

ST,MA,ET,

ST,AP,IB,AP,MC,IB

ST,AP,IB

ST,MA,ET,AP,MC,IB

ST,AP,IB

(Continued)

Revision M Deadstart Decks 3-19

Clear EST Assignment Entry

Table 3-2. Required and Optional Keywords (Continued)

Equipment

Type

Required

Keywords

Optional

Keywords

DV UN,CH ST,AP,IB

DW UN,CH ST,AP,IB

LR EQ.CH ST,ID,FC,TN,PS,IB

LS EQ,CH ST,ID,FC,TN,PS,IB

LT EQ,CH ST,ID,FC,TN,PS,IB

LX EQ,CH ST,ID,FC,TN,PS,IB

MP CH ST,IB

MT EQ,UN,CH,TF ST,IB

NC CH ST.IB

ND EQ,CH,PI,ND,NT ST,IB

NP EQ,CH,PI,ND,SA ST,IB

NT EQ,UN,CH,TF ST,IB

RM CH,PT ST,IB

RP CH ST,IB

SS EQ,UN,CH ST,IB

TT EQ.CH ST,PT,IB

** WO WI

>^^.

3-20 NOS Version 2 Analysis Handbook Revision M

y^s

yfPN

yams

Equipment Assignments: Nonmass Storage

The following EST entries are described in this subsection.

Nonstandard equipment

Pseudoequipments

Unit record equipment

Magnetic tape equipment

Mass Storage Extended Subsystem equipment

Stimulator equipment

Network processing unit

CDCNET device interface

Network access device

Two-port multiplexer

MAP III or IV equipment

6683 Satellite Coupler

CYBERPLUS ring port equipment

Nonstandard Equipment EST Entry

The nonstandard equipment EST entry allows you to deﬁne nonstandard equipment or

to add site debugging modiﬁcations.

Use the following format to enter the actual octal value that is to reside at that EST

ordinal.

EQest=**,Wo=valueo,Wi=valuei

Parameter Description

est

valuei

EST ordinal of the equipment; est can be from 5 to 777s.

The 1- to 20-digit octal value; this value is entered in the EST word i

for the speciﬁed ordinal. The word is right-justiﬁed and zero-ﬁlled if

valuei has fewer than 20 digits.

Revision M Deadstart Decks 3-21

Pseudoequipments

The system automatically deﬁnes EST ordinals 0 through 4 as pseudoequipment EST

entries; they cannot be used for other equipment deﬁnitions. You cannot declare,

change, or remove pseudoequipments.

EST Ordinal

Ordinal 0 (RD)

Ordinal 1 (DS)

Ordinal 2 (NE)

Ordinal 3 (TE)

Ordinal 4 (TT)

Description

This ordinal reserves an EST entry and MST entry to be used for

device reconﬁguration.

This ordinal represents the system display console.

The system uses 2 internally to signify that a ﬁle is assigned, but

that no space exists on the device. If a read is tried,

end-of-information (EOI) status occurs. If a write is attempted, the

data is discarded.

For example, you can use ordinal 2 with the permanent ﬁle utility

(PFDUMP) to validate the integrity of a permanent ﬁle device,

without taking the time to actually create a dump ﬁle on tape. In

this case, enter:

X.DIS.

ASSIGN,NE,TAPE.

PFDUMP.

This causes all dump data to be discarded, even though the

permanent ﬁle device is read and informative messages about the

permanent ﬁle device are issued to the system console. These

messages are described in the NOS Version 2 Operations Handbook.

If an association is established between ﬁle name and volume serial

number with an ASSIGN, LABEL, REQUEST, or VSN command,

the system automatically enters EQ3 in the ﬁle's FNT/FST entry.

When a tape having the desired volume serial number is assigned

to the ﬁle, the system replaces EQ3 in the ﬁle's FNT/FST entry

with the EST ordinal of the tape unit on which the tape is

mounted. If a ﬁle that has had the ﬁle name and volume serial

number association established by a VSN command is returned

prior to attempting to assign the tape equipment to the ﬁle, the

FNT/FST entry is canceled.

The system assigns to this equipment a ﬁle used for either input

from or output to an interactive terminal. This allows the system

to determine whether a ﬁle requires the special handling needed to

accomplish terminal input/output. Byte 4, set by the NAMIAF

IPRDECK entry, contains the number of network terminals.

^ms

3-22 NOS Version 2 Analysis Handbook Revision M

Unit Record Equipment EST Entry

A unit record equipment EST entry deﬁnes channel connected card readers, card

punches, and line printers.

Unit record equipment connected to RHF or PSU through communications ports are

deﬁned in the network conﬁguration ﬁles rather than with individual EST entries.

EQest=type,ST=status,TN=tn,EQ=eq,CH=ch,lD=id,FC=fc,PS=ps,lB=ib.

EQest=type-P,ST=status,TN=tn,EQ=eqfCH=ch,ID=id,FC=fc,PS=ps,IB=1b.

Parameter Description

est EST ordinal of the unit record equipment; from 5 to 7778.

type Unit record equipment type; NOS supports the following unit record

equipments.

type Equipment

Card reader

CR 405-3447/3649

Card punch

CP 415-3446/3644

CP 415-30

Line printer

LR 580-12

LS 580-16

LT 580-20

LX 5870

P Speciﬁes that a 580 printer is equipped with a programmable format

controller.

Revision M Deadstart Decks 3-23

Unit Record Equipment EST Entry

Parameter Description

TN=tn Print train for local batch line printer; from 1 to 7. NOS supports the

following print trains.

tn Print Train Description

0 595-1/596-1 CDC graphic 63/64-character set.

1 595-1/596-1 CDC graphic 63/64-character set.

4 595-6/596-6 ASCII graphic 95-character set or ASCII graphic

63/64-character set.

5 595-5/596-5 ASCII graphic 63/64-character set.

6 595-6/596-6 ASCII graphic 95-character set.

7 595-6/596-6 ASCII graphic 95-character set or ASCII graphic

63/64-character set.

If you set a nonsupported print train value, tn defaults to a supported

value. If you omit tn or specify 2 or 3, the actual value of tn is 1.

If an invalid external characteristic (EC)2 is speciﬁed, the queued ﬁle

processor cannot output the ﬁle. The following shows which ﬁles will

print and which ﬁles will not print for a given print train selection.

Print Train

Selected

Will Print File

With Speciﬁed EC

Will Not Print File

With Speciﬁed EC

0None, B4, B6 A4, A6, A9

1None, B4, B6 A4, A6, A9

4None, A4, A6, A9 B4, B6

5None, A4, A6 B4, B6, A9

6A9 None, B4, B6, A4, A6

7None, B4, B6, A4,

A9 A6,

ST=status Speciﬁes

enter om

whether unit record equipment

s of these values:

is available for system use;

status Description

ON Unit record equipment is available.

OFF Unit record equipment is ignored during system operation.

2. Refer to the NOS Version 2 Reference Set, Volume 3 for a discussion of-the ROUTE command EC

parameter.

3-24 NOS Version 2 Analysis Handbook Revision M

0^*S.

Unit Record Equipment EST Entry

Parameter Description

EQ=eq Controller number for equipment; from 0 to 7. To determine the number

for most unit record equipment, look at the dial switch on the controller.

To determine the equipment number for an LX printer, use the following

equation:

eq * 2 + 1 = addr

where addr is the 5870 host address obtained from the 5870 installer.

CH = ch Number of channel to which unit record equipment is connected; from 0

to 138 and from 20s to 338.

NOTE

/$P^V

If you want to use a card reader to perform a coldstart on a 66x tape

controller, the card reader must be available on channel 12s, 13s,

328, or 338.

To ensure that all printers are restored to their original states (such as

8 lines per inch and auto page eject) after a master clear has been

issued, all unit record equipment should be available on dedicated

channels. If it is not, printers revert to 6 lines per inch, and no auto

page eject status after a master clear is issued.

ID=id The 1- or 2-digit octal numeric identiﬁer assigned to the device; from 0

to 678. This id is assigned to any output created by a job. For card

readers, all jobs loaded from this card reader are assigned the

identiﬁer id.

FC = fc The 2-character optional forms code assigned to a line printer or card

punch. If the forms code is not present, the forms code ﬁeld is cleared.

The forms code must either be null (not speciﬁed) or in the range from

AA to 99.

NOTE

The forms code cannot be assigned to a card reader.

Revision M Deadstart Decks 3-25

Unit Record Equipment EST Entry

Parameter Description

PS=ps

IB=ib

Examples:

Paper size, L or S, for local batch line printer. Default is L.

ps Description

Long (11 inch) paper will be mounted.

Short (8 1/2 inch) paper will be mounted.

The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

EQ11=CR,ST=ON,EQ=4,CH=12.

EQ11=CP,EQ=5,CH=12,ST=0N.

EQ20=LR,ST=ON,CH=12,EQ=6,ID=15,FC=AA.

EQ21=CR,ID=15,ST=ON,EQ=7,CH=12.

EQ22=LT-P,TN=6,ST=0N,EQ=2,CH=12.

/A^^^L

3-26 NOS Version 2 Analysis Handbook Revision M

Aims.

0ms

Magnetic Tape Equipment EST Entry

The released tape subsystem supports a maximum of 16 magnetic tape units. The

minimum number of magnetic tape units that NOS requires is one 639 667 669 677

679, or 698. The format of the entry is: '

EQest =MT-n,ST=statusfEQ=eq,UN=un,CH=ch1/ch2,TF=tf,IB=ib.-

EQest =NT-n,ST=status,EQ=eq,UN=un,CH=ch1/ch2,TF=tfIlB=ib.

Parameter Description

est EST ordinal of the tape unit; from 5 to 777s. Refer to the MT-n or

NT-n parameter.

MT-n Equipment type; MT speciﬁes 7-track tape units, and NT speciﬁes

or 9-track tape units, n is the total number of magnetic tape units

NT-n connected to the controller, from 1 to 20s for 677, 679, and 698 units

with a 7021-31/32 controller and from 1 to 10s for 667 and 669 units

with a 7021-21/22 controller. The system automatically generates n

EST entries with consecutive EST ordinals beginning with the ordinal

speciﬁed in the est parameter. The n units begin with the unit number

speciﬁed in the UN=un parameter.

For 639 units, n should not be speciﬁed; if it is speciﬁed, it should be

given a value of 0 (zero).

NOTE

To clear an MT-n or NT-n assignment, enter an EQest=0 entry for all

n units. For example, to clear EQ50 = MT-4,ST=ON,..., enter:

EQ50=0.

EQ51=0.

EQ52=0.

EQ53=0.

Revision M Deadstart Decks 3-27

Magnetic Tape Equipment EST Entry

Parameter Description

ST=status Indicates whether the tape unit is available for access; enter one of

these values:

status Description

ON Magnetic tape unit is available for access.

OFF Magnetic tape unit is ignored during system operation.

DOWN All access to the magnetic tape unit is inhibited for the

operating system and user jobs. Specify DOWN if the

equipment is malfunctioning and access is not desirable or the

unit is to be used by NOS/VE in a dual-state environment

(669, 679, and 698 units only).3

EQ=eq Controller number for the tape unit; from 0 to 7. This number is usually

0 unless the controller is switched to a different number. Ask your

customer engineer for the correct number if you are unsure.

UN=un Number of the lowest numbered magnetic tape unit to be processed;

units must have consecutive physical unit numbers; from 0 to 7 for 667

and 669 units or from 0 to 17s for 677, 679, and 698 units. The unit

number is on the front of the tape unit, either as a switch or a label.

CH = chi Number of the channel(s) to which the tape unit is connected; from 0 to

138 and from 20s to 33s.

A controller can be connected to one or two channels, depending on the

controller model. However, a maximum of two channels can be handled

regardless of the number of controllers.

3. If a magnetic tape unit is accessible from more than one mainframe, concurrent use must be disabled by

one of these methods: set access switches on the tape controller so that only one mainframe can access the

unit at any time or make sure that the EST entry for each shared unit is deﬁned as ON in only one of the

mainframes.

3-28 NOS Version 2 Analysis Handbook Revision M

0^&t

Magnetic Tape Equipment EST Entry

Parameter Description

TF=tf Hardware features available. This parameter speciﬁes the following

hardware characteristics of the tape unit(s) being deﬁned; select one

option.

The following options apply to 677/679/698 units:

tf Description

ATS The unit(s) being deﬁned cannot process 6250-cpi group-

encoded (GE) tapes. If the system detects a unit with this

capability, it automatically changes this value to ATSGE

or CMTGE, indicating the availability of the GE feature.

ATSGE The unit(s) being deﬁned can process 6250-cpi GE tapes,

or Specify this value (ATSGE for 679; CMTGE for 698) only

CMTGE if a unit being deﬁned has the capability of processing GE

tapes but is down. In such a case, the system would not

be able to connect to the unit to determine the

availability of the GE feature.

The following options apply to 667/669 units:

tf Description

MTS The FCOs needed to implement the block identiﬁcation feature

have been installed in the 7021 controller for the unit(s) being

deﬁned. If the controller is a two-channel model, the block

identiﬁcation feature must have been implemented on both

channels.

The following options apply to 639 units:

tf Description

IST Specify TF=IST.

IB=ib The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

Revision M Deadstart Decks 3-29

Mass Storage Extended Subsystem EST Entry

The Mass Storage Extended Subsystem (MSE) hardware consists of a 7991 storage

module and a 7990 controller.

A mainframe that runs SSSLV in slave mode must have an MSE EST entry, even

though no MSE hardware is actually conﬁgured for this mainframe. A mainframe

that runs SSEXEC in master mode must have an LBC EQPDECK entry with a

controlware type of N8 or M8 speciﬁed.

The MSE EST entry format is:

EQest=SS,ST=status,EQ=eq,UN=un,CH=ch,IB=1b.

Parameter Description

est

ST=status

EQ = eq

UN=un

CH = ch

IB=ib

EST ordinal of the 7990 controller; from 10s to 777s.

Indicates a 7990 controller.

Speciﬁes whether the 7990 controller is available for use; enter one of

these values:

status Description

ON 7990 controller is available.

OFF 7990 is ignored during system operation.

Logical controller number.

This number sets the controller location in the unit device table within

SSEXEC's memory. Enter one of these values:

eq Description

First 7990 controller.

Second 7990 controller.

Third 7990 controller.

Fourth 7990 controller.

7991 storage module unit number; must be 0 (zero).

Number of the channel to which the 7990 controller is connected; from

0 to 138 and from 20s to 33s. You can specify up to two channels.

The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

<^^Y

3-30 NOS Version 2 Analysis Handbook Revision M

0ms.

Stimulator Equipment EST Entry

NOS IAF supports a stimulator called STIMULA. However, there are other stimulators

such as PACER that can use this EST entry. (PACER is a product that can be ordered

from the Special Products Library.)

The format of the EST entry used by the interactive subsystem during stimulation is:

EQest=TT,ST=status>EQ=eq,CH=ch,PT=lines,IB=ib.

Parameter Description

est

ST=status

EQ = eq

CH=ch

PT=lines

IB = ib

EST ordinal of the interactive stimulator; from 5 to 7778.

Indicates an interactive stimulator.

Speciﬁes whether the interactive stimulator is available for use; enter

one of these values:

status Description

ON Stimulator is available.

OFF Stimulator is ignored during system operation.

Number of the controller; from 0 to 7. Refer to the description of the

channel parameter.

Number of the channel; from 0 to 13s and from 20s to 33s. The

channel/controller combination must not have any equipment attached

to it.

Number of lines to stimulate; from 1 to 10008. If this parameter is

omitted, 1008 is used.

The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

0S^s.

Revision M Deadstart Decks 3-31

Network Processing Unit EST Entry

A total of 16 NPU, MDI, and MTI ESTs can be available (ST=ON) at the same time.

The format of the EST entry is:

EQest=NP,ST=status,EQ=eq,PI=plp,CH=ch,ND=node,SA=sa,IB=ib.

Parameter Description ^^^^

est

ST = status

EQ=eq

PI=pip

CH=ch

ND=node

EST ordinal of the NPU; from 5 to 7778.

Indicates a 255x NPU.

Speciﬁes whether the NPU is available for use; enter one of these

values:

status Description

ON NPU is available.

OFF NPU is ignored during system operation.

Number of the controller for the NPU; from 0 to 7.

Peripheral interface program index, which determines which copy of

the PP driver drives this NPU; from 1 to 4. Up to four EST entries

can have the same PIP index (that is, one PP can drive a total of four

front ends consisting of NPUs, MDIs, or MTIs).

Number of the channel to which the NPU is connected; from 0 to 13s

and from 20s to 33s.

Node number of the coupler associated with the NPU being deﬁned;

from 1 to 3778. This value is the same as the NODE parameter on the

COUPLER statement in the network conﬁguration ﬁle deﬁnition. For

the procedure to assign this value, refer to the Network Deﬁnition

Language Reference Manual.

NOTE

The node parameter is speciﬁed as an octal value on the NPU entry.

The NODE parameter is speciﬁed on NDL statements as a decimal

value.

3-32 NOS Version 2 Analysis Handbook Revision M

Network Processing Unit EST Entry

JpN.

/fP^V

Parameter Description

SA = sa

IB=ib

Example:

System autostart module (SAM) ﬂag (refer to CCP in the NAM Host

Application Programming Reference Manual). This parameter is required.

sa Description

ON SAM is present on the NPU.

OFF SAM is absent. The host attempts to load the system autostart

module program (SAM-P) to the NPU if the NPU is down.

If an NPU is channel-connected to a single host in a single-host

network, the NPU does not need a SAM. However, if an NPU is

channel-connected to two hosts, the NPU should have a SAM; otherwise,

both hosts might attempt to reload SAM-P if the NPU is down. This

requires operator intervention to resolve the problem. The NPU EST

entries for both hosts should specify SA=ON.

The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

Assume that three NPUs exist on channels 4, 5, and 6, all with controller 7. The

NPUs are connected to coupler nodes 2, 8, and 11, respectively. The NPU on channel

5 has a system autostart module; the others do not. The ﬁrst two NPUs are to be

driven by the same PP. The EST entries for these NPUs follow.

EQ70=NP,ST=ON,CH=4,EQ=7,PI=1,ND=2,SA=OFF.

EQ71=NP,ST=0N,CH=5,EQ=7,PI=1,ND=10,SA=ON.

EQ72=NP,ST=0N,CH=6,EQ=7,PI=2,ND=13,SA=0FF.

The NDLP input for the network conﬁguration would include the following statements

(refer to the Network Deﬁnition Language Reference Manual for a complete

description of these statements).

CPL1: C0UPLERN0DE=2,HNAME=H0ST1.

CPL2: C0UPLERN0DE=8,HNAME=H0ST1.

CPL3: COUPLERNODE=11,HNAME =HOST1.

The node parameter of the EST entry and NODE on the COUPLER statement have

the same numeric values, in this case 2, 8 (10s), and 11 (13s).

Revision M Deadstart Decks 3-33

CDCNET Device Interface EST Entry

Each mainframe device interface (MDI) or mainframe terminal interface (MTI) in a

CDCNET network that is connected to a CYBER mainframe must have an EST entry.

A total of 16 MDI, MTI, and NPU ESTs can.be available (ST=ON) at the same time.

The format of the EST entry is:

EQest=ND,ST=status,EQ=eq,PI=pip,CH=ch,ND=node,NT=node,IB=ib.

Parameter Description

est

ST=status

EST ordinal of the MDI or MTI; from 5 to 7778.

Indicates an MDI or MTI.

Speciﬁes whether the MDI or MTI is available for use; enter one of

these values:

status Description

EQ=eq

PI=pip

CH = ch

ND=node4

NT=node4

IB = ib

ON MDI or MTI is available.

OFF MDI or MTI is ignored during system operation.

Number of the controller for the MDI or MTI; from 0 to 7.

Peripheral interface program (PIP) index, which determines which copy

of the PP driver drives this MDI or MTI; from 1 to 4. Up to four EST

entries can have the same PIP index (that is, one PP can drive a total

of four front ends consisting of MDIs, MTIs, or NPUs).

Number of the channel to which the MDI or MTI is connected; from 0

to 138 and from 208 to 33s.

Node number of the coupler associated with the MDI or MTI being

deﬁned; from 1 to 377s.

Node number of the MDI or MTI; from 1 to 377s. This node number

identiﬁes the Control Data Network Architecture (CDNA) transport

function.

The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

The node numbers assigned to an MDI or MTI must be consistent with those assigned

to a 255x NPU network conﬁguration deﬁned through NDL. To minimize the impact on

existing 255x NPU networks, assign node numbers as follows:

1. Assign the lowest node numbers to NPU coupler nodes.

2. Reserve node numbers subsequent to those assigned in step 1 for NPU nodes.

3. Assign the remaining node numbers to MDI or MTI coupler nodes.

4. Reserve one node number to identify the CDNA transport function.

4. The procedure for assigning node numbers and creating network conﬁguration ﬁle deﬁnitions is described

in the Network Deﬁnition Language Reference Manual.

3-34 NOS Version 2 Analysis Handbook Revision M

Network Access Device EST Entry

The format for the network access device (NAD) EST entry is:

EQest=NC,ST=status,CH=ch,IB=ib.

Parameter Description

est

ST---status

CH=ch

IB=ib

EST ordinal of the NAD; from 5 to 7778.

Indicates a NAD.

Speciﬁes whether the NAD is available for use; enter one of these

values:

status Description

ON NAD is available.

OFF NAD is ignored during system operation.

Number of the channel to which the NAD is connected; from 0 to 138

and from 20s to 33s.

The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

j^S.

Revision M Deadstart Decks 3-35

Two-Port Multiplexer EST Entry

The two-port multiplexer (TPM) must be described in the EST to allow use of the

Remote Diagnostic Facility (RDF) on models 865 and 875 and CYBER 180-class

machines.

The format of the TPM EST entry is:

EQest=RM,ST=status,PT=pt,CH=ch,IB=ib.

Parameter Description

est

ST=status

PT=pt

CH=ch

IB=ib

The 1- to 3-digit octal EST ordinal of the TPM; 5 to 777s.

TPM equipment type.

Speciﬁes whether the TPM is available for use; enter one of these

values:

status Description

ON TPM is available.

OFF TPM is ignored during system operation.

Port number to be used by RDF; 0 or 1. RDF normally uses port 1.

RDF and the system console cannot use the same port number.

Channel number on models 865 and 875 and CYBER 180-class

machines; channel 15 is required.

The 1- to 4-digit octal value;.this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

y=**ss$v

3-36 NOS Version 2 Analysis Handbook Revision M

^PN

MAP III or IV Equipment EST Entry

The MAP III or IV equipment EST entry allows use of the matrix algorithm processor

(MAP).

The format of the MAP III or IV EST entry is:

EQest=MP,ST=status,CH=ch,IB=ib.

Parameter Description

est EST ordinal of MAP; from 5 to 777s.

MP Indicates MAP equipment.

ST=status Speciﬁes whether the MAP is available for use; enter one of these

values:

status Description

ON MAP is available.

OFF MAP is ignored during system operation.

CH=ch Number of the channel to which the MAP is connected; from 0 to 13s

and from 20s to 338.

IB=ib The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

Revision M Deadstart Decks 3-37

6683 Satellite Coupler EST Entry

The 6683 satellite coupler EST entry allows use of the NOS-SCOPE 2 Station Facility.

The format for the entry is:

EQest=CC.ST=status,EQ=eq,CH=chfIB=1b.

Parameter Description

est EST ordinal of 6683 coupler; from 5 to 777s.

CC Indicates 6683 coupler.

ST=status Speciﬁes whether the 6683 coupler is available for use; enter one of

these values:

status Description

ON 6683 coupler is available.

OFF 6683 coupler is ignored during system operation.

EQ=eq Controller number for equipment; from 0 to 7.

CH=ch Number of channel to which coupler is connected; from 0 to 138 and

from 208 to 33s.

IB=ib The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

3-38 NOS Version 2 Analysis Handbook Revision M

CYBERPLUS Ring Port EST Entry

The CYBERPLUS ring port equipment EST entry allows use of the CYBERPLUS

Subsystem. The format of the entry is:

EQest=RP,ST=status,CH=ch,lB=ib.

Parameter Description

est

ST=status

CH=ch

IB = ib

0ms

EST ordinal of the ring port; from 5 to 777s.

Indicates ring port.

Speciﬁes whether the ring port is available for use; enter one of these

values:

status Description

ON Ring port is available.

OFF Ring port is ignored during system operation.

Number of the channel to which ring port is connected; from 0 to 13s

and from 208 to 33s. This parameter is required.

The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

Revision M Deadstart Decks 3-39

Equipment Assignments: Mass Storage

This subsection describes the following EST entries and mass storage equipment

assignments.

Equipment

Disk equipment

Control module

Extended memory

Mass storage allocation control

Down channel entry

UP channel entry

Permanent ﬁles device

System library device

Alternate system library device

System checkpoint ﬁle device

Default family name

Removable device

Shared device

Independent shared device

Load buffer controllers

Extended memory allocation

UEM equipment initialization

Set access level limits

Set disk thresholds

Entry

MSAL

DOWN

SYSTEM

ASR

SCKP

FAMILY

REMOVE

ISHARE

LBC

UEMIN

ACCESS

THRESHOLD

3-40 NOS Version 2 Analysis Handbook Revision M

NOS Mass Storage Concepts

( Following are descriptions that deﬁne the NOS mass storage terminology and

summarize the kinds of mass storage equipment assignments that can be speciﬁed in

the EQPDECK. Table 3-3 summarizes the various functions that a particular mass

storage device can serve. For example, if a device is an alternate system device (listed

under the Alternate System column in table 3-3), then it cannot be a system device; it

can contain temporary ﬁles, direct access ﬁles, and indirect access ﬁles; it can be a

master device or a nonmaster device; it cannot be removable; it can be either an

auxiliary device or a family device; and it can be a shared device or a link device.

Alternate System Device

Whereas a system device contains all the routines in the system library, an alternate

system device contains copies of selected system library routines. The ASR entry in the

EQPDECK (refer to ASR - Alternate System Library Device Assignment Entry later in

this section) speciﬁes which mass storage devices are to be alternate system devices;

the *AD LIBDECK entry on the deadstart ﬁle speciﬁes which system library routines

are to reside on these alternate system devices. During system processing, the routines

on the alternate system device are used instead of the ones on the system device.

This feature allows each routine in the system library to reside on the mass storage

device that is most appropriate to the routine's use. For example, instead of using an

844 system device, a routine that is frequently used could use extended memory,

which has a faster transfer rate, as an alternate system device.

NOTE

0ms

f Use care when specifying alternate system devices. A mistake could result in no

alternate system device deﬁned. For example, suppose you specify the following in

your EQPDECK:

ASR=7.

ASR=11.

and you specify the following in the corresponding LIBDECK:

*AD,5,ABS/COMPASS

•AD.11.PP/1SJ

Since the system could not match equipment numbers between the EQPDECK and

the corresponding LIBDECK, the system library routine COMPASS will not be on

equipment 7 or 5. However, 1SJ will be on equipment 11 as speciﬁed.

Revision M Deadstart Decks 3-41

NOS Mass Storage Concepts

Table 3-3. Mass Storage Functions for Various Device Types

Other

Possible

Functions Type1 Type2 Type3 Type4 Type5 Type6 Type7 Type8 Type9

Alternate No Yes Yes Yes Yes Yes Yes Yes

system

device

System

device

No Yes Yes Yes Yes Yes Yes Yes

Contain Yes Yes Yes Yes Yes Yes Yes Yes

temporary

ﬁles

Contain

direct

access ﬁles

Yes Yes Yes mm Yes Yes Yes Yes Yes

Contain

indirect

access ﬁles

Yes Yes Yes Yes Yes Yes Yes Yes

Master

device

Yes Yes Yes Yes Yes Yes Yes Yes Yes

Nonmaster

device

Yes Yes Yes Yes No No Yes Yes Yes

Removable

device

No No No Yes Yes Yes No Yes No

Nonremov

able device

Yes Yes Yes Yes Yes Yes Yes Yes Yes

1. System.

2. Alternate system.

3. Containing temporary ﬁles.

4. Containing direct access ﬁles.

5. Containing indirect access ﬁles.

6. Auxiliary.

7. Default family.

8. Shared.

9. Link.

,A*^S

(Continued)

3-42 NOS Version 2 Analysis Handbook Revision M

yf^V

NOS Mass Storage Concepts

Table 3-3. Mass Storage Functions for Various Device Types (Continued)

Other

Possible

Functions Type1 Type2 Type3 Type4 Type5 Type6 Type7 Type8 Type9

A u x i l i a r y Y e s Y e s Y e s Y e s Y e s - N o Y e s Y e s

device

F a m i l y Y e s Y e s Y e s Y e s Y e s N o - Y e s Y e s

device

S h a r e d Y e s Y e s Y e s Y e s Y e s Y e s Y e s

device

L i n k Y e s Y e s Y e s Y e s Y e s Y e s Y e s

device

1. System.

2. Alternate system.

3. Containing temporary ﬁles.

4. Containing direct access ﬁles.

5. Containing indirect access ﬁles.

6. Auxiliary.

7. Default family.

8. Shared.

9. Link.

Revision M Deadstart Decks 3-43

NOS Mass Storage Concepts

Alternate Permanent File Family

More than one permanent ﬁle family can exist in a system: one default permanent ﬁle

family and one or more alternate permanent ﬁle families. One permanent ﬁle family is

deﬁned as the default family by the EQPDECK FAMILY entry (refer to FAMILY -

Default Family Name Assignment Entry later in this section). If another system's

permanent ﬁle family is introduced, it is an "alternate permanent ﬁle family; it can be

added without interrupting the default permanent ﬁle family's operation.

This is a useful feature if a site has more than one system or has groups of

installations. If one system fails, its permanent ﬁles can be accessed from another

system.

As an example, a site with two systems might run with the mass storage

conﬁguration shown in table 3-4.

Table 3-4. Mass Storage Conﬁguration for Two Systems at One Site

System Ordinal Device Spindles

Access

Used Contents

844

Direct access ﬁles

If system Y became inoperative, the B access could be connected to system X. This

could be done without interrupting operations of system X.

The EQPDECK entries in system X would be:

Entries Comments

EQ7=DI,ST=ON,UN = 0/l,CH = 0/3.

EQ10 = DI,ST=OFF,UN = 0/1.CH = 1/4.

REMOVE=10.

Deﬁnes access A.

Deﬁnes access B.

Allows introduction of access B into system

X during operation.

The EQPDECK entries in system Y would be:

Entries Comments

EQ7=DI,ST=ON,UN=0/1 ,CH = 1/4.

EQ10 = DI,ST=OFF,UN = 0/l,CH = 0/3.

REMOVE = 10.

Deﬁnes access B.

Deﬁnes access A.

Allows introduction of access A into system

Y during operation.

3-44 NOS Version 2 Analysis Handbook Revision M

0f&S

NOS Mass Storage Concepts

To allow for introduction of an alternate permanent ﬁle family:

1. Deﬁne the equipment to be introduced or removed in the EQPDECKs for both

systems (refer to the previous example).

2. Specify all of the equipment that may be introduced or removed during system

processing as removable.

3. When you want to introduce the equipment into a system, use the ON operator

command to indicate that the equipment that was set to the OFF position in the

system in.operation is now available. This introduces the alternate permanent ﬁle

family.

4. Enable the validation ﬁles for the family by entering X.ISF(FM=familyname).

Refer to section 20, System File Initialization.

Auxiliary Device

An auxiliary device is a mass storage device that is not part of a family. It is a

supplementary permanent ﬁle storage device, which may be privately owned (PRIVATE)

or may be shared by many users (PUBLIC). An auxiliary device resides on either a

removable or nonremovable device. On the permanent ﬁle entry (PF) for an auxiliary

device (for both a removable device and a ﬁxed device without packs), a pack name is

speciﬁed instead of a family name.

Refer to the NOS Version 2 Reference Set, Volume 3, for additional information

about private and public auxiliary devices.

As an example, four 844 spindles to be used as a public auxiliary device could be

deﬁned as follows:

EQ6=DI,ST=0N,UN=174,CH=2.

PF=6,X,name.

Private auxiliary devices can be created only after the system is up and running. An

operator can make a public device a private device by entering the INITIALIZE

command with the UN and TY=X parameters (refer to section 8, K-Display

Utilities).

Revision M Deadstart Decks 3-45

NOS Mass Storage Concepts

Family Device /^s

A family device is a mass storage device that is part of a family. It can be either a

removable device or a nonremovable device. The only difference between the two is

that a nonremovable device containing permanent ﬁles can also contain a copy of the

system library and/or temporary ﬁles. Refer to Alternate Permanent File Family earlier

in this section.

On the PF entry, the family name is important if two systems' permanent ﬁles are to

run on the same system. A user can only use one family of permanent ﬁles at one

time; if the user does not specify one, the default FAMILY entry is used.

A family device can contain direct and indirect access ﬁles. These ﬁles are deﬁned in

the NOS Version 2 Reference Set, Volume 3. The ﬁles that are allowed are set by

the device mask and secondary mask on the PF entry.

Link Device

Either extended core storage or extended semiconductor memory or STORNET is the

medium through which several computer systems are linked to form a multimainframe

operating environment (shared MMF). The link device contains the information

necessary to manage the mass storage that can be shared by more than one

mainframe. For a description of shared mass storage, refer to SHARE - Shared Device

Entry and ISHARE - Independent Shared Device Entry later in this section.

Master Device

The master device contains all of the permanent ﬁle catalog entries, indirect access

ﬁles, and ﬁle permits for a speciﬁc user. If permanent ﬁle access is required, the user's

master device must be available on the system, unless all access is to be to an

auxiliary device. The user index and family name uniquely describe a user's master

device.

Each master device is organized into ﬁve logical sections.

1. Allocation information.

A master device, like all mass storage devices, maintains device labels and track ^^^

reservation tables (TRTs). 1

The device label contains information describing the device, such as family name

and user mask, as well as locations of permit and catalog information and indirect

access ﬁles. Refer to the INITIALIZE - Initialization Entry later in this section.

The TRT is the key to allocating information on the master device and to

describing the physical layout of data on the device. Refer to APRDECK later in

this section.

2. Catalog information.

Catalog entries are used to determine the locations and attributes of permanent

ﬁles. The catalogs for a master device are allocated to contain catalog entries for

a speciﬁc group of user indexes. A particular catalog track may contain entries

for many users, the number depending upon the number of catalog tracks deﬁned

for the device. The user index provides the mechanism for differentiating between

user's ﬁles on a particular catalog track.

3-46 NOS Version 2 Analysis Handbook Revision M

NOS Mass Storage Concepts

3. Permit information.

Users can explicitly or implicitly allow other users to access their permanent ﬁles.

Refer to the PERMIT command in the NOS Version 2 Reference Set, Volume 3.

Information describing the permission for all permanent ﬁles is in the permit ﬁle.

Catalog entries contain a relative sector address within this permit ﬁle for

permissions that have been granted for the ﬁle.

4. Indirect access ﬁles.

The master device contains all of the user's indirect access ﬁles. These ﬁles can

be accessed by commands that generate working copies for manipulation by the

user.

5. Direct access ﬁles.

Direct access ﬁles can reside either on the master device or on another device in

the family, depending on the device masks speciﬁed on the PF entries. Direct

access ﬁles are ﬁles that can be accessed at their location on mass storage. A

working copy is not generated, so any updates or alterations made to the ﬁles are

permanent.

Revision M Deadstart Decks 3-47

NOS Mass Storage Concepts

Multispindle Device

To accommodate ﬁles that are larger than one device, you can specify multispindle

device assignments. Up to eight spindles of 844 disk drives or up to three spindles of

885 disk drives can be included in the equipment deﬁnition of one logical device, when

the device is ﬁrst deﬁned. All spindles must be available for access whenever the

device is accessed.

Multispindle devices are treated as one logical device, having a track size equal to n

times the single-spindle track size (n is the number of spindles in the device). The

tracks of an n-spindle device are broken down into n equally-sized segments, each

having a length equal to the single-spindle track size. Each segment is contained on

a different physical unit.

844 Expander

A nonexpanded controller can have up to eight disk drives connected to it. Each of the

connection paths is called a port and is identiﬁed by a port number ranging from 0 to ^^

7. An expander (10304 extender) is a hardware device that can be connected between j

controllers and 844 disk drives to increase the number of disk drives that each

controller can access.

The expander can be used only with 844-21 drives, although all equipment deﬁnitions

and equipment driving software support the 64-drive addressing scheme for both

844-21 (DI/DK) and 844-41 (DJ/DL) type equipment.

Each expander consists of either two or four expansion elements. An expansion

element connects to a single controller port and forms a connection path from that

port to from one through eight disk drives. The connection paths between an

expansion element and the eight possible disk drives are called ranks and are

identiﬁed by a rank number ranging from 0 to 7. Two expanders with four expansion

elements each can be connected to a single controller to allow that controller to

access a maximum of 64 disk drives. Each expansion element, however, is logically

independent and, as such, could be connected to any port of any controller.

3-48 NOS Version 2 Analysis Handbook Revision M

NOS Mass Storage Concepts

A single controller maximum conﬁguration can be visualized as an 8- by 8-square

checkerboard with each square representing one of 64 disk drives (ﬁgure 3-1).

0ms,

0H$mS

Expander Rank

Controller

2 3

Port

00 01 02 03 04 05 06 07

_^ Unit Numbers

""^ (octal)

10 11 12 13 14 15 16 17

20 21 22 23 24 25 26 27

30 31 32 33 34 35 36 37

40 41 42 43 44 45 46 47

50 51 52 53 54 55 56 57

60 61 62 63 64 65 66 67,,

70 71 72 73 74 75 76 77-*

tPort Digit

Rank Digit

Figure 3-1. Expander Addressing Map

A column of squares in ﬁgure 3-1 represents all the drives that are accessed through a

single controller port. A row of squares represents all the drives that have the same

expander rank. Each disk drive that can be accessed by the controller is addressed by

a 6-bit unit number. The rightmost 3 bits of this unit number select to which of the

eight controller ports the drive is connected. The leftmost 3 bits of the unit number

select to which of the eight ranks on an expansion element the drive is connected. This

unit number is speciﬁed as a 2-digit octal number in the mass storage equipment EST

entry (refer to EQ - Disk Equipment EST Entry later in this section). The right digit

(port digit) of the unit number is the port number and the left digit (rank digit) is the

rank of the unit in the particular expansion element.

If two disk drives are vertically adjacent on the expander addressing map (ﬁgure 3-1),

their unit numbers are considered to be vertically ordered; that is, both drives are

connected to the same expansion element, both have the same port number, and their

rank numbers differ by one (refer to example 1).

If two disk drives are horizontally adjacent on the expander addressing map

(ﬁgure 3-1), their unit numbers are considered to be horizontally ordered; that is,

both drives have the same rank number, and their port numbers differ by one (refer

to example 2). The special case of rank numbers of 0 for two horizontally adjacent

drives is equivalent to the deﬁnition of consecutive unit numbers for other equipment.

All drives connected to a controller, either directly or through an expansion element,

are supported as single-unit or multiunit logical devices. Unit numbers can range

from 0 to 778, rather than from 0 to 7, as for other equipment. Thus, a maximum of

sixty-four 844 disk drives connected to a single controller can be addressed. However,

a maximum of eight units can be speciﬁed per multiunit device. In addition, all units

of a multiunit device must be connected to the same channel and, therefore, to the

same controller.

Revision M Deadstart Decks 3-49

NOS Mass Storage Concepts

Figure 3-2 illustrates a conﬁguration in which two expansion elements and 20 disk

drives are connected through one controller. An expansion element with eight drives is

connected to port 0, an expansion element with six drives is connected to port 1, and

six drives are connected to six ports (ports 2 through 7). Each disk drive is shown as a

square with its appropriate unit number inside. This conﬁguration is used in the

following three examples to illustrate multiunit device assignments. The controller is

assumed to be connected to channel 1. Refer to EQ - Disk Equipment EST Entry later

in this section for speciﬁc information on assigning these devices.

Expander Rank

Controller Port

2 3 4

00 01 02 03 04 05 06 07

10 11

20 21

30 31

40 41

50 51

Figure 3-2. 844 Expander Conﬁguration With 20 Drives

NOTE

The following examples illustrate multiunit device assignment of devices. For a

description of the mass storage EST entry in the EQPDECK, refer to EQ - Disk

Equipment EST Entry later in this section.

3-50 NOS Version 2 Analysis Handbook Revision M

NOS Mass Storage Concepts

Example 1:

Figure 3-3 illustrates a possible conﬁguration for a three-unit vertically adjacent

multiunit device. This device could be assigned in the EQPDECK, specifying EST

ordinal 6, as:

EQ6=DI,ST=ON,UN=50/60/70,CH=1.

Expander Rank

Controller Port

2 3 4 5 6 7

00 01 02 03 04 05 06 07

10 11

20 21

30 31

40 41

50 51

70 vertically Auiaceni xnree-unu uevice

Figure 3-3. Vertically Adjacent Three-Unit Device

Revision M Deadstart Decks 3-51

NOS Mass Storage Concepts

Example 2:

Figure 3-4 illustrates a possible conﬁguration for a two-unit horizontally adjacent

multiunit device. This device could be assigned in the EQPDECK, specifying EST

ordinal 7, as:

EQ7=DI,ST=0N,CH=1,UN=40-41

Expander Rank

Controller Port

2 3 4 5 6 7

00 01 02 03 04 05 06 07

10 11

20 21

30 31

40 41

50 51

Figure 3-4. Horizontally Adjacent Two-Unit Device

3-52 NOS Version 2 Analysis Handbook Revision M

NOS Mass Storage Concepts

Example 3:

Figure 3-5 illustrates a possible conﬁguration of 20 disk drives into seven devices.

These devices could be assigned in the EQPDECK as follows:

EQ6=DI,ST=ON,UN=60,CH=1.

EQ7=DI,ST=0N,UN=70,CH=1.

EQ10=DI,ST=ON,UN=50-51,CH=1.

EQ11=DI,ST=0N,UN=10/20/30/40,CH=1.

EQ12=DI,ST=ON,UN,11/21/31/41,CH=1.

EQ13=DI,ST=0N,UN=0-3,CH=1.

EQ14=DI,ST=0N,UN=4-7,CH=1.

Devices on ordinals 6 and 7 are deﬁned as single units. These units may be

initialized online into a two-unit device if they are also deﬁned as removable.

Expander Rank

EQll

EQ10

EQ6

EQ7

Controller Port

0 1 2 3

EQ13

5 6

EQ14

00 01 02 03 04 05 06 07

10 11

EQ12

20 21

30 31

40 41

50 51

j/ms

Figure 3-5. Hardware Conﬁgured into Seven Devices

Revision M Deadstart Decks 3-53

NOS Mass Storage Concepts

Nonremovable Device

S*^S

A nonremovable device cannot be physically removed during system operation. It can 1

contain a copy of the system library, which means it is a system device; it can also be

available for temporary ﬁles. It may or may not contain permanent ﬁles.

Removable Device

A removable device can be logically or physically added or removed during system

operation without causing system malfunction.

A device is speciﬁed as removable with the REMOVE entry in the EQPDECK.

During deadstart, a removable device is recovered just as is any other mass storage

device, if the status is on. If the device is not available, then the status is displayed

for the operator (E,M display).

Removable devices can contain permanent ﬁles but cannot contain the system library

or temporary ﬁles, because a device containing active ﬁles (such as temporary or ^^

library ﬁles) cannot be removed from the system. A removable device can be either /^\

an auxiliary device or an alternate permanent ﬁle family device.

Shared Device

A shared device contains permanent ﬁles that can be accessed by more than one

mainframe. To have these permanent ﬁles accessible to the mainframe, the device must

be deﬁned as shared in the mainframe. Refer to SHARE - Shared Device Entry and

ISHARE - Independent Shared Device Entry later in this section.

A shared device can be removable. However, when unloading a shared device, it must }

be in global unload status before you remove it. Refer to the DSD commands

UNLOAD and MOUNT in section 5 for more information concerning removable

shared devices.

System Device

The system device is a nonremovable device on which the system library resides. It can

also contain permanent and temporary ﬁles.

Temporary File Device

The temporary ﬁle device is a nonremovable device on which the following temporary

system ﬁles reside:

• Library files

• Local files

• Rollout files

• System files

• Timed/event rollout files ^^s.

3-54 NOS Version 2 Analysis Handbook Revision M

Ji^V

NOS Mass Storage Concepts

Buffered Disks

Buffered disks are those disks for which the system provides an extended memory data

buffer. The following buffered disks are supported on the mainframes indicated:

• The 819 Disk Storage Subsystem is available only on model 176. The buffer is

kept in LCME.

• The 885-42 Disk Storage Subsystem is available on all models except CYBER

180-class machines and model 176. The buffer is kept in ESM.

• The 895 Disk Storage Subsystem is available on all CYBER 180-class machines

except models 810, 815, 825, and 830. The buffer is kept in UEM.

• The 887 and 9853 Disk Storage Subsystems are available only on models 860,

960, 990, and 995. The buffer is kept in UEM.

• Buffered disks are not available on models 810, 815, 825, and 830.

Using extended memory as a data buffer provides the following capabilities.

• The system treats the buffer as a disk cache so that multiple requests for a

particular disk data block can potentially be satisﬁed by doing only one disk read.

• The system buffers data to maintain maximum transfer rates regardless of the

user's buffer size.

To maintain maximum transfer rates, a CPU program that resides in CPUMTR

controls the buffered disk I/O request processing. The system maintains the standard

disk I/O interface to the user's programs.

The buffered disk error reporting process logs read/write errors in the binary

maintenance log. An unrecovered read/write error is also logged in the system error

log and an error message appears at the system control point. The binary

maintenance log must be processed by the Hardware Performance Analyzer (HPA) to

get detailed information concerning buffered device errors.

The amount of I/O buffer space in extended memory is important to system

performance. Refer to XM - Declare Extended Memory Space for Buffers or User

Access later in this section.

There may be situations when a particular user job runs slower with buffered disks.

This happens if the job issues CIO reads when a buffer is full or issues CIO writes

when a buffer is empty. (Both cases indicate inefﬁcient programming techniques.) You

can speed up the job by entering the DSD command ENABLE,LOGGING. The system

will perform faster I/O and issue warning messages to the job dayﬁle.

Revision M Deadstart Decks 3-55

NOS Mass Storage Concepts

Extended Memory Overview

Extended memory (EM) refers to the ﬁve types of additional memory used by NOS to

supplement central memory space and connect mainframes. This extended memory can

be used to store programs and data, but not to execute programs. Shown below are the

ﬁve types of extended memory and the mainframes that support each type. Also shown

are the maximum EM size supported by each mainframe, how each mainframe connects

to the extended memory, and what buffered devices are supported for each EM type.

Extended

Memory Mainframe

Maximum

Size

Mainframe

Connection

Buffered

Device 5

ECS All 2MW CPU port,6 DDP

ESM All 16MW CPU port,6 LSP 885-42

LCME 176 2MW CPU port 819

STORNET A ll " 4MW LSP

UEM 815,825,835,865,875

810,830

840,845,850,855,860,

870, 910

990,995

2MW

8MW

16MW

8MW

No connection,

part of mainframe's

main memory

895, 887,

9853

Physical Conﬁguration

Extended core storage (ECS) is an external memory available for all mainframes. ECS

connects to the mainframe using either a CPU port6 or a distributive data path (DDP)

through a peripheral processor.

Extended semiconductor memory (ESM) is an external memory available for all

mainframes. ESM connects to the mainframe using either a CPU port or a low-speed

port (LSP) through a peripheral processor.

Large central memory extended (LCME) is an external memory available only for

model 176 mainframes. LCME connects to the model 176 using a central processor

unit (CPU) port.

STORNET is an external memory available for all CYBER 170 and CYBER 180

mainframes. STORNET connects to the mainframe using a low-speed port (LSP)

through a peripheral processor.

<^^5S

5. The buffered devices shown are not necessarily supported on all of the mainframe types indicated.

6. A CPU port connection is not available on CYBER 180-class machines. CPU port on model 176 machines

can only connect to LCME.

3-56 NOS Version 2 Analysis Handbook Revision M

NOS Mass Storage Concepts

Uniﬁed extended memory (UEM) is a part of the mainframe's main memory and is

separated from central memory by a partition deﬁned at deadstart time by the site.

All extended memory except STORNET is accessed by special instructions

Instructions exist for multiple word transfers (FORTRAN: Level 1 variables;

COMPASS: RE/WE instructions) and single word transfers (FORTRAN- Level 2

variables; COMPASS: RX/WX instructions). The multiple word transfers are available

on all mainframe types; the single word transfers are available only with LCME and

UEM on models 176, 865, 875, and CYBER 180-class machines.

Logical Conﬁguration

Extended memory can be used for several purposes. The physical characteristics of the

various extended memory types differ, which affects the ways in which extended

memory can best be used. The possibilities are:

• To connect mainframes in a multimainframe environment (ESM, ECS and

STORNET).

• To store system routines and data files (all EM types).

• To supplement user central memory space (all EM types except STORNET).

• To provide a cache buffer for buffered I/O devices (LCME, ESM, and UEM).

ESM, ECS, or STORNET can be used as a link device to interconnect mainframes in a

multimainframe environment. UEM cannot be used as the link device since it is part

of a mainframe's main memory. For additional information concerning multimainframe

operations, refer to section 13, Multimainframe Operations.

Extended memory can be used as a mass storage device. The following system ﬁles

can be made extended memory resident as opposed to using central memory or disk

space.

System Files Description

Operating system

Temporary and

rollout

Secondary rollout

The alternate system residency EQPDECK entry ASR deﬁnes a

device to be an alternate system device. LIBDECK directives

specify which system library routines are to be placed on the

alternate system device.

The EQPDECK entry MSAL,T=est. or MSAL,R=est., where est

is the EST ordinal of the EM device, causes the system to use

extended memory as a mass storage device for temporary ﬁles or

rollout ﬁles.

The EQPDECK entry MSAL,S=est., where est is the EST ordinal

of the EM device, causes jobs that are less than a speciﬁed size

to be rolled out to extended memory. The IPRDECK entry SRST

deﬁnes the size of the rollout ﬁles allowed in extended memory.

Revision M Deadstart Decks 3-57

NOS Mass Storage Concepts

System ﬁles should be placed in extended memory only if a performance improvement

can be gained. Whether performance improves or not depends on your production

environment, including memory and input/output requirements, available central and

extended memory, and available disk units and channels.

The amount of memory available to a user job can be increased by deﬁning

user-accessible extended memory. NOS allows user data structures to reside in

memory external to the job's ﬁeld length. (Refer to COMPASS and FORTRAN

reference manuals.)

A portion of extended memory is reserved for users by the XM EQPDECK entry. The

user's memory is reserved in allocation units dependent on the amount of

user-accessible extended memory deﬁned at deadstart. The minimum allocation unit

(user EM block size [M = million] UEBS) varies as follows. Service limits are deﬁned

for the user in the IPRDECK in UEBS units.

At least But less than Allocation unit (UEBS)

1000 words

IM words

2M words

4M words

8M words

IM words

2M words

4M words

8M words

16M words

10008

2000s

40008

100008

20000s

This means that if you deﬁne 3 million words of user-accessible extended memory and

a user executes RFL(EC=1), that user will be assigned 40008 words of extended

memory.

Extended memory provides a cache buffer for buffered I/O devices. Space is allocated

by the system for buffered I/O with the XM EQPDECK entry. In most cases, system

performance improves as the amount of extended memory deﬁned for buffered I/O

increases. However, ample space must be available for other uses.

3-58 NOS Version 2 Analysis Handbook Revision M

00&S

NOS Mass Storage Concepts

CM and UEM Size Speciﬁcation and Determination

When UEM is deﬁned, its size is speciﬁed on the EQ EQPDECK entry for UEM. For

models 865 and 875, the remaining memory is used for central memory. On CYBER

180-class machines, the remaining memory is divided between native mode (NOS/VE or

EI) and central memory. If running in dual-state mode, the VE CMRDECK entry

speciﬁes the memory for NOS/VE. If running NOS only, central memory size equals

the total memory size minus the EM size minus the EI size.

Simultaneous Use of Two Types of Extended Memory

NOS allows the simultaneous use of two types of extended memory in the following

cases

•

Models 865 and 875 can access both UEM and ESM/ECS simultaneously. The

recommended use of this combination of extended memory is to deﬁne ESM/ECS

for system ﬁle residence and/or as a link device in a multimainframe environment

and to deﬁne UEM for user-accessible extended memory. Since the main memory

size for models 865 and 875 is limited to two million words, central memory space

is given up for any UEM deﬁned. Also, since ESM/ECS is an external device,

fewer memory conﬂicts result when it is accessed rather than UEM.

In this case, an EQ entry is required in the EQPDECK for ESM/ECS as well as

ASR and MSAL entries, as desired. User access is restricted to UEM via the XM

entry when the EM parameter is speciﬁed.

CYBER 180-class machines can also access both UEM and ESM/ECS/STORNET

simultaneously. The purpose of the ESM/ECS/STORNET access is to allow

connection of CYBER 180-class machines into a multimainframe environment. In

this case, ESM/ECS is deﬁned in the EQPDECK as the link device and UEM is

deﬁned for system ﬁles, I/O buffers, and user-accessible extended memory.

An EQ EQPDECK entry is required for both ESM/ECS/STORNET, but is not

required for UEM. The ESM/ECS/STORNET EQ entry deﬁnes the extended

memory as a link device via the LSP/DDP. The UEM EQ, if present, entry allows

the use of extended memory as a mass storage device. An XM entry allocates

extended memory for buffered I/O and user access.

Revision M Deadstart Decks 3-59

Specifying Ranges of EST Ordinals

Several EQPDECK entries allow you to specify multiple EST ordinals in the same

entry. For example, instead of entering

REMOVE=6,7,10,11,12.

you can enter:

REMOVE=6-12.

Ranges of ordinals can be mixed with individual ordinals, such as:

REMOVE=6,10-16.23.

Empty ordinals within a range are ignored. For example, specifying a range of 6

through 12 will ignore an unassigned ordinal 7. However, inappropriate devices

within a range or speciﬁed individually (such as specifying a printer as a removable

device) are diagnosed as errors.

EQ — Disk Equipment EST Entry

The purpose of the mass storage equipment (EQ) entries is to describe all mass storage

peripheral equipment. This includes disk equipment and extended memory equipment.

NOS requires that at least 6 million words of mass storage be available.

There can be up to 200 logical mass storage devices, and therefore, up to 200 mass

storage EST entries (this number does not include ordinals 0 through 4, which are

reserved for speciﬁc uses). An entry, however, can refer to more than one physical ^^\

unit. For example, two 885 spindles can be deﬁned as either two logical devices with

two EQ entries or as one logical device with one EQ entry.

A unit is a dual-access unit if it is accessed by one mainframe through two different

controller-channel access routes. To deﬁne a unit as a dual-access unit with its EQ

entry, specify two channel parameters. The channels should be from two controllers.

Only one channel of a dual-channel access controller is recommended for use on a

single mainframe, since using both channels of the controller results in a performance

degradation rather than an improvement. Therefore, if both channel accesses of a

controller are physically connected to the same mainframe, you should deﬁne only

one of them on an EQ entry.

NOTE

A device's EQ entry must precede any of the following entries for that device: ASR,

MSAL, REMOVE, PF, SYSTEM, FAMILY, INITIALIZE, SHARE, ISHARE, ACCESS,

and THRESHOLD. If you redeﬁne a device's EQ entry, then you must also redeﬁne

those entries.

y*^%.

3-60 NOS Version 2 Analysis Handbook Revision M

EQ — Disk Equipment EST Entry

0ams

The format of the EQ entry for disk units is:

EQest =type,ST=status,UN=units,CH=ch,pt1/ch2pt2.AP=ap,IB=ib,EQ=eq.

Parameter Description

est EST ordinal of the disk units; from 5 to 777s.

type Equipment type. The following disk units are supported by NOS.

type Equipment Number of Units

DB7 885-42, 7155-401 (full track) 1 to 3

D C 8 8 9 5 , 7 1 6 5 ( f u l l t r a c k ) 1 t o 2

DD9 834, 7255-1 (full track) 1 to 8

DF10 887 (4K sector; full track) 1 to 3

DG9 836, 7255-1 (full track) 1 to 3

DH10 887 (16K sector; full track) 1 to 2

DI 844-21, 7054/7154 (half track) 1 to 8

DJ 844-41/44, 7054/7154 (half 1 to 8

track), 7155-1

D K 8 4 4 - 2 1 , 7 1 5 4 ( f u l l t r a c k ) 1 t o 8

DL 844-41/44, 7154 (full track), 1 to 8

7155-1

DM 885-11/12, 7155-1 (half track) 1 to 3

DN10 9853 (2K sector, full track) 1

f^ DQ 885-11/12, 7155-1 (full track) 1 to 3

DVn Single-density 819 1

DW11 Double-density 819 1

When running NOS/VE in dual state with NOS, disks DC, DD, DF,

DG, or DH should not be deﬁned in the NOS EQPDECK if NOS/VE is

to use them.

7. Not applicable for CYBER 180-class machines.

8. Applicable for CYBER 180-class machines except models 810, 815, 825, and 830. You must also specify

the UEM EQ entry and XM entry to deﬁne the I/O buffer area in UEM.

9. Applicable for models 810, 815, 825, and 830 only. You must also specify the control module EQ entry.

10. Applicable to models 860, 960, 990, and 995 only.

11. Applicable to model 176 only.

Revision M Deadstart Decks 3-61

EQ — Disk Equipment EST Entry

Parameter Description

ST=status Speciﬁes whether or not the equipment is available for access; enter one

of the following values:

status Description

DOWN

IDLE

OFF

All access to the equipment is inhibited for the operating

system and user jobs. Specify DOWN if the equipment is

malfunctioning and access is not desirable or the equipment is

to be used by NOS/VE in a dual-state environment (not

applicable for disks DC, DD, DF, DG, or DH).

New ﬁles are not assigned to a device with IDLE status

unless no suitable alternative device exists, but you may

continue to access ﬁles already on the device as if the device

had a status of ON.

No user jobs can access the equipment; however, system

utilities are permitted to access the device, so it can be

dumped or loaded.

The speciﬁed equipment is generally available.

If the equipment is removable and is not available at deadstart, the

system determines it is unavailable, even if its EQ status entry speciﬁes

ON. If INITIALIZE is entered, the equipment is not initialized until it is

set to ON status. During system operation, the operator can initiate

access to this device by entering the ON command.

If the equipment can be used with either one of two different systems

(removable devices, not dual access), deﬁne the status of the EQ entry

as ON in the system to which it is currently available for access; deﬁne

the status of the EQ entry as OFF in the system to which it is not

currently available for access.

Use the ST=DOWN parameter to prevent NOS from using disks that

are to be used by NOS/VE. This applies to all disks except DC, DD, DF,

DG, or DH (these disks should not be deﬁned in the NOS EQPDECK if

NOS/VE is to use them).

3-62 NOS Version 2 Analysis Handbook Revision M

EQ — Disk Equipment EST Entry

Parameter Description

J ^ N

UN=units Deﬁnes unit number(s). Unit numbers are conﬁguration dependent.

For 834 or 836 units, unit numbers are generated as follows:

unit number = (control module physical equipment

number)*108 + (834 or 836 disk physical unit number)

For example, in UN = 10 the 1 is from the EQ = 1 parameter in the

control module entry and the 0 is the physical unit number of the

disk drive.

For 819 units, unit numbers are either from 0 to 3, if on FLPP channels

2 and 3, or from 4 to 7, if on FLPP channels 6 and 7.

For 844 units, unit numbers are from 0 to 77s.

For 885 units, unit numbers are from 40s to 77s.

For 887 units, unit numbers are from 0 to 7.

For 895 units, unit numbers have the following format:

unit number = shxxxx

where:

s = storage control address (0 or 1).

h = head of string controller address (0 or 1).

xxxx = spindle number (0 to 17s).

For 9853 units, unit numbers are from 0 to 7.

You can deﬁne the 834, 836, 844, 885, 887, 895, and 9853 physical units

with a separate EQ entry for each unit, or, if more continuous storage is

needed than is possible with one unit, you can deﬁne more than one

physical unit as one logical device with one EQ entry. If the EQ entry

is deﬁning more than one unit of a multispindle device, unit numbers

can be speciﬁed as UN=um-unn, if units are contiguous, or as

UN=uni/un2/.../unn,if units are noncontiguous. For 834, 836, and 9853

units, contiguous means that the units are all on the same control

module.

For example, three 885 units (0, 1, and 2) to be accessed as three units

are deﬁned as follows:

EQest=DM,ST=status,UN=0,CH=ch,AP=ap.

EQest=DM,ST=status,UN=l,CH=ch,AP=ap.

EQest =DM,CH=ch,UN=2,AP=ap,ST=stat us.

Revision M Deadstart Decks 3-63

EQ — Disk Equipment EST Entry

Parameter Description

UN=units (Continued) y-a^i\

Three 885 units (0, 1, and 2) to be accessed as one logical unit are

deﬁned as follows:

EQest =DM,ST=status,UN=0-2,CH=ch,AP=ap.

| CH=chi

I Pti

EQest=DM,ST=status,UN=0/1/2,CH=ch,AP=ap.

An advantage to accessing the three units as one logical unit is that

less space is used in CMR. A disadvantage is that if any of the units

malfunctions or is destroyed, all units are affected.

For all devices except 819s, 834s, 836s, 887s, and 9853s number of the

channel or channels to which the controller is connected; from 0 to 13s

and from 20s to 33s. For 819s, the channel pair for input/output(I/0)

multiplexer (primary access); enter one of these values:

c h i D e s c r i p t i o n

2 Channels 2 and 3.

4 Channels 4 and 5.

6 Channels 6 and 7.

For 834s and 836s, the channel number is speciﬁed on the control

module EQ entry.

For 887s and 9853s, the channel number for primary access; from 0 to

lis. The primary and secondary channel numbers must be different.

For 895s, a concurrent channel is speciﬁed by Cchi, or Cch2. For

example, CH = O2/C03 deﬁnes the channel pair as nonconcurrent channel

02 and concurrent channel 03. Use of a concurrent channel frees up two

nonconcurrent PPs for each channel that is moved to the concurrent I/O

subsystem (CIO). If an 895 is used to deadstart, one channel access to

the 895 must remain on a noncurrent channel.

For 887s and 9853s only, the channel port (A or B) for the primary

access channel. Default is port A.

A^S\

3-64 NOS Version 2 Analysis Handbook Revision M

EQ — Disk Equipment EST Entry

Parameter Description

/ch2 For all devices except 819s, 834s, 836s, 887s, and 9853s indicates

dual-access unit; cannot be 0; chi and ch2 should be connected to

different controllers. The system balances channel activity on dual-access

units. The balancing is done by alternately checking chi and ch2 for

availability. cti2 is assigned ﬁrst if it is free. However,MTR logically

reverses chi and ch2 during system operation.

For 819s, the secondary access channel pair; one of the values described

for chi.

For 834s and 836s, dual access is indicated on the control module EQ

entry.

For 887s and 9853s, the channel number for secondary access; from 0 to

118. The primary and secondary channel numbers must be different.

pt2 For 887s and 9853s only, the channel port (A or B) for the secondary

access channel. Default is port A.

AP=ap The 1- or 2-digit octal number that indicates which APRDECK to use. If

AP=ap is omitted, the ﬁrst APRDECK, APRD00, is assumed.

IB=ib The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

j0^s EQ=eq Control module number for the 9853 disk. Control module numbers are

f f r o m 0 t o 7 .

JJ0^*

Revision M Deadstart Decks 3-65

EQ — Control Module EST Entry

EQ —■ Control Module EST Entry

An 834 or 836 Disk Storage Subsystem consists of from one to eight 834 or from one

to three 836 disk units and the corresponding control modules. To be operational, both

the disk units and the control modules must be deﬁned in the EQPDECK. The EQ

entry for the 834 or 836 disk units is described under EQ - Disk Equipment EST

Entry earlier in this section. The description of the EQ entry for a control module

follows.

A control module for the 834 or 836 Disk Storage Subsystem is a controller that

drives up to four 834 or three 836 disks and that interfaces with the operating

system using one or two 7255-1 adapters (one per channel). The control module EQ

entry is used to:

• Specify the channels to be used to access the 834 or 836 drives defined on the

control module. Channel entries are not allowed on the 834 or 836 disk unit EQ

entries.

• Specify what level of controlware is to be loaded into the control module during *aS*^

deadstart or when you use a LOADBC command.

• Allow maintenance access to a control module without affecting other control

modules on the same channel(s).

The format of the control module entry is:

EQest =CM, EQ=eq, CH=ch, /ch2, CW=cw, I B=1 b.

Parameter Description a^^\

est EST ordinal of the control module; from 5 to 7778-

EQ=eq Physical equipment number of the control module; from 0 to 7. Each

control module must be assigned a unique equipment number.

CH=chi Number of the channel to which the control module is connected; from

0 to 138 and from 20s to 338.

/ch2 Indicates a dual-access control module; from 1 to 138 and from 20s j^m.

t o 3 3 s . " ^

3-66 NOS Version 2 Analysis Handbook Revision M

EQ — Control Module EST Entry

Parameter Description

0 ^ S " ■

( CW=cw Indicates the control module type and whether controlware is to be

installed in the control module during deadstart. The control module

load sequence takes at least 15 seconds for each control module. It is

advised that you load controlware only when necessary, cw can be one of

these values:

cw Description

0&&S

CM Control module for 834 disk. Install controlware (default).

NCM Control module for 834 disk. Do not install controlware.

C2 Control module for 836 disk. Install controlware.

NC2 Control module for 836 disk. Do not install controlware.

IB=ib The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

Revision M Deadstart Decks 3-67

EQ — Extended Memory EST Entry

NOS supports the following types of extended memory.

• Large Central Memory Extended (LCME)

• Extended Core Storage (ECS)

• Extended Semiconductor Memory (ESM)

• Unified Extended Memory (UEM)

• STORNET

Refer to the Extended Memory Overview earlier in this section for an explanation of

the various physical and logical conﬁgurations using extended memory. Also refer to

the Examples of EQPDECK Entries for Extended Memory to see examples of how to

use EQPDECK entries to deﬁne extended memory for both single mainframe and

multimainframe conﬁgurations.

You can use extended memory as an alternate system residency for often accessed

system routines. The faster access may result in a performance improvement.

However, you should use caution when placing system routines in UEM to avoid

possible performance degradation instead of an improvement. This can happen because

a portion of central memory is reserved for UEM and therefore that portion is

unavailable to users. Also, accessing the system routines that reside in UEM may

require additional CPU overhead. Use the ACPD and PROBE utilities (refer to

TRACER/PROBE Utilities, section 21) to determine the impact on system resources

when using UEM.

NOTE

If extended memory is not included in the hardware conﬁguration, do not make an

extended memory EST entry.

The format of the entry is:

EQest=typefST=status,MA=niorJe,ET=xmem/ddp/nc,SZ=size,CH=chi/ch2,AP=ap,MC=mc,IB=ib.

Parameter Description

est EST ordinal of the extended memory equipment; from 5 to 777s. If you

are using extended memory as a link device in a multimainframe

environment, est must be ordinal 5. Otherwise, extended memory can

be deﬁned as any EST ordinal in the given range.

type Extended memory equipment type; enter one of these values:

type Description

DE DDP or LSP is not available.

DP DDP or LSP is available (not applicable for UEM).

/*w%\

3-68 NOS Version 2 Analysis Handbook Revision M

EQ — Extended Memory EST Entry

Parameter Description

ST=status Speciﬁes whether extended memory is available for access; enter one of

these values:

status Description

ON Extended memory is available.

MA=mode

OFF Extended memory is ignored during system operation.

Maintenance mode (not applicable for UEM). If you specify MA=ON,

online extended memory diagnostics are allowed to reference the half of

extended memory that is placed in maintenance mode at the controller.

The other half of extended memory is available to the system. The size

of available physical extended memory is divided by 2 at deadstart.

When you initially place an extended memory device in maintenance

mode, all mainframes using the extended memory must initialize it.

When you place ECS in maintenance mode, you must also make the

PRESET entry for multimainframe operation. Refer to INITIALIZE and

PRESET later in this section. If you omit the MA = mode parameter, the

default is MA=OFF.

If you use part of extended memory in maintenance mode and you have

ESM, you must also use the MC=mc parameter to deﬁne which

maintenance port to use.

ET=xmem/ Type of large memory and ddp. If you specify ddp, you must also specify

ddp/nc xmem. If this parameter is not speciﬁed, El/Dl is assumed*

xmem Description

ddp

ECS I for all mainframes.

ECS II for all mainframes.

LCME only for model 176.

UEM for models 865 and 875 and CYBER 180-class machines.

Type must be DE. The system ensures that the sum of memory

words speciﬁed by the MINCM CMRDECK entry and speciﬁed

by this entry for UEM is present.

ESM for all mainframes.

STORNET for all CYBER 170 and CYBER 180 mainframes.

Description

DI DC135 DDP. Default for xmem of El or E2.

D2 DC145 (parity enhanced) DDP, or ESM low-speed port (LSP).

Default for xmem of ES and SN.

nc Description

NC Indicates type DP does not have a CPU coupler. NC is not

valid for xmem of DE. Default for xmem of SN.

Revision M Deadstart Decks 3-69

EQ — Extended Memory EST Entry

Parameter Description

I SZ=size Size of extended memory in words/10008. This value must not be larger

than the size of the physical memory that is present.

• For ECS II and LCME, size can range from a minimum of 108 to a

maximum of 100008.

• For STORNET, size can range from a minimum of 10s to a

maximum of 20000s.

• For ESM, size can range from a minimum of 108 to a maximum of

1000008.

• For UEM, size can range from a minimum of 108 to a maximum of

(1000008-CM size).

NOTE

Since CIP uses a small amount of CM, you may need to make

allowances for CIP when you enter a value for size.

• For ECS I, size is one of the values in the following table.

ECS I

Size (Octal) Available Number of Banks

400

1000

2000

4000

10000

125K12

250K

500K

1000K

2000K 16

The following table shows the value in EM words (both octal and

decimal) for a given value of size.

EM Words

Size (Octal) (Octal) (Decimal)

10 10000 4096

400 400000 131072

1000 1000000 262144

2000 2000000 524288

4000 4000000 1048576

10000 10000000 2097152

20000 20000000 4194304

40000 40000000 8388608

100000 100000000 16777216

12. K is 1000 60-bit words.

3-70 NOS Version 2 Analysis Handbook Revision M

EQ — Extended Memory EST Entry

Parameter Description

CH=chi,/ch2

AP=ap

MC=mc

IB=ib

Numbers of the channels to which the DDP is connected; from 0 to 13s

and from 20s to 33s.

If the equipment type is DE, do not specify a channel parameter. If a

channel parameter is speciﬁed, the system recognizes the DE entry as a

DP entry.

If the equipment type is DP, specify either one or two channels. The

second channel cannot be 0 (zero). If a DDP is present, the loading of

CPU programs residing in ECS or ESM still occurs via the CPU. A

DDP must be connected to a channel by itself.

The 1- or 2-digit octal number that indicates which APRDECK to use. If

you omit AP=ap, the ﬁrst APRDECK, APRD00, is assumed.

Number of the channel to which the maintenance port is connected; from

1 to 138 and from 20s to 33s. Channel 0 cannot be used for the

maintenance port. This channel is for ESM only and must be speciﬁed

for proper reporting of ESM errors.

The 1- to 4-digit octal value; this value is entered in the installation

byte for the speciﬁed EST ordinal. Refer to EST Formats in the NOS

Version 2 Systems Programmer's Instant.

Revision M Deadstart Decks 3-71

Examples of EQPDECK Entries for Extended Memory

The following examples show the EQPDECK entries that are necessary to deﬁne

extended memory in various conﬁgurations.

Single Mainframe Examples

The EST ordinal of extended memory equipment ranges from 5 to 7778. For single

mainframe conﬁgurations, extended memory can be deﬁned as any EST ordinal in the

given range. For these examples, the machine identiﬁer of the mainframe is assumed

to be .AA.

Example 1:

This example shows a CYBER 170 mainframe with ESM connected via the high-speed

port. ESM can be used as a mass storage device, user-accessible extended memory,

and for 885-42 disk I/O buffers.

EQ6=DP,ET=ES,ST=ON,S2=10000,CH=32,

MSAL,S=6.

ASR=6.

XM=AA,3000,1000.

Example 2:

(Deﬁne 2 million words of ESM.)

(Deﬁne ESM as secondary rollout device.)

(Deﬁne ESM as alternate system residency

device.)

(Allocate 3000000 octal words for 885-42

I/O buffers and 1000000 octal words for

user-accessible extended memory.)

This example shows a CYBER 180-class machine with part of its mainframe memory

deﬁned as UEM, which can be used as a mass storage device, for user-accessible

extended memory, and for 887, 895, or 9853 disk I/O buffers.

EQ7=DE,ST=ONfET=EM,SZ=2000

XM-AA,1000,200.

NOTE

(Deﬁne 1/2 million words of memory as

UEM.)

(Allocate 1000000 octal words for 895,

887, or 9853 I/O buffers and 200000 octal

words for user-accessible extended

memory.)

If you are running a dual-state system, the VE CMRDECK entry will affect the

amount of memory available for NOS and thus for UEM.

Example 3:

This example shows a CYBER 180-class machine where UEM is used only for

user-accessible extended memory, not as a mass storage device or for I/O buffers. No

EQ entry for UEM is necessary in this case.

XM=AA,,200, (Allocate 200000 octal words for

user-accessible extended memory.)

3-72 NOS Version 2 Analysis Handbook Revision M

Examples of EQPDECK Entries for Extended Memory

08&*\ Example 4:

This example shows a CYBER 170 model 865 or 875 mainframe with no ESM. Part

of the mainframe memory is deﬁned as UEM and is used as a mass storage device

and for user-accessible extended memory. UEM cannot be used for 885-42 disk I/O

buffers.

EQ11=DE,ST>ON,ET=EM,SZ=2000

MSAL,S=11.

ASR=11.

XM=AA.,200.

Example 5:

(Deﬁne 1/2 million words of UEM.)

(Deﬁne UEM as secondary rollout device.)

(Deﬁne UEM as alternate system

residency device.)

(Allocate 200000 octal words for

user-accessible extended memory.)

This example shows a CYBER 170 model 865 or 875 mainframe with no ESM (as in

example 4) except that UEM is used only for user-accessible extended memory and

not as a mass storage device. No EQ entry for UEM is necessary in this case.

XM=AA,,200.

Example 6:

(Allocate 200000 octal words for

user-accessible extended memory.)

This example shows a CYBER 170 model 865 or 875 mainframe with ESM connected

via the high-speed port. ESM is used as a mass storage device, for 885-42 I/O buffers,

and for user-accessible extended memory. No UEM is deﬁned.

EQ10=DP,ST=ON,ET=ES,SZ=10000,CH=7

MSAL,S=10.

ASR=10.

XM=AA,3000,1000.

(Deﬁne 2 million words of ESM.)

(Deﬁne ESM as secondary rollout device.)

(Deﬁne ESM as alternate system residency

device.)

(Allocate 3000000 octal words for 885-42

I/O buffers and 1000000 octal words for

user-accessible extended memory in ESM.)

Revision M Deadstart Decks 3-73

Examples of EQPDECK Entries for Extended Memory

Example 7:

This example shows a CYBER 170 model 865 or 875 mainframe with ESM connected

via the high-speed port. ESM is used as a mass storage device and for 885-42 disk

I/O buffers. Part of the mainframe's memory is deﬁned as UEM and used for

user-accessible extended memory.

You have the option on CYBER 170 model 865 and 875 mainframes to deﬁne

user-extended memory either in ESM or UEM. If ESM is present, as indicated by an

EQ entry for it, then user-accessible extended memory resides in ESM by default.

However, you can force user-accessible extended memory to reside in UEM with the

EM parameter on the XM entry. 885-42 disk I/O buffers must reside in ESM.

H^ifSgs

EQ10=DP,ST=ON,ET=ES,SZ=10000,CH=7

MSAL,S=10.

ASR=10.

XM=AA,3000,1000,EM.

(Deﬁne 2 million words of ESM.)

(Deﬁne ESM as secondary rollout device.)

(Deﬁne ESM as alternate system residency

device.)

(Allocate 3000000 octal words for 885-42

I/O buffers in ESM and 1000000 octal

words for user-accessible extended memory

in UEM.)

A**3S

3-74 NOS Version 2 Analysis Handbook Revision M

Examples of EQPDECK Entries for Extended Memory

JPV.

Multimainframe Examples

For all multimainframe conﬁgurations, the extended memory link device must be

deﬁned as the equipment with EST ordinal 5. The machine identiﬁers of the connected

mainframes are assumed to be AA and AB. The following examples do not show the

SHARE EQPDECK entries that are required to indicate multimainframe operation.

Example 1:

This example shows two CYBER 170 mainframes linked to ESM via high speed ports.

In addition to being the link device, ESM can also be used as a mass storage device,

for 885-42 disk I/O buffers, and for user-accessible extended memory.

EQ5=DP,ST=0N,ET=ES,SZ=10000,CH=5.

XM=AA,1000,200. (Deﬁne 2 million words of ESM.)

(Allocate 1000000 octal words for 885-42

I/O buffers and 200000 octal words for

user-accessible extended memory in ESM

for machine AA.)

XM=AB,,400. (Allocate 400000 octal words for user-

accessible extended memory in ESM for

machine AB.)

The EQPDECK entries for the linked mainframe would be the same as shown above;

the link device (ESM) is deﬁned as equipment 5 and both XM entries must be

present.

Example 2:

This example shows two linked CYBER 180-class machines connected to ESM via

low-speed ports. Part of the mainframe memory for machine AA is deﬁned as UEM,

which is used as a mass storage device, for 895, 887, or 9853 disk I/O buffers, and

for user-accessible extended memory. On CYBER 180-class machines, user-accessible

extended memory and 887 or 895 disk I/O buffers must reside in UEM; they cannot

reside in ESM.

EQ5=DP,ST=0N,ET=ES/D2/NC,SZ=10000,CH=4

EQ6=DE,ST=0N,ET=EM,SZ=2000

MSAL,S=6.

ASR=6.

XM=AA,1000,200

(Deﬁne ESM as the link device. NC

indicates that there is no CPU access to

ESM.)

(Deﬁne 1/2 million words of UEM; without

this EQ entry, UEM can be used only for

user-accessible extended memory.)

(Deﬁne UEM as secondary rollout device.)

(Deﬁne UEM as alternate system

residency device.)

(Allocate 1000000 octal words for 895,

887, or 9853 I/O buffers and 200000 octal

words for user-accessible extended memory

in UEM.)

Revision M Deadstart Decks 3-75

Examples of EQPDECK Entries for Extended Memory

The EQPDECK for the linked CYBER 180-class machine AB might look like the

following. This mainframe has part of its memory deﬁned as UEM used only for

user-accessible extended memory.

EQ5«DP,ST=ON,ET=ES/D2/NC,SZ=10000,CH=4.

XM=AB,,200.

(Deﬁne ESM as the link device.)

(Allocate 200000 octal words for

user-accessible extended memory in UEM.

Since no EQ entry is present for UEM, it

cannot be used as a mass storage device

or for I/O buffers.)

The XM entries appear only in the EQPDECK for the machine where the memory is

being allocated. Do not place them in the EQPDECKs of any linked mainframes.

Example 3:

This example shows a CYBER 170 mainframe connected to ESM via a high-speed

port linked to a CYBER 180-class machine connected to ESM via a low-speed port.

EQPDECK for CYBER 170

EQ5=DP,ST=ON,ET=ES,SZ=10000,CH=5.

MSAL,S=5.

ASR=5.

XM=AA,1000,400.

(Deﬁne ESM as the link device.)

(Deﬁne ESM as secondary rollout device.)

(Deﬁne ESM as alternate system residency

device.)

(Allocate 1000000 octal words for 885-42 I/O

buffers and 400000 octal words for

user-accessible extended memory in ESM.) ~*

EQPDECK for CYBER 180

| EQ5=DP,ST=ON,ET=ES/D2/NC,SZ=10000,CH=4.

| EQ6=DE,ST=ON,ET=EM,SZ=2000.

| MSAL,S=6.

I ASR=6.

XM^AA,1000,400

| XM=AB,1000,200

(Deﬁne ESM as the link device.)

(Deﬁne 1/2 million words of UEM; without

this EQ entry, UEM can be used only for

user-accessible extended memory.)

(Deﬁne UEM as secondary rollout device.)

(Deﬁne UEM as alternate system

residency device.)

(Linked 170 XM entry must be present to

deﬁne user-accessible extended memory

and 885-42 I/O buffers in ESM.)

(Allocate 1000000 octal words for 895,

887, or 9853 I/O buffers and 200000 octal

words for user-accessible extended memory

in UEM. This entry can not be in the 170

EQPDECK.)

>tf^ v

3-76 NOS Version 2 Analysis Handbook Revision M

y$$S

Examples of EQPDECK Entries for Extended Memory

Example 4:

This example shows a CYBER 170 model 865 or 875 connected to ESM via a

low-speed port linked to a CYBER 180-class machine also connected to ESM via a

low-speed port. This conﬁguration allows continued system operation if the high-speed

port to ESM from the CYBER 170 model 865 or 875 was down.

EQPDECK for CYBER 170 model 865 or 875

Note that ESM cannot be used for user-accessible extended memory or for 885-42 I/O

buffers if a high-speed port is not present.

EQ5=DP,ST=ONFET=ES/D2/NC.SZ=10000,CH=4. (Deﬁne ESM as the link device; NC

indicates that there is no CPU access.)

EQ11=DE,ST=ON,ET=EM,SZ=2000. (Deﬁne 1/2 million words of UEM. This is

allowed; in contrast to example 8 earlier,

because ESM has no CPU access.)

MSAL,S=11 (Deﬁne UEM as secondary rollout device.)

ASR=11. (Deﬁne UEM as alternate system residency

device.)

XM=AA,,200. (Allocate 200000 octal words for

user-accessible extended memory in UEM.

This entry cannot be in the 180

EQPDECK.)

EQPDECK for CYBER 180

EQ5=DP,ST=ON,ET=ES/D2/NC,SZ=10000,CH=4. (Deﬁne ESM as the link device.)

EQ6=DE,ST=ON,ET=EM,SZ=2000. (Deﬁne 1/2 million words of UEM; without

this EQ entry, UEM can be used only for

user-accessible extended memory.)

msal,S=6. (Deﬁne UEM as secondary rollout device.)

ASR=6. (Deﬁne UEM as alternate system residency

device.)

XM=AB, 1000,200. (Allocate 1000000 octal words for 895, 887,

or 9853 I/O buffers and 200000 octal words

for user-accessible extended memory in

UEM. This entry cannot be in the 170

EQPDECK.)

Revision M Deadstart Decks 3-77

MSAL — Mass Storage Allocation Control Entry

The MSAL entry assigns job ﬁles of the speciﬁed type to the mass storage devices

deﬁned by the speciﬁed EST ordinal.

The format of the MSAL entry is:

:::

I MSAL,t=esti,est2,. ..,estn.

I Parameter Description

I t File type; one of these values:

I t Description

I B L G O ﬁ l e s .

I D Job dayﬁles.

I I Input ﬁles.13

I L Local ﬁles.

O Output ﬁ l es.13

'f.

I P Primary ﬁles.

I R Rollout ﬁles.

S Secondary rollout ﬁles.

I T Temporary ﬁles.

I esti EST ordinal of a nonremovable mass storage device, from 5 to 57a

f Ranges of ordinals can be speciﬁed.

I N O T E

In most cases, using ESM or UEM as a secondary rollout device

improves system performance. On the other hand, depending on the '^]

I mainframe type, the size of the jobs being rolled out, and the

secondary rollout sector threshold (IPRDECK and DSD SRST

command), CPU overhead may increase signiﬁcantly. Using central

memory for pseudo-control points instead of using the same memory as

UEM for secondary rollout is normally preferable because of reduced

CPU overhead. Use the ACPD and PROBE utilities (refer to

TRACER/PROBE Utilities, section 21) to determine the best use of

system resources.

13. Routing a ﬁle to queues changes a ﬁle's type, not its residency. Thus specifying an MSAL,I=est or

MSAL,0=est entry does not necessarily force all input or output queued ﬁles to the speciﬁed devices.

3-78 NOS Version 2 Analysis Handbook Revision M

0ms.

MSAL — Mass Storage Allocation Control Entry

Secondary rollout ﬁles are rollout ﬁles whose size in sectors is smaller than a

threshold speciﬁed by the IPRDECK or DSD entry SRST. These ﬁles are rolled out to

devices speciﬁed by the MSAL,S= entry. All ﬁles selected for rollout that are equal to

or greater in size than the threshold are rolled out to devices speciﬁed by the

MSAL,R= entry. The default value of SRST is 0 (zero); thus, no secondary rollout ﬁles

exist unless this value is changed. A possible use of this feature is:

Entries Comments

EQPDECK

EQ5=DP,ST=ON,SZ=1000,CH=27. Specify equipment 5 as extended memory with a DDP.

MSAL,S=5. Direct secondary rollout ﬁles to extended memory.

IPRDECK

SRST=20. Set threshold count so that no large rollout ﬁles are

rolled out to extended memory.

Revision M Deadstart Decks 3-79

DOWN — Down Channel Entry

| DOWN — Down Channel Entry

I The DOWN entry disables the use of channels at deadstart before the system attempts /

I to use them to access devices connected to them. If a channel is the only remaining

| access to a device, .it may not be downed unless that device is deﬁned as DOWN. The

I DOWN entry does not apply to either the deadstart channel or the CIP channel.

| The DOWN entry format is:

| DOWN,CH=ch1fch2 chn.

| Parameter Description

| chi Number of the channel; from 0 to 13s and 208 to 338. A concurrent

channel is indicated by a C preﬁx (Cchi).

| Example:

| The following entry sets DOWN status for both nonconcurrent channel yrt_

23 and concurrent channel 4. ^

| D0WN,CH=23,C4.

| UP — Up Channel Entry

| The UP entry allows resumption of normal use of channels that have been disabled by

| a previous DOWN entry.

| The UP entry format is: ^=^

UP,CH=ch1tch2,...,chn.

| Parameter Description

I chi Number of the channel; from 0 to 13s and 20s to 33s. A concurrent

channel is indicated by a C preﬁx (Cchi).

I Example:

The following entry clears DOWN status of both nonconcurrent channel ***\

23 and concurrent channel 4.

I UP,CH=23,C4.

3-80 NOS Version 2 Analysis Handbook Revision M

jgjffe\

0n^S

PF — Permanent Files Device Assignment Entry

Before initializing a mass storage device (with the INITIALIZE entry in the

EQPDECK), add a PF entryu for that device to the EQPDECK anywhere after the

device's EQ entry. The PF entry information becomes part of the device's label when it

is initialized during deadstart; this label is recovered during subsequent deadstarts. For

subsequent deadstarts, it is not necessary that the PF entry be part of the EQPDECK

on the deadstart ﬁle; if it is, it is ignored. Refer to the INITIALIZE entry for the

default PF entries.

If the unit is a family device, the format is:

PF=est,type,dm.sm,name,device,nc.

If the unit is an auxiliary device, the format is:

PF=est,type,name,nc.

Parameter Description

est EST ordinal of the device; from 5 to 7778.

type Type of device; one of these values:

type Description

F Family device. It can contain indirect access ﬁles if the dm

parameter is from 1 to 377s. It can contain direct access ﬁles

if the sm parameter is from 1 to 377s. It is a master device

if the dm parameter is speciﬁed.

X Auxiliary device, which can contain both direct and indirect

access ﬁles. X must be speciﬁed on a unit's PF entry if any

of the auxiliary device commands are to be used for the

device.

dm Speciﬁes the unit's device mask; from 0 to 3778. Set according to

information under Device Masks in section 17, PF Utilities. Omit this

parameter if the device is an auxiliary device.

The device mask for a permanent ﬁle device deﬁnes the groups of

users whose catalogs reside on the device for a particular family.

sm Speciﬁes the unit's secondary mask; from 0 to 3778. Set according to

information under Device Masks in section 17, PF Utilities. Omit this

parameter if the device is an auxiliary device.

This parameter controls the residence of direct access ﬁles in the same

way that dm controls the residence of indirect access ﬁles.

14. Refer to table 3-5 for dependencies.

Revision M Deadstart Decks 3-81

PF — Permanent Files Device Assignment Entry

Table 3-5. PF Entry Parameter Settings (type, dm, name, device, and sm)

Type of PF Files Permitted on

Device Device type name d e v i c e s m

Auxiliary

Family

Indirect and/or X

direct

Direct only F

Indirect only F

(master device)

Indirect and F

direct (master device)

Omit pack Omit

0 family 1-778

1-3778 family 1-778

Omit

1-3778

1-3778 family l-77s 1-3778

1. If the device is to be a private auxiliary device, enter the INITIALIZE command

after deadstart and specify the user name. Refer to section 8, K-Display Utilities, for

the procedure. -

3-82 NOS Version 2 Analysis Handbook Revision M

PF — Permanent Files Device Assignment Entry

cParameter Description

name

device

Designates either the name of the family to which the unit belongs or

its pack name if it is an auxiliary device; from 1 to 7 alphanumeric

characters. Do not use the family name 0: it is reserved.

The family name describes the permanent ﬁle devices available to a

user. A family may consist of from 1 to 63 logical devices; however,

the master devices within the family must have device masks totaling

3778 if all possible user indexes are to be accommodated.

Usually a system runs with one family of permanent ﬁle devices

available. But you can activate additional families on a system, in order

to allow the users of these families to access their permanent ﬁles

through an alternate system. This might be helpful if one system

supplies backup service to another system. When more than one family

is active on a system, users with matching user indexes access the same

permanent ﬁles on a public auxiliary device. You can avoid this

situation by predetermining a range of user indexes for each family

running on a system. When a new family is introduced into a system,

its user indexes should be checked against those of the family or

families currently running and any matching indexes should be changed.

Refer to the IPRDECK entry COMLIB later in this section.

The pack name is the unique 7-character name associated with an

auxiliary device. An auxiliary device is a self-contained permanent ﬁle

device: all permanent ﬁles (whether direct or indirect access) represented

by the catalogs on the device reside on that device. To access a ﬁle on

an auxiliary device, users must specify the pack name as part of the

permanent ﬁle request. The pack name is used instead of the usual

algorithm for determining catalog location (user masks and family

name). An auxiliary device can be private or public. Any user who

knows the pack name and has the appropriate permissions and

validations can access ﬁles on an auxiliary device. Only the owner user

name can create ﬁles on a private auxiliary device (perform DEFINE,

SAVE, or REPLACE requests).

Number of the device in the family; from 1 to 77s. Omit this parameter

if the device is an auxiliary device.

A permanent ﬁle that does not reside on the master device has a device

number in the catalog entry or on the master device. The device number

speciﬁes on which alternate device within the family the ﬁle resides.

15. If not otherwise speciﬁed, the default family name becomes part of the tape label information. It is

checked and veriﬁed if the user speciﬁes the FA=A parameter on a command. Refer to the NOS Version 2

Reference Set, Volume 3 for a discussion of FA=A.

Revision M Deadstart Decks 3-83

PF — Permanent Files Device Assignment Entry

Parameter Description

Number of catalog tracks (optional) used only for master devices; from 1

to 2008. This value must be a power of 2. If you do not specify nc, one

of the following default values (based on the equipment type) is supplied.

Default nc Equipment Type

10 819 DV/DW

40 834 DD

40 836 DG

40 844-21 DI/DK

40 844-41/44 DJ/DL

10 885-11/12 DM/DQ

10 885-42 DB

10 887(4K) DF

10 887U6K) DH

10 895 DC

10 9853 DN

4Extended

memory

4ECS or ESM

with DDP

1Private device

Examples:

PF=6,F,125,125,SYSTEM,3,200

PF=17,X,PACK.

3-84 NOS Version 2 Analysis Handbook Revision M

SYSTEM — System Library Device Assignment Entry

0m**.

SYSTEM — System Library Device Assignment Entry

The SYSTEM entry speciﬁes which mass storage devices are to contain copies of the

NOS system library from the deadstart ﬁle. A system device can be any disk storage

device as well as extended memory.

Throughput can be greatly improved by specifying more than one system device For

example, if two system devices are speciﬁed and they are on different channels, the

time required to access system programs can be reduced. When the channel for one

system device is busy, the other is accessed. Also, if hardware problems occur on one

system device, the other system device can still be accessed. A good general rule is

to have one copy of the NOS system library per pair of channels with a maximum of

three copies. This allows alternate access to the system library while saving mass

storage space.

The following restrictions apply.

• The EQ entry for a system device cannot have the status set to OFF.

• A REMOVE entry cannot exist in the EQPDECK for a device being specified as a

system device.

• If more than one device is specified as a system device, all devices specified must

be of the same type and have the same number of spindles. For example, if there

are two system devices and the equipment EST ordinal for one of them speciﬁes

DI (single spindle), the equipment EST ordinal for the other one must also specify

DI (single spindle).

• If no devices are specified as system devices, the system library resides on the

ﬁrst nonremovable mass storage device.

• An ASR entry cannot exist in the EQPDECK for a device being specified as a

system device.

The SYSTEM entry format is:

SYSTEM=est1fest2 estn.

Parameter Description

esti EST ordinal of the device to contain a copy of the system library on

the deadstart ﬁle; from 5 to 37s . The EQ entry must be set to ON

status. One or more ordinals can be speciﬁed with one SYSTEM entry.

Ranges of ordinals can be speciﬁed. The maximum number of system

devices allowed depends upon the value of MXSY (refer to COMSMSC

Parameters in the NOS Version 2 Installation Handbook).

Revision M Deadstart Decks 3-85

ASR — Alternate System Library Device Assignment Entry.

ASR — Alternate System Library Device Assignment Entry

This entry speciﬁes which mass storage devices are to be alternate system devices. An

alternate system device is a mass storage device on which duplicate copies of system

routines can be placed by the system, either for faster access than is possible from a

system device or because they are frequently used programs. The following restrictions

apply.

• The device must be a mass storage device, including extended memory.

• The device cannot be a removable device.

• The device cannot be a system device.

When loading a system routine, the system will access the routine from the alternate

system device if practical. However, if the alternate system device is down or its access

is more congested than the system device, the system will access the routine from the

system device.

The procedure for selecting the records to be placed on the alternate device is in

LIBDECK.

The ASR entry format is:

ASR=est1,est2,....estn.

Parameter Description

esti EST ordinal of mass storage device to be used as an alternative

system device; from 5 to 37s. Ranges of ordinals can be speciﬁed.

SCKP — System Checkpoint File Device Assignment Entry

The SCKP entry speciﬁes on which mass storage devices the checkpoint ﬁle is to

reside. This entry is processed during a level 0, 1, or 2 deadstart. If no checkpoint

devices are deﬁned, the checkpoint ﬁle will reside on the ﬁrst default mass storage

device.

A level 0 deadstart clears the checkpoint status on all mass storage devices. To clear

the checkpoint status for a speciﬁc mass storage device, use the RESET entry.

The SCKP entry format is:

SCKP=esti,est2,...,estn.

Parameter Description

esti EST ordinal of the mass storage device to be deﬁned as a system

checkpoint ﬁle device; from 5 to 7778. Ranges of ordinals can be

speciﬁed.

3-86 NOS Version 2 Analysis Handbook Revision M

0ms.

0^s

FAMILY — Default Family Name Assignment Entry

The FAMILY entry deﬁnes the default family. The family that is to be deﬁned as the

default family may reside on more than one device. The EST ordinal of any device

within the family can be speciﬁed on the FAMILY entry, except in the following

situation. If the member of the family whose device mask will have bit 27 (200s in

mask) set is being initialized, the FAMILY entry must specify the ordinal of this

device. In all cases, the FAMILY entry must follow the EQ entry for the device

speciﬁed.

The following restrictions apply.

• The status parameter for a default family device's EQ entry cannot be set to OFF.

• A REMOVE entry cannot exist in the EQPDECK for a device being specified as a

default family device.

The FAMILY entry format is:

FAMlLY=est.

Parameter Description

est ' EST ordinal number of the mass storage device that the system

automatically uses to determine your family when you do not specify a

family name at login or job initiation; from 5 to 7778.

REMOVE — Removable Device Assignment Entry

If a mass storage device is to be considered removable, you must specify it as such at

deadstart with the REMOVE entry. This allows it to be introduced or removed during

system operation. A device speciﬁed as removable cannot also have associated with it

an ASR, SYSTEM, MSAL, FAMILY, DAYFILE, ACCOUNT, ERRLOG, or MAINLOG

entry (refer to Dayﬁle Descriptions earlier in this section for the last four entries).

The format is:

REMOVE=esti,est2....,estn.

Parameter Description

esti EST ordinal of mass storage device that is to be removable; from 5 to

7778. One or more ordinals may be speciﬁed with one REMOVE entry.

Ranges of ordinals can be speciﬁed.

Revision M Deadstart Decks 3-87

SHARE — Shared Device Entry

This entry identiﬁes the rotating mass storage devices to be shared through a link

device (ECS, ESM, or STORNET) by from two to eight mainframes in a

multimainframe complex. The tables necessary for the management of these devices

(MST, TRT, MRT, and DAT) are maintained on the link device. The presence of the

SHARE entry implies a multimainframe complex; shared status will be set for the link

device. If the link device is the only device to be shared, it must be speciﬁed in the

SHARE entry. If the SHARE entry is speciﬁed, the ISHARE entry must be omitted.

Refer to PRESET - Preset the Link Device Entry later in this section.

Following is a list of the equipment types that can be shared.

Equipment Type

ECS without DDP DE

ESM without LSP

ECS with DDP DP

or ESM with low-speed port (LSP)

or STORNET

844-21 * DI/DK

844-41/44 DJ/DL

885-11/12 DM/DQ

The format of SHARE is: >^

SHARE=esti,est2,...,estn.

Parameter Description

esti EST ordinal of the mass storage device being shared; from 5 to 777s

(EST ordinal 5 is reserved for the link device). Ranges of ordinals can

be speciﬁed.

Refer to section 13, Multimainframe Operations, for suggestions on shared device /*^v

conﬁgurations.

3-88 NOS Version 2 Analysis Handbook Revision M

ISHARE — Independent Shared Device Entry

This entry identiﬁes the rotating mass storage devices that are to be independently

shared by from 2 to 16 mainframes in a multimainframe complex. The tables necessary

for the management of these devices (MST, TRT, MRT, and DIT) are maintained on

the shared device. You cannot designate ECS or ESM in an ISHARE entry. When the

ISHARE entry is speciﬁed, the SHARE entry must be omitted. Refer to PRESET -

Preset the Independent Shared Device Entry later in this section.

Listed are the equipment types that can be independent shared devices.

Equipment Type

8 3 4 D D

8 3 6 D G

844-21 DI/DK

f^ 844-41/44 DJ/DL

885-11/12 DM/DQ

The ISHARE entry format is:

ISHARE=esti .est 2 estn

Parameter Description

esti EST ordinal of the mass storage device being shared; from 5 to 7778.

Ranges* of ordinals can be speciﬁed, as described under Specifying

Ranges of EST Ordinals earlier in this section.

Refer to section 13, Multimainframe Operations, for suggestions on shared device

conﬁgurations.

Revision M Deadstart Decks 3-89

LBC — Load Buffer Controllers Entry

This entry identiﬁes the type of controlware to be installed on the speciﬁed disk

channels. Depending on the speciﬁed parameters, this entry can identify the channels

as having half-track or full-track controlware but not install the controlware.

Unless you specify'the LBC entry, the system examines the mnemonics of the device

in the EQPDECK entry and causes the default version of controlware to be installed

as follows:

Device Type Controlware Version Number

DB16 MA722

DC17 MA464

DD18 MA462

DG18 MA462

DI MA710

DJ MA710

DK MA401

DL MA401

DM MA721

DQ MA721

Use the LBC entry to override these defaults. The LBC entry format is:

LBC,type, ci,C2,....cn.

Parameter Description

type Controlware to be installed; one of these values.

type Description ^\

CC Install 7165 controller with full-track (MA464) controlware.

CN Identify channel(s) as having full-track 7165 controller, but do

not install the controlware.

FM Install controller with full-track (MA721) controlware.

FT Install controller with full-track (MA401) controlware.

16. Not applicable for CYBER 180-class machines.

17. Applicable for CYBER 180-class machines except models 810, 815, 825, and 830.

18. Applicable for models 810, 815, 825, and 830 only. >«s%

3-90 NOS Version 2 Analysis Handbook Revision M

LBC — Load Buffer Controllers Entry

Parameter Description

type (Continued)

t y p e D e s c r i p t i o n

yms.

HT Install controller with half-track (MA710) controlware.

ID Install 7255-1 adapter with full-track (MA462) controlware.

M8 Install 7990 controller with MB466 controlware.

NF Identify channel(s) as having full-track controller, but do not

install the controlware.

NH Identify channel(s) as having half-track controller, but do not

install the controlware.

NI Identify channel(s) as having full-track 7255-1 adapter, but do

not install controlware.

NM Identify channel(s) as having full-track 7155-1 controller, but

do not install the controlware.

NN Identify channel(s) as having NADs, but do not install the

controlware.

NP Identify channel(s) as having 7155-401 controller, but do not

install the controlware.

NX Identify channel(s) as having 5870 printer, but do not install

the controlware.

N8 Identify channel(s) as having 7990 controller, but do not install

the controlware.

PH Install 7155-401 controller with MA722 controlware.

ci Disk channels; type determines if controlware is installed on these

channels.

Revision M Deadstart Decks 3-91

XM — Declare Extended Memory Space for Buffers or User Access

The controlware version number that can be loaded into the adapter or controller i***MS

t y p e s i s : f f

Adapter

Controller Controlware Version Number

7054 MA710

7 1 5 2 M A 7 1 0 , M A 4 0 1

7 1 5 4 M A 7 1 0 , M A 4 0 1

7155-1 MA721

7155-401 MA722

7165 MA464

7255-1 MA462

7990 MB466

XM — Declare Extended Memory Space for Buffers or User Access

The XM entry reserves space for I/O buffers and causes initialization of tables for user

access to extended memory.

NOTE

Except on models 865 and 875, the assignment of user access to extended memory ^^

forces jobs using user access to use CPU 0. This prevents CPUMTR from being

locked out during large block transfers to or from extended memory on dual-CPU

mainframes.

The XM entry is required if the system contains user-accessible extended memory or

buffered devices. If UEM is to be used only for user-accessible extended memory, you

do not need to specify an extended memory EST entry in the EQPDECK. To access

UEM, it must be enabled (refer to the USER EXTENDED MEMORY IPRDECK entry

later in this section).

3-92 NOS Version 2 Analysis Handbook Revision M

UEMIN — UEM Equipment Initialization

The XM entry format is:

XM=id,iob,uec,EM.

Parameter Description

id Identiﬁer of the mainframe that will access the reserved extended

memory space; id is required.

If an XM entry refers to shared extended memory (ECS or ESM as a

link device), the XM entry must be present in the EQPDECKs of all

linked mainframes. If an XM entry refers to nonshared extended

memory (UEM), the XM entry must be present only in the EQPDECK

for its own mainframe.

iob Number of words/10008 reserved for buffers for DB, DC, DF, DH, DN,

DV, or DW devices; from 40s to 37770s. This value is limited by'the '

amount of memory available; either physically present or logically

available (allocated in the CMRDECK and EQPDECK entries). The

optimum number of words to reserve is the amount left after the

alternate system library and user-accessible extended memory are

taken into account. If no buffered devices are deﬁned, enter 0.

uec Number of words/10008 of extended memory to reserve for user access;

maximum value is 777408. This value is limited by the amount of

memory available; either physically present or logically available

(allocated in the CMRDECK and EQPDECK entries).

EM For models 865 and 875 and CYBER 180-class machines,

user-accessible extended memory can be allocated in UEM without

allocating UEM as an equipment. If you want only user-accessible

extended memory in UEM, clear or do not specify an extended memory

EST entry in the EQPDECK and enter XM with the uec size and the

EM keyword.

For models 865 and 875, if you specify EM, user-accessible extended

memory is allocated in UEM regardless of the presence of ECS or

ESM; if you omit EM, user-accessible extended memory is allocated in

the device deﬁned in the extended memory EST entry.

UEMIN — UEM Equipment Initialization

The UEMIN entry enables full initialization for UEM equipment during a level 0

deadstart. It is equivalent to an INITIALIZE,AL,est entry. The UEMIN entry can be

included as part of the EQPDECK on the deadstart ﬁle or it can be entered from the

system console at deadstart time. If the UEMIN entry is being used, the EQ entry for

the DE equipment is required. If you enter UEMIN when it's already set, it disables

the automatic initialization of UEM. The UEMIN format is:

UEMIN.

H$PP5\

Revision M Deadstart Decks 3-93

ACCESS — Set Access Level Limits

The ACCESS entry allows you to specify equipment access level limits for mass

storage, magnetic tape, two-port multiplexer, stimulator multiplexer, and unit record

type equipment. This entry is invalid for other types of equipment. This entry

determines the upper and lower limits for the range of access levels of the data

allowed to be read from or written to the equipment. The default equipment access

level limits are zero; no secure data can be read from or written on the equipment.

Values for this entry should be supplied by a site security administrator.

If you want the equipment access level limits speciﬁed by this entry to become the

device access level limits for mass storage equipment, that equipment must be

initialized. Refer to the NOS Version 2 Security Administrator's Handbook for more

information about equipment access and device access level limits.

This entry is ignored if the system is running in unsecured mode.

The format of the ACCESS entry is:

ACCESS,1ower,upper,ord11st.

Parameter Description

.,rf^\

lower * Access level name speciﬁed in deck COMSMLS (refer to the NOS

Version 2 Installation Handbook) corresponding to the desired lower

limit.

upper Access level name speciﬁed in deck COMSMLS corresponding to the

desired upper limit.

ordlist One or more EST ordinals, separated by commas, or a range of

ordinals as described in Specifying Ranges of EST Ordinals earlier in

this section.

/ • ^

3-94 NOS Version 2 Analysis Handbook Revision M

THRESHOLD — Set Disk Thresholds

0ims.

THRESHOLD — Set Disk Thresholds

The THRESHOLD entry enables you to set threshold values for the disk storage

devices speciﬁed by EST ordinals. The system uses these threshold values as limits

when monitoring disk veriﬁcation failures, available disk space, and disk error

processing. If a threshold value is exceeded, the system performs a corrective action,

such as notifying the operator or restricting activity on the affected disk. The format of

the THRESHOLD entry is:

THRESHOLD,type=value,ordlist.

Parameter Description

type

value

ordlist

You can specify one of the following threshold types. Refer to the DSD

THRESHOLD command in section 5 for additional information.

type Description

Veriﬁcation failure threshold (default).

Restricted activity threshold.

Low space threshold.

Recovered error threshold.

Unrecovered error threshold.

The threshold value can range from 0 to 3777s. If the value parameter

is omitted, the following default values are used.

type value

1/8 of the number of tracks on the device.

1/16 of the number of tracks on the device.

50s

Enter one or more EST ordinals, separated by commas, or a range of

ordinals as described in Specifying Ranges of EST Ordinals earlier in

this section.

Revision M Deadstart Decks 3-95

EQPDECK Entries Made Only During Deadstart

A*ms

EQPDECK Entries Made Only During Deadstart

The following entries are valid only from the system console at deadstart time. They '

cannot be included as part of the EQPDECK on the deadstart ﬁle.

AUTOLOAD — Toggle Autoloading

The AUTOLOAD entry toggles the selection of buffer controller autoloading for all

7054/7154/7152/7155/7165 controllers and the 7255 adapter. The AUTOLOAD format is:

AUTOLOAD.

GRENADE — Clear Unit Reservations

The GRENADE entry causes unit reservations to be cleared on all 844 units physically

connected to each 7054/7154/7152/7155 controller and any 834 or 836 units connected to

a 7255 adapter. The GRENADE format is:

GRENADE. )

INITIALIZE — Initialization Entry

To use a mass storage device that is deﬁned with an EQ entry, it must have a label.

A label is written on a device when you initialize it by using either the INITIALIZE

command, during system operation, or the INITIALIZE entry in the EQPDECK, when

it is displayed at the system console at deadstart time.

A mass storage device's label is contained on a logical track (usually track 0). It *^^

contains information about the allocation and characteristics of a device (and its

units, if there is more than one unit on a device). This information is in the form of

a label sector for the ﬁrst unit, a TRT for the device, and a label sector for each

unit.

Initialization does not automatically occur at each deadstart because mass storage

device labels are recovered during all deadstarts. Therefore, initialize a device only in

the following situations.

• To add a new mass storage device (no label exists on the device) use the ^^

INITIALIZE entry.

• If parts of the label on a permanent file device have been destroyed by

maintenance operations (permanent ﬁles having been dumped to another device

before diagnostics were run), use the INITIALIZE entry during deadstart to write

a new label. Then reload the permanent ﬁles.

• If a device (usually a private auxiliary, public auxiliary, or alternate permanent

ﬁle family device) is added to a system during operation, use the DSD

INITIALIZE command to initialize it if it does not have a valid label on it when

it is added to the system.

• When an extended memory device is initially placed in maintenance mode, all

mainframes using extended memory must initialize it (the maintenance mode

parameter is described under EQ - Extended Memory EST Entry earlier in this

section). You must also enter the PRESET entry for multimainframe operation. ^a^

3-96 NOS Version 2 Analysis Handbook Revision M

0ms.

INITIALIZE — Initialization Entry

During a deadstart, the INITIALIZE entry has the following characteristics.

• During a level 0 deadstart, it can be entered at the system console only when the

EQPDECK is displayed. It can be entered anywhere after the EQ entry for the

device.

If it is placed in the deadstart ﬁle EQPDECK, the system issues the error

message INCORRECT ENTRY when the EQPDECK is read from the tape.

• A total initialization (op=AL) assumes that no valuable information exists on the

device and creates a new label. When the new label is created, all previously

existing information on the device, except CTI, CDA, HIVS, and MSL, is lost.

• If the EQ status for the device is OFF when INITIALIZE is entered, initialization

of the device occurs whenever the device is set to ON status by the operator with

the DSD ON command during normal system operation.

• If the device is not a master device, INITIALIZE (op=AL) only writes a label; if

it is a master device, then it also initializes the catalog track and writes EOIs at

the beginning of the permit track, the indirect access track (data chain), and each

catalog track.

• During a deadstart initialization (op=AL), all flaw reservations specified for a

device are lost and must be reentered, except for 844 type devices with

factory-formatted disk packs.

• During an initialize format pack (op=FP) for an 895 disk, NOS only formats

those cylinders it plans to use in large record format. NOS does not format the

disk cylinders occupied by CIP.

NOTE

If a disk deadstart is in progress, the deadstart disk cannot be initialized. An

attempt to initialize it will result in the system issuing the error message

INCORRECT ENTRY.

The format of INITIALIZE is:

INITIALIZE,op,est,,est2 estn.

Parameter Description

op Level of initialization; one of these values:

op Description

AF Initialize account dayﬁle.

AL Total initialization. For an 895 disk, the AL parameter is

equivalent to the FP parameter if an initial install of CIP was

done or if CIP was released.

DF Initialize system dayﬁle.

EF Initialize error log.

Revision M Deadstart Decks 3-97

INITIALIZE — Initialization Entry

Parameter Description

op (Continued) ^fsm%^

op Description

FP Initialize format pack (an automatic selection of AL also occurs).

This option applies to 844 and 895 disks only.

MF Initialize binary maintenance log.

PF Initialize permanent ﬁles.

QF Initialize queued ﬁles.

esti EST ordinal of mass storage device to be initialized; from 5 to 7778.

If the ordinal refers to a family permanent ﬁle device, then family

name, device number, and mask (if it is a master device) are speciﬁed

on the PF entry.

If it is an auxiliary device, the pack name is speciﬁed on the PF entry.

Total initialization (op=AL or FP) is the only initialization that is independent of the

content of the pack, if the initialization occurs during deadstart. If the initialization is

done while the system is running, it is applied to the device after the check mass

storage (CMS) routine has recovered it. If CMS cannot recover the device, the

initialization is similar to a deadstart initialization (that is, all information on the

device is lost).

The device number, family name, and device masks can only be changed during a

total initialization. Since all devices may contain permanent ﬁles, you should include

a PF entry for a device when performing a total initialization. If you do not, the

device is assigned a default family name, device number, and device masks. It is

possible that these parameters may conﬂict with other devices in the system. If a

conﬂict occurs, resolve it by using PF entries.

If you initialize a nonremovable device without a PF entry, the device mask and

secondary mask default to 377s for the smallest EST ordinal larger than 5. For all

other equipment, including equipment 5, the default masks are set to 0 (zero). The '*4S5%,

default family name is SYSTid (where id is the machine identiﬁer). The default '

device numbers begin at 1 and increase by 1 (starting with EST ordinal 5) for each

device that you initialize without a PF entry.

The INITIALIZE entry operates in conjunction with the dayﬁle entries DAYFILE,

ACCOUNT, ERRLOG, and MAINLOG (refer to Dayﬁle Descriptions earlier in this

section) to determine where the dayﬁles actually reside. The following examples

illustrate the various cases. Assume that the system has three mass storage devices

(EST ordinals 6, 7, and 10).

3-98 NOS Version 2 Analysis Handbook Revision M

INITIALIZE — Initialization Entry

Example 1:

For this example, no dayﬁle entries are made and no previous dayﬁles exist.

The following EQPDECK entry is made.

INITIALIZE,AL,6,7,10.

All dayﬁles reside on ordinal 6.

Example 2:

In this example dayﬁle entries are made, but no previous dayﬁles exist.

The following EQPDECK entries are made.

DAYFILE=6,200.

ACCOUNT=7,200.

ERRLOG=10.

MAINLOG=10,200.

INITIALIZE,AL,6,7,10.

In this case, the dayﬁles reside on the indicated devices (system dayﬁle on ordinal 6,

account' dayﬁle on ordinal 7, error log and binary maintenance log on ordinal 10).

The default buffer length is used for the error log buffer.

Example 3:

In this example, dayﬁle entries are made and previous dayﬁles do exist.

Assume that the EQPDECK entries in example 2 are used.

Since a total initialization has been done on each device, no dayﬁles are recovered.

They reside on the indicated devices.

Example 4:

In this example, dayﬁle entries are made, previous dayﬁles exist, but no dayﬁle

initialization entries are made.

The following EQPDECK entries are made.

DAYFILE=6.

ACCOUNT=7.

ERRLOG=10.

INITIALIZE,PF.6.

The dayﬁles may already reside on the speciﬁed devices, or they may reside on some

combination of the possible devices. In either case, since no dayﬁle initialization

entries are made, the old dayﬁles are recovered. The residence of these dayﬁles is

governed by the residence of the old dayﬁles. The PF initialization entry returns all

permanent ﬁle space and relabels the device based on the recovered device

parameters. The dayﬁles and queued ﬁles on this device are not affected by this

entry.

Revision M Deadstart Decks 3-99

INITIALIZE — Initialization Entry

Example 5:

In this example, dayﬁle entries are made, previous dayﬁles exist, no dayﬁle

initialization entries are made, and duplicate dayﬁles are in existence.

Assume that the EQPDECK entries in example 4 are used.

For the dayﬁles that do not have duplicates, the residence is deﬁned by the current

residence of the ﬁles, not the EQPDECK entries. But assume that an error log is

recovered from ordinals 6 and 10. In this case, the most recent ﬁle becomes the

active error log. Its previous residence overrides the EQPDECK entry. The other ﬁle

becomes an inactive error log (an entry exists in the mass storage table of the device

pointing to the inactive ﬁle, but the ﬁle is not in use by the system).

To produce an inactive error log, the site must run in the following manner.

1. Assume an 844 disk subsystem with two or more spindles is being used. Run with

unit 1 equated to EQ6 and unit 0 unused.

2. Redeadstart, equate unit 0 to EQ6, and do not use unit 1. 1

3. Redeadstart, equate unit 0 to EQ6, and unit 1 to EQ7.

Since unit 0 has the most recent copy of the error log, this copy would become an

active error log and the copy on unit 1 would become an inactive error log.

Example 6:

In this example, dayﬁle entries are made, the previous dayﬁles from example 2 exist, ^^

and initialization entries are made. )

The following EQPDECK entries are made.

DAYFILE=7.

ACCOUNT=7.

ERRLOG*10,300.

MAINLOG=10.

INITIALIZER, 6.

INITIALIZE,QF,6.

In this case, the account dayﬁle is recovered and continued on ordinal 7. The binary

maintenance log is recovered and continued on ordinal 10 with a CM buffer length of

100. The error log is recovered and continued on ordinal 10 with a CM buffer of 300s

words. The system dayﬁle space on ordinal 6 (from example 2) is released and the

new system dayﬁle starts on ordinal 7. The QF initialization entry releases all space

reserved by queued ﬁles on ordinal 6.

The CM buffer length is not affected by dayﬁle recovery. It is always speciﬁed by the

values deﬁned in the EQPDECK entries. If no buffer length entries exist, the system

default values are used.

r"5*%

3-100 NOS Version 2 Analysis Handbook Revision M

PRESET — Preset the Link Device Entry

This entry deﬁnes allocation space and initializes the tables (MST, TRT, MRT, and

DAT) on the link device that are required for management of shared multimainframe

mass storage devices. The entry is valid only for level 0 deadstarts by the ﬁrst

mainframe in the multimainframe complex to deadstart.

Once PRESET is issued, the SHARE entry is disabled. Therefore, all SHARE entries

must precede the PRESET entry. The PRESET entry has two formats:

PRESET.

PRESET,n.

Parameter Description

n Number of shared devices; from 1 to 77s.

If you specify n, space is allocated for the speciﬁed number of shared devices. Use this

entry when the total number of shared devices is greater than the number of shared

devices deﬁned in the EQPDECK of the ﬁrst mainframe in the multimainframe

complex to do a level 0 deadstart.

If you do not specify n, the link device is preset, and the amount of table space

reserved for the shared devices is determined by the number of shared entries in the

EQPDECK.

Revision M Deadstart Decks 3-101

PRESET — Preset the Independent Shared Device Entry

This entry presets the independent shared devices in a multimainframe complex. The )

MST, TRT, MRT, and DIT are maintained on the mass storage device itself and are

not affected by the PRESET entry. This entry is used in conjunction with the ISHARE

entry. It is valid only on a level 0 deadstart by the ﬁrst mainframe in the

multimainframe complex to deadstart. All ISHARE entries must precede the PRESET

entry.

The format is:

PRESET=est1,est2,...,estn.

Parameter Description

esti EST ordinal of the ISHARE device; from 5 to 777s. Ranges of ordinals

can be speciﬁed, as described under Specifying Ranges of EST Ordinals

earlier in this section.

Refer to section 13, Multimainframe Operations, for suggestions on shared device ^]

conﬁgurations.

WARNING

If a PRESET is entered for a device that is already in use by a second machine in a

multimainframe complex, system failures may occur on both machines, and data on

the device may be destroyed.

If a PRESET is entered for a device while a second machine in a multimainframe

complex is down, and if that second machine later attempts to perform a level 3 /a(S%

deadstart and use that device, system failures may occur on both machines, and data

on the device may be destroyed.

RESET — Reset Device Attributes

The RESET entry rescinds all device-related attributes resulting from entries such as

REMOVE, SYSTEM, MSAL, and so on. It restores the values speciﬁed with the last

EQest entry encountered. The format is: ^^

RESET=est 1,est z....,estn •

Ranges of ordinals are not allowed; each ordinal must be entered individually.

z^SSv

3-102 NOS Version 2 Analysis Handbook Revision M

APRDECK

The auxiliary mass storage parameter deck (APRDECK) is a text record on the

deadstart ﬁle that is processed during system initialization. APRDECK entries identify

areas of mass storage that are unusable (ﬂawed areas) and prevent the system from

accessing them. The system uses the information in the APRDECK entries to build the

TRT for each device that resides in CMR and also in the mass storage device label.

You can place up to 64 APRDECKs on the deadstart ﬁle. Placing several APRDECKs

on the same deadstart ﬁle allows you to use the same ﬁle to deadstart several

conﬁgurations.

APRDECK Format

The ﬁrst line in an APRDECK is the deck name. The format of the APRDECK name

is:

APRDnn

Parameter Description

nn Number identifying the APRDECK; from 00 to 77s.

An APRDECK must have a name and may have ﬂaw entries. The ﬁrst APRDECK

must contain the deck name APRD00 and nothing else. Subsequent APRDECKs must

be numbered consecutively and can contain ﬂaw entries.

^^ The released version of the APRDECK contains no entries. You can enter ﬂaws at

{ three different times:

• During deadstart, after entering all EQPDECK modifications.

• During system operation, using the FLAW entry (refer to section 8, K-Display

Utilities).

• During the configuration of a deadstart file.

If during deadstart you initialize a device and then enter NEXT, the system displays

f^ both the parameters on the device's EST entry and the APRDECK referenced by the

EST entry. You can then change the ﬂaws for the device. If the ﬁrst APRDECK is

referenced by the EST entry, the system displays the parameters on the device's EST

entry and the APRDECK name, APRD00. You can then enter ﬂaws for the device.

These changes to the APRDECKs remain in effect until the next deadstart.

For example, in a EQPDECK, the EST entry for an 844-21 disk is:

EQ07=DI,ST=0N,EQ=0,CH=31/33,AP=5,UN=2.

After you initialize equipment 07 and enter NEXT, the following display appears.

EQ TYPE ST EQ UNITS CHANNELS

0 7 D I - 1 O N 0 0 2 3 1 3 3

APRD05

SLF=4173.

SLF=7062.

Revision M Deadstart Decks 3-103

CAF — Clear All Flaw Reservations

The APRDECK entries described in this section are those to be used for entering /4esm.

ﬂaws during deadstart or during the conﬁguration of the deadstart ﬁle. Use the )

entries as follows.

• Use the CAF entry to clear all flaw reservations on a device.

• Use the SPF entry to specify the physical address of a flaw in extended memory.

(If a reservation for that physical address already exists, it remains in effect.)

• Use the SPF entry to specify the cylinder, track, and sector of a flaw in a disk.

(If a reservation for that physical area already exists, it remains in effect.)

• Use the CPF entry to cancel a particular SPF entry.

• Use the SLF entry to specify the logical address of a flaw. (If a reservation for

that logical address already exists, it remains in effect.)

• Use the CLF entry to cancel an SLF entry.

NOTE

All numeric values entered in the APRDECK must be entered in octal.

Either obtain ﬂaw addresses from a customer engineer, or run the MST (mass storage

test) on the device to determine the bad areas. MST speciﬁes the physical address of

ﬂaws.

The system reads the ﬂaw information recorded on the utility ﬂaw map of an

881/883/885 disk pack during the initialization of 844/885 equipment and reserves the

appropriate areas. For multiunit devices, the ﬂaw reservation is the union of all

utility ﬂaw maps. This automatic ﬂawing process occurs in addition to any

APRDECK entries. However, you cannot clear areas recorded as ﬂawed on the utility

ﬂaw map of an 881/883/885 disk pack with the CAF entry. Refer to appendix G for

information on clearing these ﬂaws.

You can list all APRDECKs on the deadstart ﬁle by accessing the system ﬁle

SYSTEM with an ASSIGN or COMMON command, then using the T parameter on

the CATALOG command. Refer to the NOS Version 2 Reference Set, Volume 3 for

more information.

CAF — Clear All Flaw Reservations

The CAF entry clears all ﬂaw reservations previously made with SLF or SPF entries.

The format is:

CAF.

3-104 NOS Version 2 Analysis Handbook Revision M

0m*s

SLF — Set Logical Flaws on Any Mass Storage Device

Use this entry to specify the logical address of a ﬂaw. If the track was previously

reserved, that reservation remains in effect. The format is:

SLF=track.

Parameter Description

track Logical track number. Refer to table 3-6 for valid ranges of track

numbers.

CLF — Clear Logical Flaws on Any Mass Storage Device

To cancel a reservation made with an SLF entry, enter the identical track number

using the CLF entry. The format is:

CLF=track.

Parameter Description

track Logical track number. Refer to table 3-6 for valid ranges of track

numbers.

SPF — Set Physical Extended Memory Track or Disk Area Flaws

Two formats exist for the SPF entry. One format prevents the system from using

blocks (tracks) of extended memory. The other format prevents the system from using

sectors on disks.

Use the following SPF format to prevent the system from using blocks (tracks) of

extended memory.

SPF=Aaddress.

SPF=Aaddressi-Aaddress2.

Parameter Description

Aaddress The 1- to 7-digit octal logical address in a track of extended memory;

track containing the absolute address is reserved. The letter A must

precede the address.

Aaddressi Lowest (Aaddressi) and highest (Aaddress2) addresses in a range of

Aaddress2 logical addresses in one or more tracks of extended memory. All tracks

in the range are reserved. The letter A must precede the addresses;

the hyphen is required.

Use the following SPF format to prevent the system from using sectors on disks.

SPF=Ccylinder,Ttrack,Ssector.

Revision M Deadstart Decks 3-105

CPF — Clear Physical Extended Memory Track or Disk Area Flaws

Refer to table 3-6 for the number of cylinders, tracks, and sectors for each device.

Parameter Description

Ccylinder Cylinder number; the letter C must precede the number.

Ttrack Track number; the letter T must precede the number.

Ssector Sector number; the letter S must precede the number.

CPF — Clear Physical Extended Memory Track or Disk Area Flaws

To cancel a reservation made with an SPF entry, enter the identical information with

a CPF entry. The format is the same as for the SPF entries.

Table 3-6. Information for Setting or Clearing Flaw Areas

Logical

Device Track Cylinders/ Tracks/ Sectors/

Device Type Range1 Device1 Cylinder1 Track1

819, single density DV 4000-5465 633 12 24

819, double density DW 4000-7153 1466 12 24

834 DD 4000-7135 1457 12 40

836 DG 4000-6565 1273 30 57

844-21, half track DI 4000-7137 630 22 30

844-21, full track DK 4000-7137 630 23 30

844-41/44, half track DJ 4000-7147 1464 23 30

844-41/44, full track DL 4000-7147 1464 23 30

885-11/12, half track DM 4000-7221 1511 50 40

885-11/12, full track DQ 4000-7221 1511 50 40

885-42, full track DB 4000-7221 1511 12 40

887, 4K sector DF 4000-7343 1562 46

887, 16K sector DH 4000-7343 1562 13

895 DC 4000-7351 1565 17

9853, 2K sector DN 4000-7726 2601 23 25

Extended memory DE/DP 4000-7620

1. Numbers are in octal.

3-106 NOS Version 2 Analysis Handbook Revision M

0^s

/0m*..

IPRDECK

The IPRDECK contains the system installation parameters that determine the system's

operation mode. From 1 to 64 IPRDECKs can exist on a deadstart ﬁle. The IPD entry

in the CMRDECK speciﬁes which IPRDECK to use. If you omit the IPD entry, the

system uses the ﬁrst IPRDECK on the deadstart ﬁle. IPRDECKs are named IPRDnn,

where nn is from 00 to 77s.

There are two IPRDECK console displays. The initial display, IPRINST, is an

instruction display. It gives a brief description of all valid IPRDECK entries. The

second display is the current IPRDECK. If either display overﬂows two screens, you

can page the display.

You can modify the IPRDECK by entering the appropriate changes or additions from

the console keyboard. Make these entries while either the IPRINST or IPRDECK is

displayed. Each console entry supersedes the value currently speciﬁed in the

IPRDECK.

NOTE

Changes made to the IPRDECK at deadstart time will be recovered across a level 3

deadstart; but not across a level 0, level 1, or level 2 deadstart. To make these

changes permanent, you must create a new deadstart ﬁle with these changes

incorporated into the IPRDECK.

You can list all IPRDECKs on your system by accessing the system ﬁle SYSTEM

with a COMMON command, then using the T parameter on the CATALOG command.

Refer to the NOS Version 2 Reference Set, Volume 3 for more information concerning

these commands.

Most of the IPRDECK entries are also valid DSD commands that can be used to

make changes during system operation. Changes to the IPRDECK using DSD

commands are retained after a level 3 deadstart, but not after a level 0, level 1, or

level 2 deadstart.

Revision M Deadstart Decks 3-107

Job Control Information

The QUEUE, SERVICE, and DELAY entries in the IPRDECK relate to job control. '

General information concerning job control follows.

Job Scheduling

Job scheduling is the control of jobs in the input (IN), executing (EX), and output (OT)

queues for each service class. Scheduling in the input and output queues is based on

the priority of a queue entry relative to all queue entries in the system. The priority

of a queue entry depends both upon how long the entry has been waiting in the queue

and upon the parameters speciﬁed on the QUEUE entry in the IPRDECK. The

following formula shows how the system computes the priority; all values are octal.

(ct-et)

p » + l p

Variable Description 1

p Priority; LP ^ p ^ UP. LP (a parameter on the QUEUE entry) is the

lowest priority and UP is the highest priority.

wf Weighting factor; WF parameter on the QUEUE entry.

ct Current time in seconds.

et Time in seconds at which the job entered the queue.

lp Lowest priority; LP parameter on the QUEUE entry.

When an input or output queue entry is created, its priority is the lowest priority (LP)

for its service class. The queue priority of the queue entry increases as time passes.

The rate at which the priority increases depends upon the weighting factor (WF). The

larger the weighting factor, the slower the priority increases. (The queue priority of an

entry with a WF of 108 increases eight times slower than an entry with a WF of 1.)

The queue priority increases either until the queue entry is selected for processing or

until the queue priority reaches the highest priority (UP). If the queue priority of an

entry reaches UP, it remains at UP until the entry is selected for processing. If the •y(!S%

queue priority is zero, the job or ﬁle is never selected by the job scheduler and stays '

in the queue until the operator either enters a DROP command or resets the priority

to a nonzero number.

Job scheduling for executing jobs determines how much execution time a job gets.

The amount of execution time depends both on the job's scheduling priority (which is

parameters speciﬁed on the QUEUE and SERVICE entries in the IPRDECK) and on

the scheduling priorities of other jobs that may be waiting for execution.

3-108 NOS Version 2 Analysis Handbook Revision M

0$ms.

Job Scheduling

When a job in the input queue is selected for execution, it gets an initial priority (IP;

a parameter on the QUEUE entry) for the execution queue. Scheduling priority

increases for a job in the execution queue in the same way as in the input and

output queues. After the job is scheduled to a control point, its scheduling priority

does not age; it is set to ﬁxed values that are changed only when speciﬁc events

occur. Initially, the job's scheduling priority is set to the upper priority bound for its

service class (the value of the UP parameter on the QUEUE entry). The job's

scheduling priority remains at this ﬁxed value until the job exceeds either its control

point time slice or its central memory time slice.

If the job exceeds its control point time slice (the CT parameter on the SERVICE

entry), its scheduling priority is lowered to the control point slice priority (the CP

parameter on the SERVICE entry). If the job exceeds its central memory time slice

(the CM parameter on the SERVICE entry) for the ﬁrst time, its scheduling priority

is lowered to the initial lower priority (the IL parameter on the QUEUE entry); if

the job exceeds its central memory time slice for a second or subsequent time, its

scheduling priority is lowered to the lower priority bound (the LP parameter on the

QUEUE entry).

Exceeding the CP or CM slice does not, by itself, force a job to roll out. A job

remains at a control point until the job scheduler determines that it needs this job's

control 'point and/or memory for a job with a higher priority. If a job rolls out, either

to mass storage or to a pseudo-control point, its scheduling priority starts at the

current value and begins to age upward (in the same fashion as input and output

queue ﬁles). When the job scheduler again selects this job to run at a control point,

its scheduling priority is again set to UP.

For interactive jobs, there is an additional execution queue priority, TP (a parameter

on the SERVICE entry). It is assigned to the execution queue entry of a job

restarting after terminal I/O. The value of TP aids response time to program

prompts. Also, for interactive jobs, the initial priority (IP) has an added signiﬁcance.

In addition to being the priority at which jobs are scheduled from the input queue to

the execution queue, IP is the priority assigned when a terminal command is entered.

Using separate TP and IP parameters allows the system to give faster responses to

users interacting with a job. To achieve this, a value for the TP parameter slightly

greater than the value for the IP parameter is recommended (refer to tables 3-7 and

3-8 and ﬁgure 3-6).

The relative values of the QUEUE and SERVICE parameters, both among service

classes and within a service class, affect system performance. For an example of

ranges of service class priorities, refer to ﬁgure 3-6. For an example set of speciﬁc

entries for the QUEUE and SERVICE parameters, refer to tables 3-7 and 3-8.

Revision M Deadstart Decks 3-109

Job Scheduling

Control Points

A job leaves a central memory control point when any of the following conditions are

present:

• A job completes, aborts, or is suspended.

• A system request causes a job to be rolled out to wait for a resource to become

available or for a speciﬁed time period to elapse.

Such system requests include a job request for a tape or disk pack, a ROLLOUT

command, and execution of the ROLLOUT macro.

• Terminal input/output is required.

A job leaves central memory when the following conditions occur:

- The system requests terminal input and typeahead input is not available.

- You request terminal output and the recall parameter is speciﬁed on the

request.

- You issue a RECALL macro after a request for terminal output that omitted

the recall parameter.

• The control point is made available for a higher priority job.

This ensures reasonable service to all users in the system. The operating system

controls the amount of time each type of job can be at a control point. This

ensures that one job does not monopolize system resources.

When a job is rolled out, the priority increases as time passes, giving that job a better

chance to be selected for execution again. When the job is selected and rolled in, its

priority is changed to the value of the UP parameter on the QUEUE entry, and the

job scheduling priority cycle begins again. This description on job rollout applies to

local batch, remote batch, and interactive jobs that are not doing interactive I/O. For

interactive jobs that do terminal I/O within a time slice, scheduling priority is slightly

different. When I/O is complete and input, for example, is available, the system assigns

the rolled-out job the terminal I/O scheduling priority (TP parameter on the SERVICE

entry). The TP parameter can be used to give the job a priority equal to the priority of ^^

jobs still within their initial time slice, an advantage over jobs in a second time slice, /^%

and a larger advantage over jobs in a third or higher time slice.

If a job at a control point exceeds the central memory time slice and it is not a

subsystem, the scheduling priority is set to the initial lower or lower bound priority

(the IL or LP parameter on the QUEUE entry) for its service class. Thus, any job in

the queue with a higher priority forces the executing job with the lower priority to

be rolled out. The rolled-out job ages normally until its priority is higher than the

priorities of either the jobs in the input queue or a job that is executing; then it is

again scheduled to a control point.

Once a job is scheduled, it is desirable to use the resources allocated before another

job forces it out. If a job maintained its scheduling priority when it was assigned to a

control point, another job could age past that job and force it to be rolled out before

it had an opportunity to use its time slice. For this reason, when a job is assigned to

a control point and its priority is within the queue aging range, it is given a priority

equal to the highest priority (the UP parameter in the QUEUE entry) for its service

class.

3-110 NOS Version 2 Analysis Handbook Revision M

/S*!\

j^^Nt

0^S

/im^s

0^S

Job Scheduling

Pseudo-control Points

If pseudo-control points are deﬁned, a job that is forced out of a control point may

remain in central memory at a pseudo-control point, rather than being rolled out to a

mass storage device.

• If the job is preempted by a higher priority job, the scheduling priority stays the

same.

• If the job's control-point time slice has expired, the scheduling priority drops to

the control-point slice priority.

Jobs with the following characteristics are excluded from using pseudo-control points:

• Connection to a system control point.

• Accumulator overflow flags set.

• Assigned to the first control point or the last control point.

• Error flag set.

Number of Control Points and Pseudo-control Points Available

Selecting the number of control points available on the system depends on the amount

of memory space available, the job mix, and the mode in which the system is being

run. Up to 348 control points can be deﬁned. Each control point needs 2008 words of

CMR space. For example, if an installation is running only TAF, then four or ﬁve

control points may sufﬁce. On the other hand, if the system is running a large number

of interactive terminals with heavy permanent ﬁle activity, 20 or more control points

may be needed. You may need to study memory and control point use in order to

correctly determine the setting of this option. The DSD W,R display and/or the ACPD

and PROBE utilities may be useful in making this determination.

• If memory use is high and control point use is low, select fewer control points.

• If control point use is high and memory use is low, select more control points.

Selecting the number of pseudo-control points depends on the same variables that are

used to determine the number of control points. Up to 34s pseudo-control points can

be deﬁned. The pseudo-control points provide more effective use of central memory in

environments where an insufﬁcient number of control points create a system

bottleneck. Like a control point, each pseudo-control point needs 200s words of CMR

space.

• If memory use is high and control point use is lower, do not select any

pseudo-control points.

• If control point use is high and memory use is low, select pseudo-control points.

Revision M Deadstart Decks 3-111

Memory Control

You can control the maximum memory allowed for job types and for service classes x /

with the parameters you specify on the SERVICE and DELAY entries in the

IPRDECK.

These parameters specify these lengths:

• Maximum field length divided by 1008 for a job in a service class.

Maximum ﬁeld length divided by 1008 for all jobs of the speciﬁed service class.

Maximum extended memory length in words divided by 10008 for a job in a

service class.

Maximum extended memory length in words divided by 1000s for all jobs of the

speciﬁed service class.

• Amount of central or extended memory to leave between job field length, if

possible.

Initially, the scheduler attempts to ﬁnd the highest priority job that meets the memory

constraints. However, if the scheduler is unable to schedule a job and has explicitly

rejected one or more jobs because the total ﬁeld length or the total extended memory

ﬁeld length for a service class would be exceeded, it attempts to schedule a job a

second time if ﬂexible partitions are enabled (refer to FLEXIBLE PARTITIONS in this

section). During this second attempt, any job that requires other jobs to be rolled out is

not scheduled; otherwise, the constraints (service class total ﬁeld length and total

extended memory ﬁeld length) are ignored, and the job is scheduled at the lower bound

priority, LP (a parameter on the QUEUE entry). This means that the constraints are

applied as long as there are enough jobs of each service class. However, if central

memory is unused and no other jobs are available, the scheduler attempts to schedule

the jobs without the constraints.

All of these parameters can be changed by using the SERVICE, QUEUE, and DELAY

entries.

Example of Job Control Parameters

An example set of entries for the job control parameters is shown in tables 3-7 and

3-8. These entries fall within the example of ranges of service class priorities shown in

ﬁgure 3-6. Neither the speciﬁc entries nor the ranges are recommended; they are

strictly examples to aid you in selecting QUEUE and SERVICE entry parameters.

The following discussion indicates the signiﬁcance of the values chosen and how they

relate to each other.

The entry (lowest) priority (LP parameter in the QUEUE entry) of the system service

class input queue is higher than all entry priorities, except the network supervision

and subsystem entry priorities, because it is assumed that an operator-initiated job

should receive prompt attention. A system job rolls out any batch job. The entry

priority (LP) of the network supervision service class input queue is set high to

ensure adequate response time from network programs and facilities such as CS, NS,

and NVF.

.•*^\

3-112 NOS Version 2 Analysis Handbook Revision M

Example of Job Control Parameters

The queue priorities for local batch, remote batch, and detached jobs are similar. The

time slice for detached jobs is shorter than for the local batch and remote batch jobs.

The assumption is that detached jobs need less CPU time than either local batch or

remote batch jobs.

The queue priorities are explained under Job Scheduling earlier in this section.

The time slices for the various service classes reﬂect the following objectives:

• To keep system jobs with their high entry priority from monopolizing system

resources.

• To keep at a minimum rollout activity caused by diagnostics running as

maintenance service class jobs.

• To allow most interactive jobs to compile, load, and begin execution in one time

slice.

• To give batch jobs a large time slice, because little is gained from rolling out

batch jobs. There is no problem with the time slices for batch jobs compared to

interactive jobs, because, with the priorities shown, an interactive job generally

causes a batch job to roll out.

• To ensure prompt service to all interactive users, without employing an excessive

number of rollouts, by setting the time slices for interactive jobs low. The time

slice parameters are critical to good interactive performance. In some cases,

depending on the system load, job size, and so forth, it may be desirable to

change these parameters during operation.

Revision M Deadstart Decks 3-113

Example of Job Control Parameters

Interactive performance is sensitive to the relative values of the QUEUE and SERVICE

scheduling parameters, both within a service class and among service classes. Tables

3-7 and 3-8 show almost no overlaps of values except that local batch, remote batch,

and detached jobs could age slightly past interactive jobs if they remained rolled out

for a very long time (about 26 minutes). For the interactive service class, the range

between entry (lowest) priorities and the highest priority is wide, so that few jobs are

at the highest priority simultaneously. If many jobs reach the highest priority, their

priorities are the same, and the order in which the jobs entered the queue is lost. The

job scheduler selects jobs with equal queue priority in a random manner. The terminal

I/O priority (TP) is set slightly higher than the initial priority (IP) to reduce response

time for the user interacting with a job as compared to the user initiating a new job

step. This parameter setting improves the perceived responsiveness of the system for

the interactive users.

The CPU priorities reﬂect the following objectives.

• The maintenance service class jobs are run at the lowest priority. This handles

the background CPU and memory diagnostics.

• All other jobs, except network supervision service class jobs, run at the same

priority. It is generally not desirable to run one class of jobs at a higher priority

than another because the system would roll in jobs that occupy memory without

executing, until they exceed the central memory time slice.

• The network supervision service class is set high to ensure adequate performance

from network programs and facilities such as CS, NS, and.NVF.

3-114 NOS Version 2 Analysis Handbook Revision M

Example of Job Control Parameters

Table 3-7. Example Set of Job Control Parameters, Part I

Service

Classs LP1 UP1 WF1 LP2 U P 2 W F 2 I P 2 IL2

4000

TP3

SY (system) 7770 7776 2000 7 0 0 0 1 7000

BC (local

batch)

10 4000 1000 4 0 0 4 1 2000 2000

RB (remote

batch)

10 4000 1000 4 0 0 4 1 2000 2000

CT 7770 7776 3000 7 0 0 0 1 7000 4000

(communica

tion task)

(interactive)

7000 "7770 3700 7 0 0 0 1 4004 3770 4024

(network

7770 7776 7770 7 7 7 6 1 7772 7772

supervision)

(detached)

10 4000 1000 4 0 0 0 1 2000 2000

(subsystem)

7770 7776 7770 7 7 7 6 1

•7772 7772

(mainte

nance)

10 1 0 1 10

(Installatio-

n)4

10 4000 1000 4 0 0 4 1 2000 2000

1. Input queue QUEUE parameter.

2. Execution queue QUEUE parameter.

3. Interactive job initial scheduling TP SERVICE parameter.

4. Installation classes 10, II, 12, and 13 all have release values the same as class BC.

5. All values are octal; DELAY parameters are JS=1, CR=30, AR=1750, and

MP = 400.

Revision M Deadstart Decks 3-115

Example of Job Control Parameters

Table 3-8. Example Set of Job Control Parameters, Part II

Service?

Class LP1 UP1 WF1 CT2 CM2 CP3 NJ4 TD4

| SY (system) 7000 7 7 7 6 1

| BC (local

| batch)

4 0 0 0 1

| RB (remote

| batch)

4 0 0 0 1

1 CT

| (communica-

| tion task)

7000 7 7 7 6 1

1 TS

| (interactive)

7 0 0 0 1

| NS (network

| supervision)

7 0 0 0 1

| DI (detached) 7 0 0 0 1

1 SS

| (subsystem)

7400 7 7 7 6 1

j M A

| (mainte-

7000 7 7 7 6 1

| nance)

j In

| (installation8)

7 0 0 0 1

1 0 • 2 0 6 7 7 0 7 7 7 7

20 200 3770 7777

10 200 6770 7777

10 200 7770 7777

1 0 2 0 7777

10 200 3770 7777

30/30

10 10 6770 7777 1139 30/30

74/74

20 20 3770 7777 34110 30/30

10 20 7770 7777 - 70/70

2/2

30/30

| 1. Output queue QUEUE parameter.

| 2. Time slice SERVICE parameters

I 3. Control point slice priority CP SERVICE parameter.

| 4. Number of jobs NJ SERVICE parameter.

j 5. Time-out delay TD SERVICE parameter.

| 6. CPU priority (lower and upper bounds) CB SERVICE parameter.

| 7. All values are octal; DELAY parameters are JS=1, CR=30, AR=1750, and

| MP=400.

I 8. Installation classes 10, II, 12, and 13 all have release values the same as class BC.

| 9. A TS suspended job times out after 10 minutes when the time-out delay is 113.

| 10. A DI suspended job times out after 30 minutes when the time-out delay is 341.

3-116 NOS Version 2 Analysis Handbook Revision M

Example of Job Control Parameters

10000-r

7400 --

7000--

6400--

6000

5400--

5000 --

4400--

4000 --

3400--

3000--

2400--

2000

1400--

1000--

400- •

-IP.IL, B-IP.IL,

C P C P

IP IP-—

-CP CP- i-CP

IU IL.

CP CP

■TP

•IP

IL B-CP

-I P.I L

IP.IL

Jk.

SY BC RB CT TS NS DI SS IMA SY BC RB CT TS NS DI SS MA SY BC RB CT TS NS DI SS MA

INPUT EXECUTION OUTPUT

Figure 3-6. Example of Ranges of Service Class Priorities

Revision M Deadstart Decks 3-117

IPRDECK Entries Made Only During Deadstart

You can enter the following IPRDECK entries only during deadstart. They cannot be

entered as DSD commands. Changes to the IPRDECK are not retained across deadstart

unless a new deadstart ﬁle is created to reﬂect those changes. The entries are

described in alphabetical order.

COMLIB

Format:

COMLIB,username,family,D.

Default:

None.

Signiﬁcance:

This entry speciﬁes a user name and family for a library of permanent ﬁles that

can be accessed by users on all other families. This feature eliminates the need to

maintain duplicate copies of the ﬁles on the families that require access to the

ﬁles-. This entry can also be used to reduce validation ﬁle accesses for user names

such as LIBRARY in a single family environment. The number of COMLIB

entries that can be speciﬁed is controlled by the CMRDECK entry CLT.

Parameter Description

username The 1- to 7-character user name to be associated with the ﬁles to

be accessed. This parameter should be unique to all families in the

system. It cannot be associated with more than one family.

family Family name to be associated with username.

D If D is speciﬁed, the associated username and family are deleted

from the common library table.

CPM

Format:

CPM, Si =11!, 52^2.

Default:

System selection.

Signiﬁcance:

This entry alters the central processor multiplier of type si, which is used in SRU

calculations. The si parameters are either 0 or 1 to indicate the multipliers SO or

Sl, respectively. Entering 0=n obtains a multiplier to be used for SO and entering

l = n obtains a multiplier to be used for Sl. (Refer to the NOS Version 2

Administration Handbook for a discussion of multiplier use.) The values of ni

range from 1 to 47s and are used as indexes to values deﬁned in COMSSRU in

order to determine the multiplier value. The default values are listed in table 3-9. ^35%

/<^^\

3-118 NOS Version 2 Analysis Handbook Revision M

CPM

Table 3-9. CPM Default Values for n\

COMSSRU Mainframe COMSSRU Default

ni (Octal) Name Model Multiplier Value

1CP62 6200 1.0

2CP64 6400 1.0

3CP65 6500 1.0

4CP66 6600 1.0

5CP67 6700 1.0

6 CP71 71 1.0

7CP72 72 1.0

10 CP73 73 1.0

11 CP74 74 1.0

12 C171 171 1.0

13 C172 172 1.0

14 C173 173 1.0

15 C174 174 1.0

16 C175 175 1.0

17 C176 176 1.0

20 C720 720 1.0

21 C730 730 1.0

(Continued)

0$ms

Revision M Deadstart Decks 3-119

CPM

Table 3-9. CPM Default Values for ni (Continued)

COMSSRU Mainframe COMSSRU Default

ni (Octal) Name Model Multiplier Value

22 C740 740 1.0

23 C750 750 1.0

24 C760 760 1.0

25 C810 810 1.0

26 C815 815 1.0

27 C825 825 1.0

30 C830 830 1.0

31 C835 835 1.0

32 C840 840 1.0

33 C845 845 1.0

34 C850 850 1.0

35 C855 855 1.0

36 C860 860/870 1.0

37 C865 865 1.0

40 C875 875 1.0

41 C960 960 1.0

42 C990 990/994/995 1.0

43 ICM1 Model on which 1.0

you are installing.

44 ICM2 Model on which 2.0

you are installing.

45 ICM3 Model on which 3.0

you are installing.

46 ICM4 Model on which 4.0

you are installing.

47 ICM5 Model on which 5.0

you are installing.

3-120 NOS Version 2 Analysis Handbook Revision M

CSM

0jms

0m\

CSM

Format:

CSM=csm.

Default:

Signiﬁcance:

This entry sets the operating system character set mode. To change the character

set mode for the products, a change must be made in IPARAMS, and the products

must be reassembled.

csm Description

63 63-character set.

64 64-character set.

The system assumes a 64-character set if there is no CSM entry in the current

IPRDECK.

NOTE

Unpredictable and possibly serious problems occur if the operating system is

operating in one character set and the products are operating in another.

Therefore, ensure that all installed products and the operating system are in the

same mode.

DISK VALIDATION

Format:

ENABLE,DISK VALIDATION.

DISABLE,DISK VALIDATION.

Default:

Enabled.

Signiﬁcance:

NOS performs hardware veriﬁcation on each mass storage device during the

deadstart process before users are allowed to access the device. The hardware

veriﬁcation sequence includes writing data to the disk, reading the data from the

disk, and ﬁnally comparing the read data with the write data to ensure integrity.

For 887 and 9853 disks, the hardware veriﬁcation includes running in-line

diagnostics. Since the diagnostics verify the quality of the disk media more

completely, it is advantageous to run them. These diagnostics take approximately

3 minutes to execute, during which time the 887 or 9853 disk is not available for

use. In order to keep deadstart time to a minimum (such as, during testing), disk

validation can be disabled using the DISABLE,DISK VALIDATION command.

Revision M Deadstart Decks 3-121

DSD

Format:

DSD, level ,commandi#command2#.. .#commandn

Default:

None.

Signiﬁcance:

This entry speciﬁes the initial command(s) to be executed by the DSD program

when the deadstart is complete. The commands can be the minimum number of

characters recognizable by DSD but must include terminating characters. Only the

last DSD entry on the IPRDECK for a given deadstart level is processed; other

DSD entries for the same level are ignored. The DSD entry cannot exceed one

line and a maximum of 60 characters can be speciﬁed.

Parameter Description )

level Level of deadstart (0, 1, or 2).

commandi DSD command to be executed for the level of deadstart speciﬁed.

Several commands can be speciﬁed by separating them with the #

(6-bit display code 60) or % (6-bit display code 63) character. These

characters may misposition parts of the console display of

IPRDECK if they appear as the upper 6 bits in a byte.

If you specify a series of commands including DSD display selection

commands, the display selection commands must be the last

commands. Failure to do so results in the system ignoring any

commands following the DSD display selection commands.

Examples:

DSD,0,MAI%X.QREC(PO=N)

DSD,0,X.QREC(PO=N)#SET,AEIQ.#QB.

3-122 NOS Version 2 Analysis Handbook Revision M

00?S

Format:

EI =nn 1, nri2,..., nnx.

Default:

None.

Signiﬁcance:

The EI entry should be used to ensure loading of the correct environment

interface for the operating system. This entry speciﬁes up to six 2-character

environment interface names, one of which must be loaded while deadstarting a

CYBER 180-class machine. The deadstart will not be allowed to continue if the

environment interface loaded by CTI is not one of those speciﬁed by the EI entry.

You will have to load the correct environment interface or modify the EI entry in

the IPRDECK. If an EI entry is not present, you will not be warned at deadstart.

EXTENDED STACK PURGING

Formats:

ENABLE,EXTENDED STACK PURGING.

DISABLE,EXTENDED STACK PURGING.

Default:

Disabled.

Signiﬁcance:

These entries specify the default action for instruction-stack purging for

nonsystem-origin jobs on CYBER 180-class machines. Refer to the MODE macro in

the NOS Version 2 Reference Set, Volume 4 for a description of instruction-stack

purging.

HARDWARE FAULT INJECTION

Formats:

ENABLE,HARDWARE FAULT INJECTION.

DISABLE,HARDWARE FAULT INJECTION.

Default:

Disabled.

Signiﬁcance:

The entries enable and disable the simulation of hardware faults by normal jobs

executing special instructions. This simulation is intended for test purposes only,

and should not be enabled on a production system.

Revision M Deadstart Decks 3-123

KEYPM

Format:

KEYPM»cc.

Default:

Signiﬁcance:

This entry speciﬁes the keypunch mode to be assumed during system operation.

cc Description

2 6 0 2 6 k e y p u n c h m o d e .

2 9 0 2 9 k e y p u n c h m o d e .

This entry is used for all batch jobs submitted if the keypunch mode is not

speciﬁed on the job command. This does not apply to RBF.

MEMORY CLEARING

Formats:

ENABLE,MEMORY CLEARING.

DISABLE,MEMORY CLEARING.

Default:

Disabled.

Signiﬁcance:

When memory clearing is enabled, central and extended memory are cleared when

released from a job (that is, when a job is rolled out, terminates, or reduces its

field length). When memory clearing is disabled, memory is cleared only when a a*z®s

job requests additional memory. ■

3-124 NOS Version 2 Analysis Handbook Revision M

/$85»s

yms.

MICRO

Format:

MICRO=namei, name2 name?.

Default:

None.

Signiﬁcance:

The MICRO entry should be used to ensure loading of the correct microcode for

the operating system. This entry speciﬁes up to seven 7-character microcode

names, one of which must be loaded while deadstarting a CYBER 180-class

machine. The deadstart will not be allowed to continue if the microcode loaded by

CTI is not one of those speciﬁed by the MICRO entry. You will have to load the

correct microcode or modify the MICRO entry in the IPRDECK. If a MICRO entry

is not present, you will not be warned at deadstart.

NAMIAF

Format:

NAMIAF=maxt.

Default:

2008

Signiﬁcance:

This entry speciﬁes the number of network terminals that can be connected to

IAF at one time. The maximum number of terminals IAF can support is 1039.

However, this value is dependent on the number of network terminals, multiplexer

ports, and stimulator ports speciﬁed in the deadstart decks.

Parameter Description

maxt Total number of network terminals; maxt can range from 1

to 14408.

PROBE

Formats:

ENABLE,PROBE.

DISABLE,PROBE.

Default:

Disabled.

Signiﬁcance:

These entries enable and disable the data gathering facility of CPUMTR.

Revision M Deadstart Decks 3-125

SCP

Formats:

ENABLE,SCP.

DISABLE,SCP.

Default:

Disabled.

Signiﬁcance:

These entries specify whether to use the system control point facility. You must

enable SCP if CDCS, IAF, MAP, MCS, MSE, NAM, NVE, PLA, RBF, RHF, SMF,

SSF, or TAF will be used. If none of these will be used, leave SCP disabled so

that more CMR space is available.

SCRSIM

Formats:

ENABLE,SCRSIM.

DISABLE,SCRSIM.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable the simulation of the status/control register using

the interlock register on CYBER 70 Computer Systems (refer to appendix D for

information on the SCRSIM simulator).

3-126 NOS Version 2 Analysis Handbook Revision M

J0$^\

SECCATS

Format:

SECCATS=cat i,cat2, -..,catn.

Default:

All categories are enabled.

Signiﬁcance:

This entry speciﬁes the security access categories that will be allowed in the

system for processing when the system is in a secured mode. The values should

be supplied by a site security administrator. Refer to the NOS Version 2 Security

Administrator's Handbook.

Parameter Description

cati Category names speciﬁed in deck COMSMLS corresponding to the

desired access categories. Initially, all categories are enabled. The

ﬁrst SECCATS entry clears all categories, then sets the speciﬁed

categories. Subsequent SECCATS entries set additional categories:

Enables all 32 categories.

Disables all 32 categories.

SECCATS=ALL.

SECCATS=NUL.

SPC

Format:

SPC,d=lines.

Defaults:

64 lines at 6 lines per inch.

85 lines at 8 lines per inch.

Signiﬁcance:

This entry speciﬁes the charge in number of lines for a page of printed output at

the speciﬁed print density on non-PFC printers.

Parameter Description

d Density in lines per inch; d can be 6 or 8 lines per inch, but must

be speciﬁed in octal (6 or 10s).

lines Lines per page; lines can range from 16 to 255, but must be

speciﬁed in octal (20s to 377s).

Revision M Deadstart Decks 3-127

SPD

Format:

SPD=d.

Default:

Signiﬁcance:

This entry speciﬁes the assumed density for printed output.

d Description

6 6 lines per inch.

108 8 lines per inch.

SPL

Format:

SPL=length.

Default:

Signiﬁcance:

This entry speciﬁes the assumed page length in number of lines for printed

output; length can range from 16 to 255, but must be speciﬁed in

octal (208 to 3778).

SPW

Format:

SPW=width.

Default:

136

Signiﬁcance:

This entry speciﬁes the assumed page width in number of characters for printed

output; width can range from 40 to 136, but must be speciﬁed in octal (508

to 2108).

■"^Sn

3-128 NOS Version 2 Analysis Handbook Revision M

SUBCP

Formats:

ENABLE,SUBCP.

DISABLE.SUBCP.

Default:

Disabled.

Signiﬁcance:

These entries specify whether CPUMTR is to be initialized to handle subcontrol

point (TAF) processing.

If SUBCP is disabled, CPUMTR is not initialized to handle subcontrol point

processing. If you are not running TAF and if no user applications use subcontrol

point processing, disable SUBCP so that CPUMTR uses less central memory.

TCVM

Format":

TCVM=mode

Default:

( ^ A S

Signiﬁcance:

This entry sets the tape conversion mode to be assumed during system operation.

mode Description

AS ASCII 9-track conversion.

US ANSI (previously known as USASI) 9-track conversion (same as AS).

EB EBCDIC 9-track conversion.

Revision M Deadstart Decks 3-129

TDEN

Format: /

TDEN=denslty.

Default:

HY for 7-track tapes.

PE for 9-track tapes.

Signiﬁcance:

This entry sets the system tape density. When the density is set, any tape unit

accessed is automatically set to this density unless speciﬁed otherwise by a

magnetic tape request. Two TDEN entries may be present, one for 7 track and

one for 9 track.

density Description

LO 200 cpi (7 track).

H I . 5 5 6 c p i ( 7 t r a c k ) .

H Y 8 0 0 c p i ( 7 t r a c k ) .

HD 800 cpi (9 track).

P E 1 6 0 0 c p i ( 9 t r a c k ) .

GE 6250 cpi (9 track).

TDTR

Format:

TDTR=tracktype.

Default:

Signiﬁcance:

This entry sets the default track type.

tracktype Description

MT 7 track.

NT 9 track.

3-130 NOS Version 2 Analysis Handbook Revision M

TMS

Formats:

ENABLE,TMS.

DISABLE,TMS.

Default:

Disabled.

Signiﬁcance:

These entries enable and disable the Tape Management System (TMS). These

entries are meaningless if the TMS binaries are not installed in the deadstart ﬁle.

TMSTO

V Format:

TMSTO,N=tO,S=tO.

Default:

N=TC,S=FC

Signiﬁcance:

( This entry speciﬁes the default values for the TMS tape options parameter for the

LABEL command and the LABEL macro. The N parameter speciﬁes the default

for nonsystem origin jobs and the S parameter speciﬁes the default for system

origin jobs.

to Description

TC Sets the default to TO=TC (TMS processing with catalog error checking).

TE Sets the default to TO=TE (TMS processing without catalog error

(^ checking).

FC Sets the default to TO=FC (non-TMS processing with catalog error

checking).

FE Sets the default to TO = FE (non-TMS processing without catalog error

checking).

This entry is meaningless if TMS is disabled.

Revision M Deadstart Decks 3-131

TRACE

Formats:

ENABLE,TRACE.

DISABLE,TRACE.

Default:

Disabled.

Signiﬁcance:

The entries enable or disable speciﬁcation of monitor functions and other system

data to be traced, the TRACE and TRAP capabilities. To use these capabilities,

ENABLE,SYSTEM DEBUG, must also be entered.

j^S^v

3-132 NOS Version 2 Analysis Handbook Revision M

IPRDECK Entries

You can enter the following IPRDECK entries during deadstart and online during

system operation using the L display or DSD commands. The reason for entering them

online is to change the system's operation between deadstarts. Entries made during

system operation are not retained across deadstarts. Entries made during deadstart are

not retained across deadstarts unless a new deadstart ﬁle is created to reﬂect the

changes.

A description of the enabling and disabling of subsystems follows. The remaining

IPRDECK entries are described in alphabetical order.

Subsystems

You can initiate a subsystem by either of the following methods:

• Enter an explicit call to the subsystem procedure using a DSD command. The

command format is:

su b fff f

Variable Description

sub 3-character mnemonic for the name of the subsystem and subffff is

the name of the subsystem procedure (refer to Subsystem Control

Commands in section 5).

• Enter the DSD command AUTO or MAINTENANCE. This will initiate calls to all

subsystems that are currently enabled. The calls will be to subsystem procedures

with the same name as the name of the subsystem. You can choose to have either

the AUTO command or the MAINTENANCE command issued automatically at

deadstart time by specifying the command on the DSD entry in the IPRDECK.

A subsystem procedure may be either a record in the deadstart ﬁle or a permanent ﬁle

under system user name SYSTEMX (user index 377777s). If a subsystem procedure

with the same name is present in both places, the permanent ﬁle copy will be used.

Note that a *PROC is required in the LIBDECK for any procedure in the deadstart

ﬁle.

NOTE

Before you initiate a subsystem that uses the system control point facility (such as

NAM or CDCS), you must enable the system control point facility with the

ENABLE,SCP IPRDECK entry. Refer to SCP earlier in this section.

All subsystems, except BIO and MAG, are disabled by default. You can enable or

disable a subsystem by using IPRDECK entries or by using the SUBSYST L-display

utility. (For information about the SUBSYST L display, refer to the NOS Version 2

Operations Handbook.)

The IPRDECK entries that enable and disable subsystems may also assign a default

control point for a subsystem. Thus, even if you do not want to enable a subsystem,

you may want to explicitly disable a subsystem with an IPRDECK entry to assign a

default control point for the subsystem.

Revision M Deadstart Decks 3-133

Subsystems

Use these IPRDECK entries to enable or disable a subsystem:

ENABLE,subsystem,cp.

DISABLE,subsystem,cp.

Parameter Description

subsystem Three characters that select the desired subsystem; one of these values:

subsystem Description

BIO

CDC

IAF

MAG

MAP

MCS

MSE

NAM

NVE

PLA

RBF

RDF

RHF

SMF

SSF

STM

TAF

Local batch I/O for central site line printers, card

readers, and card punches.

CYBER Database Control System.

Interactive Facility. Do not specify the cp parameter on

the ENABLE entry for IAF.

Magnetic Tape Subsystem. Enable MAG if removable

auxiliary packs are used. Disabling MAG frees a control

point for other use.

MAP III or MAP IV.

Message Control System.

Mass Storage Extended Subsystem.

Network Access Method.

NOS/VE dual state.

Plato-NAM Interface.

Remote Batch Facility.

Remote Diagnostic Facility. Do not specify the CP

parameter on the ENABLE entry for RDF.

Remote Host Facility.

Screen Management Facility.

NOS-SCOPE 2 Station Facility.

STIMULA. Do not specify the cp parameter on the

ENABLE entry for STM.

Transaction Facility.

r ^ ^ \

3-134 NOS Version 2 Analysis Handbook Revision M

Subsystems

Parameter Description

r^ subsystem (Continued)

NOTE

When MCS and RBF are started by the NAMI startup master ﬁle, the

ENABLE and DISABLE commands are ignored.

If an ENABLE is made for NVE and the subsystem procedure name is

also NVE, use the DSD IPRDECK entry to automatically down any

channels that NOS/VE uses.

CP Control point where the subsystem will reside. If you omit cp, the

current control point is used. If you enter 0 (zero) as the control point,

the subsystem may reside at any available control point. If you enter -n

as the control point, the subsystem's control point will be relative to the

system control point (-1 would be the last control point before the

system control point).

Omit cp for IAF, RDF, and STM.

0ms.

Revision M Deadstart Decks 3-135

^m?\

AUTORESTART

Formats:

ENABLE,AUTORESTART.

DISABLE,AUTORESTART.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable the AUTORESTART installation parameter. If

AUTORESTART is enabled and if step mode is set during an environmental

shutdown, the system automatically clears step mode when the power environment

returns to normal. All jobs waiting in the input and rollout queues are initiated

by performing the processing associated with the DSD AUTO command. This

recovery process requires no operator interaction.

CARTRIDGE PF STAGING

Formats:

ENABLE,CARTRIDGE PF STAGING.

DISABLE.CARTRIDGE PF STAGING.

Default: ■'

Disabled.

Signiﬁcance:

These entries specify whether permanent ﬁles that reside on MSE are staged to

disk. If disabled, jobs that attempt to access MSE-resident ﬁles are aborted.

3-136 NOS Version 2 Analysis Handbook Revision M

CLASS

jg^V

CLASS

Format:

CLASS,Ot ,SCi,SC2, ...,scn.

Default:

Origin Type Service Class

Batch BC

Remote batch RB

Interactive TS

Signiﬁcance:

Speciﬁes the valid service classes for each origin type. The system stores this

information in the service class table (refer to the NOS Version 2 Systems

Programmer's Instant).

Parameter Description

ot Origin type; must be BC (batch), RB (remote batch), or IA

(interactive). This parameter is required and order dependent.

sci Service classes. Each class selected causes validation for that

service class for the origin type speciﬁed by ot. Entering a service

class that already has validation clears validation for that service

class. You can select from 1 to 36 service classes for each origin

type. This command does not accept class SS (subsystem).

Service class is one of these values:

sc Description

BC Local batch.

CT Communication task.

DI Detached interactive.

In Installation class n (0 ^ n ^ 3).

MA Maintenance.

NS Network supervisor.

RB Remote batch.

SY System.

T S I n t e r a c t i v e .

ALL Causes validation of all service classes except subsystem

(SS) amd deadstart (DS).

NUL Clears validation for all service classes.

Revision M Deadstart Decks 3-137

CPTT

Format:

CPTT=n.

Default:

1008 sectors.

Signiﬁcance:

This entry speciﬁes the CPUPFM transfer threshold (CPTT). The value of CPTT

determines whether an indirect access ﬁle transfer will be processed entirely

within the PP program PFM or whether the CPU program CPUPFM will be

called to perform the transfer. Since CPUPFM can transfer large ﬁles faster than

PFM, CPUPFM transfers ﬁles that exceed the threshold value.

Depending on the conﬁguration, a site can adjust this value up or down. A site

with buffered I/O devices, for example, may ﬁnd it advantageous to set a lower "^

value. If a site wishes to disable CPUPFM transfers altogether, a value of zero

(CPTT=0) may be speciﬁed.

Parameter Description

n CPUPFM transfer threshold value that will be deﬁned in common

deck COMSPFM; 0 ^ n ^ 7777s PRUs.

DEBUG

Format:

DEBUG.

Default:

Disabled.

Signiﬁcance:

This entry selects or clears debug mode, depending upon the current status.

If enabled, debug mode is selected. The message DEBUG appears in the header of

the left screen display. Debug mode provides system origin privileges to validated

users and allows modiﬁcations to be made to the running system.

If disabled, debug mode is cleared. It is recommended that debug mode not be

allowed in a normal production environment.

On a secured system, this entry is ignored. While the console is in security

unlock status, debug mode can be set using DSD command DEBUG.

3-138 NOS Version 2 Analysis Handbook Revision M

DELAY

Format:

DELAY,ARar.CIci.CRcr.JQjq.JSjs.MPmp.

Default:

None.

Signiﬁcance:

This entry speciﬁes the system delay parameters. Refer to table 3-7 for an

example set of parameter entries. Refer to ﬁgure 3-6 for an example of ranges of

service class priorities.

Parameter Description

ARar

CIci

CRcr

JQjq

JSjs

MPmp

PP recall interval in milliseconds. This parameter speciﬁes the time

interval after which a peripheral processor unit (PP) in the PP

recall queue will be recalled, ar ranges from 1 to 77778.

CPU job priority interval in milliseconds. This parameter speciﬁes

the time interval after which the priorities of jobs in the wait

queue are incremented, ci ranges from 0 to 7777s.

CPU recall period in milliseconds. This parameter speciﬁes the

amount of time a job remains in recall (X status) when an RCL

request is placed in RA + 1. cr ranges from 1 to 77778.

Exponent used to determine the input ﬁle scheduling interval in

seconds, jq ranges from 0 to 14s. The interval in seconds between

scheduling of input ﬁles is calculated as follows:

interval = 2**jq

Job scheduler interval in seconds. This parameter speciﬁes the

interval in which the job scheduler is called. The scheduler may

also be called at other times, js ranges from 1 to 7777s.

Memory padding value expressed as 100s word blocks, mp ranges

from 0 to 7778. This parameter speciﬁes how much additional

(unassigned) memory should be allocated between the end of the

newly assigned job ﬁeld length and the beginning of the next job.

Increasing this value reduces the probability that a job will be

storage moved in response to a request for more memory.

Revision M Deadstart Decks 3-139

DDP ROLLOUT PATH

Formats:

ENABLE,DDP ROLLOUT PATH

DISABLE,DDP ROLLOUT PATH

Default:

Disabled.

Signiﬁcance

These entries determine which path a rollout to extended memory will take (CPU

or PP). The required supporting hardware (EM or DDP) must be present.

DFPT

Format:

DFPT=dtn.

Default:

DI1 (see the DFPT installation parameter in common deck COMSPFM).

Signiﬁcance:

The DFPT entry resets the system default removable pack type. When accessing a j

removable auxiliary device with a permanent ﬁle command, the system checks

that the equipment type and pack name of the device match the equipment type

(R parameter) and pack name (PN parameter) on the command. If R is not

speciﬁed, the system uses the equipment type speciﬁed by DFPT.

Parameter Description

dt Disk device type of the removable pack.

n Number of spindles in the removable pack; 1 ^ n ^ 8. '"^

3-140 NOS Version 2 Analysis Handbook Revision M

ENGR

Formats:

ENABLE,ENGR.

DISABLE,ENGR.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable engineering mode. If enabled, the ENGR message

appears in the header of the left screen display. Engineering mode allows the

peripheral processing unit (PPU)/hardware diagnostics and the 881/883 pack

reformatting utility FORMAT to run while the system is in operation.

On a secured system, these entries are ignored. When the console is in security

unlock status, engineering mode can be enabled using the DSD command

ENABLE,ENGR.

FLEXIBLE PARTITIONS

Formats:

ENABLE,FLEXIBLE PARTITIONS.

DISABLE,FLEXIBLE PARTITIONS.

Default:

Enabled.

Signiﬁcance:

This entry enables or disables ﬂexible memory partitioning. The job scheduler

attempts to use memory space to the greatest extent possible when ﬂexible

partitions are enabled. Some service classes may be allotted more total memory

space than memory partitioning constraints normally allow. If ﬂexible partitions

are disabled, the total memory used by jobs of a given service class are never

allowed to exceed the maximum speciﬁed on the SERVICE command (AM and EM

parameters).

Revision M Deadstart Decks 3-141

LOCK

Format:

LOCK.

Default:

Unlocked.

Signiﬁcance:

This entry speciﬁes the system is locked. This software function prevents entry of

restricted commands; all other DSD commands can be entered. The console is

normally locked when the system is being used in a production environment.

LOGGING

Formats:

ENABLE,LOGGING.

DISABLE,LOGGING.

Default:

Disabled.

Signiﬁcance:

These entries specify whether dayﬁle messages intended for system analysis are

logged in the dayﬁle. Typical messages deal with informing the user that the

program is making inefﬁcient CIO calls, such as reading to a full buffer or

writing from an empty buffer. The dayﬁle messages are documented in an

appendix of the NOS Version 2 Operations Handbook.

3-142 NOS Version 2 Analysis Handbook Revision M

MASTER MSE

Formats:

ENABLE,MASTER MSE

DISABLE,MASTER MSE

Default:

Disabled.

Signiﬁcance:

These entries specify whether the MSE executive program (SSEXEC), when

initialized, is to run in master (enabled) or slave (disabled) mode.

MS VALIDATION

Formats:

ENABLE,MS VALIDATION.

DISABLE,MS VALIDATION.

Default:

Disabled.

Signiﬁcance:

This entry enables or disables mass storage validation. If enabled, CMR is

increased by 48 words, and the system veriﬁes that, for each mass storage device,

the sum of the counts of unreserved tracks and preserved ﬁles equals values

speciﬁed in the device's mass storage table.

If the device is a master device (contains user catalogs), the system also veriﬁes

these conditions:

• The device's track reservation table (TRT) specifies that the first tracks of the

indirect access ﬁle chain and the permit area are reserved and preserved.

The label track is linked to the ﬁrst catalog track.

The number of catalog tracks is a power of 2.

The catalog chain is reserved, of correct length, and contiguous if ﬂagged as

such in the device's MST.

To enable/disable mass storage validation with a DSD command entry, enable the

validation in the IPRDECK during a level 0, 1, or 2 deadstart.

Revision M Deadstart Decks 3-143

OQSH

Format:

OQSH=level.

Default:

None.

Signiﬁcance:

This entry speciﬁes the value of the output queue special handling (OQSH) level.

Output ﬁles with an access level equal to or greater than the output queue

special handling level will remain in the queue until released by the DSD

command RELEASE. When no OQSH level or access level name LVLO is selected,

all ﬁles will be processed. The value for this entry is supplied by the site security

administrator.

Parameter Description

Access level name speciﬁed in deck COMSMLS (refer to the NOS

Version 2 Installation Handbook) that corresponds to the desired

output queue special handling level.

level

PCLASS

Format:

CLASS, SCO,SCi , ....SC7 .

Default:

None.

Signiﬁcance:

Speciﬁes the service class associated with each priority level (PO through P7) for

selection on the Job command (refer to the NOS Version 2 Reference Set,

Volume 3).

Parameter Description

SCi The 2-character service class symbol. Because parameters are

positional, a comma must appear for any null parameter. The

default for a null parameter is that no service class will be

associated with the priority level represented by the null

parameter's position in the string. This command does not accept

the SS (subsystem) service class. Refer to the SERVICE IPRDECK

entry later in this section for a list of service classes.

.imfejjjV

3-144 NOS Version 2 Analysis Handbook Revision M

PF VALIDATION

Formats:

ENABLE,PF VALIDATION.

DISABLE,PF VALIDATION.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable preserved ﬁle (PF) validation. If enabled, the

system aborts an attach of a direct access permanent ﬁle if its end-of-information

(EOI) was altered during recovery of the ﬁle. If NA (no abort) is speciﬁed on the

attach request, the system attaches the ﬁle.

If mass storage validation is also enabled, TRT veriﬁcation of preserved ﬁles

takes place during a level 3 deadstart as follows:

• For all files, the system ensures that all tracks are reserved and that no

circular linkage exists.

• For all queued, permanent direct-access, and fast-attach files, the system also

ensures that the ﬁrst track is preserved.

If mass storage validation is enabled on a level 1 or 2 deadstart, TRT veriﬁcation

takes place automatically, regardless of the status of PF VALIDATION.

PRIVILEGED ANALYST MODE

Formats:

ENABLE,PRIVILEGED ANALYST MODE.

DISABLE,PRIVILEGED ANALYST MODE.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable privileged analyst mode operations. If enabled, a

user validated with AW = CPAM (refer to the NOS Version 2 Administration

Handbook for information on MODVAL validation) is permitted to read system

status information (such as the system dayﬁle, account ﬁle and error log) using a

nonsystem-origin job. On a secured system, this entry is ignored. Refer to the

DSD ENABLE/DISABLE command in section 5 for more details.

Revision M Deadstart Decks 3-145

PRIVILEGED RDF

Formats:

ENABLE,PRIVILEGED RDF.

DISABLE,PRIVILEGED RDF.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable privileged mode of RDF. If enabled, a user's

commands are checked to ensure that a maintenance function is being performed.

QUEUE

Format:

QUEUE,sc,qt,IL11,IPip,LPlp,UPup,WFwf.

Default:

None.

Signiﬁcance:

This entry speciﬁes the queue priorities associated with the input, executing, and

output queues for each job service class. Refer to table 3-7 for an example set of

parameter entries and to ﬁgure 3-6 for an example of ranges of service class

priorities.

Parameter Description

sc Service class.

sc Description

BC Local batch.

CT Communication task.

DI Detached interactive.

In Installation class n (0

MA Maintenance.

NS Network supervisor.

RB Remote batch.

SS Subsystem.

SY System.

TS Interactive.

n *- 3).

<<<3^\

3-146 NOS Version 2 Analysis Handbook Revision M

QUEUE

Parameter Description

ILil

rIPip

LPlp

rUPup

WFwf

Job queue type.

qt Description

EX Executing jobs.

IN Input queued ﬁles.

OT Output queued ﬁles.

Priority a job receives when it initially exhausts its CM time slice

(refer to the SERVICE command CM parameter). Thereafter,

whenever the CM time slice is exhausted the job's priority will be

set to the value of lp. il ranges from 0 to 7777s, but must be in the

range of values for lp and up. This parameter is valid only for

executing jobs (EX).

Initial priority only for an executing batch job or for an interactive

job when a terminal command is entered. Online interactive jobs

with terminal I/O available are scheduled at tp priority (refer to the

SERVICE command TP parameter), ip ranges from 0 to 7777s, but

must be in the range of values for lp and up.

Lowest priority. For an input or output queue ﬁle, the priority

assigned to a ﬁle or job entering the queue. For an executing job,

the priority assigned to job which has exceeded a non-initial CM

slice. The IL value is used for the ﬁrst CM slice, lp ranges from 0 to

77778, but must be less than the value of up.

Highest priority. For input and output queues, this is the highest

priority a ﬁle can reach in that queue; aging stops when this priority

is reached. For the execution queue, this is the priority assigned to a

job when initially assigned to a control point, up ranges from 0 to

77778, but must be greater than the value of lp.

Weighting factor for queue priority calculation; wf must be a power

of 2, from 1 to 40008. The smaller the weighting factor, the faster

the queue entry reaches its highest priority.

jp»\

Revision M Deadstart Decks 3-147

REMOVABLE PACKS

Formats:

ENABLE,REMOVABLE PACKS.

DISABLE,REMOVABLE PACKS.

Default:

Enabled.

Signiﬁcance:

These entries enable or disable automatic label checking for mass storage devices

that are deﬁned as removable.

If enabled, automatic label checking occurs. This status must be available to

perform label veriﬁcation before removable devices can be accessed.

If disabled, any removable devices introduced into the system will not be

recognized.

y*^\

,>«^SK

3-148 NOS Version 2 Analysis Handbook Revision M

RESIDENT RDF

Formats:

ENABLE,RESIDENT RDF.

DISABLE.RESIDENT RDF.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable resident mode of RDF. While in resident mode,

RDF remains active, regardless of terminal inactivity, until RDF is disabled.

When resident mode is disabled, RDF becomes inactive if no one is logged on at

the remote diagnostic terminal for 15 minutes.

SECONDARY USER COMMANDS

Formats:

ENABLE.SECONDARY USER COMMANDS.

DISABLE,SECONDARY USER COMMANDS.

Default:

Disabled.

Signiﬁcance:

The enable option allows jobs to issue more than one USER command. These

entries are ignored on a secured system since secondary USER commands are not

allowed on a secured system.

Revision M Deadstart Decks 3-149

SECURES

Format:

SECURES,0t,LA=1a,UA=ua.

Default:

The security access limits of the system origin type (SY) are set to access level

name LVLO. All security access limits for other origin types are set to system

limits.

Signiﬁcance:

This entry speciﬁes system limits for the system origin type (ot equals SY) or

limits for other origin types (ot equals BC, IA, or RB). System limits deﬁne the

highest and lowest levels allowed in the system. No job may execute and no ﬁle

may be accessed or created at a level outside this range.

Limits for origin types other than SY must be within the system limits. If origin /*%

type SY is speciﬁed, limits for all origin types are set to the selected values.

Both la and ua must be entered; they can be set to the same value, restricting

system access or a particular origin type access to a single level. These values are

supplied by a site security administrator.

This entry is not meaningful in an unsecured system.

Parameter Description

ot Origin type.

ot Description

S Y S y s t e m .

BC Batch.

IA Interactive.

RB Remote batch.

la Access level name speciﬁed in deck COMSMLS (refer to the NOS

Version 2 Installation Handbook) corresponding to the desired lower

access level limit.

ua Access level name speciﬁed in deck COMSMLS corresponding to the

desired upper access level limit.

3-150 NOS Version 2 Analysis Handbook Revision M

SERVICE

Format:

SERVICE,sc.AMam.CBlpup,CMcm.CPcp.CScs.CTct.DSds.DTsc.ECec.EMem.FCfc.FLﬂ,

FSfs.NJnj.SDsd.TDtd.TPtp.

Default:

None.

Signiﬁcance:

This entry speciﬁes the service limits associated with each service class. Refer to

table 3-7 for an example set of parameter entries and to ﬁgure 3-6 for an

example of ranges of service class priorities.

Parameter Descri

Service

iption

sc 5 class; one of these values:

sc Description

•BC Local batch.

CT Communication task.

DI Detached interactive.

In Installation class n (0 -£ n -£ 3).

MA Maintenance.

NS Network supervisor.

RB Remote batch.

SS Subsystem.

SY System.

TS Interactive.

AMam Maximum ﬁeld length divided by 100s for all jobs of the speciﬁed

service class. This parameter partitions central memory by limiting the

ﬁeld length available to each service class. For example, if scheduling

a job to a control point would cause the AM value for its service class

to be exceeded for its service class, it may not be scheduled until the -

required ﬁeld length is available. This means that a lower priority job

from a different service class may be scheduled ﬁrst. However, a job

that would cause the AM value for its service class to be exceeded,

can be scheduled to a control point if not enough jobs in other service

classes exist to ﬁll central memory and FLEXIBLE PARTITIONS is

enabled. The system attempts to use central memory to its greatest

capacity, am ranges from 0 to 77777s.

Revision M Deadstart Decks 3-151

SERVICE

Parameter Description

CBlpup

CMcm

CPcp

Upper and lower bound CPU priorities.

Value Description

Lower bound CPU priority, lp ranges from 0 to 77s.

Upper bound CPU priority, up ranges from 0 to 77s, but

must be equal or larger than the value of lp.

Central memory time slice in seconds. This parameter speciﬁes the

maximum amount of time a job of the speciﬁed service class can remain

in central memory either at a control point or at a pseudo-control point

before its priority is set to the lower boundary. When the job initially

exhausts its CM time slice, this lower boundary is the value of il.

Subsequently, when the job exhausts its CM time slice, this lower

boundary is the value of lp (refer to the QUEUE command IL and LP

parameters), cm ranges from 0 to 7777s.

Control point slice priority. This parameter speciﬁes the value of the

scheduling priroity which will be set after a job has been at a control

point for ct seconds. The priority speciﬁed must be greater than or equal

to the lower bound and less than or equal to the upper bound execution

queue scheduling priorities speciﬁed by the CB parameter

(lp ^ cp ^ up). Refer to the QUEUE command LP and UP parameters.

3-152 NOS Version 2 Analysis Handbook Revision M

/ms

0ps

SERVICE

Parameter Description

CScs Cumulative size in PRUs allowed for all indirect access permanent ﬁles.

cs indicates a limit value for the cumulative size.

cs Limit Value

0 Unlimited

1 1000s

2 5000s

3 500008

4 1000008

5 200000s

6 4000008

7 Unlimited

CTct Control point time slice in seconds. This parameter speciﬁes the

maximum amount of time that a job of the speciﬁed priority can remain

at a control point before its priority is changed to the control point slice

priority (cp). ct ranges from 1 to 7777s.

DSds Size in PRUs allowed for individual direct access permanent ﬁles, ds

indicates the limit value for the size of the ﬁles.

ds Limit Value

0Unlimited

110008

250008

350000s

41000008

52000008

64000008

7Unlimited

DTsc Service class to which a detached job is assigned if the job is detached.

The default value for sc is DI (detached interactive).

ECec Maximum user-accessible extended memory ﬁeld length in words divided

by UEBS for any job of the speciﬁed service class. This parameter

performs the same function for extended memory ﬁeld length that the

FL parameter does for central memory ﬁeld length, ec ranges from 0 to

37778.

19. Refer to Extended Memory Overview earlier in this section to determine the value of UEBS.

Revision M Deadstart Decks 3-153

SERVICE

EMem

FCfc

Parameter Description

Maximum extended memory length in words divided by UEBS for all

jobs of the speciﬁed service class. This parameter performs the same

function for extended memory ﬁeld length that the AM parameter does

for central memory ﬁeld length, em ranges from 0 to 37778.

Number of permanent ﬁles allowed, fc indicates a limit value that is the

maximum number of permanent ﬁles allowed.

fc Limit Value

Unlimited

108

408

FLﬂ

3 1 0 0 8

4 2 0 0 8

5 10008TBL

6 40008

7 Unlimited

Maximum ﬁeld length divided by 1008 for any job of the speciﬁed

service class. If more memory is requested, the job is aborted. You

typically use this parameter to limit the memory requirement for jobs of

a speciﬁc service class during certain hours of the day. For example,

you may use the FL parameter to specify a maximum ﬁeld length for

all batch service class jobs between the hours of 2 p.m. and 4 p.m. ﬂ

ranges from 0 to 7777s.

3-154 NOS Version 2 Analysis Handbook Revision M

SERVICE

Parameter Description

FSfs

NJnj

SDsd

TDtd

TPtp

Size in PRUs allowed for individual indirect access permanent ﬁles, fs

indicates a limit value for the size of the ﬁles.

fs Limit Value

Unlimited

10s

30s

1008

3008

1000s

2000s

Unlimited

Maximum number of jobs. For each service, this parameter speciﬁes the

number of jobs that can be executing in the system, nj ranges from 0 to

7777s.

CPU job switch delay in milliseconds, sd ranges from 0 to 7777s.

Suspension time-out delay. A suspended job will not be timed out for

td*108 seconds. The maximum delay is approximately 9 hours, td ranges

from 0 to 7777s.

Terminal job priority assigned to a terminal job that is rolled out

waiting for terminal input, tp ranges from 0 to 7777s, but must be in

the range of values for the execution queue parameters lp and up (refer

to the QUEUE command LP and UP parameters).

Revision M Deadstart Decks 3-155

SPINDOWN

Formats:

ENABLE,SPINDOWN.

DISABLE,SPINDOWN.

Default:

Enabled.

Signiﬁcance:

These entries enable or disable the spindown of 834, 836, 887, and 9853 disk

units. When spindown is enabled, entering a CHECK POINT SYSTEM command

will cause all 834, 836, 887, and 9853 disk units that are on and not globally

unloaded to automatically spin down. When spindown is disabled, entering a

CHECK POINT SYSTEM command will not spin down the 834, 836, 887, and

9853 disk units. System checkpoints that are initiated by mainframe errors do not

affect the state of these disk units.

SRST

Format:

SRST=n.

Default:

Signiﬁcance:

This entry speciﬁes the secondary rollout sector threshold. Any rollout ﬁle smaller

than n sectors (0 ^ n ^ 7777s) is considered a secondary rollout ﬁle for the

purpose of equipment selection.

NOTE

The size of the rollout ﬁle for any job must be at least seven sectors larger than

the combined size in sectors of the job's central memory and extended memory

ﬁeld lengths.

•■•"**%

3-156 NOS Version 2 Analysis Handbook Revision M

SYSTEM DEBUG

Formats:

ENABLE,SYSTEM DEBUG.

DISABLE,SYSTEM DEBUG.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable the system debug mode of operation. When the

system is in system debug mode, it is less tolerant of system errors; that is, it is

more likely to hang upon experiencing errors. When the system is not in system

debug mode, it rates system errors as critical or noncritical. For critical errors,

rthe system partially or totally interrupts system operation to tend to the errors.

For noncritical errors, the system logs them in the binary maintenance log (BML)

and inasmuch as possible allows system operation to proceed. You can initiate the

system debug mode with the DSD ENABLE command or the corresponding

IPRDECK entry.

There is another system state called debug mode, which is conceptually different

from that of system debug mode. Debug mode is the state of the system where a

user with system origin privileges can make modiﬁcations to the running system.

You can initiate this mode of operation with the DSD command DEBUG. The left

screen header of the system console indicates whether the system is in debug

mode or not.

Revision M Deadstart Decks 3-157

TAPE PF STAGING

Formats:

ENABLE,TAPE PF STAGING.

DISABLE.TAPE PF STAGING.

Default:

Disabled.

Signiﬁcance:

These entries specify whether permanent ﬁles that reside on tape alternate

storage are staged to disk. If disabled, jobs that attempt to access tape alternate

storage resident ﬁles are aborted.

UNLOCK

Format:

UNLOCK.

Default:

Unlocked.

Signiﬁcance:

This entry speciﬁes the system console is unlocked. All DSD commands can be

entered when the console is unlocked. The console is usually locked when the

system is being used in a production environment. Refer to LOCK earlier in this

section.

USER EXTENDED MEMORY

Formats:

ENABLE,USER EXTENDED MEMORY. /

DISABLE,USER EXTENDED MEMORY.

Default:

Disabled.

Signiﬁcance:

These entries enable or disable scheduling of jobs that access user extended

memory.

3-158 NOS Version 2 Analysis Handbook Revision M

/P*\

LIBDECK

LIBDECK is a SYSEDIT directive record on the deadstart ﬁle. SYSEDIT reads

LIBDECK during the system load. LIBDECK speciﬁes program residence, ﬁeld length,

record type, and parameter format.

Up to 64 LIBDECK records can be placed on the deadstart ﬁle. LIBDECKs are

named LIBDnn, where nn is 00 to 77s. A speciﬁc record can be selected with a

LIB = n entry in CMRDECK. The multiple deadstart deck capability enables the use

of a single deadstart ﬁle on virtually any system conﬁguration.

You can list all LIBDECKs on the deadstart ﬁle by accessing the system ﬁle

SYSTEM with a COMMON command, then using the T parameter on the CATALOG

command. Refer to the NOS Version 2 Reference Set, Volume 3 for more information

concerning these commands.

The following list provides brief descriptions of SYSEDIT directives acceptable in

LIBDECK. Complete descriptions of all SYSEDIT directives are in section 19,

SYSEDIT. A list of valid record types follows the directives.

Directive Format

*AD,nn,tyi/reci,ty2/rec2,...,tyn/recn

*CM,tyi/reci,ty2/rec2,...,tyn/recn

*FL,tyi/reci-ﬂi,ty2/rec2-ﬂ2,...,

tyn/recn-ﬂn

*MS,tyi/reci ,ty2/rec2,... ,tyn/recn

Signiﬁcance

Speciﬁes the alternate device to be

used in addition to the system

device(s) for storing ABS, OVL, PP,

and REL type records, nn is either

the EST ordinal or the equipment

type, tyi is the record type, and reci

is the record name.

Deﬁnes record reci of type tyi as

being central memory resident; valid

only for record types ABS, OVL, or

PP. Like any other job in NOS,

SYSEDIT has a ﬁeld length

restriction of 3760008 central

memory words. You cannot use more

than approximately 326000s central

memory words (3760008-500008 words

for SYSEDIT's program FL) when

deﬁning records to be central

memory resident.

Record reci of type tyi is loaded with

a ﬁeld length speciﬁed by ﬂi (ﬂi is

ﬁeld length divided by 100s).

Deﬁnes record reci of type tyi as

being mass storage resident. This is

the default residence for routines

with no storage area speciﬁed in

LIBDECK.

Revision M Deadstart Decks 3-159

LIBDECK

Directive Format Signiﬁcance

♦PROC,reci,rec2,...,recn Defines record reci of type PROC as a

procedure ﬁle.

*SC,tyi/reci,ty2/rec2,..., Deﬁnes record reci of type tyi as

tyn/recn product set format commands. The

command parameters are processed in

product set format (refer to the NOS

Version 2 Reference Set, Volume 3).

*/ comments Deﬁnes comment lines.

The following record types may be speciﬁed in SYSEDIT directives. Some directives do

not allow all types.

Type (tyi) Description

ABS Multiple entry point overlay.

CAP Fast dynamic load capsule.

OPL Modify old program library deck.

OPLC Modify old program library common deck.

OPLD Modify old program library directory.

OVL Central processor overlay.

PP 12-bit peripheral processor program. )

PPL 16-bit peripheral processor program (CYBER 180 models only).

PPU First-level peripheral processor (FLPP) program (model 176 only).

PROC Procedure.

REL Relocatable central processor program.

TEXT Unrecognizable as one of the other types.

ULIB User library program.

3-160 NOS Version 2 Analysis Handbook Revision M

f*»

DIS Operations 4

DIS Job Dayﬁle Display (A) 4.3

DIS Job Status Display (B) 4.4

DIS Memory Displays (C, D, F, G) . 4-6

DIS Exchange Package Display (X) 4.8

DIS Directory Display (Z) 4.9

Console Operation 4-10

Display Selection Commands 4.13

DIS Keyboard Entries 4-14

Memory Entry Commands 4-18

PP Call Commands , 4-20

*■)■

..*%

DIS Operations

DIS displays information about a single job. Under DIS, the X display shows the

exchange package area for the job. Central memory addresses relative to the job's

reference address are used for the data and program displays.

Initiate DIS at a control point to monitor the progress of a job by using any of the

following methods:

•You can call DIS by entering X.DIS,ﬂ (ﬂ=ﬁeld length desired) or X.DIS (ﬁeld

length of 600008 assumed by default). This brings DIS to an empty control point

to initiate utility programs.

A job can execute a command in the form DIS (the job must be system origin or

have system origin privileges).

You can call DIS to an existing job by entering the command:

DIS,jsn.

Parameter Description

jsn Job sequence name to which you want DIS assigned.

On either a secured or unsecured system, before you can call DIS to an existing job,

the console must be unlocked (refer to the DSD UNLOCK command in section 5). For

a secured system, the system must be set to security unlock status by the security

administrator.

You can toggle between DSD and DIS. DIS permanently returns control to DSD when

you enter DROP.; the job is not dropped unless no commands remain.

When DIS is called to a control point, automatic command processing stops and the A

and B displays for DIS appear on the left and right display screens, respectively.

Keyboard entry is necessary to begin processing subsequent commands. Unless

automatic command processing is reenabled,* the job is stopped after each command

is processed. That is, only one command can be processed at a time. Under DIS, the

#^ B display shows only the condition of the job to which it is assigned, including

^ upcoming commands. When the job is not using the central processor, a copy of its

exchange package is displayed on the X display. Displays available under DIS are

selected in the same manner as DSD displays. Refer to Console Operation later in

this section for information concerning display selection commands and to DIS

Keyboard Entries for information about other DIS commands.

1. You can initiate automatic command processing by entering either a period (.) or the RCS command (refer

to Console Operation and to DIS Keyboard Entries later in this section).

Revision M DIS Operations 4-1

DIS Operations

The displays available under DIS are:

Display

Identiﬁer Description

A Job dayﬁle. Messages and ﬁles (local FNT entries) attached to the job.

B Job status. Individual job status, equipment assigned, current messages,

and command buffer.

C,D Central memory. Contents of central memory words (selectable 8-word

groups) in ﬁve columns of four octal digits with display code

equivalents (same as the DSD C and D displays).

F Central memory. Contents of central memory words (selectable 8-word

groups) in four columns of ﬁve octal digits with display code

equivalents.

G Central memory. Contents of central memory words (selectable 8-word

groups) in four columns of ﬁve octal digits with COMPASS instruction "*^

mnemoric equivalents.

H File status. All ﬁles assigned to the job as well as equipment assigned

to ﬁles.

M Extended memory. Contents of 60-bit words of extended memory

(selectable 8-word groups) in ﬁve columns of four octal digits with

display code equivalents (same as the DSD M display).

N Blank screen. You may wish to select the N display or one screen to /0m\

reduce the screen ﬁlter on the other screen.

T,U Text display. Displays text from central memory in coded lines (up to

60 characters per line). The display terminates after 256 words have

been displayed.

V Central memory buffer. Displays directly from central memory. The

display terminates after 512 words have been displayed.

X E x c h a n g e p a c k a g e . B r e a k p o i n t a d d r e s s a n d t h e e x c h a n g e p a c k a g e . / - " * \

Y Monitor functions. Displays mnemonics and the values of all monitor

functions (same as the DSD Y display).

Z Directory. DIS displays available.

NOTE

Although all displays listed may appear on the left screen, only the B,

C, D, and N displays may appear on the right screen. If you attempt to

bring any other display to the right screen, the message INCORRECT

COMMAND is issued to the job dayﬁle and is displayed in the message

buffer of the B display.

4-2 NOS Version 2 Analysis Handbook Revision M

DIS Job Dayﬁle Display (A)

Figure 4-1 shows the DIS job dayﬁle display (A). The ﬁgure shows the job dayﬁle

messages for the control point to which DIS is currently assigned and as many ﬁles

attached to that control point as will ﬁt in the display. All ﬁles attached to the job can

be observed by using the ﬁle status display (H).

DIS A. XB DAYFILE.

09.46.34.RFL(060000)

09.46.34.DIS.

09.47.31.USER.ABC1234,.

09.47.31.ABSC. S.

Dayﬁle 09.48.43.PACKNAM,721C.R-DL.

sessages 09.49.39.NOS.

09.50.06. 6ET.N0S/PN«PACKC.R"DJ.UN«KRONM0D.

09.50.06. DEVICE UNAVAILABLE.

09.50.29. EXIT.

09.50.41. 6ET.NOS/PN-PACKV2.R»DJ.UN-KRCNM0D.

09.50.41. DEVICE UNAVAILABLE.

09.55.21.COMMON.SYSTEM.

FNT NAM E TY PE EST TR ACK FS S TATUS LEVEL

0 I N P U T - I N » 7 4 0 0 4 N D 5 LVLO Files (local fnt

5 S Y S T E M L l » 6 4 0 1 0 1 LVLO entries) attacned

1 0 Z Z Z Z Z C O L I > 7 4 2 1 3 N D 5 LVLO to the joo

1 1 Z Z Z Z Z C 2 L O 6 5 6 5 2 N D 3 0 7 LVLO

Figure 4-1. DIS Job Dayﬁle Display (A)

The level ﬁeld, shown in the ﬁgure, is displayed only on a secured system.

yms

Revision M DIS Operations 4-3

DIS Job Status Display (B)

Aa!S^S.

DIS Job Status Display (B)

The DIS B display shows the status of a speciﬁc job executing at a control point. It

also shows: any equipment assigned exclusively to the job by EST ordinal, message 1

and message 2 from the control point area, and the current command buffer, allowing

you to anticipate future job requirements.

Figure 4-2 shows the DIS job status display (B).

DIS B. JOB STATUS.

J S N = A A A C E J T O = 5 P = 2046

S R U A = 1 R A » 1741

U I = 2 7 5 5 S R U L = 7 7 7 7 7 7 F L = 26 Job status

F M = M L S T E S T C S » B A T C H R A E = 160

P N = 7 2 1 C C O N N = 0 F L E =

L E V E L = L V L O C P A = 600

C P U = X

EST - 1 } Equipment assigned

MS1 = DEVICE UNAVAILABLE.

MS2 = Current messages

NOS.PACKV2.

RETURN,NOS.

REVERT,NOLIST.

EXIT.

REVERT,ABORT.PACKC OR PACKV2 NOT FOUND. Command b u f f e r

REVERT,NOLIST.

EXIT.

REVERT.CCL

EXIT.CCL

REVERT,ABORT.CCL

Figure 4-2. DIS Job Status Display (B)

4-4 NOS Version 2 Analysis Handbook Revision M

DIS Job Status Display (B)

A^S.

/^eJm^v

The job status portion of the display shows three columns of information. Each item

has the form item=value. The items are described next in the order that they appear

on the display.

Item Description

JSN Job sequence name of the DIS job.

UI User index.

FM Current family name.

PN Current pack name.

CPU CPU status.

EJTO Executing job table ordinal.

SRUA System resource units accumulator.

SRUL Account block limit for system resource units.

CS Connection status.

CONN Connection number (interactive jobs only).

LEVEL Job access level (secured systems only).

P P register address from exchange package.

RA Central memory reference address.

FL Central memory ﬁeld length.

RAE Extended memory reference address.

FLE Extended memory ﬁeld length.

CPA Control point area address.

Revision M DIS Operations 4-5

DIS Memory Displays (C, D, F, G)

Figure 4-3 shows the DIS data storage display (F). The contents of each central

memory word is displayed in columns of ﬁve octal digits along with the display code

equivalent. Only the memory locations currently assigned to the job can be displayed.

The message ****SECURED AREA**** is displayed for all other locations.

The DIS C and D displays have the same format as the DSD C display, except that

the DIS memory displays may show a managed table bias word preceding the groups

of central memory words.

DIS F. CENTRAL MEMORY.

00000060 00000 00000 OOOOO OOOOO

00000061 ooooo 00000 OOOOO 00000

00000062 00000 00000 OOOOO OOOOO

00000063 OOOOO OOOOO OOOOO 00000

00000064 03171 51517 16000 00001 COMMON A

00000065 40000 00000 00010 00507 5 A E G

00000066 40000 00000 00020 00131 5 B A Y

00000067 40000 OOOOO OOOOO OOOOO

00000070 03171 51517 16562 33123 COMMON,SYS

00000071 24051 55700 00000 00000 TEM.

00000072 00000 00000 OOOOO OOOOO

00000073 00000 00000 OOOOO 00000

00000074 00000 00000 OOOOO OOOOO

00000075 00000 00000 OOOOO OOOOO

00000076 00000 OOOOO 00000 00000

00000077 OOOOO 00000 OOOOO OOOOO

Figure 4-3. DIS Data Storage Display (F)

4-6 NOS Version 2 Analysis Handbook Revision M

DIS Memory Displays (C, D, F, G)

Figure 4-4 shows the DIS program storage display (G). The G display shows the

contents of central memory and the COMPASS mnemonic translation.

OIS G. CENTRAL E8EMORY.

OOOOOOOO

00000001

00000002

00000003

00000004

00000005

00000006

00000007

00000010

00000011

00000012

000000.13

00000014

00000015

00000016

00000017

Address

00000 00000 00000 00000

05160 42000 00000 00000

23312 32405 15000 00001

00000 00000 00000 00000

00000

ooooo

00Q00

ooooo

NE BB

AX XB

DXX+X

BXX°-X PS

Memory

contents

COMPASS aneaontc

translation

Figure 4-4. DIS Program Storage Display (G)

Revision M DIS Operations 4-7

DIS Exchange Package Display (X)

The DIS X display shows the breakpoint address (BKP=addr) if a breakpoint was set

at address addr. The breakpoint address is followed by the job's exchange package.

Figure 4-5 shows the DIS exchange package display (X).

DIS X. EXCHANGE PACKAGE.

BKP I00.

P=452 AO = 600 BO =

RA 174100 A1 = B1 =

FL 600 A2 = B2 » 777776

EM = 74070000 A3 = B3 -

RAE A4 = B4 =

FLE A5 = B5 =

MA 600 A6 = B6 =

EEA A7 = 375 B7 =

XO = 7777 7777 7777 7700 0000

X1 = oooo 0000 0000 0000 0000

X2 = oooo 0000 0000 0000 0375

X3 = oooo 0000 0000 0000 0000

X4 = oooo 0000 oooo oooo 0003

X5 = 2331 2324 0515 0000 0001 SYSTEM A

X6 = 0516 0420 oooo oooo 0000 ENDP

X7 = 0000 0000 oooo oooo 0000

+ AND - WILL SET *BKP* TO (P) +/- 1.

Figure 4-5. DIS Exchange Package Display (X)

4-8 NOS Version 2 Analysis Handbook Revision M

DIS Directory Display (Z)

Figure 4-6 shows the DIS directory display (Z). The Z display lists all displays

available under DIS control. If the letter entered to select the left screen display is not

a valid display identiﬁer, the Z display is selected automatically.

DIS Z. DIRECTORY. LEFT SCREEN ONLY DISPLAY

A JOB DAYFILE.

F CENTRAL MEMORY.

G CENTRAL MEMORY.

H FILE STATUS.

M EXTENDED MEMORY.

T TEXT DISPLAY.

U TEXT DISPLAY.

V CM BUFFER.

X EXCHANGE PACKAGE.

Y MONITOR FUNCTIONS

Z DIRECTORY.

LEFT AND RIGHT SCREEN DISPLAYS

FOUR GROUPS OF FIVE

BJOB STATUS.

CCENTRAL MEMORY. FIVE GROUPS OF FOUR

DCENTRAL MEMORY. FIVE GROUPS OF FOUR

NBLANK SCREEN.

Figure 4-6. DIS Directory Display (Z)

0ms

Revision M DIS Operations 4-9

Console Operation

Unlike DSD, DIS is not interpretive. You must complete every entry and signal DIS to

act upon the message by pressing CR or NEXT. The following rules apply to all DIS

commands.

• For input, spaces in an octal field are ignored but can be inserted for readability.

• For output, all octal fields are right-justified with leading zero fill; excess octal

digits are ignored.

In addition to the command entries, the following keys have special meaning to DIS

when entered as the ﬁrst character. The corresponding special keys for the CC545,

CC598B, and CC634B console types are listed along with a description of the use of

each key.

CC545 CC598B CC634B Description

F15

Up arrow, +, +

or grey +

Down arrow,

-, or grey -

P g U p o r ( (

P g D n o r ) )

/ /

If DSD has relinquished the main

display console to DIS, this key acts as

a quick hold, and DIS drops the display

channel so that DSD can use it.

Pages the left screen forward for the A,

C, D, F, G, H, M, T, and U displays.

Increments the breakpoint address for

the X display.

Pages the left screen backward for the

C, D, F, G, M, T, and U displays.

Decrements the breakpoint address for

the X display. Resets the A and H

displays to the beginning.

Pages the right screen forward for the C

and D displays.

Pages the right screen backward for the

C and D displays.

Advances the left screen memory display

address by the value in the lower 18

bits of the ﬁrst word displayed

(applicable only to memory displays C,

D, F, G, and M).

Sets auto mode (initiates automatic

command processing). This key performs

the same function as the RCS command

described under DIS Keyboard Entries

later in this section.

X ^ X

/<^lsN.

4-10 NOS Version 2 Analysis Handbook Revision M

Console Operation

rCC545 CC598B CC634B Description

8 8 8

Right blank

(display key)

CR (carriage

return)

Tab

Enter/Return NEXT

Advances the pointer indicating the ﬁrst

address of managed tables for the left

screen (applicable only to memory displays

C, D, F, and G).

Decrements the pointer indicating the ﬁrst

address of managed tables for the left

screen (applicable only to memory displays

C, D, F, and G).

Advances the left screen display sequence

established by the SET,screen. command

(refer to Display Selection Commands later

in this section).

Sets the repeat entry ﬂag; the message

REPEAT ENTRY is displayed on the error

message line of the left screen. The

subsequent command entry is processed

but is not erased after completion. That

command is processed each time this key

is pressed. To clear the repeat entry mode,

press the left blank (erase) key on the

CC545 or the |< key on the CC634B.

The following keys are interpreted as control characters by DIS.

CC545

Left blank

(erase)

BKSP

(backspace)

CR (carriage

return)

CC598B CC634B Description

Esc |< or Clears current DIS keyboard entry and

ERASE any resultant error message; auto mode

(automatic command processing) is also

cleared.

Back Space <

Enter/Return NEXT

The following keyboard messages may appear above your entry

Deletes last character displayed and

clears error message (if one exists).

Initiates processing of an entered

command.

Message Description

yms

AUTO MODE.

COMMAND BUFFER

FULL.

COMMAND TOO

LONG.

The command buffer is read automatically. Automatic

command processing can be selected by the RCS command

or by entering a period.

The ELS command has been entered and there is

insufﬁcient room in the command buffer to add the

characters requested.

The command you entered is too long.

Revision M DIS Operations 4-11

Console Operation

Message Description

COMMANDS ON

FILE.

DIRECT CPU INPUT.

DISK BUSY.

EXTENDED MEMORY

NOT AVAILABLE.

The ELS command has been entered and commands are

currently located in a ﬁle instead of in a command buffer. It

is not possible to add a command to the ﬁle.

The N command has been entered and all data entered from

the keyboard is being passed directly to central memory.

DIS is waiting for an overlay to be loaded from a mass

storage device.

You entered the ENFLE command, but user extended

memory is not deﬁned.

INCORRECT ENTRY. The command cannot be processed.

INCORRECT FL

REQUEST.

INCORRECT

PRIORITY.

INSUFFICIENT FIELD

LENGTH.

JOB ACTIVE.

OUT OF RANGE.

PP BUSY.

REPEAT ENTRY.

REQUEST EXCEEDS

MAXIMUM FLE.

STORAGE NOT

AVAILABLE.

WAITING FOR

EXTENDED

MEMORY.

WAITING FOR

STORAGE.

You entered an ENFL command to set the ﬁeld length. The

requested ﬁeld length must be greater than 100008 and less

than 3777008 including negative ﬁeld length.

The priority you entered using the ENPR command is not

correct. CPU priority must be greater than 1 and less than

718.

Your ﬁeld length is not long enough to support either

extended memory or 026.

The previous request has not completed. The command must

be reentered when the job is not active.

The memory entry address is greater than the ﬁeld length.

DIS is waiting for a PP to be assigned in order to process a

keyboard entry.

The command in the command buffer is repeated each time

you press CR, Enter/Return, or NEXT. This can be cleared

by pressing the left blank, Esc, or |< key.

You requested more extended memory than your system can

support.

The amount of central memory requested by the ENFL

command or required by the 026 ﬁle editor is not available.

DIS is waiting for extended memory after you entered the

ENFLE command.

DIS is waiting for central memory after you entered the

ENFL or 026 command.

4-12 NOS Version 2 Analysis Handbook Revision M

Display Selection Commands

Specify the displays you want shown on the console's left and right screens by using

the following commands. Follow the entry with a carriage return.

Command

xy.

xz,loc.

x.addr.

SET.xxxx.

Description

Brings the x and y displays to the left and right screens, respectively.

Note that although all DIS displays may appear on the left console

screen, only the B, C, D, and N displays may appear on the right

screen. If you attempt to bring any other display to the right screen,

the entry is interpreted as a command and the message INCORRECT

ENTRY appears. In addition, if the letter entered to select the left

screen display (x) is not a valid display identiﬁer, the Z display is

selected automatically.

Brings speciﬁed memory display to the left screen, if not currently

selected, and provides display modiﬁcations as follows:

Parameter Description

loc

Display identiﬁer (C, D, F, G, or M).

Type of display modiﬁcations: 0-3, 4, 5, or 6

Use 0-3 to display speciﬁed group (8 words) starting at

location loc.

Use 4 to display all 8-word groups in contiguous

locations starting at location loc.

Use 5 to advance the display by loc locations.

Use 6 to decrement the display by loc locations.

Location parameter (maximum of eight digits for central

memory address or seven digits for extended memory

address).

If x speciﬁes one of the memory displays (C, D, F, G, or M), addr is

the address used to obtain the bias address for the managed table

display. (The bias address is the lower 18 bits of the word at addr.)

Sets the left screen display sequence; xxxx consists of one to four

display identiﬁers. Pressing the right blank key on the CC545, the

Tab key on the CC598B, or the —> key on the CC634B after this

command is entered causes the ﬁrst display to appear on the left

console screen. Pressing the key again selects the second display. The

next display in the speciﬁed sequence appears on the left screen each

time the key is pressed, for example, SET,ACFD.

Revision M DIS Operations 4-13

DIS Keyboard Entries

You can enter the following commands while in DIS.

If a job is currently active (CPU active, waiting, on recall, or PP active), many

commands are not accepted and the message JOB ACTIVE is displayed.

Command Description

BEGIN ,pname,pﬁle.

BKP,addr.

BKPA,addr.

CEF.

DCP.

DDF.

DIS.

DROP.

ELS.commandstring.

ENAi,addr.

ENBi,addr.

ENEM,n.

ENFL,ﬂ.

Sets auto mode and calls the procedure pname that is on ﬁle

pﬁle.

Breakpoint to address addr in the program. Central processor

execution begins at the current value of P and stops when

P=addr, and DIS is the only PP active at the control point.

Breakpoint to address addr in the program with assigned

PPs. Central processor execution begins at the current value

of P and stops when P = addr. PPs attached to the control

point can still be active. DIS clears addr to stop the program

at that point. The breakpoint may be cleared by setting the

breakpoint address to a new value.

Clears the skip-to-exit ﬂag. This allows command processing

to continue with the next command instead of skipping to

the EXIT command after an error.

Drops the central processor and displays the exchange

package area on the X display.

Calls the display disk ﬁle (DDF) utility to the control point.

Reloads the main DIS overlay.

Drops DIS, but normal processing of the job continues (it

does not drop the job until all commands are processed).

Allows entry of the commandstring command after the last

command in the command buffer, if there is space. This

command is valid only when auto mode is not set.

Sets register Ai = addr in the exchange package area.

Sets register Bi=addr in the exchange package area.

Sets CPU program exit mode to n (0 ^ n ^ 7).

Sets central memory ﬁeld length FL=ﬂ in the exchange

package area (0 ^ ﬂ ^ 777777s). ﬂ must be at least 100008

if user extended memory is assigned.

sfMS

4-14 NOS Version 2 Analysis Handbook Revision M

DIS Keyboard Entries

rCommand Description

ENFLE,ﬂe.

ENP,addr.

ENPR,pr.

ENPR,*.

ENS.commandstring.

ENTERVcommandl/

command2.

ENTL,timlmt.

ENXi,cont.

ENXi,Lcont.

ENXi,Dcharacters.

ENXi,b,value.

ERR.

GO.

HOLD.

M.characters.

N.characters.

Sets extended memory ﬁeld length FLE to ﬂeOOO in the

exchange package area (1 ^ ﬁe ^ 7777s). If user extended

memory is assigned (ﬂe = 0), central memory FL, set by the

ENFL command, must be greater than or equal to 10000s.

Sets P=addr (next instruction address).

Sets CPU priority to pr (2 ^ pr ^ 70s).

Sets CPU priority to the minimum non-decrementing

value, 60s.

Allows entry of the commandstring command as the next

unprocessed command in the command buffer. The command

can then be processed using RNS, RSS, or DROP. Use of

ENS with CCL procedure ﬁles produces unexpected results.

This command is valid only when auto mode is not set.

Allows entry of the commandl and command2 commands

from the keyboard and sets auto mode.

Sets the job time limit to timlmt (77777s speciﬁes no limit).

Sets register Xi = cont in the exchange package area.

Sets register Xi = cont, left-justiﬁed, in the exchange package

area.

Sets register Xi to characters in display code.

Sets byte b of register Xi to value.

Sets forced error ﬂag (FSE), terminates program execution,

and clears auto mode if set.

Restarts a program that has paused.

DIS relinquishes the display console, but the job is held at

the present status. The console must be reassigned to

continue use of DIS.

Enters characters as a CPU program command. Data is

stored at RA + CCDR.

Sets direct CPU input mode. Characters entered from

keyboard are passed one character at a time, right-justiﬁed,

directly into central memory at RA + CCDR. Pressing the left

blank key on the CC545, the Esc key twice on the CC598B,

or the K key twice on the CC634B clears direct CPU

input mode.

Revision M DIS Operations 4-15

DIS Keyboard Entries

Command Description

OFFSWs.

ONSWs.

026.

RCP.

RCS.

RNS.

ROLLOUT.

ROLLOUT,spr.

RSS.

RSS,commandstring.

SCS.

SULuserindex.

T,addr.

Turns off sense switch s for the job (1 ^ s ^ 6).

Sets sense switch s for the job (1 ^ s ^ 6).

Calls the 026 ﬁle editor to a control point. Refer to the NOS

Version 2 Systems Programmer's Instant for a description of

the editing commands.

Requests central processor. Depending on job priority,

execution begins at the next program address for a job

suspended by a DCP request.

Sets auto mode, which initiates automatic command processing.

All succeeding commands are read from the command buffer

and processed automatically until an SCS command, an erase

function, or an error is encountered. A period (.) may also be

used to initiate automatic command processing.

Reads and processes the next command in the DIS command

buffer.

Allows the job to roll out. This command should be issued

when the message ROLLOUT REQUESTED appears (or the *

or F15 key may be used).

Places job in rollout status for spr job scheduler delay

intervals. The job is automatically rolled back in after this

period of time. If a number greater than 7778 is speciﬁed for

spr, 7778 is used.

Reads the next command from the command buffer and stops

prior to CPU execution. This is used to initiate breakpointing

of a program.

Reads the commandstring command and stops prior to

execution. The action is similar to ENS followed by RSS

except that the command buffer is not cleared.

Clears auto mode, which stops automatic command processing.

Allows access to a user index above AUIMX (377700s). Any

permanent ﬁle activity that is to be done on such user indices

must be done through system origin jobs. This command is not

accepted by a secured system.

Changes the T display to start at address addr.

4-16 NOS Version 2 Analysis Handbook Revision M

DIS Keyboard Entries

Command

U,addr.

UCC=c

V,addr.

X.commandstring.

* commandstring.

commandstring.

Description

Changes the U display to start at address addr.

Sets the uppercase character c. This command does not

terminate with a period.

Changes the V display to start at address addr.

Processes commandstring as the next command. Only the ﬁrst

50 characters following X are used. This may be used to enter

a leading slant or a command that is the same as DIS

display.

If an asterisk (*) followed by a blank and commandstring is

encountered during automatic command processing (auto

mode), commandstring is interpreted as a direct DIS display

selection command. For example, * C4,100. will set the left

screen display to the central memory C display at address

100. Using this feature, it is possible to set up procedure ﬁles

that use DIS to breakpoint a program to a desired stopping

point.

Processes commandstring as a command if it is not a

recognizable DIS command.

Calls the 026 ﬁle editor to a control point. Refer to the NOS

Version 2 Systems Programmer's Instant for a description of

the editing commands. (This command is the same as the

026 command.)

Revision M DIS Operations 4-17

Memory Entry Commands

The following commands are used in conjunction with the C, D, F, G, and M memory

displays to change the contents of central memory and extended memory. Only

locations relative to the reference address (RA) of the job to which DIS is assigned can

be changed. When changing the contents of memory relative to a job, the negative ﬁeld

length area of the job can be accessed by specifying a negative address. For example,

to change the content of RA-3, enter the address 777775s.

On a secured system no memory entry commands are allowed unless the security

unlock status is set. The memory display shows the message

••••SECURED AREA****

instead of the contents of the memory locations to prevent you from examining these

locations; you may not alter the contents of these locations.

Character values or numeric data can replace the current word contents. Either one

12-bit byte, one 15-bit parcel, one 30-bit parcel, or 60 bits can be changed. A single ""^

byte can be changed by inserting the byte number after the location to be changed;

bytes are numbered 0 through 4 from left to right. The address and contents are

assembled right-justiﬁed with leading zero ﬁll. Leading zeros may be omitted in the

entry. Only words within the ﬁeld length of the job may be changed.

CAUTION

Improper use of these commands may result in damage to the system or to user jobs.

Formats and descriptions of the memory entry commands follow. )

Command Description

addr,cont. Changes the contents of memory location addr to

o r c o n t . T h e s e c o n d f o r m o f t h e c o m m a n d p e r f o r m s

addr+cont. essentially the same function but leaves the address at

addr + 1, allowing immediate entry for the next memory

location.

a d d r , b , c o n t . C h a n g e s t h e c o n t e n t s o f b y t e b a t m e m o r y l o c a t i o n / ^ ^

or addr (eight digits) to cont. Each location

addr+b,cont. consists of ﬁve 12-bit bytes, numbered 0 through 4 from left

to right. The contents are octal characters. The second form of

the command performs essentially the same function but

leaves the address at addr+1, allowing immediate entry for

the next memory location.

2. If the message REPEAT ENTRY is displayed above the entry line, the cont ﬁeld is not cleared and may ^^S

be entered in successive memory locations as many times as desired by pressing CR or NEXT. The repeat

entry mode is enabled by pressing CR or NEXT before initial entry of the command. This is also applicable

to the b and n ﬁelds of the second, ﬁfth, and seventh commands.

4-18 NOS Version 2 Analysis Handbook Revision M

Memory Entry Commands

Command

igffrSi

Description

addr,Dcont.

addr + Dcont.

addr,Lcont.

addr + Lcont.

addr,In,cont.

addr + In,cont.

Eaddr ,cont.

Eaddr + cont.

Eaddr,b,cont.

Eaddr+ b,cont.

Eaddr,Dcont.

Eaddr + Dcont.

Changes the contents of memory location addr (eight digits)

to display code characters cont (left-justiﬁed and

zero-ﬁlled). The second form of the command performs

essentially the same function but leaves the address at

addr+1, allowing immediate entry for the next memory

location.

Changes the contents of memory location addr (eight digits) to

left-justiﬁed cont. The second form of the command performs

essentially the same function but leaves the address at

addr + 1, allowing immediate entry for the next memory

location.

Changes the contents of instruction n (0 through 3 from

left to right) at memory location addr (eight digits) to

cont; cont may be a 15- or 30-bit instruction. However, one or

more bits must be set in the upper 15 bits of a 30-bit

instruction or the entry will be treated as a 15-bit instruction.

The second form of the command performs essentially the

same function but leaves the address at addr + 1, allowing

immediate entry for the next memory location.3

Changes the contents of extended memory location addr to

cont. The second form of the command performs essentially

the same function but leaves the address at addr + 1, allowing

immediate entry for the next extended memory location.

Changes the contents of byte b at extended memory location

addr to cont. Each location consists of ﬁve 12-bit bytes,

numbered 0 through 4 from left to right. The contents are

four octal characters. The second form of the command

performs essentially the same function but leaves the address

at addr + 1, allowing immediate entry for the next extended

memory location.

Changes the contents of extended memory location addr to

display code characters cont (left-justiﬁed and

zero-ﬁlled). The second form of the command performs

essentially the same function but leaves the address at

addr + 1, allowing immediate entry for the next extended

memory location.

3. If the message REPEAT ENTRY is displayed above the entry line, the cont ﬁeld is not cleared and may

be entered in successive memory locations as many times as desired by pressing CR or NEXT. The repeat

entry mode is enabled by pressing CR or NEXT before initial entry of the command. This is also applicable

to the b and n ﬁelds of the second, ﬁfth, and seventh commands.

Revision M DIS Operations 4-19

PP Call Commands

Any PP program having a name that begins with a letter may be initiated by DIS.

However, before entering any of these commands, it is necessary to have a working

knowledge of the PP program to be called. This ensures correct use of the speciﬁed

program.

CAUTION

Improper use of these commands may result in damage to the system or to user jobs.

In table 4-1, prg denotes the name of the PP program and n is the control point

number.

Table 4-1. PP Call Formats

Command Description

prg. Calls PP program prg to the control

point.

prg,pl.* Calls PP program prg to the control

point; pl is an octal parameter required

by prg.

prg,pl,p2. Calls PP program prg to the control

point; pl and p2 are octal parameters

required by prg.

Format of PP

Call Initiated

18/3Lprg,6/n,36/0

18/3Lprg,6/n,18/0,18/pl

18/3Lprg,6/n,18/pl,18/p2

/^®Sk

4-20 NOS Version 2 Analysis Handbook Revision M

0ms

0^ DSD Commands 5

Display Selection Commands 5.2

Dayﬁle Commands 5.3

Queued File Utility Commands 5.4

Job Processing Control Commands 5.4

General Job Control Commands 5-4

Interactive Job Control Commands 5-8

Job Communication Commands 5-9

Job Flow Commands 5-10

Peripheral Equipment Control Commands 5-24

Subsystem Control Commands 5-50

System Control Commands 5-56

Secured System Control Commands 5-70

Memory Entry Commands 5-72

Channel Control Commands 5-74

Extended Memory Flag Register Commands 5-75

r^ Breakpoint Package Commands , 5-75

- PP Breakpoint Commands 5-75

V Display 5-77

PP Breakpoint Precautions 5-79

CPUMTR Breakpoint Commands 5-80

CPUMTR Breakpoint Display (C Display) 5-81

CPUMTR Breakpoint Precautions 5-83

f^ Debugging Commands 5-84

V^ J

DSD Commands

After the system has been deadstarted successfully, you can enter the DSD commands

necessary to provide optimum performance and reliability for users. The following

general categories of DSD commands are available for this purpose.

Category Description

Display selection

Dayﬁle

Queued ﬁle utility

Job processing control

Peripheral equipment control

Subsystem control

System control

Memory entry

Channel control

Extended memory ﬂag register

Breakpoint package

Debugging commands

Selects DSD displays.

Dumps the system, account, or error log

dayﬁle to a speciﬁed device.

Provides control over selected queued ﬁles.

Provides added control over job scheduling and

processing.

Controls the peripheral equipment available to

the system.

Schedules a subsystem to a control point or

terminates a current subsystem.

Maintains system integrity in a normal

production environment or debugs a system

that is in an abnormal state.

Changes the contents of central memory and

extended memory.

Controls activity on a speciﬁed data channel

in circumstances where abnormal hardware

and/or system operation is detected.

Clears and sets bits in the extended memory

ﬂag register.

Provides control over PP breakpoint processing

and CPUMTR breakpoint processing.

Traps certain conditions and traces selected

pool PP to CPUMTR and MTR to CPUMTR

functions.

Although all DSD commands are generally available, many of them are seldom used in

a normal production environment. Many DSD commands are used only by the system

analyst for maintenance or debugging. These commands include all memory entry and

channel control commands as well as several commands in the other categories listed

on an unsecured system. Memory entry and several other commands are restricted on a

secured system (refer to the UNLOCK command later in this section).

When unusual problems arise, do not attempt corrective action unless you have

considerable experience relating to the current problem. Misguided attempts to correct

a system problem can often destroy information required to successfully analyze the

problem.

Revision M DSD Commands 5-1

Display Selection Commands

To assist customer engineers and software analysts in tracking problems, the system

enters the ﬁrst characters (up to 25) of the following commands into the error log ^^\

after each execution.

DOWN,param.

FORM,param.

IDLE,param.

INITIALIZE,param.

LOG,param.

OFF,param.

ON,param.

REDEFINE.param.

99.

Memory entry commands

Channel control commands

Any other commands that you specify at installation time

These commands are preﬁxed by the characters DS in the error log but otherwise

appear exactly as they are entered. This feature can be enabled and disabled using the "^)

99 command (refer to System Control Commands later in this section).

The manner in which the DSD commands are entered and the use of special keyboard

characters are described in the NOS Version 2 Operations Handbook. Command

formats are ﬁxed ﬁeld; that is, the ﬁelds in the command format must be speciﬁed as

shown. Embedded blanks are allowed in octal ﬁelds. Leading spaces in command

entries are not allowed.

Display Selection Commands ^

The system display program DSD generates system-oriented displays. You can select

any of the DSD displays with the following DSD command:

xy.

where x and y represent the letter designation (screen identiﬁer) of the displays.

Display x appears on the left screen and display y appears on the right screen. If x

and y are identical, both screens display the same information, except for the B and ^^

P displays when using the CC598B or CC634B console. Refer to the NOS Version 2 '

Operations Handbook for details concerning the DSD display formats and keyboard

operating instructions. You can specify a sequence of DSD displays that you want

displayed on the left screen. To preselect the left screen display sequence, enter the

following DSD command:

SET,screens.

where screens represents a string of four letters (screen identiﬁers) designating any

four DSD displays. Usually you specify four different displays although DSD accepts

any four valid screen identiﬁers. Refer to the NOS Version 2 Operations Handbook for

DSD displays and keyboard operating instructions.

5-2 NOS Version 2 Analysis Handbook Revision M

Dayﬁle Commands

The system saves messages in ﬁve types of dayﬁles.

• Account dayfile

• Binary maintenance log

• Error log

• Job dayfile

• System dayfile

The account dayﬁle keeps a record of all resources charged to a job. This dayﬁle can

be used for customer billing and other accounting purposes. The binary maintenance

log records the information used in Control Data maintenance in binary format. The

error log records system error messages, such as disk errors. Job dayﬁles keep entries

for individual jobs. The system dayﬁle keeps a history of all commands for all jobs

processed.

The following commands dump the account dayﬁle, system dayﬁle, or error log to a

systemrdeﬁned mass storage device. The resultant mass storage ﬁle is put in the

output queue for printing. The system automatically prints the job dayﬁle at the end

of the job's output.

Command Description

^wn X.AFD. Requests that the account dayﬁle be dumped to a

system-deﬁned mass storage device. The resultant mass

storage ﬁle is put in the output queue for printing.

X.DFD. Requests that the system dayﬁle be dumped to a

system-deﬁned mass storage device. The resultant mass

storage ﬁle is put in the output queue for printing.

X.ELD. Requests that the error log be dumped to a system-deﬁned

mass storage device. The resultant mass storage ﬁle is put

in the output queue for printing.

Refer to section 18, Queue/Dayﬁle Utilities, for more information on dayﬁle dumps.

The binary maintenance log is designed to be processed through an interpreter

program, and therefore is normally dumped to tape or disk.

Refer to the NOS Version 2 Operations Handbook for descriptions of dayﬁle displays.

Revision M DSD Commands 5-3

Queued File Utility Commands

The following commands are used to control queued input and output ﬁles.

Command Description

X.QREC Deactivates or activates selected queued ﬁles and purges selected

inactive queued ﬁles.

X.QMOVE Moves queued ﬁles from one mass storage device to another.

Refer to section 18, Queue/Dayﬁle Utilities, for more information on queued ﬁle utility

commands.

Job Processing Control Commands

Under normal circumstances, control over job processing is performed automatically by

the system. Although the following commands may not be used frequently, they provide ^_

an added measure of control over job processing. ^

General Job Control Commands

Several of the commands described here change internal system parameters which

control job scheduling and processing. Give careful consideration to their use since job

ﬂow and overall system performance can be affected. Refer to the individual command

descriptions for further information.

Command Description >*^\

CKPJsn. Checkpoints the job with job sequence name jsn. The

checkpoint information includes a copy of the job's ﬁeld length,

the system information used for job control, and the name and

contents of all local ﬁles currently assigned to the control

point. It is the responsibility of the user's job to establish a

magnetic tape or mass storage permanent ﬁle to receive the

checkpoint information. Otherwise, checkpoint information is

automatically written to a local ﬁle named CCCCCCC and is

not available if a restart becomes necessary (restart is ■^s\

user-controlled and is accomplished using the RESTART

command). Refer to the NOS Version 2 Reference Set, Volume

3 for supplementary information concerning the

checkpoint/restart feature available to users.

NOTE

If the current job command has secure system memory (SSM)

status set, the job cannot be checkpointed. SSM status is set

for certain jobs to prevent dumping of the job's ﬁeld length.

Subsystems cannot be checkpointed.

^3ss!\

5-4 NOS Version 2 Analysis Handbook Revision M

General Job Control Commands

Command

DROP,jsn,qt,ujn.

ENPR,jsn,pr.

ENPRJsn,*.

ENQP,jsn,pr.

Description

Drops the job with job sequence name jsn from the queue qt

where it currently resides. If a jsn is not speciﬁed, a 1- to

7-character user job name ujn can be speciﬁed after the queue

type.

The console must be unlocked to use this form of the DROP

command. If no jsn or ujn is speciﬁed, all jobs in the speciﬁed

queue type are dropped. If the queue type is not speciﬁed, the

default is the executing queue.

The DROP command cannot be used to terminate a subsystem.

Also the DROP command will not terminate an interactive

user's session (this can be done only with a KILL command).

The queue type is one of the following:

qt Queue Type

ALL

All jobs and queued ﬁles.

All jobs in the executing queue (including the rolled

out jobs).

All jobs in the input queue.

All jobs in the plot queue.

All jobs in the print queue.

All jobs in the punch queue.

All jobs in the wait queue.

Sets CPU priority to pr (2 ^ pr ^ 70s).

Sets CPU priority to the minimum non-decrementing

value, 60s.

Enters queue priority of pr for a queue type ﬁle with job

sequence name jsn. The value of pr can range from the

minimum to the maximum for the job's service class. If the

priority is zero, an input ﬁle is not scheduled back to a

control point automatically. The value speciﬁed overrides the

current queue priority for the ﬁle. The current queue priority

can either be increased or decreased using this command.

Revision M DSD Commands 5-5

General Job Control Commands

Command

KILLjsn.

Description

Drops the job with job sequence name jsn from the executing

job table (EJT) without exit processing. If you want the job to

proceed with exit processing, use the DROP command. The

KILL command cannot be used to drop a subsystem.

KILL is the only command that immediately terminates an

interactive user's session. The DROP and OVERRIDE

commands only terminate the current job step and do not log

the user out of the system. There will be no recoverable job

after you issue the KILL command.

NOTE

Before initiating the command, ensure that the correct job

sequence name has been speciﬁed.

OVERRIDEJsn.

In some cases, a KILL command will be intercepted by a job's

reprieve processing. If the job does not terminate after

ﬁnishing its reprieve processing, issue another KILL command

to terminate the job.

Certain types of job processing are unaffected by the DROP,

KILL, or STOP commands. These include operations such as

setting permanent ﬁle device interlocks, interlocking

ﬁles/tracks, clearing VSN entries, and waiting for certain

types of tape/PP activity to end. The OVERRIDE command

terminates this type of processing and drops the job with job

sequence name jsn from the executing job table regardless of

queue priority. The OVERRIDE command will not terminate

an interactive user's session (this can be done only with a

KILL command). Unlock the console (refer to the UNLOCK

command later in this section) to enter this command. Use of

this command is recorded in the error log.

NOTE

Exercise extreme caution in using the OVERRIDE command.

Undesirable situations (such as interlocks being left set, VSN

entries remaining uncleared, or certain tape/PP activities left

outstanding) could occur that would cause potentially

damaging system activity. Never use this command during

normal operations.

5-6 NOS Version 2 Analysis Handbook Revision M

General Job Control Commands

Command

RERUNJsn.

ROLLIN,jsn,L.

ROLLOUT,jsn,sd.

Description

Terminates the job with job sequence name jsn, then reruns

the job from the beginning. The job must be in rerun status

as set by the RERUN command or macro.

Allows the job deﬁned by job sequence name jsn to be

scheduled to an available control point. L is optional; if

omitted, the job can be selected by the scheduler for rollout.

Removes the currently executing job with job sequence name

jsn and makes it a rolled out job. A subsystem cannot be

rolled out. sd is the number of scheduler intervals before the

job can be scheduled again. The acceptable range for sd is

between 0 and 777s. If sd is not present or is 0, the job is not

scheduled back to a control point automatically. That is, your

action is required to return the job to a control point. This

can be done by using the ROLLIN command.

The amount of time required for one job scheduler interval is

initially set in the IPRDECK but may be changed by using

the DELAY command (JS parameter) described later in this

section. Normally, it is a 1-second interval.

Revision M DSD Commands 5-7

Interactive Job Control Commands

a&$S

Interactive Job Control Commands

The following commands apply only to interactive online jobs. The interactive facility

subsystem must be active at control point 1. Refer to the CDCNET Network Operations

Manual for information about sending messages to terminal users that are connected

through CDCNET.

Command

WARN,messagetext.

Description

DIAL,jsn,messagetext. Sends message messagetext (48 characters maximum) to the

terminal currently assigned to the job with job sequence name

jsn. Examine the T display to determine the appropriate job

sequence name. The message is sent to the terminal

immediately except when output is being sent to the terminal.

In that case, the message follows the output data.

Sends message messagetext (48 characters maximum) to all

terminals currently logged into IAF. The message is received

at a terminal upon completion of the current command or at

the end of a job step. Each subsequent terminal that logs into

IAF also receives this message. This continues until either a

new message is entered or the message is cleared by using

the WARN command. In addition, the current message also

appears at the IAF Subsystem control point on the B,0

display.

When sent to an interactive terminal, the message

messagetext is always preceded by the statement

hh.mm.ss. WARNING

specifying the time (hours.minutes.seconds) when you entered

the WARN command.

For example, if you enter

WARN,SYSTEM SHUTDOWN AT 1500.

the following information would be transmitted to all

interactive terminals.

hh.mm.ss. WARNING

WARN.

SYSTEM SHUTDOWN AT 1500.

This command is typically used to notify interactive users of

an interruption in service or system shutdown.

Clears the message entered by the WARN,messagetext

command. Unless this command is entered, the existing

message (if any) continues to be transmitted to each new

terminal that logs into IAF.

5-8 NOS Version 2 Analysis Handbook Revision M

Job Communication Commands

The following job communication commands are used to communicate with a job

currently in the executing job table.

Command Description

CFO,jsn,messagetext.

COMMENTjsn,

messagetext

GO, jsn.

OFFSW,jsn,si,S2,...,se.

ONSW,jsn,si,s2,...,S6.

PAUSEJsn.

Sends a message messagetext (36 characters maximum) from

the console to the job with job sequence name jsn. Bit 14 of

the job's RA must be set before the CFO command is accepted.

The message is placed in locations RA + 70s through RA + 74s

of the program's ﬁeld length.

Enters comment messagetext (48 characters maximum) in the

dayﬁle for the job with job sequence name jsn.

Clears the pause bit of the job with job sequence name jsn. A

job may set the pause bit if an error is encountered or if an

operator response is required. If jsn is not speciﬁed,, the

command applies to the system control point.

Turns off one or more sense switch(es) si (1 ^ si ^ 6) of the

job with job sequence name jsn. Refer to Subsystem Control

Commands later in this section for deﬁnitions of the sense

switches that can be set for the BIO, IAF, MSE, and TAF

Subsystems.

Turns on sense switch(es) si (1 ^ si ^ 6) of the job with job

sequence name jsn. Refer to Subsystem Control Commands

later in this section for deﬁnitions of the sense switches that

can be set for the BIO, IAF, MSE, and TAF Subsystems.

Sets the pause bit of the job with job sequence name jsn. If

jsn is not speciﬁed, the command applies to the system control

point.

Revision M DSD Commands 5-9

Job Flow Commands

The following job ﬂow commands affect scheduling and execution of jobs in the system. )

Improper use of these commands can drastically hamper job ﬂow as well as system

performance. In certain cases, jobs may be lost.

The priority associated with each priority parameter in the following commands is

established by an entry in the IPRDECK for each service class. The value of each

priority parameter for each service class is listed on the system control display(s)

(refer to the NOS Version 2 Operations Handbook for more information on the S

display).

Command Description

CLASS,ot,sci,sc2,.-.scn. Deﬁnes the authorized service class for each origin type.

Refer to the SCTD L display in section 9, L-Display

Utilities, for further information.

Parameter Description Ams

ot Origin type (must be batch, remote batch, or

interactive). This parameter is order

dependent. Types with brief descriptions

follow:

BC Batch.

RB Remote batch.

IA Interactive. "^%^

sci Authorizes service classes for the speciﬁed

origin type. Entering a service class that

already has validation clears validation for

that service class. Service classes with brief

descriptions (in parentheses) follow:

BC (Local batch).

CT (Communication task). ""^

DI (Detached interactive).

In [Installation class n (0 ^ n ^ 3)].

MA (Maintenance).

NS (Network supervisor).

RB (Remote batch).

SY (System).

TS (Interactive).

ALL (Sets all classes except SS and DS).

NUL (Clears all service classes). )

5-10 NOS Version 2 Analysis Handbook Revision M

Job Flow Commands

0^S

rCommand

CPTT=n

DELAY,pi,p2,...,pn.

J0^!s\

Description

Changes the CPUPFM transfer threshold (CPTT) to the value

n (0 *s n ^ 77778 PRUs). The value of CPTT determines

whether an indirect access ﬁle transfer will be processed

entirely within the PP program PFM or whether the CPU

program CPUPFM will be called to perform the transfer. Since

CPUPFM can transfer large ﬁles faster than PFM, CPUPFM

will transfer ﬁles that exceed the threshold value.

The default value is 100s sectors, as deﬁned in common deck

COMSPFM. Depending on the conﬁguration, you may want to

adjust this value up or down. A site with buffered I/O devices,

for example, may ﬁnd it advantageous to set a lower value.

To disable CPUPFM transfers altogether, enter CPTT=0. The

current value of CPTT is displayed on the W,M display.

Alters current system delay parameters. Examine the W,M

display to determine the current delay parameter values.

ARar

CIci

CRcr

JQjq

JSjs

Description

PP recall interval in milliseconds. This parameter

speciﬁes the time interval after which a peripheral

processor unit (PP) in the PP recall queue will be

recalled, ar ranges from 1 to 7777s.

CPU job priority interval in milliseconds. This

parameter speciﬁes the time interval after which

the priorities of jobs in the wait queue are

incremented, ci ranges from 0 to 7777s.

CPU recall period in milliseconds. This parameter

speciﬁes the amount of time a job remains in recall

(X status) when an RCL request is placed in

RA+1. cr ranges from 1 to 7777s.

Exponent used to determine the input ﬁle

scheduling interval in seconds, jq ranges from 0 to i

148. The interval in seconds between scheduling of

input ﬁles is calculated as follows:

interval = 2**jq

Job scheduler interval in seconds. This parameter

speciﬁes the interval in which the job scheduler is \

called. The scheduler may also be called at other

times, js ranges from 1 to 7777s.

Revision M DSD Commands 5-11

Job Flow Commands

Command Description

| DELAY,pi,p2,...,pn. (Continued)

Pi Description

MPmp Memory padding value expressed as 100s word

blocks, mp ranges from 0 to 777s. This

parameter speciﬁes how much additional

(unassigned) memory should be allocated

between the end of the newly assigned job ﬁeld

length and the beginning of the next job.

Increasing this value reduces the probability

that a job will be storage moved in response to

a request for more memory.

The value for each system delay parameter may be established

by using a DELAY command entry in the IPRDECK. If no

DELAY entry is present, default values are provided.

Figure 5-1 provides space to record the original values

(speciﬁed in the W,M display) in the event that any are

altered temporarily. For additional information concerning the

DELAY command, refer to section 3, Deadstart Decks.

DELAY VALUES

Figure 5-1. Record of Original Values in W,M Display (Delay Values)

5-12 NOS Version 2 Analysis Handbook Revision M

Job Flow.Commands

rCommand Description

PCLASS,sco,sci,...,sc7. Selects the service class associated with each priority level

(pothrough P7) on the Job command. This parameter is order

dependent; a comma must appear for any parameters not

speciﬁed. The default for any parameters not speciﬁed is 0

(zero). Subsystem (SS) service class is not allowed with this •

command. Refer to the SCTD L display in section 9, L-Display

Utilities, for more information.

sci is a 2-character service class parameter. Entries are listed

and brieﬂy described below.

sci Description

BC Local batch.

CT Communication task.

DI Detached interactive.

In Installation class n (0 *s n ^ 3).

MA Maintenance.

NS Network supervisor.

RB Remote batch.

SY System.

TS Interactive.

Revision M DSD Commands 5-13

Job Flow Commands

Command

QUEUE,sc,qt,qpi,

qp2„...,qpn.

Description

Alters the queue priorities qpi associated with the input,

executing, and output queues qt for each service class sc.

Examine the L display SDSPLAY to determine the priority

values currently associated with each service class.

sc Description

BC Local batch.

CT Communication task.

DI D eta ch ed i n te ra c ti ve.

In Installation class n (0 ^ n =£ 3).

MA Maintenance.

NS Network supervisor.

R B R e m o t e b a t c h .

SS Subsystem.

SY System.

T S I n t e r a c t i v e .

There are three job queue types.

qt Description

Executing.

Input.

Output.

5-14 NOS Version 2 Analysis Handbook Revision M

Job Flow Commands

Command Description

{' QUEUE,sc,qt,qpi, (Continued)

qp2„...,qpn.

There are ﬁve priorities.

qpi Description

ILil Priority a job receives when it initially exhausts I

its CM time slice (refer to the SERVICE

command CM parameter). Thereafter, whenever |

the CM time slice is exhausted, the job's

priority will be set to the value of lp. il ranges |

from 0 to 77778, but must be in the range of

values for lp and up. This parameter is valid

only for executing jobs (EX).

IPip Initial scheduling priority for an executing batch!

job or for an interactive job when a terminal

command is entered. Online interactive jobs withf

terminal I/O available are scheduled at tp

priority (refer to the SERVICE command TP

parameter), ip ranges from 0 to 7777s, but mustf

be in the range of values for lp and up.

LPlp Lowest priority. For an input or output queue

ﬁle, the priority assigned to a ﬁle or job

entering the designated queue. For an executing!

job, the priority assigned to a job which has

exceeded a non-initial CM slice. The IL value is I

used for the ﬁrst CM slice, lp ranges from 0 to |

77778, but must be less than the value of up.

UPup Highest priority. For input and output queues,

this is the highest priority a ﬁle can reach in

that queue; aging stops when this priority is

reached. For the execution queue, this is the

priority assigned to a job when initially assigned;:

to a control point, up ranges from 0 to 7777s,

but must be greater than the value of lp.

WFwf Weighting factor wf for queue priority

calculation, wf must be a power of 2, from 1

to 40008. The smaller the weighting factor, the 1

faster the queue entry reaches its highest

priority.

The priority associated with each queue is established by

using QUEUE command entries in the IPRDECK for each

service class. These entries normally reﬂect the ideal queue

priorities for the job mix of the particular installation. The

values speciﬁed in the IPRDECK are considered critical to

optimum system performance and are not normally altered.

Figure 5-2 provides space to record the original values

speciﬁed in the L display SDSPLAY. For additional

information on the QUEUE command, refer to section 3,

Deadstart Decks.

Revision M DSD Commands 5-15

Job Flow Commands

JOB

SERVICE

CLASS

PRIORITIES FOR EACH QUEUE TYPE

INPUT QUEUE EXECUTIN6 QUEUE OUTPUT QUEUE

LP UP UF IL IP LP UP UF LP UP UF

Figure 5-2. Record of Original Values in L Display SDSPLAY

yS^s.

>«C^v

5-16 NOS Version 2 Analysis Handbook Revision M

Job Flow Commands

JP*\

Command Description

SERVICE,sc,pi,p2,. ..,pn. Alters the service limits pi associated with each service

class sc.

sc Description

BC Local batch.

CT Communication task.

DI Detached interactive.

In Installation class n (0 ^ n ^ 3).

MA Maintenance.

NS Network supervisor.

RB Remote batch.

SS Subsystem.

SY System.

TS Interactive.

Pi Description

AMam Maximum ﬁeld length/100s for- all jobs of the

speciﬁed service class. This parameter is used

to partition central memory by limiting the

total ﬁeld length available to each service class.

For example, if scheduling a job to a control

point would cause the AM value to be exceeded

for its service class, it may not be scheduled

until the required ﬁeld length is available.

This means that a lower priority job from a

different service class may be scheduled ﬁrst.

However, a job that would cause the AM value

for its service class to be exceeded can be

scheduled to a control point if there are not

enough jobs in other service classes to ﬁll the

central memory and if FLEXIBLE PARTITIONS

is enabled. The system always attempts to use

central memory to its greatest capacity, am

ranges from 0 to 77777s.

Revision M DSD Commands 5-17

Job Flow Commands

Command Description

SERVICE ,sc,pi ,p2,... ,pn. (Continued)

pi Description

CBlpup

CMcm

CPcp

Upper and lower bound CPU priorities.

lp is the lower bound CPU priority,

lp ranges from 0 to 77s.

up is the upper bound CPU priority,

up ranges from 0 to 77s, but must be

equal to or larger than the value of lp.

Central memory time slice in seconds. This

parameter speciﬁes the maximum amount of time

a job of the speciﬁed service class can remain in

central memory, either at a control point or at a

pseudo-control point, before its priority is set to

the lower boundary. When the job initially

exhausts its CM time slice, this lower boundary

is the value of il. Subsequently, when the job

exhausts its CM time slice, this lower boundary

is the value of lp (refer to the QUEUE command

IL and LP parameters), cm ranges from 0 to

77778.

Control point slice priority. This parameter

speciﬁes the value of the scheduling priority

which will be set after a job has been at a conrol

point for ct seconds. The priority speciﬁed must

be greater than or equal to the lower bound and

less than or equal to the upper bound execution

queue scheduling priorities speciﬁed by the CB

parameter (lp ^ cp ^ up). Refer to the QUEUE

command LP and UP parameters.

'"<%Y

^^S

5-18 NOS Version 2 Analysis Handbook Revision M

Job Flow Commands

jjgfPS

Command Description

SERVICE,sc,pi ,p2,...,pn. (Continued)

CScs

J&Ray

CTct

Description

Cumulative size in PRUs allowed for all indirect

access permanent ﬁles; cs indicates a limit value

for the cumulative size:

0 Unlimited

1 10008

2 5000s

3 50000s

4 1000008

5 2000008

6 400000s

7 Unlimited

Control point time slice in seconds. This

parameter speciﬁes the maximum amount of time:

that a job of the speciﬁed priority can remain at j

a control point before its priority is changed to

the control point slice priority (cp). ct ranges

from 1 to 77778.

JfP*V

0ms

Revision M DSD Commands 5-19

Job Flow Commands

Command Description

SERVICE,sc,pi,p2,...,pn. (Continued)

DSds

DTsc

ECec

EMem

Description

Size in PRUs allowed for individual direct

access permanent ﬁles; ds (0 through 7) is used

to specify the corresponding values shown next:

0 Unlimited

1 10008

2 50008

3 500008

4 100000s

5 2000008

6 4000008

7 Unlimited

Service class to which a detached job is

assigned. The default value for sc is DI

(detached interactive).

Maximum user-accessible extended memory ﬁeld

length in words divided by UEBS* for any job

of the speciﬁed service class. This parameter

performs the same function for extended memory

ﬁeld length that the FL parameter does for

central memory ﬁeld length, ec ranges from 0 to

3777s.

Maximum extended memory divided by UEBS

for all jobs of the speciﬁed service class. This

parameter performs the same function for

extended memory ﬁeld length that the AM

parameter does for central memory ﬁeld length,

em ranges from 0 to 3777s.

1. Refer to Extended Memory Overview in section 3 to determine the value of UEBS.

5-20 NOS Version 2 Analysis Handbook Revision M

Job Flow Commands

Command Description

SERVICE,sc,pi,p2,...,pn. (Continued)

FCfc

FLﬂ

FSfs

Description

Number of permanent ﬁles allowed; fo

(0 through 7) is used to specify the

corresponding values (limit value in octal) shownf

0 Unlimited

1 108

2 40s

3 1008

4 2008

5 10008

6 40008

7 Unlimited

Maximum ﬁeld length divided by 100s for any

job of the speciﬁed service class. If more

memory is requested, the job is aborted. You

typically use this parameter to limit the

memory requirement for jobs of a speciﬁc

service class during certain hours of the day.

For example, you may use the FL parameter to I

specify a maximum ﬁeld length for all batch

service class jobs between the hours of 2 p.m.

and 4 p.m. ﬂ ranges from 0 to 7777s.

Size in PRUs allowed for individual indirect

access permanent ﬁles; fs (0 through 7) is used |

to specify the corresponding values (limit value |

in octal) shown next:

0 Unlimited

1 108

2 30s

3 1008

4 300s

5 10008

6 20008

7 Unlimited

Revision M DSD Commands 5-21

Job Flow Commands

Command Description

SERVICE ,sc,pi ,p2,... ,pn. (Continued)

SRST,t.

Description

NJnj Maximum number of jobs. For each service

class, this parameter speciﬁes the number of

jobs that can be executing in the system, nj

ranges from 0 to 77778.

SDsd CPU job switch delay in milliseconds, sd ranges

from 0 to 77778.

TDtd Suspension time-out delay. A suspended job will

not be timed out for td*10s seconds, td ranges

from 0 to 77778. The maximum delay is

approximately 9 hours.

TPtp Terminal job priority assigned to a terminal job

that is rolled out waiting for terminal input, tp

ranges from 0 to 7777s, but must be in the

range of values for the execution queue

parameters lp and up (refer to the QUEUE

command LP and UP parameters).

The service limits associated with each service class are

established by using SERVICE command entries in the

IPRDECK. These entries normally reﬂect the ideal service

limits for the job mix of the particular installation. The

values speciﬁed in the IPRDECK are important to optimum

system performance and are not normally altered. However,

when changes are necessary they are usually temporary and

the original values will be reset. Figure 5-3 provides space

to record the original values speciﬁed in the L display

SDSPLAY. For additional information concerning the

SERVICE command, refer to section 3, Deadstart Decks.

Changes the secondary rollout sector threshold to the value t

(0 s£ t ^ 77778). Any rollout file smaller than t sectors is

considered a secondary rollout ﬁle for the purpose of

equipment selection.

NOTE

The size of the rollout ﬁle for any job must be at least

seven sectors larger than the combined size in sectors of the

job's central memory and extended memory ﬁeld lengths.

/^^\

A^^S.

■^^s

5-22 NOS Version 2 Analysis Handbook Revision M

Job Flow Commands

vﬁtftafy

/0^S

JOB

SERVICE

CLASS

SERVICE LIMITS

CB/L CP CT CM NJ TD CB/U FL AM TP AJ DT EC EM DS FC CS FS

Figure 5-3. Record of Original Values in L Display SDSPLAY

Revision M DSD Commands 5-23

Peripheral Equipment Control Commands

The commands described in this category provide control of the channel-connected

peripheral equipment available to the system. To control unit record equipment that is

connected through communications ports, use the RBF K display described in section 8

and the PSU K display described in the NOS Version 2 Operations Handbook. You

should become familiar with the following DSD displays, which are closely associated

with the use of these and other commands described throughout this section.

• Equipment status display (E,A).

• Disk configuration display (E,C).

• Disk errors display (E,E).

• Family status display (E,F).

• Disk thresholds display (E,H).

• Disk status display (E,M).

• Resource requests display (E,P).

• Tape status display (E,T).

• BIO status display (I).

A complete description of each of these displays is given in the NOS Version 2

Operations Handbook.

Command Description

ASSIGN,jsn,est. Assigns equipment deﬁned by EST ordinal est (normally a

tape unit) to the job with job sequence name jsn. This

command is entered in response to a ﬂashing REQUEST

message. Use of this command for assignment of a tape unit

should not normally be required because tape assignment is

performed automatically when a volume serial number (VSN)

is speciﬁed in the job request. However, if a VSN is not ^.

speciﬁed in the job request for a labeled or unlabeled tape, the j

REQUEST message appears at the job's control point (on the

B,0 display), and the ASSIGN command must be entered to

assign a tape unit to the job.

BKSP,est,rr. Backspaces rr octal logical records on the print ﬁle for the

BIO equipment deﬁned by EST ordinal est. When rr is not

speciﬁed, the default is one record.

BKSPF,est,ff. Backspaces ff octal ﬁles on the print ﬁle for the BIO

equipment deﬁned by EST ordinal est. When ff is not

speciﬁed, the default is one ﬁle.

" * )

2. Operation of peripheral equipment is described in the NOS Version 2 Operations Handbook.

5-24 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command Description

BKSPRU,est,ss.

CONTINUE,est.

DOWN,CH = ch,

EQ=est.

DOWN,CCH=cch,

EQ=est.

DOWN,MCH = ch,

EQ=est.

DOWN,EQ=est.

Backspaces ss physical record units (PRUs) on the print ﬁle

for the BIO equipment deﬁned by EST ordinal est. The PRU

count, ss, must be speciﬁed. There is no default setting.

Printing resumes at the beginning of a line.

Resumes printing on the BIO equipment deﬁned by EST

ordinal est.

The ﬁrst two forms of the command disallow the use of

channel ch or concurrent channel cch for the mass storage

device speciﬁed by EST ordinal est. If the EQ = est

parameter is omitted, channel ch or cch is disallowed for

all devices in the EST. The EQ=est parameter cannot be

speciﬁed if the CH=ch or CCH=cch parameter speciﬁes a

magnetic tape channel. Use of all ports of a concurrent

channel is disallowed.

If channel ch or cch is the only channel available to a mass

storage device, its use is disallowed providing the device is

down. If channel ch or cch is

• not defined as a magnetic tape or mass storage channel

• the last active channel on an equipment that is not

globally unloaded (status N on the E,M display)

• a nonimpact printer channel and any equipment on the

channel is logically ON or assigned to a control point

the command is ignored and the following message appears on

the left screen.

INCORRECT ENTRY

Correct the entry and retry the command.

The third form of the command disallows the use of MUX

channel ch for the equipment speciﬁed by EST ordinal est

(this form applies only to model 176). If the EQ=est

parameter is omitted, MUX channel ch is disallowed for all

equipments in the EST.

The fourth form of the command disallows the use of

equipment with EST ordinal est for all channels. It disallows

all access to the equipment by the operating system and user

jobs. This device state is used to allow diagnostic routines

exclusive use of the device for problem isolation and for

preventing access to a device that is being repaired.

Revision M DSD Commands 5-25

Peripheral Equipment Control Commands

Command Description

DOWN,CH=ch,

EQ=est.

(Continued)

This command also clears the SUSPCT state for a tape unit

that is currently unavailable (but not OFF or DOWN) because

an unrecovered hardware or tape I/O error has occurred on

the tape or unit. This command can be entered followed by

the ON,EQ=est command to restore the unit to service after a

customer engineer has checked it out, or the unit has been

tested following Control Data's recommended procedures:

You receive an error message if you attempt to DOWN

equipment:

• That already has a DOWN status.

• That is assigned to a job.

• That is the last system device not DOWN.3

• That contains the system dayfile, account file error log, or

maintenance log.

In these cases, the following message appears.

INCORRECT ENTRY

Correct the entry and retry the command.

NOTE

This command should be used with caution since it directly

affects the operation of system peripheral equipment.

DOWN,MID=mid. Conﬁrms that a machine running in low-speed port

multimainframe mode with machine identiﬁer mid is down.

Enter this command in response to a request at the system

control point of the form:

IF mid DOWN ENTER *D0WN,MID=mid*

Make this entry if the machine is down; otherwise, clear the

message by entering:

GO,SYS.

,A^^S

3. Applies only to mass storage equipment.

<^fe\

5-26 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command

END,est,rc.

yms

FORM,est,fc.

ID,est,id.

Description

Terminates the current operation on the BIO equipment

deﬁned by EST ordinal est. If est deﬁnes a line printer or

card punch, BIO then assigns the next available ﬁle to that

equipment. If est deﬁnes a card reader that is actively reading

cards when END,est is entered, the job terminates at the last

card read. The next card is treated as the beginning of a new

job. If another card deck follows the end-of-information card

(multipunch 6/7/8/9), it is processed normally.

If est deﬁnes a card reader that is stopped because of a

compare error when END,est. is entered, perform the following

steps:

1. Remove the remainder of the card deck, except the

end-of-information card, from the card reader input hopper.

2. Ready the card reader to read the end-of-information card.

The job terminates; and if another card deck follows the

end-of-information card, it is processed normally.

The re parameter cancels a portion of the repeat count

speciﬁed for that equipment (refer to the REPEAT command

later in this section). For example, if the current operation on

equipment est had been set to be repeated ﬁve times

(operation performed six times), entering a value of 4 for re

would permit the operation to be performed just two times. If

the repeat count is zero, this command performs the END

operation once.

Assigns a 2-character forms code fc to the BIO line printer or

card punch deﬁned by EST ordinal est. Only those ﬁles in the

output queue assigned the forms code fc are directed to

equipment est. A user can assign a forms code to an output

ﬁle using the ROUTE command. (For a description of the

ROUTE command, refer to the NOS Version 2 Reference Set,

Volume 3.) The value of the forms code can range from AA to

99. If the forms code is not present, the current forms code

ﬁeld is cleared (value is null).

Assigns a numeric identiﬁer id to the card punch, card reader,

or printer deﬁned by EST ordinal est. The value of the

identiﬁer can range from 00 to 67s. Only those ﬁles in the

proper queue with an identiﬁer equal to id are directed to the

card punch or printer deﬁned by est. Refer to the LOAD

command in this section to assign an identiﬁer to a queue

type ﬁle. All subsequent jobs loaded from card reader est are

assigned the identiﬁer id.

Revision M DSD Commands 5-27

Peripheral Equipment Control Commands

Command

IDLE,EQ=est.

Description

Use this command when you want to continue using the disk

storage device speciﬁed by EST ordinal est even though it is

failing occasionally. When the IDLE command is used in

conjunction with the THRESHOLD command, it allows some

degree of ﬂexibility in deciding when to repair such a device.

New ﬁles will not be assigned to a device with IDLE status

unless no suitable alternative device exists, but ﬁles already

on the device may continue to be accessed as if the device had

a status of ON. If the device is a spun down 834, 836, 887, or

9853 drive, it will be automatically spun up. It may take up

to 3 minutes to spin up the device.

Suppose a failing disk storage device causes the system to set

the device's status to OFF. Suppose also that the device is

failing infrequently, for example, once or twice per hour, and

contains several permanent ﬁles that must remain available

during peak production periods. In this situation, you may use

the THRESHOLD command to increase the veriﬁcation failure

threshold and then use the IDLE command to allow continued

use of the device. This device could then be repaired after the

peak production period has passed, causing less disturbance to

system users.

NOTE

The decision to continue using a failing device should be made

under the advisement of a customer engineer. If there is a

good chance of data corruption due to the nature of the

failure, it may be best to leave the device OFF until it can be

repaired.

If you enter an IDLE command for a device with a status of

OFF or DOWN, the system performs a device veriﬁcation to

ensure the hardware is usable and the disk pack contains

valid data. If this veriﬁcation fails and the veriﬁcation failure

threshold is exceeded, the device state remains unchanged.

However, if the device veriﬁcation ﬁnds no errors the device's

state changes to IDLE, the veriﬁcation failure count (displayed

on the E,H display) is set to zero, and any jobs that were

rolled out waiting for the device are allowed to continue

processing.

5-28 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

yms

Command

IDLE,EQ = est.

IDLEFAMILY,est.

Description

(Continued)

If you attempt to change a device's state to IDLE, the message

INCORRECT ENTRY appears if one of the following

conditions is true:

• The device is already IDLE.

• The device is DOWN and assigned to a job.

• The device is DOWN and no channels are UP for the

device.

If you enter IDLE for a device with a status of ON, the

veriﬁcation process is not performed; the device state is simply

changed to IDLE.

If you enter IDLE for a device other than a disk, the IDLE

command is ignored.

This command performs one of the two following functions,

depending on the status of the family.

• If the family is active, the IDLEFAMILY command

deactivates it by aborting all new jobs and USER

commands for the family on the equipment speciﬁed by

EST ordinal est. Jobs in progress are allowed to complete.

ISF and permanent ﬁle utilities, such as PFDUMP, can be

used on the inactive family.

• If the family is inactive, the IDLEFAMILY command

activates it. Jobs can then access the family on the

equipment speciﬁed by EST ordinal est.

The FAMC column of the E,F display shows the number of

jobs in progress on the equipment.

Revision M DSD Commands 5-29

Peripheral Equipment Control Commands

Command Description

INITIALIZE,op,esti,

est2,...,est5.

Reverses current setting of initialize option op for mass

storage devices deﬁned by EST ordinals est (maximum of ﬁve

devices). Examine the E,A display to determine correct EST

ordinals.

op Level of Initialization

AF Initialize inactive account dayﬁle.

AL Total initialization.

DF Initialize inactive system dayﬁle.

EF Initialize inactive error log.

FP Format pending.4

FT Total initialization as full-track device.5

HT Total initialization as half-track device.5

MF Initialize binary maintenance log.

PF Initialize permanent ﬁles.

QF Initialize inactive queue ﬁles.

This command provides the capability to initialize and ﬂaw

tracks on any mass storage device during normal system

operation. Each time this command is entered it is logged in

the error log.

If local unload (L) status is set for the device, the INITIALIZE

command is ignored and the following message appears on the

left console screen.

INCORRECT ENTRY

However, the INITIALIZE command is allowed to execute

while local unload status is set if an error code (CE, IL, LE)

is set.

4. Sets format pending bit in the MST for 881/883 pack reformatting. Refer to the description of the

FORMAT program in appendix G.

5. Device must be removable.

5-30 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

rCommand Description

INITIALIZE,op,esti,

est2,...,est5.

(Continued)

Entry of this command does not in itself initialize the

speciﬁed device. It merely sets initialize status for the device

so that it may be initialized. However, if fast attach ﬁles

(special system ﬁles) are active on the speciﬁed device and

initialization level of AL or PF is speciﬁed, initialize status

cannot be set until these ﬁles are returned. In this case, the

message

FAST ATTACH FILES ON DEVICE.

appears at the system control point on the system status

display (B,0). Refer to the description of this message in the

NOS Version 2 Operations Handbook for additional

information.

The procedure involved in initializing a mass storage device is

described in section 8, K-Display Utilities, under the heading

Initialize K Display. The following describes system activity

when initialization occurs.

If the device is shared in a multimainframe environment,

initialization does not proceed until all other mainframes

sharing the device have processed an UNLOAD command for

the device and user counts on all machines are zero. If all

machines have not unloaded the device, this control point

message is displayed:

EQest BUSY ON 10=1d.

Variable Description

est

EST ordinal.

Machine identiﬁer of the ﬁrst machine found

without unload status set.

Revision M DSD Commands 5-31

Peripheral Equipment Control Commands

Command Description

INITIALIZE,op,esti,

est2,...,est5.

LOAD,est,id.

(Continued)

If initialize status is set on this device for another mainframe,

the INITIALIZE command is ignored and this message appears

at the system control point on the system status display (B,0):

INITIALIZE PENDING ON THIS DEVICE.

When the AL initialization option is speciﬁed, the label on the

device to be initialized is either updated or a new label is

created. If the label on the device is bad or cannot be

recognized, the new label is created and all current data on

the device is lost. If the label is found to be good, it is

updated and all permanent ﬁle information is cleared. In this

case, system library or temporary ﬁles (local, rollout, and so

forth) residing on the device are not disturbed. If the device

being initialized is a master device, the system scans all other

mass storage devices in the family that contain direct access

ﬁles and releases the space for ﬁles with catalogs on this

device. If the device being initialized contains direct access

ﬁles, the system scans all other master devices and sets the

catalog entries on those devices to indicate that the ﬁles were

purged. All or part of the permanent ﬁle system can be

initialized and then reloaded if necessary (refer to the

description of permanent ﬁle utilities in section 17, Permanent

File Utilities.

Requests that a job be loaded from equipment deﬁned by EST

ordinal est (normally tape unit). The job is assigned a numeric

identiﬁer id ranging from 00 to 67s. If id is not speciﬁed, it

defaults to 00. Examine the E,A display to determine the EST

ordinal of the equipment to be used.

/s?v

5-32 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

rCommand

MOUNT,est,P.

Description

Clears the local unload (L) and global unload (N) status for a

mass storage device and reactivates the device. If the device is

a spun down 834 or 836 drive, it will be automatically spun

up. The device is deﬁned by EST ordinal est (examine the E,A

display to determine the EST ordinal).

When you specify P in the MOUNT command for an

independent shared device in a multimainframe environment,

the system presets the device with EST ordinal est. The preset

(P) option can be speciﬁed only on the ﬁrst mainframe to

access the device.

If the device deﬁned by EST ordinal is not a mass storage

device, the MOUNT command is ignored and this message

appears on the left console screen:

INCORRECT EQUIPMENT.

If the device is shared in a multimainframe environment and

another mainframe has an unsatisﬁed initialize request

pending for that device, the MOUNT command is ignored and

this message appears at the system control point on the

system status display (B,0):

INITIALIZE PENDING ON THIS DEVICE.

Revision M DSD Commands 5-33

Peripheral Equipment Control Commands

Command

MSAL,t=esti,

est2,...,estn.

Description

Assigns job ﬁles of type t to mass storage devices deﬁned by

EST ordinals esti. The speciﬁed mass storage device must be

nonremovable. A limit of 47 devices may be speciﬁed on a

single entry. Examine the disk status display (E,M) to

determine if the device is nonremovable.

t Description

BLGO.

D Job dayﬁle.

I Input.

L Local.

O Output.

P Primary.

R Rollout.

S Secondary rollout.

T Temporary.

If no devices are speciﬁed for a ﬁle type, the system selects a

temporary device.

Secondary rollout ﬁles do not exist until the command

SRST=t (where t is the size in sectors of the ﬁle, from 0 to

77778) deﬁnes a threshold; the default value of a threshold is

0. All rollout ﬁles smaller than the threshold are secondary

rollout ﬁles. These ﬁles are assigned to the devices speciﬁed

with the MSAL,S command.

All ﬁles greater than or equal to the threshold are assigned to

the rollout ﬁle devices (MSAL,R command).

The following example illustrates a use for secondary rollout

ﬁles. Equipment 5 is extended memory via DDP. Secondary

rollout ﬁles to extended memory is assigned by

MSAL,S=5.

5-34 NOS Version 2 Analysis Handbook Revision M

Command

MSAL,t=esti,

est2,...,estn.

Peripheral Equipment Control Commands

Description

(Continued)

The threshold count is set by the SRST command so that

rolled out ﬁles under 20 sectors long are assigned to extended

memory. For additional information concerning the SRST

command, refer to section 3, Deadstart Decks.

EQPDECK

EQ005=DP,ST=0N,SZ=4000,CH=30.

MSAL,S=5.

IPRDECK

SRST=20

NEXTREEL,est. Indicates to the system that you have veriﬁed that the correct

tape has been mounted (or is about to be mounted) on the

speciﬁed tape drive. You enter this command in response to a

CHECK AND MOUNT message on the E,P display.

The system issues the CHECK AND MOUNT message

whenever it unloads an unlabeled tape as a result of a ring

conﬂict or an end-of-reel condition. This message reminds you

that the speciﬁed tape unit is still assigned to a job and is

not available for general use. The system only issues this

message if the COMSMTX installation parameter PONR is 1.

0ms

Revision M DSD Commands 5-35

Peripheral Equipment Control Commands

Command Description

OFF,EQ=est. Logically turns off the device deﬁned by EST ordinal est. This

command allows you to logically remove a device from the

operating environment so that no user access of the device is

permitted. If a user job attempts to access the device, it is

rolled out to wait for the device status to become ON or

IDLE. However, system utilities are permitted to access the

device, so it can be dumped or loaded. If the device is a spun

down 834, 836, 887, or 9853 drive, it will be automatically

spun up. It may take up to 3 minutes to spin up the device.

If you enter the OFF command for a device with a status of

DOWN, the system performs a veriﬁcation of the device label.

This provides assurance that the correct pack is mounted on

the drive and the controller is accessing the right drive after

a repair action.

If you attempt to change a device's state to OFF, the message

INCORRECT ENTRY appears if one of the following

conditions is true:

• The device is already OFF.

• The device is DOWN and assigned to a job.

• The device is DOWN and no channels are UP for the

device.

Examine the E,A display to determine the EST ordinal and

current status (ON, OFF, IDLE, or DOWN) of the device. If

est speciﬁes a disk storage device and the system library or

temporary ﬁles (local, rollout, and so forth) reside on that

device, it should not be turned off. Examine the disk status

display (E,M) to determine which disk storage devices have

system residency or allow system allocation of temporary ﬁles.

In addition, if an MSAL entry is currently speciﬁed for a disk

storage device, it is cleared when that device is turned off.

The MSAL designation is not reset automatically when the

device is turned back on and must be reset manually (if

necessary) using the DSD command MSAL.

This command also clears the SUSPCT state for a tape unit

that is currently unavailable (but not OFF or DOWN) because

an unrecovered hardware or tape I/O error has occurred on

the tape or unit. This command can be entered followed by

the ON,EQ=est command to restore the unit to service after a

customer engineer has checked it out, or the unit has been

tested following Control Data's recommended procedures.

yS&V

5-36 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command

ON,EQ=est.

PRSIZE,est,ps.

Description

Logically turns on the device deﬁned by EST ordinal est. This

command allows you to activate a device currently having

OFF status in the EST. If the device is a spun down 834, 836,

887, or 9853 drive, it will be automatically spun up. Examine

the E,A display to determine the EST ordinal and current

status (ON, OFF, IDLE, or DOWN) of the device.

Use this command to bring a disk storage device back into

use following its repair. You can also use the ON command

after repairing a controller that provides the only access to a

disk storage device.

When you enter ON for a device with a status of OFF or

DOWN, the device's state does not change to ON immediately.

First the system attempts to verify that the drive and

channels accessing the device are usable and the data on the

disk pack is in the proper format. If this veriﬁcation process

fails and the veriﬁcation failure threshold is exceeded, the

system sets the device's state to OFF. If the veriﬁcation ﬁnds

no errors, the device's state changes to ON, the veriﬁcation

failure count (displayed on the E,H display) is set to zero, and

any jobs that were rolled out waiting for the device are

allowed to continue processing.

If you enter ON for device with a status of IDLE, the

veriﬁcation process is not performed; the device's state simply

changes to ON.

If you attempt to change a device's state to ON, the message

INCORRECT ENTRY appears if one of the following

conditions is true:

• The device is already ON.

• The device is DOWN and assigned to a job.

• The device is DOWN and no channels are UP for the

device.

Sets the paper status ps to short (S) or long (L) paper for the

printer with EST ordinal est. Short paper has a form length of

8.5 inches and long paper has a form length of 11 inches.

Revision M DSD Commands 5-37

Peripheral Equipment Control Commands

Command

REPEAT,jsn,rc.6

REPRINT,est,pr.

REPUNCH,est,pr.

RETRY,est.

Description

Alters the repeat count for the queue ﬁle speciﬁed by job

sequence name jsn. If the ﬁle is still in the queue, the repeat

count for the ﬁle is set to re. If the ﬁle is already being

processed by BIO, the remaining repeat count for the ﬁle is

increased by re. The maximum value that can be entered for

re is 778. The default value for re is 1.

Terminates the current operation on the BIO printer

equipment deﬁned by EST ordinal est and reenters the job in

the print queue with a queue priority speciﬁed by prOO

(service class minimum ^ prOO ^ service class maximum;

refer to the S display in the NOS Version 2 Operations

Handbook; pr value is multiplied by 100s internally). If pr is

not speciﬁed, the service class default priority is assigned.

Terminates the current operation on the BIO card punch

equipment deﬁned by EST ordinal est and reenters the job in

the punch queue with a queue priority speciﬁed by prOO

(service class minimum ^ prOO ^ service class maximum;

refer to the S display in the NOS Version 2 Operations

Handbook; pr value is multiplied by 1008 internally). If pr is

not speciﬁed, the service class default priority is assigned.

Reissues a tape operation that previously aborted with one of

the following load point errors appearing in the E,P display:

CLEANER FAULT

READ ID BURST

WRITE ID BURST

est represents the EST ordinal of the tape unit where the load

point error occurred. Examine the E,P display to determine

the EST ordinal. The tape unit must be ready before you

enter the RETRY command. Examine the E,T display to

ensure that the tape is loaded and the tape unit is ready.

The RETRY command is to be used after the system detects a

load point error on a tape. Upon detecting such an error, the

system issues an error message to the E,P display and unloads

the tape. The operator then has the opportunity to ﬁx the

failing tape or tape drive. Upon completing this step, the

operator reloads the tape on the same tape unit and enters

the RETRY command. The system then reissues the tape

operation. If the failure cannot be ﬁxed at this time, the

operator should enter the TERMINATE command to abort the

recovery attempt.

6. When the current BIO operation is repeated, maximum line and card limits are reinitialized prior to

printing or punching of the ﬁle being processed. User control limits apply individually to each output ﬁle

copy produced.

5-38 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command

SCRATCH,est.

SECUREQ,est,

LA = lowerlevel,

UA=upperlevel.

SKIP,est,rr.

SKIPF,est,ff.

Description

Declares the tape mounted on an unassigned magnetic tape

unit, deﬁned by EST ordinal est, to be a scratch tape. This

command enables a tape to be available to satisfy scratch

VSN requests and still be assigned by its original VSN Thus

the VSN deﬁned on the tape (in VOL1 label) is not redeﬁned'

as scratch although the VSN will appear as SCRATCH on the

tape status display (E,T).

Scratch status is retained for only one job assignment. This

allows a tape to be used for scratch purposes on a temporary

basis. For example, a job requests a tape mounted on the tape

unit deﬁned in this command by specifying the current VSN

for that tape in the request. The tape is then assigned to the

job as a scratch tape (the original VSN is retained and not

made scratch). When that job releases the tape, SCRATCH

status is cleared, and unless this command is entered again,

that tape would not be assigned as a scratch tape in future'

requests. To determine if SCRATCH status is in effect for a

tape, monitor the tape status display (E,T).

Changes the equipment access level limits for the unit record

equipment with EST ordinal est. On a secured system, only

ﬁles that have access levels within these limits can be printed

on the speciﬁed equipment. The original limits are set during

deadstart by the ACCESS command entries in the EQPDECK.

The parameters LA = lowerlevel and UA = upperlevel specify

the lower and upper access level limits for the equipment.

NOTE

On a secured system this command is accepted only if a

security administrator has placed the system in security

unlock status.

Skips forward rr octal logical records on the print ﬁle for the

BIO equipment deﬁned by EST ordinal est. When rr is not

speciﬁed the default is one record.

Skips forward ff octal ﬁles on the print ﬁle for the BIO

equipment deﬁned by EST ordinal est. When ff is not speciﬁed

the default is one ﬁle.

Revision M DSD Commands 5-39

Peripheral Equipment Control Commands

Command

SKIPRU,est,ss.

SPINDOWN,est.

SPINUP,est.

Description

Skips forward ss PRUs on the print ﬁle for the BIO

equipment deﬁned by EST ordinal est. All parameters must be

speciﬁed; there are no default settings. The PRU count, ss, is

limited to 10s PRUs (the current buffer size) plus the number

of PRUs remaining in the buffer. If the buffer was empty, ss

would be limited to 20s PRUs. If ss is larger than the number

of PRUs remaining in the buffer, the buffer is ﬁlled again and

the additional PRUs are skipped on the new print ﬁle. For

example, if ﬁve PRUs remained in the print ﬁle and ss was

speciﬁed as 10, the remaining ﬁve PRUs would be skipped,

the buffer ﬁlled again, and ﬁve additional PRUs skipped.

Printing resumes at the beginning of a line.

Spins down the 834, 836, 887, or 9853 disk storage device

deﬁned by EST ordinal est. To enter this command, the

console must be unlocked (refer to the UNLOCK command

later in this section).

Spins up the 834, 836, 887, or 9853 disk storage device

deﬁned by EST ordinal est.

STOP,est.

NOTE

Spinning down an 834, 836, 887, or 9853 disk storage device

that does not have global unload status (N), can cause mass

storage device status errors or permanent ﬁle errors when the

device is spun up.

Stops printing on the BIO equipment deﬁned by EST ordinal

est.

5-40 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command

SUPPRESS,est.

TEMP=esti,

est2,...,estn.

jp^s.

TERMINATE,est.

Description

Suppresses automatic printer carriage control on the BIO line

printer deﬁned by EST ordinal est. This command stops the

page eject function on the line printer to provide a continuous

listing for the current job.

Reverses current set or clear condition of temporary ﬁle status

for mass storage devices deﬁned by EST ordinals esti. When

temporary ﬁle status is set, the system can use the speciﬁed

device for allocation of temporary ﬁles. This command is not

valid if the device speciﬁed is deﬁned as removable. Examine

the disk status display (E,M) to determine:

• The EST ordinal of the device.

• If the device is defined as removable.

• If temporary file status is currently selected (set) for the

device.

Aborts a tape load point error recovery attempt (refer to the

RETRY command in this section), est represents the EST

ordinal of the tape unit where the load point error occurred.

Examine the E,P display to determine the EST ordinal.

The TERMINATE command also clears the error message in

the E,P display and issues the error message to the system

dayﬁle.

Revision M DSD Commands 5-41

Peripheral Equipment Control Commands

Command Description

THRESHOLD, You can use the THRESHOLD command to set disk storage

type=value, device thresholds for the devices deﬁned by EST ordinals esti.

esti,est2,...,estn. You must specify at least one EST ordinal. Use this command

in conjunction with the IDLE command. You can specify one

of the following threshold types. If no threshold type is

speciﬁed, VF is assumed.

type Description

VF Veriﬁcation failure threshold. Each time an

unrecovered error occurs on a device, the system

performs a device veriﬁcation. If the error occurs

again during the veriﬁcation process and the error is

in the device or in the only available channel or in

controlled access to the device, the veriﬁcation failure

count is incremented and compared with the

veriﬁcation failure threshold. If the threshold is

exceeded, the system sets the device's state to OFF.

If the threshold is not exceeded, the device's state is

set to IDLE to restrict allocation of new ﬁles on the

device. Because a disk storage error can cause a job

to abort, it is important to understand that the higher

this threshold is set the more jobs that may abort

before the system sets the device's state to OFF. By

setting the veriﬁcation failure threshold to a value

greater than zero, you can continue to use a failing

device. Do this only when the failures are relatively

infrequent, the failures do not cause corruption of

permanent ﬁles, and the device is essential during the

next several hours of production. After the peak hours

of production, the device can be repaired during a

planned corrective maintenance period.

RA Restricted activity threshold. When the available

space on a device falls below this threshold, new ﬁles

are not assigned to the device unless no other

suitable device is available.

LS Low space threshold. When the available space on a

device falls below this threshold, the operator is

notiﬁed on the A.OPERATOR display.

RE Recovered error threshold. Each time a recovered

error occurs on a disk storage device a count is

incremented and checked against the recovered error

threshold for that device. If the threshold is exceeded,

the operator is notiﬁed on the A,OPERATOR display.

UE Unrecovered error threshold. Each time an

unrecovered error occurs on a disk storage device a

count is incremented and checked against the

unrecovered error threshold for that device. If the

threshold is exceeded, the operator is notiﬁed on the

A,OPERATOR display.

v^^v

5-42 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command

THRESHOLD,

type = value,

esti,est2,.-->estn.

TRAIN,est,t.

Description

(Continued)

The threshold value can range from 0 to 3777s. Specify a

threshold value of * to obtain the default value for the

speciﬁed type, for example, THRESHOLD,RA=*,21*. The

following default values are used.

type value

1/8 of the number of tracks on the device

1/16 of the number of tracks on the device

50s

Assigns or changes print train identiﬁer t of the line printer

deﬁned by EST ordinal est. This command can set the

identiﬁcation if it was not speciﬁed in the EQ entry of the

EQPDECK, or change an identiﬁcation previously included in

the EQPDECK. An LR designation in the EQ entry indicates

a 580-12 line printer, LS is a 580-16 line printer, LT is a

580-20 line printer, and LX is a 5870 printer. Print trains

supported for the 580 printers are 595-1/596-1, 595-5/596-5,

and 595-6/596-6. The print train supported for the 5870 printer

is t = 7. The t ﬁeld speciﬁes the print train.

t Description

0 595-1/596-1 (CDC graphic 63/64-character set);

default.

595-1/596-1 (CDC graphic 63/64-character set).

Reserved for future use.

Reserved for future use.7

595-6/596-6 (ASCII graphic 63/64-character set or

ASCII graphic 95-character set).

595-6/596-6 (ASCII graphic 63/64-character set).

595-6/596-6 (ASCII graphic 95-character set).

595-6/596-6 (ASCII graphic 63/64-character set or

ASCII graphic 95-character set).

7. These values are allowed but will default to 595-1/596-1.

Revision M DSD Commands 5-43

Peripheral Equipment Control Commands

Command

TRAIN,est,t.

Description

(Continued)

If an invalid external characteristic (EC)8 is speciﬁed, the

queued ﬁle processor cannot output the ﬁle. The following

shows which ﬁles will print and which ﬁles will not print

for a given print train selection.

Will Print File Will Not Print

Print Train With the FUe With the

Selected Speciﬁed EC Speciﬁed EC

0None, B4, B6 A4, A6, A9

1None, B4, B6 A4, A6, A9

4None, A4, A6,

A9 B4, B6

5None, A4, A6 B4, B6, A9

6A9 None, B4, B6, A4, A6

7None, B4, B6,

A4, A6, A9

y^BB\

8. Refer to the NOS Version 2 Reference Set, Volume 3 for a discussion of the ROUTE command EC

parameter.

5-44 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command Description

UNLOAD,est. Physically unloads a tape or logically removes a removable

disk storage device from the operating system. The device to

be unloaded is deﬁned by EST ordinal est (examine the E,A

display to determine the EST ordinal). Also, in a

multimainframe environment, the UNLOAD command must

be issued if another mainframe wants to initialize a shared

mass storage device, whether the device is removable or

nonremovable (refer to the INITIALIZE command earlier in

this section). If a magnetic tape is speciﬁed, the tape is

physically unloaded. If a removable disk storage device is

speciﬁed, you can dismount the disk pack on that device

after all activity on the device has been completed.

NOTE

If a nonremovable shared mass storage device is to be

speciﬁed, the console must be unlocked (refer to the

UNLOCK command later in this section).

Magnetic tape units:

If a tape is currently assigned to a job, it cannot be

unloaded. If you attempt to unload it, the UNLOAD

command is ignored and this message appears on the left

console screen:

UNIT NOT AVAILABLE

Examine the tape status display (E,T) before entering the

UNLOAD command to determine if the tape to be unloaded

is currently assigned to a job. If the tape is not currently

assigned, entering this command unloads the speciﬁed tape.

Mass storage devices:

The UNLOAD command is valid for any shared mass

storage device in a multimainframe environment for the

purpose of initialization. Otherwise, the command is valid

only for removable devices. (Only removable devices may be

physically removed by unloading.) After entering the

UNLOAD command, monitor the disk status display (E,M).

Execution of this command immediately causes local unload

(L) status to appear in the STATUS ﬁeld for that device.

While L status is displayed, no new users are permitted to

access ﬁles on the device. A user currently accessing ﬁles on

the device can continue while at least one direct access ﬁle

from the device is attached to the job. When the user count

is zero and there are no checkpoint requests pending, one of

the following two actions occurs.

Revision M DSD Commands 5-45

Peripheral Equipment Control Commands

Command

UNLOAD,est.

Description

(Continued)

• If the device is removable and the L status is set in all

machines accessing the device, global unload (N) status is

displayed. This indicates that the device may now be

physically dismounted.

NOTE

If a multispindle family is mounted on a single spindle

device, only the ﬁrst device shows the global unload status.

• If an initialize is pending on the device and all other

machines accessing the device have L status set, the

initialization proceeds. However, initialization cannot take

place if the device has been unloaded.

NOTE

A device should be physically dismounted only if global

unload status (N) is displayed on all machines accessing

the device.

If a removable pack is dismounted before the N status is

displayed, the following may occur.

• Mass storage device status errors.

• Permanent file errors when the pack is remounted at some

later date.

• If another pack has been mounted, accesses made by a

previously attached user may destroy information on the

new pack or the user may retrieve information from the

new device which he is not necessarily privileged to access.

Mass storage device status errors are also possible in this

situation.

NOTE

If the Mass Storage Extended Subsystem (MSE) is active,

it must be idled before unloading a removable family pack

which has MSE ﬁles. After dismounting the family pack,

MSE can be initialized again.

^^S

5-46 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command Description

UP,CH=ch.

UP,MCH=ch.9

UP,CCH=cch,

EQ=est.

VALIDATE,est.

VSN,est,.

The ﬁrst two forms of the command allow resumption of

normal use of channel ch by magnetic tape units or

mass storage devices, reversing the effects of the

DOWN command.

The third form of the command allows resumption of normal

use of both ports of concurrent channel cch for the mass

storage device speciﬁed by EST ordinal est. If the EQ=est

parameter is omitted, channel cch resumes normal use for

all devices in the EST.

If you enter UP for a channel (ch or cch) that is already

UP, the following message appears on the console:

INCORRECT ENTRY

If the channel is already up, no further action is necessary.

Causes validation of mass storage tables associated with the

equipment deﬁned by EST ordinal est. The equipment must

be available mass storage and the MS VALIDATION option

must have been selected in the IPRDECK at deadstart.

Declares the tape mounted on an unassigned magnetic tape

unit, deﬁned by EST ordinal est, to be a scratch tape. This

command is similar in function to the SCRATCH command

in that it enables a tape to be available to satisfy scratch

VSN requests. However, if the tape is labeled and a write

function is performed, the VSN speciﬁed in the VOL1 label

will be rewritten as a scratch VSN, destroying the original

VSN and making the tape available for future scratch VSN

requests. The VSN also appears as ***est (est is the ordinal

of the est) on the tape status display (E,T). Refer to the'

VSN,est,vsn command, next, for a discussion of the

INCORRECT ENTRY message which is also applicable to

VSN,est,.

9. Model 176 only.

Revision M DSD Commands 5-47

Peripheral Equipment Control Commands

Command Description

VSN,est,vsn.10 Assigns VSN vsn to an unassigned magnetic tape unit deﬁned

by EST ordinal est. This command allows you to specify a

1- to 6-character VSN for a mounted, unlabeled tape so it may

be assigned and referenced automatically. For example, when

a job speciﬁes a VSN in the request for an unlabeled tape, an

entry for that job appears in the resource requests display

(E,P). This display indicates the job sequence name of the job,

the type of tape unit, 7 track (MT) or 9 track (HD, PE, or

GE), on which the tape is to be mounted, the required VSN,

user name of the job, and the required write ring status (IN

or OUT). If the correct tape is not currently mounted, mount

the tape on an available unit (ensuring that track type and

write ring status are correct), ready the unit, and enter this

command. The system equates the VSN you enter with that

speciﬁed by the job and assigns the tape automatically upon

demand. ^,

If the tape mounted on the tape unit deﬁned by EST ordinal

est is a labeled tape, has already had a VSN assigned by

console command, or has not yet been checked for a label by

the magnetic tape subsystem, this command is ignored. The

message

INCORRECT ENTRY

a p p e a r s o n t h e l e f t c o n s o l e sc r e e n . To c h a n g e a V S N _ ^ ~

previously assigned by this command, clear the ﬁrst VSN by 1

entering:

VSN,est.

est Description

est EST ordinal of the tape unit.

/<ss^\

10. Special characters cannot be entered using this command. If a special character is encountered in vsn,

the VSN entered is truncated at the character preceding the special character.

5-48 NOS Version 2 Analysis Handbook Revision M

Peripheral Equipment Control Commands

Command Description

VSN,est,vsn. (Continued)

This command can then be entered:

VSN,est,vsn.

Parameter Description

e s t E S T o r d i n a l o f t h e t a p e u n i t .

vsn New VSN.

00^\

(The preceding message and procedure for changing a VSN are

also applicable to the VSN,est,. command.)

If a job speciﬁes a VSN in the request for a labeled tape,

assignment occurs automatically, without your intervention,

unless the correct tape is not mounted. In this case, an entry

is formed in the resource requests display (E,P), which

describes the tape to be mounted. When the tape is mounted

and the tape unit made ready, assignment occurs

automatically without additional intervention by you. For

multireel ﬁles, automatic tape assignment occurs only if the

tape units on which the tapes are mounted are similar and on

the same channel(s). That is, if the ﬁrst reel of the ﬁle is on

a 669 tape unit on channels 13 and 33, all subsequent reels

must be on a 669 unit on channels 13 and 33. When assigning

tapes, models 679-2, 679-3, and 679-4 drives (800/1600 cpi) are

similar. Also models 679-5, 679-6, and 679-7 drives (1600/6250

cpi) are similar.

If two or more unassigned tapes having identical VSNs are

mounted on units of the same track type, the message

REQUEST,dt,vsn

appears on the B,0 display. The dt ﬁeld is either the device

type MT or the density requirement HD, PE, or GE; vsn is

the VSN required.

You must assign one of the tapes using the ASSIGN

command. If the duplicate VSNs are SCRATCH, the resource

executive routine assigns one automatically.

NOTE

It is not possible to specify a VSN of SCRATCH with this

command since only six characters may be used to deﬁne a

VSN. To deﬁne a scratch tape (used to satisfy scratch VSN

requests), refer to the description of the SCRATCH command

earlier in this section.

Revision M DSD Commands 5-49

Subsystem Control Commands

The commands that follow provide control over which subsystems are to be used. When

a system deadstart is performed, parameters speciﬁed in the IPRDECK determine

which subsystems initially are available. Scheduling other subsystems to a control point

or terminating a current subsystem depends on your action.

When a subsystem is scheduled to a speciﬁc control point, any job currently assigned

to that control point is rolled out if it is not another subsystem or special system job.

However, if the job cannot be rolled out, the command used to call the subsystem

would not be valid. In this case, either terminate the job (if the subsystem required

that control point) or specify another control point for the subsystem using the

SUBSYST L-display utility (refer to section 9, L-Display Utilities, for more

information).

Command

BIO.

Description

Calls the Batch Input/Output (BIO) Subsystem. BIO must be

active in order to use any of the local batch peripheral

devices. BIO equipment activity can be monitored on the I

display (refer to the NOS Version 2 Operations Handbook).

Response to line printer errors is controlled by setting sense

switch 1 or 2 as follows:

Entry Response

CDCffff. n

IAFffff.11

ONSW,BIO,1. Lines causing print error messages are not

ﬂagged or retried.

ONSW,BIO,2. Lines causing print error messages are

ﬂagged and retried. The printer halts and is

placed on hold until you correct the error

and enter CONTINUE,est. The line reprints

and printing continues.

Calls the CYBER Database Control System (CDCS). CDCﬁff

calls a procedure ﬁle named CDCﬁff under the system user

index to initiate the CDCS Subsystem.

Calls the Interactive Facility (IAF) Subsystem to control

point 1. IAFffff calls a procedure ﬁle named IAFffff under

the system user index to initiate the IAF Subsystem. The

IAF Subsystem must be active before jobs can be entered

from an interactive terminal. Use the T display to monitor

activity at the IAF Subsystem control point.

11. The characters ffff are optional; if required, installation personnel must supply the necessary 1 to 4

alphanumeric characters.

5-50 NOS Version 2 Analysis Handbook Revision M

Subsystem Control Commands

0^S

ysRS.

Command Description

IAFffff. (Continued)

Several options are available to control the operation of IAF.

Select these options by setting sense switches after the

subsystem is initiated. Although normal operation does not

require these sense switches to be set, sense switch 5

(ONSW,LAF,5) is set by default when the IAF Subsystem is

initiated. Once set, the sense switches may be turned off again

by entering OFFSW,IAF,s (s is the sense switch number.)

Entry

/tf^^V,

Response

ONSW,IAF,l. When the IAF Subsystem is terminated (by

an IDLE,IAF command), enter users into a

detached state and inhibit restarting

operations.

ONSW,IAF,2. The IAF Subsystem will not reload after an

abort. This option releases the output ﬁle to

the output queue.

ONSW,IAF,3. Aborts the IAF Subsystem on all abnormal

conditions. This ensures that the subsystem

does not continue to operate if an internal

malfunction occurs. This may be used with

sense switch 5 to enable an analyst to

determine the problem. Some conditions cause

the IAF Subsystem to abort even though

sense switch 3 is not set.

ONSW,IAF,4. Enables a dump on normal termination. This

option should be activated whenever a dump

is desired after the IDLE.IAF command.

ONSW,IAF,5. Dumps information to the output ﬁle after the

IAF Subsystem is dropped or aborted. This

option provides a listing that may assist an

analyst in determining the problem that

existed when the subsystem dropped or

aborted. This sense switch is set by default

when the IAF Subsystem is activated.

ONSW,IAF,6. Releases the output ﬁle containing dump

information written after the IAF Subsystem

is dropped or aborted (refer to the

ONSW,IAF,5 option) immediately after the

dump is taken. When sense switch 6 is not

set, the output ﬁle is not released until the

IAF Subsystem is terminated. It is then

printed with the job dayﬁle. This switch

applies only when sense switch 5 is set.

Revision M DSD Commands 5-51

Subsystem Control Commands

Command

IDLE,sub.

Description

Sets idledown status for subsystem sub.

NOTE

MAGffff.12

Do not use this command to idle the NAM subsystem. NAM is

idled by using the procedure shown in NAM Shutdown (refer

to section 15, Network Operations).

The subsystem terminates when idledown conditions are met.

MAG terminates when no tapes are assigned. MSE terminates

when no requests are outstanding and no MSE utilities are

connected. BIO terminates when no active equipments remain.

TAF terminates when no active transactions are outstanding.

For all other subsystems, there are no idledown conditions;

they terminate immediately.

The system does not initiate new activity, such as assigning

tapes and beginning print jobs, when idledown status is set. It

is recommended that you use this command for terminating all

subsystems.

Calls the Magnetic Tape Subsystem (MAG) to the control point

speciﬁed in the IPRDECK or by the SUBSYST L-display

utility. MAG must be active in order for tape processing and

resource allocation to take place and to make the resource

requests available. Tape use can be monitored on the tape

status (E,T) and resource requests (E,P) displays. MAGffff

executes a procedure ﬁle named MAGffff under the system

user index to initiate the MAG Subsystem.

NOTE

You must not change unit numbers or power down tape units

after MAG is initialized. Also, MAG should not be stopped

(via the STOP command) while tapes are assigned. The IDLE

command should be used instead.

MAPffff.12

MCSffff. 12

Calls the Matrix Algorithm Processor (MAP) Subsystem to the

control point speciﬁed in the IPRDECK or by the SUBSYST

L-display utility. MAPffff calls a procedure ﬁle named MAPffff

under the system user index to initiate the MAP Subsystem.

Calls the Message Control System (MCS) to the control point

speciﬁed in the IPRDECK or by the SUBSYST L-display

utility. MCSffff calls a procedure ﬁle named MCSffff under the

system user index to initiate the MCS Subsystem. NAM must

be active at a control point to use MCS.

12. The characters ffff are optional; if required, installation personnel must supply the necessary 1 to 4

alphanumeric characters.

5-52 NOS Version 2 Analysis Handbook Revision M

Subsystem Control Commands

Command Description

MSEffff. 13

NAMffff.13

yiﬂppSi,

PLAffff.13

RBFffff.13

RDFffff.13

Calls the Mass Storage Extended Subsystem (MSE) processing

program to the control point speciﬁed in the IPRDECK or by

the SUBSYST L-display utility. MSEffff calls a procedure ﬁle

named MSEffff under the system user index to initiate the

MSE.

You can turn the trace mode on or off by using the following

sense switch entries after the subsystem is initiated.

Entry Response

ONSW,MSE,l.

OFFSW,MSE,l.

Turn the trace mode on. The trace mode

issues all MSE messages to the job dayﬁle.

Turn the trace mode off. Default is trace

mode off.

Calls the network access method (NAM) to the control point

speciﬁed in the IPRDECK or by the SUBSYST L-display

utility. NAMffff calls a procedure ﬁle named NAMffff under

the system user index to initiate several NAM supervisor

modules. These modules appear momentarily at different

control points as needed during NAM activity.

The release versions are NAM and NAMNOGO. Use NAM if

you want to initialize the network with no operator

intervention. Use NAMNOGO if you want to alter the memory

ﬁle. Refer to section 15, Network Operations, for more

information.

Additional commands that you can enter during NAM

initiation are also described in section 15, Network Operations.

Calls the Plato-NAM Interface (PLA) Subsystem to the control

point speciﬁed in the IPRDECK or by the SUBSYST L-display

utility. PLAffff calls a procedure ﬁle named PLAffff under the

system user index to initiate the PLA Subsystem.

Calls the Remote Batch Facility (RBF) Subsystem to the

control point speciﬁed in the IPRDECK or by the SUBSYST

L-display utility. RBFffff calls a procedure ﬁle named RBFffff

under the system user index to initiate the RBF Subsystem.

Calls the Remote Diagnostic Facility (RDF) Subsystem to

control point 1. RDFffff calls a procedure ﬁle named RDFffff

or IAFffff (depending upon whether the site has the IAF

enabled) under the system user index to initiate the RDF

Subsystem.

If your site uses IAF, ensure that NAM is active at a control

point (refer to the B,0 or B,A display) before you enter the

RDFffff command.

13. The characters ffff are optional; if required, installation personnel must supply the necessary 1 to 4

alphanumeric characters.

Revision M DSD Commands 5-53

Subsystem Control Commands

Command

RHFffff.14

SMFffff. 14

SSFffff.14

STMffff. 14

STOP,sub.

Description

Calls the Remote Host Facility (RHF) Subsystem to the

control point speciﬁed in the IPRDECK or by the SUBSYST

L-display utility. RHFffff calls a procedure ﬁle named RHFffff

under the system user index to initiate the RHF Subsystem.

Calls the Screen Management Facility (SMF) Subsystem to the

control point speciﬁed in the IPRDECK or by the SUBSYST

L-display utility. SMFffff calls a procedure ﬁle name SMFffff

under the system user index to initiate the SMF Subsystem.

IAF must be active to use SMF.

Calls the SCOPE 2 Station Facility (SSF) Subsystem to the

control point speciﬁed in the IPRDECK or by the SUBSYST

L-display utility. SSFffff calls a procedure ﬁle named SSFffff

under the system user index to initiate the SSF Subsystem.

Calls the stimulator to the last control point. STMffff calls a

procedure ﬁle named STMffff under the system user index to

initiate the stimulator.

Drops (terminates) the subsystem sub. Any acceptable

three-letter subsystem name can be speciﬁed. Unlock the

console to enter this command. This command can cause

termination errors in the subsystem being dropped. The

IDLE,sub command does the same thing without the

termination errors.

14. The characters ffff are optional; if required, installation personnel must supply the necessary 1 to 4

alphanumeric characters.

5-54 NOS Version 2 Analysis Handbook Revision M

Subsystem Control Commands

Command

TAFffff.15

Description

Calls the Transaction Facility (TAF) Subsystem to the control

point speciﬁed in the IPRDECK or by the SUBSYST L-display

utility. TAF must be active for transaction processing to take

place.

TAFffff calls a procedure ﬁle named TAFffff under the system

user index to initiate the TAF Subsystem. NAM must be

active at a control point for interactive access to TAF. Batch

jobs can access TAF without NAM.

Response to an operator drop or fatal error is controlled by

setting sense switches as follows:

Entry Response

ONSW,TAF,3. Attempt recovery after TAF is dropped or

aborted. To successfully drop the transaction

subsystem once this entry has been made,

the command OFFSW,TAF,3 must be entered

prior to the IDLE,TAF command.

ONSW,TAF,4. Automatically restart TAF after a recovery.

ONSW,TAF,5. Dump the entire ﬁeld length and release the

output ﬁle after.the transaction subsystem is

dropped or aborted.

ONSW,TAF,6. Print the job dayﬁle upon termination.

Activity at the transaction control point can be monitored on

the O displays (refer to the NOS Version 2 Operations

Handbook).

Refer to the AUTO and MAINTENANCE commands in this section and the SUBSYST

L display in the NOS 2 Operations Handbook for additional information concerning

subsystem control.

15. The characters ffff are optional; if required, installation personnel must supply the necessary 1 to 4

alphanumeric characters.

Revision M DSD Commands 5-55

System Control Commands

The following DSD commands control the operating system as well as the subsystems

that run under the system. Several of these commands are typically used only for

debugging purposes when the system is in an abnormal state. You may use others

frequently to maintain system integrity in a normal production environment.

Appropriate cautions are included with individual command descriptions in cases where

you do not normally use the command or it is not recommended for use in a

production environment.

y^^^V

Command Description

AUTO.

CHECK POINT

SYSTEM.

Calls enabled subsystems to control points, initiates

automatic job processing, and spins up any 834, 836, 887, or

9853 disk drive that is ON and not unloaded. The IPRDECK

used at deadstart time determines which subsystems will be

activated by default. However, any of those subsystems not

currently assigned to a control point can be disabled or

others enabled through use of the SUBSYST L display (refer

to the NOS Version 2 Operations Handbook). Individual

subsystems can also be called to a control point or removed

independent of the AUTO command by using the Subsystem

Control Commands described earlier in this section. For

additional information concerning the AUTO command, refer

to Initiating Job Processing in section 2, Deadstart.

Provides for termination of job processing and writes the

contents of central memory tables to mass storage. The

SCKP EQPDECK entry speciﬁes which mass storage devices

will contain a copy of the checkpoint ﬁle. This command is

typically entered in preparation for recovery deadstart. The

following sequence of operations takes place:

1. Sense switch 1 is set for the IAF Subsystem, which

causes all users to be placed in detached job status when

an operator drop occurs. When all users are in detached

job status, the IAF Subsystem is dropped and the

checkpoint continues.

2. All job scheduling is inhibited. (This has the same effect

as if the IDLE command was entered.)

3. Jobs with a zero subsystem identiﬁer are rolled out. All

of these jobs are recovered on a level 1 or level 2

recovery deadstart.

4. The system moves the system dayﬁle buffers maintained

in CMR to disk.

5. All subsystems except the magnetic tape executive (MAG)

are aborted.

sA^^%\

^i^S

5-56 NOS Version 2 Analysis Handbook Revision M

System Control Commands

Command

CHECK POINT

SYSTEM.

0& S

DATE.yy/mm/dd.

Description

(Continued)

6. MAG is rolled out when no other jobs are active. The

rolling out of MAG allows recovery of all tape ﬁles

associated with jobs rolled out if the tapes are not

repositioned prior to the level 1 or level 2 recovery

deadstart. Because of this, if a tape unit is used for

deadstart, it should be unassigned prior to the checkpoint

command.

7. The system is left in an idle state. If SPINDOWN is

enabled, all 834, 836, 887, or 9853 disk drives that are ON

and not unloaded are spun down. Normal processing may

be continued with an AUTO command. If this is done, no

attempt should be made to later perform a level 1 or level

2 recovery deadstart unless another checkpoint command is

performed.

During the processing of the checkpoint, the message

PROCESSING SYSTEM CHECKPOINT

is issued at the system control point. It is also possible that

under certain circumstances the checkpoint routine will not be

able to properly abort a job (such as one that has NOEXIT

selected and has a nonzero subsystem identiﬁer). If such a job

continues processing after it has been aborted by the system,

the operator must then abort the job for the checkpoint to

continue. A more desirable approach to this situation is to

ensure that jobs such as permanent ﬁle dumps are ﬁnished

prior to the checkpoint. For additional information concerning

the CHECK POINT SYSTEM command, refer to Preparing for

Recovery Deadstart in section 2, Deadstart.

Changes the current system date. Unlock the console before

entering this command (refer to the UNLOCK command later

in this section).

Parameter Description

Year; 00 through 99.

Month; 01 through 12.

Day; 01 through 31.

zgp?\

Revision M DSD Commands 5-57

System Control Commands

Command

DEBUG.

Description

Reverses the current status of the debug mode. When the

debug mode is set, the message DEBUG appears in the header

of the left screen display. The debug mode provides system

origin privileges to validated users and allows modiﬁcations to

be made to the running system. Unlock the console before

entering this command (refer to the UNLOCK command later

in this section). Use of the debug mode is not commonly

allowed in a normal production environment.

NOTE

DIS.jsn.

On a secured system, the console must be in security unlock

status to accept this command (refer to the

UNLOCK,username,password command later in this section).

Calls the job display package (DIS) to the job with job

sequence name jsn. The A and B displays for DIS

automatically appear on the left and right console screen,

respectively. Unlock the console before entering this command

(refer to the UNLOCK command later in this section). Refer

to section 4, DIS Operations, for complete information

concerning the DIS package.

NOTE

On a secured system, the console must be in security unlock

status to accept this command (refer to the

UNLOCK,username,password command later in this section).

■j**l^\

5-58 NOS Version 2 Analysis Handbook Revision M

System Control Commands

J$M&rrnz\^

Command

DISABLE,op.

ENABLE,op.

Description

Disables or enables option op.

If you enter the ENABLE command and op is currently

enabled, the system ignores the command. The system also

ignores the DISABLE command if you enter it and op is

already disabled.

op is one of the following options:

AUTORESTART

CARTRIDGE PF

STAGING

DDP ROLLOUT

PATH

ENGR16

Description

Enables or disables the AUTORESTART^

installation parameter. If

AUTORESTART is enabled and if step |

mode is set during an environmental

shutdown, the system automatically

clears step mode when the power

environment returns to normal. All jobsl

waiting in the input queue are initiated!

by performing the processing associated |

with the AUTO command described

earlier in this section. This recovery

process requires no operator action.

Enables or disables the staging of

MSE-resident permanent ﬁles to disk.

Disabling CARTRIDGE PF STAGING

causes jobs to abort if they attempt to I

access ﬁles which only reside on MSE.

If the SSEXEC is running, enabling

CARTRIDGE PF STAGING allows

MSE-resident ﬁles to be staged to disk |

when accessed.

Enables or disables the use of the DDP I

for job rollout and rollin. If DDP

equipment is not present, this entry hasl

no effect.

Enables or disables engineering mode.

When engineering mode is set, the

message ENGR appears in the header §

of the left screen display. Engineering |

mode allows PP/hardware diagnostics

and FORMAT/FDP to be executed if the!

user has system origin privileges.

Unlock the console before entering this |

command (refer to the UNLOCK

command later in this section).

16. On a secured system, the console must be in security unlock status to accept this command (refer to the

UNLOCK,username,password command later in this section).

Revision M DSD Commands 5-59

System Control Commands

Command Description

DISABLE,op.

ENABLE,op.

(Continued)

FLEXIBLE

PARTITIONS

LOGGING

MASTER MSE

MS VALIDATION

PF VALIDATION

Description

Enables or disables ﬂexible memory

partitioning. The job scheduler attempts

to use memory space to the greatest

extend possible when ﬂexible partitions

are enabled. Some service classes may

be allotted more total memory space

than memory partitioning constraints

normally allow. If ﬂexible partitions are

disabled, the total memory used by jobs

of a given service class is never allowed

to exceed the maximum speciﬁed on the

SERVICE command (AM and EM

parameters).

Enables or disables logging of dayﬁle

messages concerned with program

efﬁciency.

Enables or disables master mainframe

mode for MSE processing. When MSE

is brought to a control point, the

SSEXEC program runs if master

mainframe mode is enabled. The SSSLV

program runs if master mainframe

mode is disabled. This entry has no

effect unless MSE processing is

activated.

Enables or disables automatic

veriﬁcation of mass storage tables. This

command cannot be used unless the MS

VALIDATION option has been selected

in the IPRDECK used at deadstart. The

validation that occurs for each level of

recovery deadstart is described under

Preparing for Recovery Deadstart in

section 2, Deadstart.

Enables or disables veriﬁcation of

BOI/EOI on preserved ﬁles. Selection of

this option causes BOI/EOI information

to be veriﬁed when a direct access ﬁle

is attached or purged, and when an

existing local ﬁle is deﬁned as a direct

access permanent ﬁle.

•»"SfS\

5-60 NOS Version 2 Analysis Handbook Revision M

System Control Commands

Command Description

DISABLE,op.

ENABLE,op.

(Continued)

PRIVILEGED

ANALYST MODE

/|Ife\

PRIVILEGED RDF

REMOVABLE

PACKS

RESIDENT RDF

Description

Enables or disables PRIVILEGED

ANALYST MODE operations. When

enabled, a user validated with

AW-CPAM (refer to the NOS Version 2

Administration Handbook for

information on MODVAL validation) is

permitted to read system status

information (such as the system dayﬁle,

account ﬁle, and error log) using a

nonsystem-origin job. Unlock the

console (refer to the UNLOCK

command later in this section) to enable

or disable this option. This command is

not accepted on a secured system.

Enables or disables privileged mode of

the Remote Diagnostic Facility (RDF).

When enabled, a user's commands are

checked to ensure that a maintenance

function is being performed.

Enables or disables automatic label

checking for mass storage devices

deﬁned as removable. Examine the disk

status display (E,M) (refer to the NOS

Version 2 Operations Handbook) to

determine which mass storage devices

(if any) are deﬁned as removable. If

this option is disabled, all removable

devices subsequently introduced into the

system cannot be accessed. This option

must be enabled to perform label

veriﬁcation before those devices can be

accessed.

Enables or disables resident mode of

RDF. While in resident mode, RDF

remains active, regardless of terminal

activity, until you disable RDF. When

resident mode is disabled (the default

condition), RDF becomes inactive if no

one logs into the remote diagnostic

terminal for 15 minutes. To allow

terminal activity to resume, you must

reactivate RDF with a call to RDFffff.

/5^s

Revision M DSD Commands 5-61

System Control Commands

Command Description

DISABLE,op.

ENABLE,op.

(Continued)

op Description

SECONDARY

USER COMMANDS

SPINDOWN

Enables or disables the acceptance of

more than one USER command in a

job. If disabled, any USER command

encountered after the ﬁrst one causes

the job to be aborted. Unlock the

console (refer to the UNLOCK

command later in this section) to enable

or disable this option. This command is

not accepted on a secured system.

Enables or disables the spinning down

of 834, 836, 887, and 9853 disk storage

devices during a system checkpoint.

System checkpoints that are initiated by

mainframe errors do not affect the state

of these devices.

r-^^K

V**Bfﬂ^\

>ssev

5-62 NOS Version 2 Analysis Handbook Revision M

System Control Commands

Command Description

DISABLE,op.

ENABLE,op.

(Continued)

SYSTEM DEBUG

Description

Enables or disables the system debug

mode of operation. Unlock the console

(refer to the UNLOCK command later

in this section) to enable or disable this

option.

When the system is in system debug

mode, it is less tolerant of system

errors; that is, it is more likely to hang

upon experiencing errors. When the

system is not in system debug mode, it

rates system errors as critical or

noncritical.

For critical errors, the system partially

or totally interrupts system operation to

tend to the errors. For noncritical

errors, the system logs them in the

binary maintenance log (BML) and

inasmuch as possible allows system

operation to proceed. You can initiate

the system debug mode with the DSD

ENABLE command or the corresponding

IPRDECK entry.

There is another system state called

debug mode which is conceptually

different from that of system debug

mode. Debug mode is the state of the

system where a user with system origin

privileges can make modiﬁcations to the

running system. You can initiate this

mode of operation with the DSD

command DEBUG. The left screen

header of the system console indicates

whether the system is in debug mode

or not.

0ims

Revision M DSD Commands 5-63

System Control Commands

Command Description

DISABLE,op.

ENABLE,op.

IDLE.

K.messagetext

L.messagetext

LOCK.

(Continued)

op Description

TAPE PF STAGING

USER EXTENDED

MEMORY

Enables or disables the staging of tape

alternate storage resident ﬁles to disk.

Disabling TAPE PF STAGING causes

jobs to abort if they attempt to access

ﬁles which only reside on tape alternate

storage.

Enables or disables use of the user

accessible area of extended memory. If

disabled, no job can access the user

area of extended memory and, other

than subsystems, all jobs currently

accessing the area are rolled out.

Prevents any new jobs from being scheduled to a control point

but does not terminate the jobs currently assigned. If a job is

rolled out while this command is in effect, it will not be

scheduled back to a control point until the AUTO,

MAINTENANCE, or SCHEDULE command is entered.

Allows entry of data messagetext in the user- or

system-deﬁned CPU buffer for control when the K display is

active. Refer to section 8, K-Display Utilities, for information

about the K display.

Allows entry of data messagetext in the user- or

system-deﬁned CPU buffer for control when the L display is

active. Refer to section 9, L-Display Utilities, for information

about the L display.

Locks the console keyboard. This command prevents entry of

restricted commands (refer to the UNLOCK command later in

this section for a list of restricted commands). All other DSD

commands can be entered when the console is locked. The

console is normally locked when the system is being used in a

production environment.

^"^^\

5-64 NOS Version 2 Analysis Handbook Revision M

System Control Commands

Command

MAINTENANCE.

SCHEDULE

STEP.

Description

This command performs the same functions as the AUTO

command but additionally starts several maintenance routines.

Refer to Initiating Job Processing in section 2, Deadstart, for

complete information concerning this command.

Initiates automatic job processing (reversing the effect of a

previous IDLE command), but does not initiate subsystems as

the AUTO command does.

Sets the monitor in step mode. Setting the monitor in step

mode stops all central memory I/O operations and prevents the

system from processing PP requests when the next monitor

function is encountered. Pressing the space bar releases the

present step and stops again for each subsequent monitor

function. If a DSD command is entered while the system is

in step mode and all PPs are assigned, it is possible the

request cannot be satisﬁed. In that case, the screen(s) blinks

until you correct the condition by clearing the entry.

When step mode is set, the message STEP appears in the

header of the left screen display. Unlock the console before

entering this command (refer to the UNLOCK command later

in this section).

This command is generally used for debugging purposes and

should not be used in a normal production environment. In

addition, the system may set step mode automatically upon

detection of a main power failure or abnormal environmental

condition.

17. Monitor functions are described in the NOS Version 2 Systems Programmer's Instant.

Revision M DSD Commands 5-65

System Control Commands

>*^^v

Command

STEP,jsn,ff,b,v.

STEP„ff,b,v.

TIME.hh.mm.ss.

Description

Sets step mode for the job with job sequence name jsn

(ﬁrst form of the command) or all jobs (second form of the

command) at the next occurrence of monitor function ff when

output register byte b has value v.

If jsn is speciﬁed (ﬁrst form of the command), the monitor

function ff, the output register byte number b, and the output

control of when the step mode is executed. The values for b

can be from 0 to 4; the values for v can be from 0 to 7777s.

If jsn is not speciﬁed (second form of the command), the

monitor function ff must be speciﬁed. The output register byte

number b and value v can be optionally speciﬁed for more

precise control of when the step mode is executed. The values

for b can be from 0 to 4; the values for v can be from 0

to 77778.

For either form of the command, pressing the spacebar

releases the current step and steps the job or system at the

next occurrence of the monitor function ff, if speciﬁed, or the

next monitor function from the job if the ﬁrst form is used

and no monitor function is speciﬁed.

Using the second form of the command may stop all central

memory I/O operations and prevent the system from processing

PP requests. If a DSD command is entered while the system

is in step mode and all PPs are assigned, it is possible the

request cannot be satisﬁed. In this case, the screens blink

until you clear the entry. The message STEP followed by the

monitor function number ff (if speciﬁed) appears in the header

of the left screen displays while this command is in effect.

The console must be unlocked before entry of this command is

permitted (refer to the UNLOCK command later in this

section).

These commands are generally used only for debugging

purposes. Do not enter these commands if the system has

automatically set step mode because of a power failure or

abnormal environmental condition.

Changes the current system time. Unlock the console before

entering this command (refer to the UNLOCK command later

in this section).

Parameter Description

Hour; 00 through 23.

Minute; 00 through 59.

Second; 00 through 59

y ^ ^ \

5-66 NOS Version 2 Analysis Handbook Revision M

System Control Commands

Command Description

UNLOCK. Unlocks the console keyboard. When this command is active,

the message UNLOCK appears in the header of the left screen

display. Although all DSD commands can be entered when the

console is unlocked, the following commands are restricted to

entry only when the console is unlocked.

DATE.yy/mm/dd.18

DEBUG.

DISABLE,PRIVILEGED ANALYST MODE.19

DISABLE.SECONDARY USER COMMANDS.19

DISABLE,SYSTEM DEBUG.

DROP„qt,ujn.

ENABLE,ENGR.18

ENABLE,PRIVILEGED ANALYST MODE.19

ENABLE,SECONDARY USER COMMANDS.19

ENABLE,SYSTEM DEBUG.

OVERRIDEJsn.

SPINDOWN,est.

STEP.

STEP,jsn,ff,b,v.

STOP,sub.

TIME.hh.mm.ss.18

UNLOAD,est. (est speciﬁes a nonremovable shared mass

storage device)

UNSTEP.

All memory entry commands18

All channel control commands

All extended memory ﬂag register commands

All breakpoint package commands18

18. Refer to Secured System Control Commands later in this section for information on unlocking the console

keyboard on a secured system.

19. Not allowed on secured systems.

Revision M DSD Commands 5-67

System Control Command's

Command

UNLOCK.

UNSTEP.

X.MDD(p)

Description

(Continued)

Always lock the console when the system is being used in a

production environment. However, you can unlock the console

to enter the STEP command if a system failure requiring a

recovery deadstart occurs (refer to Preparing for Recovery

Deadstart in section 2, Deadstart).

Clears step mode. This command clears the effect of any

format of STEP command. Unlock the console before entering

this command (refer to the UNLOCK command later in this

section). If the system has set step mode because of a main

power failure or abnormal environmental condition, do not

enter this command until the conditions that caused the

automatic setting have returned to normal.

Refer to Secured System Control Commands later in this

section for information on unlocking the console keyboard on a

secured system.

Initiates the monitor display driver (MDD). MDD is a PP

program and is independent of the operating system. Refer to

the CYBER Initialization Package User's Handbook for more

information.

Parameter Description

X.name.

X.name(parameters)

X.name,ﬂ.

p Port number parameter. If you enter 1, the ﬁrst

port is connected. If you enter 2, the second

port is connected. If you do not specify this

parameter, the second port is connected by

default.

Calls a system program or utility speciﬁed by name

to an available control point. If parameters are to

be passed to the program, the second form of the command

is used where (parameters)the parameters. In both the

ﬁrst and second form of the command, the ﬁeld length

speciﬁed in the library for the command is used. If no ﬁeld

length is speciﬁed in the library, a value of 600008 is

assumed. If a ﬁeld length different from the default is

required, the third form of the command is used. The ﬁeld

length is speciﬁed by ﬂ. Only the ﬁrst 58 characters following

X. are used.

■^^s

■^\

5-68 NOS Version 2 Analysis Handbook Revision M

System Control Commands

Command Description

99. Disables or enables syntax overlay processing and logging of

DSD commands in the system dayﬁle and error log. That is,

depending upon current status, syntax overlay processing and

logging are reversed. When this mode is enabled, 99 appears

on the left screen header.

When syntax overlay processing and logging of DSD commands

is disabled, DSD does not allow you to enter any command

requiring the loading of an overlay from mass storage. Disable

overlay loading only when the system is in an abnormal state

to prevent PPs from being requested when they cannot

perform the necessary tasks (for example, when a system disk

channel is hung). A 99 command that enables logging will be

logged itself, but a 99 command which disables logging will

not be logged.

This command is normally used only for debugging purposes.

When the system is in an abnormal state, the commands that

require entry of the 99 command depend on which syntax

overlays for DSD are in central memory.

Revision M DSD Commands 5-69

Secured System Control Commands

The following commands are allowed only on a secured system.

Command

OQSH=level.

RELEASEjsn.

SECURES,ot,

LA=lowerlevel,

UA = upperlevel.

Description

Speciﬁes the output queue special handling (OQSH) level on a

secured system. The (OQSH) level is set initially during

deadstart by the OQSH IPRDECK entry. The current OQSH

level appears on the DSD S display. The OQSH command can

be entered at any time from the system console to change the

current level. Output ﬁles with an access level greater than or

equal to the OQSH level speciﬁed in this command are not

printed but remain in the queue until released by the operator

(refer to the RELEASE command, next). If level is set to the

lowest access level or no level is speciﬁed, no ﬁles are held in

the queue.

Allows the operator to release a ﬁle from the output queue

whose access level is equal to or above the OQSH level on a

secured system. Output queue ﬁles and their access levels can

be examined using the DSD Q display. The output ﬁle with

job sequence name jsn is released from the output queue and

is processed by the Batch Input/Output Subsystem. The

RELEASE command can be entered at any time from the

system console. Other restrictions based on device access levels

and ﬁle access levels set by your site continue to apply.

Sets the security access level limits for the system

(ot=SY) or the security access level limits for a

particular origin type (ot=IA, BC, or RB). The system limits

determine the range of access levels allowed in the system;

jobs may not execute at an access level outside this range,

and ﬁles may not be created or accessed at an access level

outside this range. Origin type limits determine the range as

it applies to jobs of one particular origin type. By default, all

origin type limits are the same as the system limits.

The options available for using this command to change the

system limits (that is, whether they can be raised, lowered, or

both) are controlled by the OPSECM CMRDECK entry (refer

to section 3, Deadstart Decks). The origin type limits for

ot=IA, BC, and RB can be changed to any values within the

system limits. When the system limits are changed (ot=SY),

all origin type limits are reset to the new system limits.

The parameters LA and UA specify the lov/er and upper

access level limits. Both parameters must be speciﬁed. The

value corresponding to the lower access level must be less

than or equal to the value corresponding to the upper access

level.

5-70 NOS Version 2 Analysis Handbook Revision M

Secured System Control Commands

y^^^s

Command Description

UNLOCK,username,

password.

y^$s

Unlocks the console keyboard for special commands when

the system is running in secured mode. When this command is

entered, the message SECURITY UNLOCK appears in the

header on the left screen display. If the system is running in

secured mode, the following commands are restricted to entry

only when the console is in security unlock status.

DATE.yy/mm/dd.

X.DDF.

DEBUG.

DISJsn.

ENABLE,ENGR.

QDSPLAYJsn.

SECURES,ot,LA=lowerlevel,UA=upperlevel.

SECUREQ,est,LA = lowerlevel,UA = upperlevel.

TIME.hh.mm.ss.

All memory entry commands.

All PP breakpoint commands.

All CPUMTR breakpoint commands.

In addition, the memory displays are partially disabled on a

secured system unless the console is in security unlock status.

If you attempt to display anything outside of the system

tables, a memory display shows the following message instead

of the contents of the memory locations.

****SECURED AREA****

Revision M DSD Commands 5-71

Memory Entry Commands

The following commands are used to change the contents of central memory and

extended memory. Either absolute locations or those relative to a speciﬁc job's

reference address (RA) can be changed.

CAUTION

Use these commands with extreme caution to avoid damage to the system or to user

jobs.

Before entering these commands you must unlock the console keyboard (refer to the

UNLOCK command earlier in this section). On a secured system, the console must be

in security unlock status. To change central memory, you must bring a central memory

display (C, D, F, or G) to the left screen. To change extended memory, you must bring

the M display to the left screen. It is this display that controls whether the memory

entry commands change absolute memory locations or relative memory locations.

For example, if you enter

C , .

to bring up the C display, any memory entry commands entered make changes to

absolute memory locations. If you enter

C.jsn.

where jsn is a valid job sequence name, any memory entry commands entered make ■'***%

changes to memory locations relative to the job's RA.

Character values or numeric data can replace the current word contents. Either one

12-bit byte or a 60-bit word can be changed. A single byte can be changed by

inserting the byte number after the location; bytes are numbered 0 through 4 from

left to right. The address and contents are assembled right-justiﬁed with zeros ﬁlling

unused leading positions. Leading zeros may be omitted in the entry.

When you are changing the contents of memory relative to a speciﬁc job (a valid job

sequence name was speciﬁed when the memory display was brought to the left ""^

screen), the negative ﬁeld length area of the job can be accessed. This area is

accessed by specifying a negative address in the memory entry command. For

example, to change the contents of word RA-3, you enter the address 77777775s.

On an unsecured system when you are changing a memory location relative to a

speciﬁc job's RA, the system checks for ﬁeld length violations. If you specify an

address larger than your job's RA plus ﬁeld length, you cannot change its contents.

On a secured system no memory entry commands are allowed unless the security

unlock status is set. The memory display shows the message

♦••"SECURED AREA****

instead of the contents of the memory locations to prevent you from examining these

locations; you may not alter the contents of these locations. When you are displaying

absolute memory you can see only the system tables.

/**S\

5-72 NOS Version 2 Analysis Handbook Revision M

Memory Entry Commands

Formats and descriptions for these commands are:

Command

addr,cont.

addr+cont.20

addr,b,cont.

addr+b,cont.20

addr,Dcont.

addr+Dcont.20

Eaddr,c.ont.

Eaddr + cont.20

Description

Changes the contents of central memory location addr

(maximum of eight digits) tc cont (maximum of 20

digits). The second form of the command performs essentially

the same function but is used when it is necessary to change

successive memory locations.

Changes the contents of byte b at central memory location

addr (maximum of eight digits) to cont. Each location

consists of ﬁve 12-bit bytes, numbered 0 through 4 from left

to right. The contents are four octal characters. The second

form of the command performs essentially the same function

but is used when it is necessary to change successive memory

locations.

Changes the contents of central memory location addr

(maximum of eight digits) to display code characters cont

(left-justiﬁed and zero-ﬁlled). The second form of the command

performs essentially the same function but is used when it is

necessary to change successive memory locations.

Changes the contents of extended memory location addr

(maximum of seven digits) to cont (maximum of 20 digits).

The second form of the command performs essentially the

same function but is used when it is necessary to change

successive extended memory locations.

Eaddr,b,cont.

Eaddr+ b,cont.20

Eaddr,Dcont.

Eaddr+ Dcont.20

Changes the contents of byte b at extended memory location

addr (maximum of seven digits) to cont. Each location

consists of ﬁve 12-bit bytes, numbered 0 through 4 from left

to right. The contents are four octal characters. The second

form of the command performs essentially the same function

but is used when it is necessary to change successive extended

memory locations.

Changes the contents of extended memory location addr

(maximum of seven digits) to display code characters

cont (left-justiﬁed and zero-ﬁlled). The second form of the

command performs essentially the same function but is used

when it is necessary to change successive extended memory

locations.

20. If the + sign is speciﬁed, the address is incremented by one location (addr + 1) after the initial entry is

processed; the - sign causes the address to be decremented by one location (addr-1). This allows immediate

entry for the next (or previous) memory location. If the message REPEAT ENTRY is displayed above the

entry, the cont ﬁeld is not cleared and can be entered in successive memory locations as many times as

desired.

Revision M DSD Commands 5-73

Channel Control Commands

The following commands enable control activity on a speciﬁed data channel in

circumstances where abnormal hardware and/or system operation is detected. Extreme

caution must be exercised if any of these commands are entered during normal system

operation. In addition, the console keyboard must be unlocked before entry of any of

these commands is permitted (refer to the UNLOCK command earlier in this section).

DSD does not reserve the channel speciﬁed in any of the channel control commands.

The channels are numbered 0 to 13s in a 10 PP system and 0 to 13s, 20s to 33s in a

20 PP system.

CAUTION

Extreme caution must be exercised when using the following channel control

commands.

Command Description

ACN,cc.

DCH,cc.

DCN,cc.

FCN,cc,func.

IAN,cc.

LDC,nnnn.

MCH,cc.

OAN,cc.

Activates channel cc. This command alerts and prepares

peripheral equipment on channel cc for the exchange of data.

Drops channel cc. This is a software function to release the

current reservation of channel cc.

Deactivates channel cc. As a result, peripheral equipment on

channel cc stops and any current I/O operation is terminated.

Outputs a function code func to channel cc. This releases all

equipment selections on that channel. If func is not speciﬁed,

a zero function code (no activity) is output.

Inputs data to the pseudo A register from channel cc.

Loads the pseudo A register with nnnn (normally a

peripheral equipment function code). The current value of

nnnn is shown in the channel status display (W,C).

Master clears and removes all 3000 Computer Systems

peripheral equipment selections on channel cc (6681 function

code 17008 is issued).

Outputs the contents of the pseudo A register to channel cc.

5-74 NOS Version 2 Analysis Handbook Revision M

0$m\

00!&\

Extended Memory Flag Register Commands

The extended memory ﬂag register commands are used to set or clear speciﬁc bits in

the extended memory ﬂag register. These commands are typically used when debugging

a problem in a multimainframe environment. The console must be unlocked (refer to

the UNLOCK command earlier in this section).

CAUTION

Extreme caution must be exercised when using these commands.

Command Description

CFR,bb. Clears the extended memory ﬂag register bit bb

(0 ^ bb ^ 17).

SFR,bb. Sets the extended memory ﬂag register bit bb

j p v ( 0 ^ b b ^ 1 7 ) .

Breakpoint Package Commands

The breakpoint package consists of a set of DSD commands that controls PP breakpoint

processing and CPUMTR breakpoint processing.

PP Breakpoint Commands

y^\ The following DSD commands control PP breakpoint processing. These commands and

V the DSD V display enable you to communicate with the PP breakpoint package. The V

display is described following the command descriptions. The PP breakpoint commands

require that the system console be unlocked (refer to the UNLOCK command earlier in

this section); on a secured system, the console must be in security unlock status.

CAUTION

Use these commands with extreme caution to avoid damage to the system or to user

jobs. Incorrect usage may cause a system to hang up and require a deadstart.

NOTE

Breakpoint can be set for only one PP program at a time. Before setting breakpoint

for another program, it must be cleared on the ﬁrst program.

Revision M DSD Commands 5-75

PP Breakpoint Commands

Command Description

| BKP.

| BKP,prog,cp.

| PPnn.A,val.

I PPnn.C,addr.

| PPnn.C,addr,val.

| PPnn.D,addr.

| PPnn.D,addr,val.

§ PPnn.EXR.

I PPnn.FTN.

Clears an existing PP breakpoint.

Sets breakpoint for PP program prog at control point cp. The

control point parameter cp is optional. The PP program prog

can be any PP program or overlay except the following:

• PPR overlays (1DD, OSE, etc.).

• Mass storage drivers.

• Mass storage error processors.

• Any PP program that overwrites PP memory greater than

75008 with instructions or data. The breakpoint program

uses PP memory locations 75008 to 77778. If the PP

program is allowed to modify those locations on load,

execution, or overlay load, unpredictable results may occur.

Refer to the PP breakpoint precautions in the following

section.

• Concurrent (16-bit) PP programs.

The PP program need not be a main program, and it can

reside in central memory, disk, or extended memory

(ECS/ESM/UEM).

Sets the A register to the value val for PP number nn. Up to

18 bits can be speciﬁed for the value parameter val.

Causes the V display for PP number nn to display 100s bytes

of PP memory, starting at base address addr, as the C block.

Changes the contents of PP memory word addr for PP number

nn to value val. The speciﬁed address must point to one of the

PP memory words displayed in the C block of the V display.

Causes the V display for PP number nn to display 1008 bytes

of PP memory, starting at base address addr, as the D block.

Changes the contents of PP memory word addr for PP number

nn to value val. The speciﬁed address must point to one of the

PP memory words displayed in the D block of the V display.

Resumes execution in PP number nn with a breakpoint set in

the PP-resident subroutine EXR (after the next overlay load

completes, but before execution of the overlay begins).

Resumes execution in PP number nn with a breakpoint set in

the PP-resident subroutine FTN + 1 (to trap the next monitor

function before it is issued to MTR/CPUMTR).

^<«S5sV

5-76 NOS Version 2 Analysis Handbook Revision M

V Display

00^S

0&%S

Command

PPnn.G.

PPnn.G,addr.

PPnn.P,addr.

PPnn.S.

PPnn.SX.

V,nn.

V,HNG.

V,PS.

V Display

Description

Resumes normal PP program execution for PP number nn

with no breakpoint set.

Resumes PP program execution for PP number nn with a

breakpoint set at address addr. The address addr must be

within executable code and lie on a PP instruction boundary,

but the validity of the address is not checked.

Resets the P register to address addr for PP number nn.

When the PP program resumes execution, it will start at

P = addr.

Steps the PP program one instruction for PP number nn.

Indexed jump instructions, channel jump instructions, and

instructions that jump to the current program address or

current program address minus one cannot be stepped. If you

try to step these instructions, the following message appears

on the system console:

CAN-T STEP.

Steps the PP program one instruction for PP number nn. This

command performs the same function as the PPnn.S. command

except that if an RJM instruction is encountered, a breakpoint

is set to the instruction after the RJM and the entire

subroutine is executed.

Brings up the DSD V display for PP number nn.

Brings up the DSD V display for a hung PP.

Brings up the DSD V display for the pseudo PP.

The DSD V display complements the PP breakpoint commands to provide an analyst's

tool for PP breakpoint processing. The information displayed is for the PP number

speciﬁed in the V,nn command. Figure 5-4 shows the format of the V display. The

display shows the PP communications area for the speciﬁed PP (10s CM words), the

contents of the P and A registers, and two 1008 word blocks of PP memory (C and D).

The base addresses of the C and D blocks are set by PPnn.C,addr and PPnn.D,addr

commands.

If the speciﬁed PP is not at breakpoint when the V display is brought up, the display

shows only the communications area and the following message:

NOT AT BREAK.

Revision M DSD Commands 5-77

V Display

V,nn PP BREAKPOINT.

7420 0310 1702 oooo 0000 1234

0000 0001 OOOO OOOO OOOO

xxxx xxxx xxxx xxxx xxxx

xxxx xxxx xxxx xxxx xxxx = 0460

xxxx xxxx xxxx xxxx xxxx

xxxx xxxx xxxx xxxx xxxx A = 220346

xxxx xxxx xxxx xxxx xxxx

C(1100) 0200 2754 2000 3465 5400 3472 1006 3401

(1110) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(1120) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(1130) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(1140) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(1150) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(1160) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(1170) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

D(0000) oooo 0002 0056 4567 oooo 0003 4157 0234

(0010) OOOO 0001 0000 OOOO oooo OOOO oooo oooo

(0020) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(0030) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(0040) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(0050) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(0060) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

(0070) xxxx xxxx xxxx xxxx xxxx xxxx xxxx xxxx

y^*feav

Figure 5-4. PP Breakpoint V Display

-*^\

5-78 NOS Version 2 Analysis Handbook Revision M

PP Breakpoint Precautions

The following precautions and information are important to the successful use of the

PP breakpoint package.

• The breakpoint must be set in executable code and must lie on a PP instruction

boundary.

• When a breakpoint is set, two words of PP memory are replaced with the

following:

RJM BKP

When the breakpoint is hit, the two replaced words are immediately restored.

However, 16-bit instructions are not restored properly since they revert to 12-bit

instructions.

• The PP breakpoint package uses PP locations 7500s through 7777s. If these

locations are altered by the running PP, the system will probably fail.

• If the PP program being breakpointed encounters a mass storage error, the mass

storage driver's error processor will overlay the PP breakpoint code, and will

probably crash the operating system.

• If the PP program being breakpointed performs code modification on the

instruction at the address at which the breakpoint has been set, unpredictable

results may occur.

• The PP breakpoint package cannot be used on DIS or 026. However, it can be

used on other PPs at the same control point as DIS or 026.

• The PP breakpoint information (BKP,prog) remains set until BKP is entered or a

SYSEDIT or any level of deadstart is performed.

Revision M DSD Commands 5-79

CPUMTR Breakpoint Commands

The following DSD commands control CPUMTR breakpoint processing. These commands

and the DSD C display enable you to communicate with the CPUMTR breakpoint

package. The C display is described following the command descriptions. The CPUMTR

breakpoint commands require that the system console be unlocked (refer to the

UNLOCK command earlier in this section); on a secured system, the console must be

in security unlock status.

CAUTION

Use these commands with extreme caution to avoid damage to the system or to user

jobs. Incorrect usage may cause a system to hang up and require a deadstart.

NOTE

Only one CPUMTR breakpoint can be set at a time. Any existing breakpoint is

cleared when you enter a CPB,BKP,... command.

Command Description

CPB,BKP,addr.

CPB,BKPy/addr.

CPB,BKP,/bbbb/addr.

CPB,DSP.

CPB,DSP,addr.

CPB,DSP,//addr.

CPB,DSP,/bbbb/addr.

CPB,DSP +

CPB,DSP-

Sets breakpoint at absolute address addr. Restarts execution,

if the CPU is waiting for a GO.

Sets breakpoint at CPUMTR relative address addr. Restarts

execution, if the CPU is waiting for a GO.

Sets breakpoint at address addr in CPUMTR block bbbb.

Restarts execution, if the CPU is waiting for a GO.

Brings up the CPUMTR breakpoint variant of the C display

on the left screen if the C display is not already active.

Causes the memory block in the C display to begin at the

breakpoint address.

Brings up the CPUMTR breakpoint variant of the C display

on the left screen if the C display is not already active.

Causes the memory block in the C display to begin at

absolute address addr.

Brings up the CPUMTR breakpoint variant of the C display

on the left screen if the C display is not already active.

Causes the memory block in the C display to begin at

CPUMTR relative address addr.

Brings up the CPUMTR breakpoint variant of the C display

on the left screen if the C display is not already active.

Causes the memory block in the C display to begin at

address addr in CPUMTR block bbbb.

Brings up the CPUMTR breakpoint variant of the C display

on the left screen if the C display is not already active.

Advances each of the memory block addresses by 10s.

Brings up the CPUMTR breakpoint variant of the C display

on the left screen if the C display is not already active.

Decrements each of the memory block addresses by 10s. ^ ^ 5 \

5-80 NOS Version 2 Analysis Handbook Revision M

0$m\

CPUMTR Breakpoint Display (C Display)

Command Description

CPB,GO. Clears any breakpoint that is set and allows CPUMTR to

restore its registers and continue execution.

CPB,P+ Sets breakpoint at the address following the current

breakpoint address and restarts execution, if the CPU is

waiting for a GO.

CPB,P- Sets breakpoint at the address preceding the current

breakpoint address and restarts execution, if the CPU is

waiting for a GO.

CPUMTR Breakpoint Display (C Display)

The CPUMTR breakpoint display is a variant of the DSD C display that is formatted

for use in CPUMTR breakpoint processing. This display is initiated by entering one of

the CPB,DSP commands. Figure 5-5 shows the format of the CPUMTR breakpoint

display. The display consists of the 208-word exchange package from the current or last

breakpoint (displayed in 5 groups of 4 octal digits per line). All exchange package

information is displayed except RA, FL, RAE, FLE, and MA. The P address shown is

the actual breakpoint address.

The exchange package is followed by two 108-word blocks of central memory

(displayed in 4 groups of 5 octal digits per line). The address of the memory block to

be displayed can be speciﬁed using one of the CPB,DSP commands. Even if the

address speciﬁed in the CPB,DSP command is a relative address, the addresses shown

in the display are absolute; thus, giving the actual location in memory of the

displayed data.

Revision M DSD Commands 5-81

CPUMTR Breakpoint Display (C Display)

C. CENTRAL MEMORY.

00062662 5 4 7 2 6 1 0

00062663 0 1 0 6 1

00062664 0 55730 75230

00062665 0 4 3 5 5 7 1 3

00062666 0 5 5 7 3 1 4

00062667 0 1 0 2 2 0

00062670 0 54533 55676 Exchange

00062671 0 4 3 7 2 0 0 Package

00062672 7777 7777 7777 7777 0000

00062673 0000 0000 0000 0000 0003

00062674 3423 1200 0000 0000 0000 1SJ

00062675 0000 0000 0000 0000 0001

00062676 OOOO OOOO OOOO OOOO 0003

00062677 OOOO OOOO 0002 OOOO oooo

00062700 3423 1200 0002 OOOO OOOO 1SJ B

00062701 4000 0002 0001 0040 0344 5 B A 5C9

00054726 51200 00052 43324 20360 (P )8Z7C

00054727 11232 51300 00111 20303 ISUK IJCC

00054730 03330 54777 61000 46000 COE* D

00054731 43724 20760 51700 55010 8 7G ( E/H

00054732 02000 64037 61000 46000 B F54 D

00054733 10322 43405 51200 00057 HZT1E(P .

00054734 76770 20451 20735 15664 BD(P (. Memory

00054735 20251 12667 11242 43422 PUIV ITT1R Block

00054736 13227 03020 54776 46000 KR XP=

00054737 03030 54777 61000 46000 CCE* D

00054740 61600 54746 07410 55735 E*-G6E.2

00054741 51200 00144 51140 55010 (P A9(LE/H

00054742 21244 73110 03020 54776 QT*YHCBE*

00054743 03010 55735 43301 53110 CAE.28XMYH

00054744 20352 15114 36713 15773 P2Q(L3 Y.

00054745 04000 71362 61000 46000 D GK D

,^ms

Figure 5-5. CPUMTR Breakpoint Display

5-82 NOS Version 2 Analysis Handbook Revision M

j|^V

CPUMTR Breakpoint Precautions

The following precautions and information are important to the successful use of the

CPUMTR breakpoint package.

• The breakpoint must be set in executable code.

breakpoint address is hit, CPUMTR saves all of the registers in the display area,

clears the breakpoint, and then waits to receive a GO. The message CPUMTR

BREAKPOINT HIT is displayed at the system control point.

a From the time that the breakpoint is hit until the CPU continues, the system

clock stops. Therefore, you should reset the system clock (with the DSD TIME

command) after you have ﬁnished breakpoint processing.

alternating between the modes is not advised. Going from monitor to program

mode is probably safe, but going from program to monitor mode is almost certain

to fail causing CPUMTR to abort with the following message issued to the system

console:

CPUMTR ERROR EXIT.

« Since the C display is resident in RPL, changing displays is allowed (before the

breakpoint is hit) because you can always return to the C display. However, once

the breakpoint is hit until the CPU continues, CPUMTR is hung. This means that

all activity in the system stops until the CPU restarts. If DSD attempts to

request a monitor function, the following message appears in the B display:

NO MONITOR RESPONSE.

This is normal since CPUMTR is hung in the breakpoint code and the message

may be ignored. In this case, the DSD request that required the monitor function

is not processed.

• If you attempt to reference a CPUMTR block that is not loaded, the following

message appears on the system console:

INCORRECT ENTRY.

• The C display retains the last addresses displayed. For example, if you change to

another display and then come back to the C display it will still show the same

absolute addresses.

• Once a breakpoint has been hit and CPUMTR is waiting for a CPB,GO (or

another CPB,BKP) command, the contents of any register or memory location can

be changed by using the normal memory entry commands. To change a register,

simply change the corresponding memory location on the C display (as initiated by

the CPB,DSP command). The exchange package shown on the C display will be

reloaded to display the actual registers when execution resumes.

NOTE

Since CPUMTR has been halted, any memory entries made at this time must be

preceded by a 99 command to disable DSD syntax overlay processing and the

logging of the command to the error log.

Revision M DSD Commands 5-83

Debugging Commands

The following commands trap certain conditions and at the same time trace selected PP

and MTR to CPUMTR functions.

TRACE

TRAP

These debugging commands can be used to solve speciﬁc system problems. Care must

be exercised in using the commands in a production environment as performance may

be impacted signiﬁcantly.

The following entry must be made in the IPRDECK at deadstart.

ENABLE.TRACE.

In addition, system debug mode must be enabled at the system console (the console

must be unlocked to do this) in order to execute the TRAP and TRACE commands.

Command Description

TRACE.type, Permits speciﬁcation of monitor functions and other system

parameters. data. Entering the command without any parameters

(TRACE,PPU. or TRACE,MTR.) clears the trace function.

The following types may be traced.

type Description

PPU Traces PP to CPUMTR requests. Each trace

buffer entry consists of the following words:

• Time stamp.

• PP input register.

• PP output register.

• PP message buffer (6 words).

• Absolute address 1 (default is SMRL).

• Absolute address 2 (default is CMCL).

The format of the time stamp word is

•••dhmslll

In this format, dhms is bits 23-0 (day, hour,

minute, and second) of PDTL and 111 is bits

17-0 of RTCL.

The trace buffer can hold 20 entries. The

oldest entry in the buffer is overwritten by a

new entry once the buffer is ﬁlled. The DSDI

directive TBDUMP,PPU displays the buffer, as

well as the functions traced, the absolute

addresses saved, and the buffer IN pointer.

^^5V

5-84 NOS Version 2 Analysis Handbook Revision M

Debugging Commands

Command Description

TRACE,type,

parameter.

(Continued)

type

MTR

00S.

SET,ABS

SET,CPA

Description

Traces MTR to CPUMTR functions. Each trace

buffer entry consists of the following words:

Time stamp.

Function.

Absolute address 1 (default is SMRL).

Absolute address 2 (default is CMCL).

*MCT*

MCT* expands to the following for each entry

in the memory control table (MCT):

MCT word 1.

MCT word 2.

CPA/PCPA address 1 (default is STSW).

CPA/PCPA address 2 (default is FLSW).

Word TFSW from the CPA/PCPA.

Word JSNE from the EJT.

Depending on the number of pseudo-control

points deﬁned, there is room for approximately

10 entries in the trace buffer. The oldest entry

in the buffer is overwritten by a new entry

once the buffer is ﬁlled.

The DSDI directive TBDUMP,MTR displays the

buffer, as well as the functions traced, the

absolute and CPA/PCPA addresses saved, and

the buffer IN pointer.

Sets the absolute addresses of the CM words to

be saved in the trace buffer entry.

Sets the CPA/PCPA addresses of the CM words

to be saved in the trace buffer entry.

yHﬂmm&y

Revision M DSD Commands 5-85

Debugging Commands

I Command Description

I TRACE,type,

| parameters.

(Continued)

The following parameters are used with TRACE,MTR and

TRACE,PPU. Trace data is collected by CPUMTR immediately

before the selected function is processed.

parameters Description

* Traces all functions of the selected type.

Fi,F2,...,F3 Traces up to ﬁve functions. Functions are

speciﬁed by their symbolic names.

The following parameters are used with TRACE,SET,ABS and

TRACE,SET,CPA.

parameters Description

addri,addr2 Sets the absolute (TRACE,SET,ABS) or relative

(TRACE,SET,CPA) addresses of the words to be

saved in each trace buffer entry. The addresses

must be octal digits.

The following examples illustrate various forms of the TRACE

command.

.^S^y

Example Description

TRACE, PPU, ROCM, j ACM. Causes a trace entry to be created each

time a JACM or ROCM monitor function

is issued.

TRACE,MTR,ACSF. Causes a trace entry to be created each

time an ACSF function is issued.

TRACE,SET,ABS,21,152. Causes words 21s and 152s to be saved in

each trace buffer entry.

TRACE, SET, CP A, 21,57. Causes words 21s and 57s of each CPA and

PCPA to be saved for each MTR trace

buffer entry.

/*^™^*K

><5^v

5-86 NOS Version 2 Analysis Handbook Revision M

Debugging Commands

Command Description

TRAP,type,parameters. Halts the system immediately (in monitor mode) upon

detection of a speciﬁed condition. Only one type of trap can be

active at once. Entering the command without any parameters

(TRAP.) clears the trap.

The following types do not accept parameters.

type Description

jHv&tizis

NFL Checks for erroneous data in negative ﬁeld

length of executing jobs.

MCT Checks for invalid data in the memory control

table.

EJT Checks for invalid data in the executing job

table.

DJB Checks for jobs that are no longer executing,

but are failing to advance.

The following types accept parameters.

type Description

CPA Checks a speciﬁed control point area ﬁeld or

pseudo-control point area ﬁeld for a given

value.

MEM Checks a ﬁeld speciﬁed by an absolute address

(central memory only) for a given value.

The following parameters, which are associated with CPA amd

MEM, indicate the ﬁeld and value being checked.

parameters Description

W Absolute address (type = MEM) or offset

(type = CPA).

V Value (octal). If V=N, any nonzero ﬁeld value

halts the system.

H Left-most bit of ﬁeld. Decimal value ranges

from 0 to 59.8

L Right-most bit of ﬁeld. Decimal value ranges

from 0 to 59.

Revision M DSD Commands 5-87

Debugging Commands

Command Description

| TRAP,type,parameters. (Continued)

The following examples illustrate various forms of the TRAP

command.

Example Description

TRAP,CPA,20,N,47,36. Checks for any nonzero

value in bits 47 to 36 of

word 20 in each control

point area.

TRAP,MEM,37461,36,7,3 Checks for a value of 36 in

bits 7 to 3 of 37461.

5-88 NOS Version 2 Analysis Handbook Revision M

(* Express Deadstart Dump Interpreter

(DSDI)

j^N

f^'

C a l l i n g t h e E x p r e s s D e a d s t a r t D u m p I n t e r p r e t e r 6 - 3

Input Directives 6-6

Directive Format 6-7

List Control Directives 6-8

EJ — Force Page Eject 6-8

EJOFF — Turn Off Auto Page Eject 6-8

EJON — Turn On Auto Page Eject 6-8

PD — Reset Print Line Density . 6-8

*. — Comment in Subtitle Line . 6-9

File Manipulation and Control Directives 6-10

DISPOSE — Dispose Alternate List File to Print Queue 6-10

OUTPUT — Assign Output to Alternate List File 6-10

READ — Read Alternate Directives File 6-11

REWIND — Rewind File 6-11

Central Memory/Extended Memory Dump Directives 6-12

Dump Control Directives 6-12

ALLMEM — Extend Central Memory Dumps 6-12

CM — Set Memory Type to Central Memory 6-12

EC — Set Memory Type to Extended Memory • 6-13

RA — Reset Reference Address 6-13

RAC — Reset Reference Address to RA of Control Point 6-14

. UEC — Set Memory Type to User-Access Extended Memory f>14

Memory Dump Directives ... . 6-15

C — Dump Memory in Instruction Parcel Format . 6-15

D — Dump Memory in Byte Format 6-16

E — Dump Memory in Word Format 6-16

I — Dump 64-Bit Memory in Instruction Parcel Format 6-17

W — Dump 64-Bit Memory in Word Format 6-18

PP Dump Directives 6-19

AP — Dump Analysis of PP and PP Memory in Octal Line Format . 6-19

M P P — M o v e P P 6 - 2 0

P — Dump PP Memory in Octal Block Format 6-20

PF. — Dump FLPP Memory in Octal Block Format 6-21

PMS — Read PP Select Switches 6-21

PO — Dump 16-Bit PP Memory in Octal Block Format 6-22

PX — Dump 16-Bit PP Memory in Hexadecimal Block Format 6-22

Q — D u m p P P M e m o r y i n O c t a l L i n e F o r m a t 6 - 2 3

QF — Dump FLPP Memory in Octal Line Format 6-23

QOA — Dump 16-Bit PP Memory in Octal/ASCII Line Format 6-24

QOD — Dump 16-Bit PP Memory in Octal/Display Line Format 6-24

QXA — Dump 16-Bit PP Memory in Hexadecimal/ASCII Line Format 6-25

QXD — Dump 16-Bit PP Memory in Hexadecimal/Display Line Format 6-25

CMR Dump Directives 6-26

ACCOUNT — Dump Account Dayfile Buffer 6-26

CBT — Dump Control Buffer Table 6-27

CCT — Dump Channel Control Table 6-27

CP — Dump Active Control Point Areas 6-28

CPO — Reset Default list Options 6-30

CT — Dump Channel Tables 6-30

DAYFILE — Dump System Dayfile Buffer 6-30

DB — Dump Disk Buffers 6-30

DBW — Dump Buffered Device/Buffer Statistics/PP-I/O Buffer Link Tables .. 6-31

DDB — Dump Dayfile Dump Buffer 6-31

DP — Dump Dayfile Buffer Pointers 6-31

EICB — Dump Environment Interface Communication Buffer 6-31

EJT — Dump Executing Job Table .... 6-31

EPB — Dump Extended Memdry/PP Buffer 6-32

ERRLOG — Dump Error Log Buffer ..... 6-32

EST — Dump Equipment Status Table 6-32

FNT —- Dump System File Name/File Status Table .... 6-32

FOT — Dump Family Ordinal Table 6-32

HAT— Dump Hash Table 6-33

J C — D u m p J o b C o n t r o l A r e a f o r E a c h S e r v i c e C l a s s 6 - 3 3

LC— Dump Low Central Memory ....... 6-33

LDIS — Dump L-Display Buffer 6-34

LIDT — Dump Logical Identifier Table 6-34

MAINLOG — Dump Binary Maintenance Log Buffer 6-34

MCT — Dump Memory Control Table . .... 6-34

MST — Dump Mass Storage/Track Reservation Tables 6-35

MTR — Dump CPU Monitor . ..... 6-35

MTRQUEUE — Dump CPUMTR Request and Recall Queues . 6-36

ODIS — Dump Operator Display Buffer 6-36

PCP — Dump Pseudo-control Point ... 6-36

PLD' — Dump Peripheral Library Directory ..... 6-37

PP — Dump PP Communication Areas 6-37

PROBE — Dump PROBE Data Tables 6-37

PST— Dump Program Status Table and Entry Point Directory .... 6-37

PUT — Dump Physical Unit Table 6-38

QFT — Dump Queued File Table ........-. .. 6-38

RCL — Dump Resident Central Library 6-38

RPL — Dump Resident Peripheral Library 6-39

SAB — Dump System Attribute Block ... 6-39

SDA — Dump Extended Statistical Data Area .......;.. 6-39

SECDED — Dump SECDED Identifier Table................................ 6-39

SST — Dump Subsystem Control Point/Subsystem Assignment Tables 6-40

TBDUMP — Dump Trace Buffer ..... 6-40

Subsystem Dump/Analysis Directives .... 6-41

BATCHIO (BIO) — Dump Associated Memory for Analysis 6-41

IAF — Dump Associated Memory for Analysis ...... 6-42

MAGNET (MAG) — Dump Associated Memory for Analysis 6-44

RHF — Dump Associated Memory for Analysis .... 6-45

Hardware Register Dump Directives 6-46

FMFREG ■— Dump PP Registers . 6-46

IOUCR — Dump Concurrent PP Channel Registers .. ... . 6-46

IOUMR — Dump IOU Maintenance Registers . .,..• 6-46

LP VA — L oa d C en tr al M e mo ry P VA i n t o P se ud o R e g is te r 6 - 4 7

MEMMR — Dump Memory Maintenance Registers 6-47

PROCA — Dump Processor Operand Cache 6-48

PROCW — Dump Processor Controlware Part Number and Revision Level... 6-48

PROMR — Dump Processor Maintenance Registers .. 6-48

PROPM — Dump Processor Page Map . 6-49

PRORF — Dump Processor Register File .... ...., ..."... 6-49

PROSM— Dump Processor Segment Map 6-49

PROXP — Dump Processor Exchange Package 6-50

SC — Dump S/C Register 6-50

SETCPU— Set CPU Number 6-50

SETIOU — Set PP Dump Defaults 6-51

SETJPS— Change the JPS Register Value 6-51 ^*%

SETRMA — Convert Address into RMA 6-52

SETVEP — Set Virtual Address Parameters 6-52

TRACEBK — Traceback Stack Frame Save Area . 6-53

XP - Dump Deadstart Exchange Package 6-53

Buffer Controller Directive 6-54

Interactive Use of DSDI "6-55

Terminal Output Directives 6-57

C — Dump Memory in Instruction Parcel Format 6-57

CP — Dump Active Control Point Areas 6-58

D — Dump Memory in Byte Format 6-59

PP — Dump PP Communication Areas 6-60

Q — Dump PP Memory in Line Format 6-61

QOA, QOD, QXA, QXD •— Dump 16-Bit PP Memory in Line Format 6-62

Example of DSDI Terminal Use 6-63

Printer Output Listing Examples 6-65

/p*

(P*

j ^ v

^ Express Deadstart Dump Interpreter

f ( D S D I ) 6

The deadstart dump interpreter (DSDI) is a utility program that converts selected

portions of the binary information on an express deadstart dump (EDD) ﬁle into reports

to be listed. The EDD ﬁle is generated on magnetic tape by the EDD utility, which

may be run at deadstart time after a system malfunction has occurred. Refer to the

CIP User's Handbook for complete information concerning the use of EDD.

Selection of data to be listed by DSDI is provided through input directives, either on

an input ﬁle or on the DSDI command. Normal octal dumps of central memory,

extended memory, and PP memory can be produced by these directives as well as

specially formatted dumps of speciﬁc system tables and buffers.

EDD and DSDI offer these features:

® The EDD tape file contains a dump of memory, the executing exchange packages,

the CYBER 170 status and control (S/C) registers, where applicable; maintenance

registers for models 865 and 875 and CYBER 180-class machines, register ﬁle,

exchange package, processor control store, processor operand cache, processor page

map, processor segment map, where applicable; and all PPs except for PPO, which

can be saved by either copying it to another PP or by biasing the PP switch on

the deadstart panel. EDD can optionally dump all or part of extended memory and

the selected buffer controllers. EDD and DSDI permit analysis of a system

malfunction to be performed entirely ofﬂine.

• Because DSDI copies the EDD file to a word-addressable random file on mass

storage, dump data can be accessed in any order.

• EDD requires only a small amount of time during deadstart because the data is

transferred in binary form to magnetic tape.

until it is no longer needed. Thus, a selective listing can be generated at any

time.

y™srw?v

General information from the EDD ﬁle appears in the title and subtitle line of

each page of listed output. The ﬁrst 50 columns of each input directive are

included in the title line of the output list it produces. An input directive is

provided that enables insertion of comments into the subtitle line.

Use of DSDI is possible from an interactive terminal as well as from the batch

environment. The output produced by several directives is formatted for terminal

output (72 columns). From a batch environment, output is formatted for a

136-column printer.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-1

Express Deadstart Dump Interpreter (DSDI)

DSDI can also be used to create a message ﬁle containing the contents of the

maintenance registers (input/output unit, memory, and processor) that were dumped to

the EDD ﬁle for CYBER 180-class machines. This message ﬁle, which is similar in

format to the binary maintenance log (BML), can then be analyzed by an interpreter

progam, such as the Hardware Performance Analyzer (HPA), to produce maintenance

later in this section for additional information.

DSDI writes up to three messages to this message ﬁle: one message each for

input/output unit errors, memory errors, and processor errors. If the status summary

does not write the corresponding message to the message ﬁle.

In addition, the message ﬁle contains messages for any errors that were detected by

the dedicated fault tolerance (DFT) processor but not yet written to the BML at the

time the EDD ﬁle was created. These messages are identical in format to the ones that

would have been written to the BML by the normal logging process.

J^MW^V

6-2 NOS Version 2 Analysis Handbook Revision M

/0m»._

Calling the Express Deadstart Dump Interpreter

Processing of the EDD ﬁle is initiated with the DSDI command. The format of the

command is:

DSDI,plFp2 pn.

Each parameter pi is either a keyword or a keyword equated to a value. All

parameters are optional and order independent.

B=bmlﬁle

B omitted

D omitted

DMB

DMB omitted

Description

The 1- to 7-character name of the message ﬁle on which the

maintenance register messages and unlogged DFT error

messages are to be written. The message ﬁle is similar in

format to the BML and can be analyzed by HPA. This

parameter is valid only if maintenance register messages or

unlogged DFT error messages have been dumped to the EDD

ﬁle.

No message ﬁle is to be created.

This parameter creates a random dump ﬁle with the same name

as the EDD ﬁle and returns the original EDD ﬁle. The created

ﬁle can then be used as the dump ﬁle on subsequent executions

of DSDI, eliminating the need to read the entire dump tape on

each call.

No random dump ﬁle is to be created.

A binary dump ﬁle produced by the DMB or LOADBC command

is to be analyzed instead of an EDD ﬁle.

EDD ﬁle is to be analyzed.

When the DMB parameter is speciﬁed for a DMB-created dump,

DSDI uses directives as though the job were running at control

point 1. The DMB command dumps the exchange package,

central memory, and extended memory in binary (refer to the

NOS Version 2 Reference Set, Volume 3). Use the following

directives to dump selected portions of memory to be analyzed.

Directive Description

CP,1

RAC,1/C,fwa,lwa

RAC,1/D,fwa,lwa

RA,0/EC/C,fwa,lwa

'RA,0/EC/D,fwa,lwa

Dump exchange package.

Dump central memory in C format.

Dump central memory in D format.

Dump extended memory in C format.

Dump extended memory in D format.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-3

Calling the Express Deadstart Dump Interpreter

Pi Description

DMB omitted

F=dumpﬁle

F omitted

I=inﬁle

I omitted

L=outﬁle

L omitted

NR omitted

P omitted

PD = n

(Continued)

fwa and lwa are the ﬁrst and last word addresses of memory to

be dumped. When this parameter is speciﬁed to analyze a

LOADBC controlware dump, the only DSDI directive that can be

used is:

BCDUMP/ops.

Refer to the BCDUMP directive described later in this section

for a description of the ops parameter.

The 1- to 7-character name of the EDD ﬁle that DSDI will

interpret.

Same as F=DUMP.

The 1- to 7-character name of ﬁle on which input directives are

written.

Same as I=INPUT.

The 1- to 7-character name of ﬁle on which list output is to be

written. The user must save or print the ﬁle.

Same as L=OUTPUT, except that the ﬁle is automatically

printed.

EDD ﬁle is not to be rewound.

EDD ﬁle is to be rewound before processing.

CMR pointers from the running system are to be used. Selecting

this parameter causes the CMR pointers from the running

system to be used to locate tables and buffer areas on the EDD

ﬁle. This parameter is typically used when it is known that the

CMR pointers on the EDD ﬁle were destroyed by system

malfunction. Directives used to dump CMR will dump the

pointers contained on the EDD ﬁle, not those from the running

system.

This parameter should be used only when the conﬁguration of

the running system is the same as the system in use when the

EDD ﬁle was created.

CMR pointers from EDD ﬁle are to be used.

Print density in number of lpi (3, 4, 6, or 8).

, ^ ^ \

6-4 NOS Version 2 Analysis Handbook Revision M

Calling the Express Deadstart Dump Interpreter

Pi Description

PD omitted

Z omitted

Same as PD = 8.

Same as the job's default print density.

Input directives are contained on the command after the

terminator. The I parameter is ignored. This eliminates the need

to use a separate input ﬁle for the directives when only a few

directives are needed.

Input directives are not contained on the command. The system

uses the I parameter.

When input directives appear on the DSDI command, the ﬁrst

character following the command terminator is the separator

character for all directives on the command. Any display code

character that is not used in any of the directives, including a

space, can be used as the separator character. Each directive

must be preceded by a separator and terminated by a period.

The directives can extend to column 72 on the command.

Continuation lines are not permitted.

For example, (slant used for separator):

DSDI,Z./SC./XP./P./D,0,20000./EC./D,0,10000.

If the directives are included in the input ﬁle, the following

equivalent job would appear.

DSDI.

—EOR—

SC.

XP.

D,0,20000.

EC.

D,0,10000.

—EOI—

Dump CYBER 170 S/C register.

Dump executing exchange packages.

Dump all PPs.

Dump the ﬁrst 20000s locations of central memory.

Set memory type to extended memory.

Dump the ﬁrst 10000s locations of extended

memory.

A request for the EDD tape must precede the DSDI command. Since EDD writes

information on a labeled, 7- or 9-track tape at a density of 800 characters per inch

(cpi) for 7 track and 1600 or 6250 cpi for 9 track, the request should appear as follows:

LABEL,DUMP,D=density,tape,F=S,LB=KL,VSN=DUMP.

Parameter Description

density

tape

800 cpi for 7-track tape.

1600(PE) or 6250(GE) cpi for 9-track tape.

MT for 7-track tape.

NT for 9-track tape.

The request is presented in the resource mounting preview display and the job is rolled

out until the tape is mounted and assigned. Although the default express dump ﬁle

name (DUMP) is used in this example, a different ﬁle name can be speciﬁed, provided

the same ﬁle name is also speciﬁed on the DSDI command (F parameter).

Revision M Express Deadstart Dump Interpreter (DSDI) 6-5

Input Directives

DSDI input directives provide the capability to selectively dump only those portions of

the EDD ﬁle that are of interest. The input directives are grouped into the following

categories.

Type of Directive Description

List control

File manipulation and control

Central memory/extended

memory dump

PP dump

CMR dump

Subsystem dump/analysis

Hardware register dump

Enables you to control line printer page eject and

print density, and to specify comments in subtitle line

of the output listing.

Enables you to specify alternate ﬁles for DSDI input

directives and listing output.

Provides octal dumps of speciﬁed portion of central

memory or extended memory. Absolute or relative

addresses can be printed (three print formats are

available). Two additional dump formats are available

to print 64-bit memory in hexadecimal for CYBER

180-class machines.

Provides octal memory dumps of all or selected PPs

(two print formats are available). PP analysis data can

be included in the dumps. Six additional formats are

available to print 16-bit PP memory for CYBER

180-class machines.

Provides specially formatted dumps of selected areas

in central memory resident. These areas are speciﬁed

by name rather than by address.

Provides specially formatted dumps of subsystem

control points and associated tables and buffers.

Provides dumps of speciﬁed hardware registers.

y*^Hj!y

Several of the DSDI input directives have interactive capabilities. The output produced

by these directives is specially formatted for listing at an interactive terminal. Refer to

Interactive Use of DSDI later in this section for information concerning directive entry

and use of DSDI from an interactive terminal.

Refer to Printer Output Listing Examples, later in this section, for examples of

listings produced by several of the DSDI input directives.

6-6 NOS Version 2 Analysis Handbook Revision M

Directive Format

A directive has the following format.

directive, Pi ,P2,.•-.Pn-comments

Field Description

directive The directive name starts in column 1. It is terminated by a separator

or terminator character.

pi Parameter(s) for the directive. Depending on the requirements of the

directive, the directive may have no parameters or a number of

parameters.

comment Any characters following the directive terminator are considered

comments and are ignored by DSDI. However, the comments are

included with the directive in the title line of each page of the output

listing (a combined total of 50 characters appear in the title line).

separator You can use any character, including a space, to separate the ﬁelds of a

directive. The exceptions are:

A to Z

0 to 9

Refer to Interactive Use of DSDI later in this section for additional information

concerning directive entry from an interactive terminal.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-7

List Control Directives

The list control directives provide the capability to specify print density and page eject

options, and to add comments in the subtitle line of each page listed.

EJ — Force Page Eject

The EJ directive forces DSDI to issue a page eject function before listing the output

produced by the next directive processed. The EJ directive can also force a page eject

upon reaching a speciﬁed point on the page being printed. The page eject function is

performed automatically unless disabled by the EJOFF directive.

Format:

EJ,nn.

Parameter

Description

Force page eject only if less than nn decimal lines remain on the

current page. If nn is omitted, page eject is forced before listing the

output from the next directive processed.

EJOFF — Turn Off Auto Page Eject

The EJOFF directive disables auto page eject. Until this directive is processed, DSDI

automatically issues a page eject function before listing the output produced by each

new directive.

Format:

EJOFF.

EJON — Turn On Auto Page Eject

The EJON directive enables auto page eject (default condition). DSDI automatically

issues a page eject function before listing the output produced by each new directive

processed. This directive has no effect unless auto page eject has been disabled by the

EJOFF directive.

Format:

EJON.

PD — Reset Print Line Density

The PD directive resets the print line density to a value other than that speciﬁed on

the DSDI command.

Format:

PD.n.

Parameter Description

New print line density in number of lpi (3, 4, 6, or 8). If n is omitted or

an incorrect value is speciﬁed, a diagnostic message is issued.

6-8 NOS Version 2 Analysis Handbook Revision M

*. — Comment in Subtitle Line

The *. directive speciﬁes a comment that appears in the subtitle line of each page

listed.

Format:

*.ccc...ccc

Parameter Description

ccc...ccc Comment; up to 36 characters are printed.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-9

File Manipulation and Control Directives

File manipulation and control directives provide the capability to specify alternate ﬁles

for DSDI input directives and listing output.

DISPOSE — Dispose Alternate List File to Print Queue

The DISPOSE directive causes the alternate list ﬁle speciﬁed by the OUTPUT directive

to be disposed to the print queue. DSDI listing output then resumes on the original

output ﬁle. This directive has no effect unless output has previously been assigned to

an alternate list ﬁle (refer to the OUTPUT directive, next). Refer to Interactive Use of

DSDI later in this section for additional information concerning use of this directive

from an interactive terminal.

Format:

DISPOSE,username.

Parameter Description

username User name under which the remote batch terminal to receive the

listing is logged in. If username is omitted, the listing is printed at a

central site line printer.

OUTPUT — Assign Output to Alternate List File

The OUTPUT directive temporarily assigns DSDI listing output to a ﬁle other than

that speciﬁed on the DSDI command. When the alternate ﬁle is disposed to the print

queue (refer to the DISPOSE directive, above), output resumes on the original output

ﬁle. If the alternate ﬁle is not disposed, both the original and the alternate output ﬁles

remain at the job control point as local ﬁles. Refer to Interactive Use of DSDI later in

this section for additional information concerning use of this directive from an

interactive terminal.

Format:

OUTPUT.ﬁlename.

Parameter Description

ﬁlename Name of alternate list ﬁle (from 1 to 7 characters). Only one alternate

output ﬁle may be active at a time; ﬁlename cannot be the same name

as the normal output ﬁle. If ﬁlename is omitted, the system assumes

ﬁle name ALTRNT.

■<-=5\

6-10 NOS Version 2 Analysis Handbook Revision M

READ — Read Alternate Directives File

The READ directive causes DSDI to temporarily stop reading the current directives ﬁle

and begin reading directives from the speciﬁed record on the named alternate ﬁle or

from the current position if the record name is omitted. DSDI reads from the speciﬁed

alternate directives ﬁle until an end-of-record is encountered (end-of-ﬁle or empty

record if * is speciﬁed) and then resumes with the next directive on the original input

directives ﬁle.

Format:

READ.ﬁlename,rec,*.

Parameter Description

ﬁlename Name of alternate directives ﬁle (local ﬁle).

rec Optional record name. If rec is speciﬁed, ﬁle ﬁlename is searched for

record rec from the current position to end-of-ﬁle or an empty record. If

rec is not found, DSDI issues an error message. If rec is not speciﬁed,

DSDI reads directives from the current position to end-of-record. Records

must be in text format where the ﬁrst word of the record is the record

name, unless the ﬁle is assigned to an interactive terminal; then,

directives may be entered directly.

* Optional character specifying that DSDI is to read directives from all

records until an end-of-ﬁle or an empty record is encountered.

REWIND — Rewind File

The REWIND directive repositions the speciﬁed ﬁle to beginning-of-information.

Format:

REWIND,ﬁlename.

Parameter Description

ﬁlename Name of ﬁle to be rewound.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-11

Central Memory/Extended Memory Dump Directives

The central memory/extended memory dump directives provide the capability to dump

any portion of central memory or extended memory in instruction parcel, byte, or word

format. There are 6-bit display code character equivalents included with each format.

Either absolute or relative memory locations may be dumped. Refer to CMR Dump

Directives later in this section for directives used to dump speciﬁc portions of NOS

central memory resident (CMR).

Dump Control Directives

Dump control directives select the type of memory to be dumped (central memory or

extended memory) and the addressing mode to be used (absolute or relative).

ALLMEM — Extend Central Memory Dumps

The ALLMEM directive enables central memory dumps to extend past the central

memory boundary on machines with central memory extension. ^^

Format:

ALLMEM.

CM — Set Memory Type to Central Memory

The CM directive speciﬁes that subsequent C, D, and E directives dump central

memory locations. Unless the EC or TJEC directive is speciﬁed, central memory ^^

locations are dumped by default. '^^j

Format:

CM.

6-12 NOS Version 2 Analysis Handbook Revision M

Dump Control Directives

EC — Set Memory Type to Extended Memory

The EC directive speciﬁes that subsequent C, D, and E directives dump extended

memory or uniﬁed extended memory (UEM) locations. Unless the EC or UEC directive

is speciﬁed, central memory locations are dumped by default.

Format:

EC,emid.

EC.emid.fwa.

Parameter Description

emid Extended memory identiﬁcation; emid can be UEM or ESM. If emid is

omitted, UEM is assumed. This parameter is useful only on models 865,

875, and CYBER 180-class machines since all other models have only

one type of extended memory available.

fwa First word address (fwa) divided by 1000s. fwa is added to the fwa and

last word address (lwa) parameters in subsequent C, D, and E directives

when dumping UEM. If fwa is omitted, DSDI determines the beginning

of uniﬁed extended memory from the pointer in CMR and adds it to the

fwa and lwa parameters of the C, D, and E directives.

This parameter is ignored on any machine other than models 865 and

875 and CYBER 180-class machines.

RA — Reset Reference Address

The RA directive speciﬁes that subsequent C, D, and E directives dump memory

locations relative to a speciﬁed reference address. Unless the RA or RAC directive is

entered, absolute memory locations are dumped by default.

Format:

RA.nnnnnnn.

Parameter Description

nnnnnnn Reference address; addresses speciﬁed on subsequent C, D, and E

directives are relative to this address.

Clearing the reference address speciﬁed on the most recent RA or RAC directive

reenables absolute addressing. This is done by entering the RA directive in this format:

RA.O.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-13

Dump Control Directives

RAC — Reset Reference Address to RA of Control Point

The RAC directive speciﬁes that subsequent C, D, and E directives dump memory 1

locations relative to the reference address of a speciﬁed control point. Unless the RA or

RAC directive is entered, absolute memory locations are dumped by default.

Format:

RAC,nn.

Parameter Description

nn Control point number; addresses speciﬁed on subsequent C, D, and E

directives are relative to the reference address of this control point.

Refer to the description of the RA directive to reenable absolute addressing.

UEC — Set Memory Type to User-Access Extended Memory

The UEC directive speciﬁes that subsequent C, D, and E directives dump 7

user-accessible extended memory locations. Unless the EC or UEC directive is speciﬁed,

central memory locations are dumped by default.

Format:

UEC,fwa.

Parameter Description

fwa First word address divided by 1000s for user-accessible extended memory. ^\

fwa is added to the fwa and lwa parameters in subsequent C, D, and E

directives. If fwa is omitted, DSDI determines the beginning of

user-accessible extended memory from the pointer in CMR and adds it to

the fwa and lwa parameters of the C, D, and E directives.

.as^S

6-14 NOS Version 2 Analysis Handbook Revision M

yms

Memory Dump Directives

Memory dump directives specify the area of memory to be dumped and determine the

format of the output listing (refer to Printer Output Listing Examples later in this

section for a sample output listing). The CM and EC or UEC directives determine the

type of memory to be dumped (default is central memory). Absolute memory locations

are dumped unless relative addressing has been enabled (refer to RA and RAC

directives earlier in this section).

C — Dump Memory in Instruction Parcel Format

The C directive causes the speciﬁed locations of central memory or extended memory to

be dumped in four groups of ﬁve octal digits (three words per line) with 6-bit display

code character equivalents. Repetitive data is suppressed.

Format:

C,fwa,lwa.

jpfev

Parameter Description

fwa

lwa

First word address to be dumped (mandatory).

Last word address, plus one location, to be dumped. If lwa is omitted,

fwa+1 is assumed by default.

The output listing is read from top to bottom by column rather than across the page.

Refer to Interactive Use of DSDI later in this section for additional information

concerning use of this directive from an interactive terminal.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-15

Memory Dump Directives

D — Dump Memory in Byte Format

The D directive causes the speciﬁed locations of central memory or extended memory

to be dumped in ﬁve groups of four octal digits (three words per line) with 6-bit

display code character equivalents. Repetitive data is suppressed.

Format:

D,fwa,lwa.

Parameter Description

fwa

lwa

First word address to be dumped (mandatory).

Last word address, plus one location, to be dumped. If lwa is omitted,

fwa+1 is assumed by default.

The output listing is read from top to bottom by column rather than across the page.

Refer to Interactive Use of DSDI later in this section for additional information

concerning use of this directive from an interactive terminal.

E — Dump Memory in Word Format

The E directive causes the speciﬁed locations in central memory or extended memory

to be dumped in word format (four words per line) with 6-bit display code character

equivalents.

Format:

E,fwa,lwa.

Parameter Description

fwa First word address to be dumped (mandatory).

lwa Last word address, plus one location, to be dumped. If lwa is omitted,

fwa+1 is assumed by default.

.-^^SivX

6-16 NOS Version 2 Analysis Handbook Revision M

>fﬁiﬁfﬁtfSV

Memory Dump Directives

/ — Dump 64-Bit Memory in Instruction Parcel Format

The I directive is valid on CYBER 180-class machines only. It prints the speciﬁed

locations of central memory in four groups of four hexadecimal digits (three words per

line) with 7-bit ASCII code character equivalents. The second format is for NOS/VE

dumps.

Format:

I,fba,lba,ei.

I.fba,lba,asid.

Parameter Description

fba First byte address (fba), hexadecimal, to be printed (mandatory).

lba Last byte address (lba), plus one location, to be printed. If lba is

omitted, fba+8 is assumed by default.

ei If ei is 1, fba and lba specify byte addresses relative to byte 0 of the

environment interface. If ei is omitted, fba and lba specify absolute byte

addresses.

asid Speciﬁes an actual segment identiﬁer (ASID) for NOS/VE dumps. If you

specify this parameter, the fba and lba parameters must represent byte

offsets and have values in the range of from 0 to 7FFFFFFF

hexadecimal. This parameter is optional.

When you call DSDI from a batch job the output from this directive is printed in three

columns, read from top to bottom, one column at a time. When you call DSDI from an

interactive job the words are displayed one word per line.

For NOS/VE dumps, the addresses given for the fba and lba parameters must be real

memory addresses (RMAs) or, when translated to system virtual addresses (SVAs), the

actual segment identiﬁers (ASIDs) must match. Refer to appendix H for NOS/VE

address formats to use with the DSDI utility.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-17

Memory Dump Directives

W — Dump 64-Bit Memory in Word Format

The W directive is valid on CYBER 180-class machines only. It prints the speciﬁed

locations of 64-bit memory in hexadecimal (4 words of 16 digits per line) with 7-bit

ASCII code character equivalents. The second format is for NOS/VE dumps.

Format:

W,fba,lba,el.

W,fba,lba,asid.

Parameter Description

fba First byte address, hexadecimal, to be printed (mandatory).

lba Last byte address, plus one location, to be printed. If lba is omitted,

fba+8 is assumed by default. ^

ei If ei is 1, fba and lba specify byte addresses relative to byte 0 of the

environment interface. If ei is omitted, fba and lba specify absolute byte

addresses.

asid Speciﬁes an actual segment identiﬁer (ASID) for NOS/VE dumps. If you

specify this parameter, the fba and lba parameters must represent byte

offsets and have values in the range of from 0 to 7FFFFFFF

hexadecimal. This parameter is optional.

For NOS/VE dumps, the addresses given for the fba and lba parameters must be real

memory addresses (RMAs), or when translated to system virtual addresses (SVAs), the

actual segment identiﬁers (ASIDs) must match. Refer to appendix H for NOS/VE

address formats to use with the DSDI utility.

6-18 NOS Version 2 Analysis Handbook Revision M

PP Dump Directives

PP dump directives provide the capability to obtain a memory dump of all or selected

PPs. Two dump formats are available, block format and line format.

AP — Dump Analysis of PP and PP Memory in Octal Line Format

The AP directive causes PP memory to be dumped in line format with 6-bit display

code character equivalents (same format as the Q directive). Repetitive lines are

suppressed and zero bytes are represented by hyphens (—). An analysis of the PP is

printed before the memory dump. Analysis data includes the associated PP

communications area, resident entry point call addresses, read-only variables in direct

cells, and the values of the P, Q, K, and A registers (only applies to certain CYBER

180 class machines). Certain direct cell variables are veriﬁed and those in error are

indicated.

NOTE

Correct operation of this directive requires that the PP communication area on the

EDD ﬁle be intact.

Format!

P,ni,n2 nm.

Parameter Description

n» Number of PP to be dumped or a program name. If a program name is

speciﬁed, all PPs executing that program are dumped. A warning

message is issued if an incorrect number is speciﬁed or the program

name is not found in any PP. If ni is omitted, all active PPs are

dumped.

Refer to Printer Output Listing Examples for a sample of the printer output listing

produced by this directive.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-19

MPP — Move PP

The MPP directive causes the correct logical PP to be dumped if the logical position of

PPOO has been changed prior to the full dump to tape. If the PPS-O/PPS-1 toggle

switch has been toggled, the n parameter should not be speciﬁed. If PPOO has to be

moved to another PP via a deadstart panel program, the n parameter should be

speciﬁed.

Format:

MPP.n.

Parameter Description

Number of PP to which PPOO was moved, n cannot equal 0. If n is

omitted, 10 is assumed (CYBER 170 and CYBER 70 Computer Systems

with 20 PPs).

P — Dump PP Memory in Octal Block Format

The P directive causes PP memory to be dumped in block format, where each block

represents 64 words of memory. The blocks are read by column (top to bottom), where

each column contains eight 12-bit words in octal format numbered 0 through 7. There

are eight columns in each block, numbered 0 through 7. Repetitive data is not

suppressed and zero words are represented by hyphens (—). For models 865 and 875,

bytes 77768 and 77778 contain the PP's R register.

Format:

P»ni,n2,...,nm.

Parameter Description

ni Number of PP to be dumped. If omitted, all PPs are dumped.

r^^S,

6-20 NOS Version 2 Analysis Handbook Revision M

0$**\.

yims

PF — Dump FLPP Memory in Octal Block Format

The PF directive causes ﬁrst-level peripheral processor (FLPP)* memory to be dumped

in block format, where each block represents 64 words of memory. The blocks are read

by column (top to bottom), where each column contains eight 12-bit words in octal

format numbered 0 through 7. There are eight columns in each block, numbered 0

through 7. Repetitive data is not suppressed and zero words are represented by

hyphens (—).

Format:

PF,n1,n2,...,nII).

Parameter Description

m Number of FLPP to be dumped. If omitted, all FLPPs are dumped.

PMS — Read PP Select Switches

The PMS directive causes the dump of the S/C register (maintenance register for

models 865 and 875) to be read to determine the current value of the PP memory

select switches and the correct logical PP to be dumped, if the logical position of PPOO

has been changed prior to the full dump to tape. If the PP memory select switches

have been changed, this directive should be speciﬁed with the binary value of the

switches prior to the change. This directive is meaningful only on a CYBER 170

Computer System with the exception of models 815, 825, 835, 845, and 855.

Format:

PMS,n.

Parameter Description

n Previous select switch setting; 0 through lis. If n is the same value as

that read from the S/C register, this directive is not meaningful.

1. Hardware manuals deﬁne peripheral processors making up a peripheral processor subsystem as PPs, and

the ﬁrst-level peripheral processors as peripheral processing units (PPUs). In this manual, ﬁrst-level

peripheral processors are referred to as FLPPs. FLPPs are available only on model 176.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-21

PO — Dump 16-Bit PP Memory in Octal Block Format

The PO directive is valid on CYBER 180-class machines only. It prints PP memory in

block format, where each block represents 64 words of memory. DSDI represents each

16-bit word by six octal digits. Words whose upper 4 bits are zero are represented by

four octal digits. Blocks are read by columns (top to bottom), where each column

contains eight 16-bit words numbered 0 through 7. There are eight columns in each

block, numbered 0 through 7. Repetitive data is not suppressed and zero words are

represented by hyphens (—). The PP's R register is displayed preceding the first line

of the dump.

,*^\

Format:

P0,ni,n2,...,nm.

Parameter Description

ni Number of PP to be dumped. If omitted, all PPs are printed.

PX — Dump 16-Bit PP Memory in Hexadecimal Block Format

The PX directive is valid on CYBER 180-class machines only. It prints 16-bit memory

in block format, where each block represents 256 words of memory. Each 16-bit word is

represented by four hexadecimal digits. Blocks are read by columns (top to bottom),

where each column contains sixteen 16-bit words in hexadecimal format numbered 0

through F. There are 16 columns in each block, numbered 0 through F. Repetitive data

is not suppressed and zero words are represented by hyphens (—). The PP's R register

is displayed preceding the ﬁrst line of the dump.

Format:

PX,n1fn2 nm.

Parameter Description

ni Number of PP to be printed. If omitted, all PPs are printed.

,^^y

6-22 NOS Version 2 Analysis Handbook Revision M

yams

Q — Dump PP Memory in Octal Line Format

The Q directive causes PP memory to be dumped in line format with 6-bit display code

character equivalents. Each line contains 16 bytes (PP words) printed in two sets of 8

bytes in octal format. Each set consists of an address, 8 bytes, and display code

character equivalents. Repetitive lines are suppressed and zero bytes are represented by

hyphens (—). Refer to Interactive Use of DSDI later in this section for additional

information concerning use of this directive from an interactive terminal.

Format:

Q,ni,n2,...,nm.

Parameter Description

ni Number of PP to be dumped. If omitted, all PPs are dumped.

QF — Dump FLPP Memory in Octal Line Format

The QF directive causes FLPP memory to be dumped in line format with 6-bit display

code character equivalents. Each line contains 16 bytes printed in two sets of 8 bytes

in octal format. Each set consists of an address, 8 data bytes, and 6-bit display code

character equivalents. Repetitive lines are suppressed and zero bytes are represented by

hyphens (—).

Format:

QF,n1fn2, . - -,nm.

Parameter Description

ni Number of FLPP to be dumped. If omitted, all FLPPs are dumped.

2. Hardware manuals deﬁne peripheral processors making up a peripheral processor subsystem as PPs, and

the ﬁrst-level peripheral processors as peripheral processing units (PPUs). In this manual, ﬁrst-level

0^s peripheral processors are referred to as FLPPs. FLPPs are available only on model 176.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-23

QOA — Dump 16-Bit PP Memory in Octal/ASCII Line Format

The QOA directive is valid on CYBER 180-class machines only. It prints PP memory "l

in line format with 7-bit ASCII code character representation. Each line contains 16

words printed in two sets of eight. Each set consists of an octal address, eight words

(six digits per word if the upper 4 bits are nonzero, four digits otherwise), and 7-bit

ASCII code character equivalents. Repetitive lines are suppressed and zero words are

represented by hyphens (—). The PP's R register is displayed preceding the first line

of the dump. Refer to Interactive Use of DSDI later in this section for additional

information concerning use of this directive from an interactive terminal.

Format:

QOA,n1fn2,. ...nm.

Parameter Description

m Number of PP to be printed. If omitted, all PPs are printed.

QOD — Dump 16-Bit PP Memory in Octal/Display line Format

The QOD directive is valid on CYBER 180-class machines only. It prints PP memory

in line format with 6-bit display code character representation. Each line contains 16

words printed in two sets of eight. Each set consists of an octal address, eight words

(six digits per word if the upper 4 bits are nonzero, four digits otherwise), and 6-bit

display code character equivalents. Repetitive lines are suppressed and zero words are

represented by hyphens (—). The PP's R register is displayed preceding the first line

of the dump. Refer to Interactive Use of DSDI later in this section for additional

information concerning use of this directive from an interactive terminal.

Format:

Q0D,n,,n2 nm.

Parameter Description

m Number of PP to be printed. If omitted, all PPs are printed.

6-24 NOS Version 2 Analysis Handbook Revision M

QXA — Dump 16-Bit PP Memory in Hexadecimal/ASCII Line Format

The QXA directive is valid on CYBER 180-class machines only. It prints PP memory

in line format with 7-bit ASCII code character representation. Each line contains 16

words printed in two sets of eight. Each set consists of a hexadecimal address, eight

words (four hexadecimal digits per word), and 7-bit ASCII code character equivalents.

Repetitive lines are suppressed and zero words are represented by hyphens (—). The

PP's R register is displayed preceding the ﬁrst line of the dump. Refer to Interactive

Use of DSDI later in this section for additional information concerning use of this

directive from an interactive terminal.

Format:

QXA,n,,n2,...,nm.

Parameter Description

ni Number of PP to be printed. If omitted, all PPs are printed.

QXD — Dump 16-Bit PP Memory in Hexadecimal/Display Line Format

The QXD directive is valid on CYBER 180-class machines only. It prints PP memory

in line format with 6-bit display code character representation. Each line contains 16

words printed in two sets of eight. Each set consists of a hexadecimal address, eight

words (four hexadecimal digits per word), and 6-bit display code character equivalents.

Repetitive lines are suppressed and zero words are represented by hyphens (—). The

PP's R register is displayed preceding the ﬁrst line of the dump. Refer to Interactive

Use of DSDI later in this section for additional information concerning use of this

directive from an interactive terminal.

Format:

QXD,n,,n2,...,nm.

Parameter Description

ni Number of PP to be printed. If omitted, all PPs are printed.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-25

CMR Dump Directives

The CMR dump directives provide the capability to selectively dump speciﬁed areas of

central memory resident. Refer to the NOS Version 2 Systems Programmer's Instant

for detailed illustrations of central memory resident.

Successful use of the CMR dump directives is dependent upon the integrity of central

memory resident at the time EDD was performed. Most important is the integrity of

the CMR pointers on the EDD ﬁle. If these pointers are not intact, the dump

produced by DSDI may prove meaningless. Thus, if it is suspected that the CMR

pointers are not intact, specifying the P keyword on the DSDI command allows DSDI

to use the CMR pointers from the running system. This option should be used only

when the conﬁguration of the running system is the same as the system in use at

the time the EDD ﬁle was created. If the CMR pointers on the EDD ﬁle are not

intact, the integrity of the other areas of central memory is also questionable. In this

case, the output produced by the CMR dump directives may be unpredictable.

ACCOUNT — Dump Account Dayfile Buffer

The ACCOUNT directive causes the account dayﬁle pointers and buffer to be dumped

in word format (four words per line) with 6-bit display code character equivalents. This

format is the same format as that for the E memory dump directive. This directive

also dumps the buffer in a line-by-line format, as on the DSD A display.

Format:

ACCOUNT.

.<b3bSX

6-26 NOS Version 2 Analysis Handbook Revision M

00S.

00&S

CBT — Dump Control Buffer Table

The CBT directive causes all control buffer table entries to be dumped unless ordinal

numbers are speciﬁed, then only the speciﬁed control buffer table entries are dumped.

The dump is formatted to reﬂect the appropriate parameter ﬁelds for each word.

Control buffer table entries that are output to a printer are listed in two columns. The

format is the same format as that for the MST directive. The control buffer table

entries can also be output to a terminal, but only when ordinal numbers are speciﬁed.

Control buffer table entries that are output to a terminal are listed in a single column

format.

Format:

CBT,ni,n2,...,nra.

Parameter Description

ni Ordinal number of the control buffer table entry to be dumped. If ni is

omitted, all control buffer table entries are dumped.

CCT — Dump Channel Control Table

The CGT directive causes all nonzero channel control table entries to be dumped unless

ordinal numbers are speciﬁed, then only the speciﬁed channel control table entries are

dumped. The dump is formatted to reﬂect the appropriate parameter ﬁelds for each

word. Channel control table entries that are output to a printer are listed in two

columns. The format is the same format as that for the MST directive. The channel

control table entries can also be output to a terminal, but only when ordinal numbers

are speciﬁed. Channel control table entries that are output to a terminal are listed in

a single column format.

Format:

C C T, n i , n 2 , . . . , n r a .

Parameter Description

ni Ordinal number of the channel control table entry to be dumped. If m is

omitted, all nonzero channel control table entries are dumped.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-27

CP — Dump Active Control Point Areas

The CP directive causes all active control point areas, or a selected subset, to be

dumped. List options provide the capability to dump only desired portions of the control

point area and other control point related data. Speciﬁed without any parameters, the

CP directive dumps all active control point areas and pseudo-control point areas. Refer

to Interactive Use of DSDI later in this section for additional information concerning

use of this directive from an interactive terminal.

Format:

CP.

CP,ni/opSi,n2/ops2,.. .tnm/oj3Sm.

Parameter Description

m Control point number.

0Psi List options; a string of up to 10 characters indicating the portion of the

control point area, or control point related data, to be dumped. If list

options are speciﬁed, they apply only to the control point number with

which they are associated. Valid options are:

opsi Description

/*^V

X Dumps the exchange package, parameter summary, control

point's entry in the executing job table, and contents of the

memory locations indicated by the values in each exchange

package register. These register values are treated as

relative addresses within the ﬁeld length; invalid addresses

are represented with contents of zero.

The current command and special entry point parameters are

not printed for the system control point. Refer to Printer

Output Listing Examples later in this section for a sample of

the printer output listing produced.

T Provides a detailed dump of the control point area with a

description of each parameter ﬁeld and NOSTEXT symbol for

each word.

A Dumps the job dayﬁle pointers and buffer in word format

with 6-bit display code character equivalents. This format is

the same format as the E memory dump directive. This

option also dumps the buffer in a line-by-line format, as on

the DSD A display.

F3 Provides a dump of the 3-word FNT entry, EST, and mass

storage track chain, if one exists, for all ﬁles attached to the

speciﬁed control point.

aX^s

3. This option does not dump information for the system control point.

6-28 NOS Version 2 Analysis Handbook Revision M

CP — Dump Active Control Point Areas

Parameter Description

opsi (Continued)

opsi Description

C Dumps the ﬁeld length of a speciﬁed control point in

instruction parcel format with 6-bit display code character

equivalents. This format is the same format as the C

memory dump directive. Repetitive data is suppressed.

D Dumps the ﬁeld length of a speciﬁed control point in byte

format with 6-bit display code character equivalents. This

format is the same format as the D memory dump directive.

Repetitive data is suppressed.

E Dumps the ﬁeld length of a speciﬁed control point in word

format with 6-bit display code character equivalents. This

format is the same format as the E memory dump directive.

G Dumps the control point area in instruction parcel format

with 6-bit display code character equivalents. This format is

the same format as the C memory dump directive.

H Dumps the control point area in byte format with 6-bit

display code character equivalents. This format is the same

format as the D memory dump directive.

I Dumps the control point area in word format with 6-bit

display code character equivalents. This format is the same

format as the E memory dump directive.

M Dumps the user-accessible extended memory of a speciﬁed

control point in byte format with 6-bit display code character

equivalents. This format is the same format as the D

memory dump directive.

N Dumps the negative ﬁeld length of a speciﬁed control point.

This format is the same format as the D memory dump

directive.

P Provides a dump and analysis of all active PPs associated

with control point n.

default If n is speciﬁed with no corresponding list options, options

XTAF are selected automatically. Refer to Printer Output

Listing Examples later in this section for a sample of the

printer output listing produced. The default options selected

can be changed with the CPO directive.

All of the list options speciﬁed for a particular control point area (up to 10) are

processed. For example, if the C and D options are both speciﬁed, the control point

ﬁeld length is dumped twice, once in instruction parcel (C) format and again in byte

(D) format.

4. This option does not dump information for the system control point.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-29

CPO — Reset Default List Options

The CPO directive selects a new string of default list options for the CP directive.

Format:

CPO,ops.

Parameter Description

ops New default list options for CP directive, a string of up to 10

characters. Refer to the description of the CP directive for a list of valid

option characters.

CT — Dump Channel Tables

The CT directive causes the channel status table, EJT assignment table, and channel

controlware table to be dumped in byte format with 6-bit display code character

equivalents. This format is the same format as that for the D memory dump directive.

Format:

CT.

DAYFILE — Dump System Dayfile Buffer

The DAYFILE directive causes the system dayﬁle pointers and buffer to be dumped in

word format (four words per line) with 6-bit display code character equivalents. This

format is the same format as that for the E memory dump directive. This directive

also dumps the buffer in a line-by-line format, as on the DSD A display.

Format:

DAYFILE.

DB — Dump Disk Buffers

The DB directive causes all disk buffers, or a selected subset, to be dumped to a

printer in byte format with 6-bit display code character equivalents. This format is the

same format as that for the D memory dump directive.

Format:

DB,id,ni,n2 nm.

Parameter Description

id Speciﬁes the machine identiﬁer of the mainframe from which to dump

the disk buffers. If id is omitted, the default machine identiﬁer is taken

from CMR of the mainframe that is dumping the disk buffers.

ni Number assigned to a disk buffer to be dumped. If ni is omitted, all

active disk buffers are dumped.

. ^ " s ^ V

6-30 NOS Version 2 Analysis Handbook Revision M

DBW — Dump Buffered Device/Buffer Statistics/PP-I/O Buffer Link Tables

The DBW directive causes the list control words from the buffered device table, the

buffer statistics table, and the PP-I/O buffer link table to be dumped. The dump is

formatted to reﬂect the appropriate parameter ﬁelds for each word. Output to a printer

is listed in two columns. Output to a terminal is listed in one column. The format is

the same format as that for the MST directive.

Format:

DBW.

DDB — Dump Dayfile Dump Buffer

The DDB directive causes the dayﬁle dump buffer to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive.

Format:

DDB.

DP — Dump Dayfile Buffer Pointers

The DP directive causes the dayﬁle buffer pointers to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive.

Format:

DP.

EICB — Dump Environment Interface Communication Buffer

The EICB directive causes the environment interface (EI) communication buffer to be

dumped. The dump is formatted to reﬂect the appropriate parameter ﬁelds and

COMSDST symbols for each word. This directive also dumps the dedicated fault

tolerance/operating system (DFT/OS) buffer in instruction parcel format (four groups of

four hexadecimal digits with 7-bit ASCII code character equivalents).

EJT — Dump Executing Job Table

The EJT directive causes the executing job table to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive.

Format:

EJT.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-31

EPB — Dump Extended Memory/PP Buffer

The EPB directive causes the extended memory/PP buffer to be dumped in byte format

with 6-bit display code character equivalents. This format is the same format as that

for the D memory dump directive.

Format:

EPB.

ERRLOG — Dump Error Log Buffer

The ERRLOG directive causes the error log pointers and buffer to be dumped in word

format (four words per line) with 6-bit display code character equivalents. This format

is the same format as that for the E memory dump directive. This directive also

dumps the buffer in a line-by-line format, as on the DSD A display.

Format:

ERRLOG.

EST — Dump Equipment Status Table

The EST directive causes the equipment status table to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive.

Format:

EST.

FNT — Dump System File Name/File Status Table

The FNT directive causes the system ﬁle name/ﬁle status table (FNT/FST) to be

dumped in byte format with 6-bit display code character equivalents. This format is the

same format as that for the D memory dump directive.

Format:

FNT.

FOT — Dump Family Ordinal Table

The FOT directive causes the family ordinal table to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive.

Format:

FOT.

-/°fiS%

6-32 NOS Version 2 Analysis Handbook Revision M

y^PS^v

HAT — Dump Hash Table

The HAT directive causes all nonzero hash table entries to be dumped unless ordinal

numbers are speciﬁed, in which case only the speciﬁed hash table entries are dumped.

The dump is formatted to reﬂect the appropriate parameter ﬁelds for each word. Hash

table entries that are output to a printer are listed in two columns. The format is the

same format as that for the MST directive. The hash table entries can also be output

to a terminal, but only when ordinal numbers are speciﬁed. Hash table entries that are

output to a terminal are listed in a single-column format.

Format:

H AT, n i , n 2 , . . . , n m .

Parameter Description

ni Ordinal number of the hash table entry to be dumped. If ni is omitted,

all nonzero hash table entries are dumped.

JC — Dump Job Control Area for Each Service Class

The JC directive causes the job control area for each job service class to be dumped.

The dump is formatted to reﬂect the appropriate parameter ﬁelds and NOSTEXT

symbol for each word. This directive also dumps the service class control table (SCT) in

a line-by-line format.

Format:

JC.

LC — Dump Low Central Memory

The LC directive causes DSDI to dump the contents of low central memory (that is,

central memory locations 0 through 177s). Each word is divided into the appropriate

parameter ﬁelds. Each ﬁeld is listed on a separate line with a description of the

parameter. The absolute address and 6-bit display code character equivalents are also

listed for each word.

Format:

LC.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-33

LDIS — Dump L-Display Buffer

The LDIS directive causes the L-display buffer to be dumped in instruction parcel

format with 6-bit display code character equivalents. This format is the same format as

that for the C memory dump directive.

Format:

LDIS.

LIDT — Dump Logical Identifier Table

The LIDT directive causes the logical identiﬁer table to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive.

Format:

LIDT.

MAINLOG — Dump Binary Maintenance Log Buffer

The MAINLOG directive dumps the binary maintenance log pointers and buffer in word

format, four words per line, with 6-bit display code character equivalents.

Format:

MAINLOG.

MCT — Dump Memory Control Table

The MCT directive dumps the contents of the memory control table (MCT). For each

control point and pseudo-control point, there are two lines. The ﬁrst line displays the

control point (or pseudo-control point) having the next lower reference address, the one

having the next higher reference address, the unassigned CM Fl/lOOs between the

control point (or pseudo-control point) having the next higher reference address, and

the job's reference address and ﬁeld length. The second line displays the corresponding r^\

information for extended memory. '

Format:

MCT.

6-34 NOS Version 2 Analysis Handbook Revision M

0ms

yrtftsfﬂm^K

MST — Dump Mass Storage/Track Reservation Tables

The MST directive causes all mass storage and track reservation tables to be dumped

unless EST ordinals are speciﬁed, in which case only the speciﬁed MSTs are dumped.

The dump format for the mass storage tables reﬂects the appropriate parameter ﬁelds

and NOSTEXT symbol of each word. The portion of the dump describing the track

reservation tables is presented in byte format with 6-bit display code character

equivalents. The track link byte ordinal and status bits (three groups of 4 bits) are

indicated for each word. Refer to Printer Output Listing Examples later in this section

for a sample of the printer output listing produced by this directive.

Format:

MST,esti,est2,....estn -

Parameter Description

esti EST ordinal of equipment whose mass storage table is to be dumped. If

esti is omitted, all mass storage and track reservation tables are

dumped.

MTR — Dump CPU Monitor

The MTR directive causes the CPU monitor to be dumped. Exchange packages are

dumped in exchange package format while the program area is dumped, using relative

addressing, in instruction parcel format with 6-bit display code character equivalents.

This format is the same format as that for the C memory dump directive.

Format:

MTR.

j#^V

Revision M Express Deadstart Dump Interpreter (DSDI) 6-35

MTRQUEUE — Dump CPUMTR Request and Recall Queues

The MTRQUEUE directive causes all CPUMTR request and recall queues, or a selected

subset, to be dumped.

Format:

MTRQUEUE,idents.

Parameter Description

idents A string of up to four characters indicating the queues to be dumped.

The queue identiﬁers can be speciﬁed in any order and are dumped in

the order speciﬁed. Default is to dump all four queues. Valid queue

identiﬁers are:

idents Queue Name

C CPUCIO request.

P PP request.

R Recall.

W CPU request.

ODIS — Dump Operator Display Buffer

The ODIS directive causes the operator display buffer (used for communication between

an operator and a remote diagnostic user) to be dumped in instruction parcel format "a^\

with 6-bit display code character equivalents. This format is the same format as that

for the C memory dump directive.

Format:

ODIS.

PCP — Dump Pseudo-control Point

The PCP directive causes all active pseudo-control point areas, or a selected subset, to

be dumped. The format is the same as that for the CP control point dump directive.

,'*^^\

6-36 NOS Version 2 Analysis Handbook Revision M

0SS

PLD — Dump Peripheral Library Directory

The PLD directive causes the PP library directory to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive.

Format:

PLD.

PP — Dump PP Communication Areas

The PP directive causes the contents of all PP communication areas to be dumped in

byte format with 6-bit display code character equivalents. This format is the same

format as that for the D memory dump directive. The control point assignment,

channel assignment, and monitor function are listed with each communication area.

Refer to Printer Output Listing Examples later in this section for a sample of the

printer output listing produced by this directive. Refer to Interactive Use of DSDI later

in this section for additional information concerning use of this directive from an

interactive terminal.

Format:

PP.

PROBE — Dump PROBE Data Tables

The PROBE directive causes the PROBE data tables to be dumped in byte format with

6-bit display code character equivalents.

Format:

PROBE.

PST — Dump Program Status Table and Entry Point Directory

The PST directive causes the program status table and the entry point directory to be

dumped in byte format with 6-bit display code character equivalents. This format is the

same format as that for the D memory dump directive.

Format:

PST.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-37

PUT — Dump Physical Unit Table

The PUT directive causes all nonzero physical unit table entries to be dumped unless

ordinal numbers are speciﬁed, then only the speciﬁed physical unit table entries are

dumped. The dump is formatted to reﬂect the appropriate parameter ﬁelds for each

word. Physical unit table entries that are output to a printer are listed in two columns.

The format is the same format as that for the MST directive. The physical unit table

entries can also be output to a terminal, but only when ordinal numbers are speciﬁed.

Physical unit table entries that are output to a terminal are listed in a single column

format.

Format:

PUT,n!,n2 nm.

Parameter Description

ni Ordinal number of the physical unit table entry to be dumped. If ni is

omitted, all nonzero physical unit table entries are dumped.

QFT — Dump Queued File Table

The QF.T directive causes the queued ﬁle table to be dumped in byte format with 6-bit

display code character equivalents. This format is the same format as that for the D

memory dump directive.

Format:

QFT.

RCL — Dump Resident Central Library

The RCL directive causes the resident CPU library to be dumped in instruction parcel

format with 6-bit display code character equivalents. This format is the same format as

that for the C memory dump directive. A warning is printed if the resident central

library is empty.

Format:

RCL.

6-38 NOS Version 2 Analysis Handbook Revision M

0m**~

/|fBN

RPL — Dump Resident Peripheral Library

The RPL directive causes the resident PP library to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive. A header line for each PP program dumped indicates the

name of the program and its length in bytes. Each succeeding line contains 10 bytes

(two central memory words) of the PP program. The PP address of the ﬁrst byte in

each line, relative to address zero of the PP, is also listed.

Format:

RPL.

SAB — Dump System Attribute Block

The SAB directive causes the system attribute block to be dumped in byte format with

6-bit display code character equivalents. This format is the same format as that for the

D memory dump directive.

SDA — Dump Extended Statistical Data Area

The SDA directive causes the extended statistical data area to be dumped. Each word

is divided into the appropriate parameter ﬁelds and is listed with a description of the

parameter. The absolute address and 6-bit display code character equivalents are also

listed for each word.

Format:

SDA.

SECDED — Dump SECDED Identifier Table

The SECDED directive causes the single error correction double error detection

(SECDED) identiﬁer table to be dumped in byte format with 6-bit display code

character equivalents. This format is the same format as that for the D memory dump

directive.

Format:

SECDED.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-39

SST — Dump Subsystem Control Point/Subsystem Assignment Tables

The SST directive causes the subsystem control point table and subsystem assignment

table to be dumped in byte format with 6-bit display code character equivalents. This

format is the same format as that for the D memory dump directive.

Format:

SST.

TBDUMP — Dump Trace Buffer

The TBDUMP directive dumps the speciﬁed trace buffer. The functions traced are

listed, followed by the absolute and control point area addresses that were saved, and

the buffer IN pointer. The trace buffer data is displayed in the same format as that for

the D memory dump directive.

Format:

TBDUMP,MTR.

or .

TBDUMP,PPU,

Parameter Description

MTR

PPU

MTR to CPUMTR function trace buffer.

PP to CPUMTR function trace buffer.

>*^v

6-40 NOS Version 2 Analysis Handbook Revision M

0$ms

y$*s

Subsystem Dump/Analysis Directives

The subsystem dump/analysis directives provide the capability to selectively dump

portions of central and/or PP memory associated with a speciﬁc subsystem (BIO, IAF,

MAG, and RHF). Although many other directives already described in this section can

be used to dump the same areas of memory, these directives dump those areas most

frequently analyzed when subsystem-related malfunctions occur. In addition, many of

the dumps are specially formatted to provide a detailed description of the area being

dumped. If the speciﬁed subsystem was not active at the time the EDD ﬁle was

created, an error message is issued.

BATCHIO (BIO) — Dump Associated Memory for Analysis

The BATCHIO directive causes the areas of central and/or PP memory that are most

frequently analyzed when BATCHIO malfunctions are indicated to be dumped. The

areas and type of memory dumped are determined by the list options speciﬁed.

Formats:

BATCH10,ops.

BIO,ops.

Parameter Description

°PS List options; a string of characters indicating the areas of memory to be

dumped. If no options are speciﬁed, all three options (PBN) are selected

by default and are processed in order as listed.

ops Description

P Provides analysis and full memory dump of PPs having

resident copies of 1CD, 110, QAP, QAC, or DSP. The output

listing generated is the same (in format and content) as that

produced by entering the AP directive in this format:

AP,1CD,1I0,QAP,QACfDSP.

Refer to the description of the AP directive earlier in this

section for additional information.

B Provides specially formatted dumps of each active BATCHIO

buffer point. Included with the dump of each buffer point is

the associated equipment type and FET, as well as EST and

FNT/FST entries.

N Provides a dump of the negative ﬁeld length associated with

the BATCHIO control point in byte format with 6-bit display

code character equivalents. This format is the same format as

that for the D memory dump directive.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-41

IAF — Dump Associated Memory for Analysis

The IAF directive causes the areas of central and/or PP memory that are most

frequently analyzed when IAF malfunctions to be dumped. The areas and type of

memory are determined by the list options speciﬁed. The IAF current entry word

(SSPA) is always printed at the beginning of the listing, in byte format, regardless of

which list options are speciﬁed.

Format:

IAF,ops.

Parameter Description

ops List options; a string of up to ﬁve characters indicating the areas of

memory to be dumped. If no options are speciﬁed, four options (ETLP)

are selected by default and processed in order as listed.

ops Description

C Provides analysis of the IAF command table.

E Dumps the IAF reentry table in byte format (two words per

line) with 6-bit display code character equivalents. The ﬁrst

word in each line is preceded by its ordinal within the

table.

T Provides a specially formatted dump of the IAF terminal

table in which each word reﬂects the appropriate parameter

ﬁelds. The message status table entry is included when "*^\

dumping network terminal tables. In addition, each word is

preceded by a description of the parameter ﬁelds and its

COMSREM symbol. Terminal table entries that are empty,

except for having status bit 0 in word 3 (VROT entry) set,

are not printed.

L Provides a dump of pot link table and all pots. The pot link

table is dumped in byte format with pot link byte ordinals

indicated for each word, but no 6-bit display code character ^^

equivalents. Repetitive pot link table entries are suppressed. '^7

The pots are dumped in word format, three lines per pot,

with the ﬁrst line containing only the pot number.

6-42 NOS Version 2 Analysis Handbook Revision M

IAF — Dump Associated Memory for Analysis

Parameter Description

ops (Continued)

ops Description

Provides analysis and full memory dump of all PPs having

resident copies of TLX, 1TA, ITN, and 1T0. This option also

provides an analysis and dump of all PPs having resident

copies of 1R0 and 1RI that are associated with control points

of interactive origin. The output listing generated is the same

(in format and content) as that produced by entering the AP

directive in this format:

AP.TLX,1TA,1TN,1T0,1R0,1RI.

The exception is that the AP directive also dumps all PPs

having copies of 1RO and 1RI rather than only those

associated with control points of interactive origin. Refer to

the description of the AP directive, earlier in this section, for

additional information.

/gS*y

Revision M Express Deadstart Dump Interpreter (DSDI) 6-43

MAGNET (MAG) — Dump Associated Memory for Analysis

The MAGNET directive causes the areas of central and/or PP memory that are most 1

frequently analyzed when a malfunction within the Magnetic Tape Subsystem is

indicated to be dumped. The areas and type of memory dumped are determined by the

list options speciﬁed.

Format:

MAGNET,ops.

MAG,ops.

Parameter Description

ops List options; a string of characters indicating the areas of memory to

be dumped. If no options are speciﬁed, all four options (UQSP) are

selected by default and are processed in order as listed. ***h

ops Description

U Provides a specially formatted dump of the Magnetic Tape

Subsystem unit descriptor tables with associated FET, EST,

and FNT/FST. Each word of a unit descriptor table is

formatted to reﬂect appropriate parameter ﬁelds. In addition,

each word is preceded by a description of the parameter

ﬁelds and its COMSMTX symbol. If extended labels are

present, they appear with the FET in the output listing. The "^

FET also indicates the address and control point number of

the user.

Q Provides a dump of the Magnetic Tape Subsystem queue

table in byte format (two words per line) with 6-bit display

code character equivalents. The ﬁrst word in each line is

preceded by its ordinal within the table.

S Provides a dump of the Magnetic Tape Subsystem active

stage job table, staging tape VSN list, and stage request ^^)

table, if they are deﬁned. The tables are dumped in byte

format (two words per line) with 6-bit display code character

equivalents. The ﬁrst word in each line is preceded by its

ordinal within the table.

P Provides analysis and full memory dump of all PPs having

resident copies of IMT. The output listing generated is the

same (in format and content) as that produced by entering

the AP directive in this format:

AP.1MT.

Refer to the description of the AP directive earlier in this

section for additional information.

6-44 NOS Version 2 Analysis Handbook Revision M

RHF — Dump Associated Memory for Analysis

The RHF directive causes the areas of central and/or PP memory that are most

frequently analyzed when RHF malfunctions are indicated to be dumped. The areas and

type of memory dumped are determined by the list options speciﬁed.

Format:

RHF,ops.

Parameter Description

ops List options; a string of characters indicating the areas of memory to be

dumped. If no options are speciﬁed, all three options (ACP) are selected

by default and are processed in order as listed.

ops Description

A Provides a dump of the RHF dayﬁle buffer in word format

/^ with 6-bit display code character equivalents. This format is

the same format as that for the E memory dump directive.

This option also dumps a standard dayﬁle.

C Provides a dump of the RHF ﬁeld length in byte format with

6-bit display code character equivalents. This format is the

same format as that for the D memory dump directive.

>sffff\

Provides analysis and full memory dump of all active PPs

associated with the control point.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-45

Hardware Register Dump Directives

The hardware register dump directives provide the capability to dump speciﬁed

hardware registers.

FMFREG — Dump PP Registers

The FMFREG directive is valid on CYBER 180-class machines only. For each PP

PP register contents.

Format:

FMFREG.

IOUCR — Dump Concurrent PP Channel Registers

The IOUCR directive is valid only on CYBER machines that support concurrent

channels and concurrent PPs. For each concurrent PP channel speciﬁed, DSDI prints

the channel number, channel type, IOU status register contents, T register contents,

fault status mask, and test mode operand generator. The channel number and channel

type are octal; all of the other ﬁelds are hexadecimal.

Format:

IOUCR,chi,ch2,...,chn.

Parameter Description

chi Channel number of the concurrent PP channel register to be printed. If

omitted, all concurrent PP channel registers are printed.

IOUMR — Dump IOU Maintenance Registers

The IOUMR directive is valid on CYBER 180-class machines only. For each

input/output unit (IOU) maintenance register speciﬁed, DSDI prints the hexadecimal

containing error indicators, DSDI prints a description of each error.

Format:

IOUMR,ﬁrst,last.

Parameter Description

ﬁrst First register (hexadecimal) to be printed. If omitted, printing begins

with register 00.

last Last register +1 (hexadecimal) to be printed. If omitted, printing ends

with register ﬁrst+1.

If no parameters are speciﬁed, DSDI prints all maintenance registers.

^WsSky

6-46 NOS Version 2 Analysis Handbook Revision M

0*®**..

LPVA — Load Central Memory PVA into Pseudo Register

The LPVA directive is valid on CYBER 180-class machines only. You use this directive

when you are doing an analysis of a NOS/VE dump tape. It retrieves a copy of 48 bits

starting at the speciﬁed address and loads them into the pseudo register named process

virtual address (PVA). This value is printed at your terminal and saved for use by

subsequent directives.

Format:

LPVA,address,offset.

Parameter Description

address Speciﬁes the beginning byte address of the 48 bits loaded into PVA.

You can enter this parameter in any of the acceptable NOS/VE address

formats. Refer to appendix H for NOS/VE address formats to use with

the DSDI utility. This parameter is required.

offset Speciﬁes the hexadecimal number of bytes from the start-address to

begin loading the 48-bit address. This parameter is optional.

MEMMR — Dump Memory Maintenance Registers

The MEMMR directive is valid on CYBER 180-class machines only. For each memory

maintenance register speciﬁed, DSDI prints the hexadecimal register number, the

DSDI prints a description of each error.

Format:

MEMMR,ﬁrst,last.

Parameter Description

ﬁrst First register (hexadecimal) to be printed. If omitted, printing begins

with register 00.

last Last register+ 1 (hexadecimal) to be printed. If omitted, printing ends

with register ﬁrst+1.

If no parameters are speciﬁed, DSDI prints all maintenance registers.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-47

PROCA — Dump Processor Operand Cache

The PROCA directive is valid on CYBER 180 model 990 and 995 machines only. This ^

directive prints the contents of the processor operand cache (POC) record from the EDD

ﬁle. The dump format has three columns. The ﬁrst column contains the speciﬁed word

numbers, the second column contains the contents of the corresponding words from the

control memory part of the POC record in hexadecimal, and the third column contains

the contents of the corresponding words from the data memory part of the POC record

in hexadecimal.

Format:

PROCA,ﬁrst,last.

Parameter Description

ﬁrst First word number (hexadecimal) to be printed. If omitted, printing

begins with word 0.

last Last word number (hexadecimal) to be printed. If you specify ﬁrst but

omit last, printing ends with word ﬁrst+1. If you omit both ﬁrst and

last, printing ends with word FFFi6.

PROCW — Dump Processor Controlware Part Number and Revision Level

The PROCW directive is valid on CYBER 180-class machines only. It prints the

processor controlware part number and the revision level. For model 990 and 995

machines, DSDI extracts the controlware revision level and date from the last 256-bit

word of the processor control store (PCS) record from the EDD ﬁle. <*^

Format:

PROCW.

PRO MR — Dump Processor Maintenance Registers

The PROMR directive is valid on CYBER 180-class machines only. For each processor

maintenance register speciﬁed, DSDI prints the hexadecimal register number, the

descriptions of each error are printed.

Format:

PROMR,ﬁrst,last.

Parameter Description

ﬁrst First register (hexadecimal) to be printed. If omitted, printing begins

with register 00.

last Last register+1 (hexadecimal) to be printed. If omitted, printing ends

with register ﬁrst-I-1.

If no parameters are speciﬁed, all maintenance registers for both processors are

printed. If ﬁrst and last are speciﬁed, then the speciﬁed registers are dumped for all

processors.

6-48 NOS Version 2 Analysis Handbook Revision M

PROPM — Dump Processor Page Map

The PROPM directive is valid on CYBER 180 model 990 and 995 machines only. The

directive prints the contents of the processor page map (PPM) record from the EDD

ﬁle. The listing consists of 128 data lines plus column headings. Each line contains a

set number (0 to 3), the entry number (0 to lFi6) within its set, and the contents of

that entry in hexadecimal.

Format:

PROPM.

PRORF — Dump Processor Register File

The PRORF directive is valid on CYBER 180-class machines only. For all CYBER

180-class machines except models 990 and 995, it prints the maintenance channel

interface port number followed by the hexadecimal register number and contents of

each processor register speciﬁed.

Format:

PRORF,ﬁrst,last.

Parameter Description

ﬁrst First register to be printed. If omitted, printing begins with register 00.

last Last register +1 to be printed. If omitted, printing ends with register

ﬁrst+1.

If no parameters are speciﬁed, the entire register ﬁle for both processors is printed. If

ﬁrst and last are speciﬁed, then the speciﬁed registers are dumped for all processors.

For CYBER 180 model 990 and 995 machines, enter the PRORF directive without

parameters. DSDI produces a listing of the entire contents of each of the following

records from the EDD ﬁle: PRF, PIS, PRG, and PRH. Refer to Printer Output Listing

Examples later in this section for a sample of the listing DSDI produces for model 990

and 995 machines.

PROSM — Dump Processor Segment Map

The PROSM directive is valid on CYBER 180 model 990 and 995 machines only. The

directive prints the contents of the processor segment map (PSM) record from the EDD

ﬁle. The listing consists of 64 data lines plus column headings. Each line contains a

set number (0 or 1), the entry number (0 to lFi6) within its set, and the contents of

that entry in hexadecimal.

Format:

PROSM.

yms

Revision M Express Deadstart Dump Interpreter (DSDI) 6-49

PROXP —- Dump Processor Exchange Package

PROXP — Dump Processor Exchange Package

The PROXP directive is valid on CYBER 180-class machines only. It prints the

maintenance channel interface port number followed by the hexadecimal register

number, the contents of each processor register, and (for most registers) a description

of the register's contents. The exchange packages for all processors are printed.

Format:

PROXP.

For NOS/VE dumps, use the following format:

PROXP,address.

Parameter Description

address Speciﬁes the ﬁrst byte address of the exchange package. You can enter

this parameter in any of the acceptable NOS/VE address formats. Refer

to appendix H for NOS/VE address formats to use with the DSDI

utility. If you omit this parameter, the current contents of the central

processor registers are printed. If the memory address you specify is

for the stack frame save area, the only registers printed are the

registers recorded in the save area.

SC — Dump S/C Register

The SC directive is valid only on CYBER 170 Computer Systems with the exception of

models 815, 825, 835, 845, and 855; it causes the S/C registers (maintenance registers a^s

for models 865 and 875) to be dumped. ■■ '

Format:

SC.

SETCPU — Set CPU Number

The SETCPU directive is valid on any CYBER model that supports 170 mode in both

CPUs. It sets up processor pointers for the CPU speciﬁed in the directive. "^l

Format:

SETCPU,number.

Parameter Description

number Speciﬁes the number of the CPU (0 or 1). Default is 0.

6-50 NOS Version 2 Analysis Handbook Revision M

0ms

SETIOU — Set PP Dump Defaults

The SETIOU directive presets the parameter defaults for processing PP memory dumps

to allow printing of concurrent PP memory.

Format:

SETIOU,opt ion.

Parameter Description

option Speciﬁes whether to process concurrent PPs or nonconcurrent PPs. If no

option is speciﬁed, option N is assumed.

option Description

Process concurrent PPs to allow printing 8192 words of

concurrent PP memory. This option is valid only on CYBER

machines that support concurrent channels and concurrent

PPs.

N Process nonconcurrent PPs to allow printing 4096 words of PP

memory.

>#^N

SETJPS — Change the JPS Register Value

The SETJPS directive is valid on CYBER 180-class machines only. You use this

directive when you are doing an analysis of a NOS/VE dump tape. It sets the pseudo

Format:

SETJPS,address.

Parameter Description

address Speciﬁes the location of, or the actual new value for, the JPS register.

You can enter this parameter in any of the acceptable NOS/VE address

formats. Refer to appendix H for NOS/VE address formats to use with

the DSDI utility. If you specify address as a process virtual address

(PVA) or system virtual address (SVA), the system loads the value stored

at that location into the pseudo JPS register. If you specify address as a

real memory address (RMA), the system loads that value into the pseudo

JPS register. This parameter is required.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-51

SETRMA — Convert Address into RMA

The SETRMA directive is valid on CYBER 180-class machines only. You use this

directive when you are doing an analysis of a NOS/VE dump tape. It converts the

speciﬁed address into a real memory address (RMA) and saves that value in a pseudo

Format:

SETRMA,address.

Parameter Description

address Speciﬁes the address to be converted to an RMA. You can enter this

parameter in any of the acceptable NOS/VE address formats. Refer to

appendix H for NOS/VE address formats to use with the DSDI utility.

This parameter is required.

>^™^v

SETVEP — Set Virtual Address Parameters

The SETVEP directive is valid on CYBER 180-class machines only. You can use this

directive when you are doing an analysis of a NOS/VE dump tape. It changes the

virtual .memory parameters used for the NOS/VE DSDI memory display directives (I

and W).

Format:

SETVEP,address,lengthtmask,mps.

Parameter Description

address

length

mask

mps

Speciﬁes the beginning address of the page table as a hexadecimal

number of from 1 to 6 digits. This parameter is required.

Speciﬁes the length of the page table as a hexadecimal number of from

1 to 6 digits. This parameter is required.

Speciﬁes the value of the page size mask as a hexadecimal number of

from 1 to 6 digits. This parameter is required.

Speciﬁes the contents of the pseudo register named monitor process state

(MPS) as a hexadecimal number of from 1 to 6 digits. This parameter is

required.

6-52 NOS Version 2 Analysis Handbook Revision M

TRACEBK — Traceback Stack Frame Save Area

The TRACEBK directive is valid on CYBER 180-class machines only. You use this

directive when you are doing an analysis of a NOS/VE dump tape. It displays the

exchange package found at the speciﬁed address; and, for each stack frame, the module

name and contents of the registers saved.

Format:

TRACEBK,address.

Parameter Description

address Speciﬁes the address of the exchange package where the traceback

starts. You can enter this parameter in any of the acceptable NOS/VE

address formats. Refer to appendix H for NOS/VE address formats to use

with the DSDI utility. If you omit this parameter, the system uses the

exchange package currently in the central processor.

0ms.

( If you enter this directive interactively, the output includes just the minimum save

area (the P register and address registers AO, Al, and A2). If you enter this directive

in a batch job with line printer output, the output contains the entire stack frame as

saved by the task. The actual contents depends on the information saved by the task.

XP - Dump Deadstart Exchange Package

The XP directive causes the CPU exchange package executing at the time of deadstart

to be dumped. If there are two CPUs in the system, both exchange packages in

0s execution at the time of deadstart are dumped.

Format:

XP.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-53

Buffer Controller Directive

The buffer controller directive provides the capability to dump speciﬁed buffer

controllers and list veriﬁcation information for controlware dump ﬁles produced by the

LOADBC utility. The BCDUMP directive causes the selected buffer controllers to be

dumped.

Format:

BCDUMP/ops.

BCDUMP ,CCi,CC2 ccn.

BCDUMP,CCi/OPSi ,CC2/0PS2, . . . ?CCn/0PSn.

Parameter Description "^

cci Channel number of buffer controller to be dumped. If no channels are

speciﬁed, all buffer controllers are dumped.

opsi List options; a character string indicating the line format, interpretation,

and veriﬁcation of the selected buffer controllers to be dumped. If no

options are speciﬁed, options H and D are selected by default.

opsi Line Format

H Hexadecimal line format (default).

0 Octal line format.

Options A and D indicate interpretation.

opsi Interpretation

A 7-bit ASCII code interpretation.

D 6-bit display code interpretation (default). ^^\

The following options indicate veriﬁcation.

opsi Veriﬁcation

V List veriﬁcation information (used with the controlware

dump ﬁle produced by the LOADBC utility).

V omitted Do not list veriﬁcation information.

6-54 NOS Version 2 Analysis Handbook Revision M

0^s.

jgpsv

Interactive Use of DSDI

Incorporated within DSDI is an interactive facility that allows several of the directives

already described in this section to be entered interactively from a terminal. This

interactive facility is designed to provide the following additional capabilities.

• Allows preliminary examination of the EDD file to determine which areas should

be listed in detail at a line printer.

• Allows examination of certain areas of the EDD file not listed during normal

operational procedures following a system malfunction. Typically, predeﬁned

portions of the EDD ﬁle are listed following a system malfunction.

• Allows online examination of the EDD file from a remote location.

Refer to Example of DSDI Terminal Use later in this section for an example showing

interactive use of DSDI.

When the DSDI command is entered from an interactive terminal, there will be a 10-

to 60-second delay before input directives can be entered. During this delay, DSDI is

copying the EDD ﬁle to a random-access mass storage ﬁle. The length of the delay

depends on device speed and current system activity. When DSDI is able to accept

input directives, it will issue the following prompt to the terminal.

ENTER DIRECTIVES- -

Directives are entered following the question mark prompt. Only one directive can be

entered at a time, and each directive is restricted to one line. The format is the same

as described for batch input (refer to Directive Format earlier in this section).

Generally, any of the DSDI input directives can be entered at an interactive terminal.

However, the output produced by many of the directives is formatted for listing only

at a line printer (136 columns) and cannot be listed at the terminal (72 columns).

The L parameter on the DSDI command initially determines the disposition of the list

output. If a ﬁle name is not speciﬁed, list output is assigned to ﬁle OUTPUT by

default (that is, the terminal). In this case, entry of directives that produce output

that cannot be listed at the terminal results in the message:

DIRECTIVE RESTRICTED TO PRINTER OUTPUT.

If a list output ﬁle name is speciﬁed on the DSDI command, all input directives can be

entered at the terminal. All list output (including error messages) is written to the

speciﬁed ﬁle.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-55

Interactive Use of DSDI

These two input directives are provided to further control the disposition of list output:

Directive Description

OUTPUT,ﬁlename. Assigns output to alternate ﬁle ﬁlename (ﬁle name OUTPUT

is not allowed; that is, alternate list output cannot be assigned

to the terminal). If ﬁlename is omitted, the system assumes

ﬁle name ALTRNT. While this directive is active, all input

directives can be entered at the terminal. All list output

(except error messages) is written to ﬁle ﬁlename and is

formatted for transmission to a line printer. Error messages

are written directly to the terminal.

DISPOSE. Disposes the alternate list ﬁle (speciﬁed in the OUTPUT

directive) to the print queue. Output will be printed at the

central site line printer. All subsequent list output resumes on

the original output ﬁle speciﬁed on the DSDI command.

Refer to File Manipulation and Control Directives earlier in this section for additional

information concerning use of these directives.

6-56 NOS Version 2 Analysis Handbook Revision M

yms

Terminal Output Directives

The following directives produce output formatted for listing at an interactive terminal.

C — Dump Memory in Instruction Parcel Format

The C directive causes the speciﬁed locations of central memory or extended memory to

be dumped in four groups of ﬁve octal digits (one word per line) with 6-bit display code

character equivalents. No pagination is processed for terminal output. The CM, EC, and

UEC directives (refer to Central Memory/Extended Memory Dump Directives earlier in

this section) determine the type of memory to be dumped; default is central memory.

The RA or RAC directive (refer to Central Memory/Extended Memory Dump Directives

earlier in this section) must be entered to dump relative addresses; default is absolute

addressing.

Format:

C, fwa, l wa.

Parameter Description

fwa

lwa

First word address to be dumped (mandatory).

Last word address, plus one location, to be dumped. If omitted, fwa+l

is assumed by default.

Figure 6-1 shows the terminal output produced by the C directive.

? C,6230,6240

0006230 34240 10100 00012 50036 1TAA AU 3

0006231 00764 70000 ooooo 10113 AAK

0006232 04154 70000 ooooo 10113 DM* AAK

0006233 00004 67446 74000 10005 - A E

0006234 05153 05700 ooooo ooooo EMX.

0006235 ooooo ooooo ooooo ooooo

0006236 ooooo ooooo ooooo ooooo

0006237 ooooo 00005 05111 14422 EEII9R

Figure 6-1. C Directive Output

Revision M Express Deadstart Dump Interpreter (DSDI) 6-57

CP — Dump Active Control Point Areas

The CP directive causes the job sequence name and control point area address for each 1

control point to be dumped.

Format:

CP.

Figure. 6-2 shows the terminal output produced by the CP directive.

CP 01 C P 02 C P 0 3 C P 0 4 CP 0 5 CP 0 6 C P 07 C P 1 0

IAF AALT NAM AAKW AAAG AAAF AAAE

0200 0400 0600 1000 1200 1400 1600 2000

CP 11 C P 1 2 CP 1 3 C P 1 4 CP 1 5 CP 16 C P 1 7 CP 20

AAAD AAAC AAAB AALN AALU

2200 2400 2600 3000 3200 3400 3600 4000

CP 21 C P 22 C P 2 3 C P 2 4 CP 2 5 CP 2 6 C P 27 C P 3 0

RBF

4200 • 4400 4600 5000 5200 5400 5600 6000

CP 31 CP 32 CP 33 CP 34 •<—Control Point Number

" A G B I O S Y S « * — J o b S e q u e n c e N a m e a t C o n t r o l P o i n t

6200 6400 6600 7000 •<— Control Point Area Address

Figure 6-2. CP Directive Output

6-58 NOS Version 2 Analysis Handbook Revision M

D — Dump Memory in Byte Format

The D directive causes speciﬁed locations of central memory or extended memory to be

dumped in ﬁve groups of four octal digits (one word per line) with 6-bit display code

character equivalents. No pagination is processed for terminal output. The CM, EC, and

UEC directives (refer to Central Memory/Extended Memory Dump Directives earlier in

this section) determine the type of memory to be "dumped; default is central memory.

The RA or RAC directive (refer to Central Memory/Extended Memory Dump Directives

earlier in this section) must be entered to dump relative addresses; default is absolute

addressing.

Format:

D,fwa,lwa.

Parameter Description

fwa

lwa

First word address to be dumped (mandatory).

Last word address, plus one location, to be dumped. If omitted, fwa+1

is assumed by default.

J ^ W y,

Figure 6-3 shows the terminal output produced by the D directive.

? D,6230,6240.

0006230 3424 0101 0000 0125 0036 1TAA AU 3

0006231 0076 4700 0000 0001 0113 * AAK

0006232 0415 4700 0000 0001 0113 DM* AAK

0006233 0000 4674 4674 0001 0005 - - A E

0006234 0515 3057 0000 0000 0000 EMX.

0006235 0000 0000 0000 0000 0000

0006236 0000 0000 0000 0000 0000

0006237 0000 0000 0505 1111 4422 EEII9R

Figure 6-3. D Directive Output

Revision M Express Deadstart Dump Interpreter (DSDI) 6-59

PP — Dump PP Communication Areas

The PP directive causes the PP number, executing program name, control point

assignment, and input register address for each PP communication area to be dumped.

Format:

pp.

Figure 6-4 shows the terminal output produced by the PP directive.

PP Number

Program in Execution 1 Control

\ /

Point Assignment

PPOO PP01 PP02 PP03 PP04 \^PP05 /PP06 PP07 PP10 PP11

WTR-34 OSD-34 QAC-16 1RO-05 1SJ-34 1WT-31 OAC-14

7200 7210 7220 7230 7240 7250 7260 7270 7300 7310

PP20 PP21 PP22 PP23 PP24 PP25 \ PP26 PP27 PP30 PP31

PIP-03 CPD-34 QAC-02 110-32 1MA-01

7320 7330 7340 7350 7360 7 3 7 0 \ 7 4 0 0 7410 7420 7430

PPOO Input Register Address

7440

y*3i|y

Figure 6-4. PP Directive Output

"**%*

6-60 NOS Version 2 Analysis Handbook Revision M

00ms

00S

Q — Dump PP Memory in Line Format

The Q directive causes the speciﬁed locations of PP memory to be dumped in line

format. Each line contains 8 bytes (PP words) with 6-bit display code character

equivalents. Repetitive lines are suppressed and zero bytes are represented by

hyphens (—).

Format:

Q,n,fwa,lwa.

Parameter Description

fwa

lwa

NOTE

Number of PP to be dumped.

First word address to be dumped.

Last word address, plus one location, to be dumped.

fwa and lwa are automatically adjusted so that the dump limits fall within a multiple

of 108 words.

This format is valid only for terminal output. If attempted from a job of batch origin

or while an alternate list ﬁle is active, the fwa and lwa parameter will be interpreted

as PP numbers.

Figure 6-5 shows the terminal output produced by the Q directive.

? Q . 5 ,0,100

0000 0003 2020 3340 0614 4334 0117 CPPo5 FL 81A0

0010 0064 0001 7772 0100 0006 1073 1401 6072 A F H L A

0020 2250 3225 0027 0012 4402 5747 5751 R/2U W J93.*.(

0030 0011 7646 0001 0141 0600 I - AA6F

0040 1501 1116 2014 0074 0203 MAINPL BC

0050 3404 2330 0035 6213 1707 4000 6675 1DSX 2 KOG 5

0060 4521 6101 0001 0153 0001 0532 +Q A AAS AE2

0070 0001 0100 1000 0003 6000 6250 6251 6252 AA H C / ( )

Figure 6-5. Q Directive Output

Revision M Express Deadstart Dump Interpreter (DSDI) 6-61

QOA, QOD, QXA, QXD — Dump 16-Bit PP Memory in Line Format

QOA, QOD, QXA, QXD — Dump 16-Bit PP Memory in line Format

These four directives are valid on CYBER 180-class machines only. Each directive

prints speciﬁed locations of 16-bit PP memory in line format. Each line contains eight

PP words in octal or hexadecimal with 6-bit display code or 7-bit ASCII.code character

representations. Repetitive lines are suppressed and zero bytes appear as hyphens (—).

Numeric Character

Directive Format Representation Representation

QOA,n,fwa,lwa,R. Octal 7-bit ASCII code

QOD,n,fwa,lwa,R. Octal 6-bit display code

Q X A , n , f w a , l w a , R . H e x a d e c i m a l 7 - b i t A S C I I c o d e

QXD,n,fwa,lwa,R. Hexadecimal 6-bit display code

Parameter Description

n Number of PP to be printed.

fwa First word address to be printed.

lwa Last word address, plus one location, to be printed.

R If speciﬁed, the R register is printed.

This directive format is valid only for terminal output. If it is used in a job of batch

origin or while an alternate list ﬁle is active, DSDI interprets the fwa and lwa

parameters as PP numbers.

6-62 NOS Version 2 Analysis Handbook Revision M

Example of DSDI Terminal Use

0ms

Example of DSDI Terminal Use

This example illustrates how DSDI might be used following a system malfunction to

analyze portions of the EDD ﬁle from an interactive terminal. It is assumed that an

EDD ﬁle was created during normal system recovery procedures. Vertical spacing has

been expanded to permit commentary. The example begins after the login sequence has

been completed.

batch

RFL.O.

/label,dump,vsn=dump,lb=kl,f=s,mt,d=800

/get.altdir

/ d s d i .

ENTER DIRECTIVES - -

? output,a 1 tout

? read.altdir

? dispose.

EXPRESS DUMP COMPLETE (FL USED XXXXXXB)

Enter Batch subsystem.

The LABEL command is entered to

assign the EDD dump tape to this job.

Use of the vsn parameter allows the job

to be rolled out while the tape is

mounted and assigned.

Retrieve alternate directives ﬁle ALTDIR

(refer to the example in ﬁgure 6-6).

Calls DSDI, which copies EDD tape to a

random mass storage ﬁle.

Enter terminal output directives (refer to

Terminal Output Directives earlier in

this section) to list any portion of the

EDD ﬁle at the terminal. DSDI issues

the prompt (?) when it is ready to accept

a new directive.

List output produced by subsequent

directives is written to local ﬁle

ALTOUT. This allows entry of directives

that produce line printer formatted

output.

All input directives in alternate

directives ﬁle ALTDIR are read and

processed. List output is written to local

ﬁle ALTOUT. DSDI does not request

terminal input until the last directive on

ALTDIR is processed.

Dispose local ﬁle ALTOUT to the print

queue for listing at the central site line

printer. Output produced by subsequent

directives is listed at the terminal.

DSDI is terminated by entering a

carriage return in response to the ?

prompt.

In summary, the following operations were performed by DSDI. First, small areas of

the dump ﬁle were listed at the terminal for preliminary examination. This was done

both to analyze the cause and effect of the system failure, as well as determine the

extent of line printer listings required. An appropriate comment may be placed in the

list ﬁle subtitle at this time via the *.ccc...ccc directive.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-63

Example of DSDI Terminal Use

Next, directives were entered to generate the necessary line printer listings. These

listings are generally extensive, or contain specially formatted output that cannot be

listed at the terminal. Thus, output was written to an alternate list ﬁle named

ALTOUT.

After all necessary directives had been entered from the terminal, an alternate

directives ﬁle (ALTDIR) was read. ALTDIR is a permanent ﬁle containing input

directives necessary to obtain a printer listing of speciﬁc areas in the dump ﬁle that

are frequently examined following a system failure (for example, CMR tables and

buffers, PP memory, and so forth). Figure 6-6 illustrates a typical alternate directives

ﬁle. Comments describing areas of the dump ﬁle to be listed appear, along with the

directive, in the title line of the output listing.5

When DSDI ﬁnished processing the last directive in ﬁle ALTDIR, it again issued the

? prompt to the terminal requesting further directive input from the keyboard. At

this time, the DISPOSE directive was entered causing ﬁle ALTOUT to be printed at

the central site line printer. Refer to Printer Output Listing Examples later in this

section for examples of printer output listings.

DUMP

LC.LOW CENTRAL MEMORY POINTERS

PP.PP-COMMUNICATION AREA

CPO.H.

EJOFF.

EJ.

CP.CONTROL POINT AREAS

CPO.XTAF.

DP.DAYFILE BUFFER POINTERS

EJ.

EST.EQUIPMENT STATUS TABLE

FNT.SYSTEM FILE NAME TABLE

EJ.

MST.MASS STORAGE TABLES

JC.JOB CONTROL PARAMETERS

CP.30/T.SYSTEM CONTROL POINT AREA

ACCOUNT.ACCOUNT FILE BUFFER

ERRLOG.ERROR LOG BUFFER

EJON.

AP.ACTIVE PPS

Figure 6-6. Sample Alternate Directives File

5. All records in an alternate directives ﬁle must be in text format; the ﬁrst word in each record contains

only the record name.

6-64 NOS Version 2 Analysis Handbook Revision M

Printer Output Listing Examples

The listings illustrated in this section are samples of the line printer output listings

produced by several of the input directives described under Input Directives earlier in

this section.

Each page of output listings begins with two header lines: a title line and a subtitle

line. The header lines are formatted as follows.

RA = current input directive DUMP nn DSDI yy/mm/dd. hh.mm.ss.

0000000 dump type hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM. NOS 2 comments PAGE n

Header Description

RA = 0000000

current input directive

DUMP nn

yy/mm/dd. hh.mm.ss.

dump type

hh.mm.ss. yy/mm/dd.

CDC NETWORK

OPERATING

SYSTEM. NOS 2

Indicates absolute addressing is in effect. If relative

addresses were being dumped (RA or RAC directive

entered), a nonzero reference address would appear in

this ﬁeld.

Input directive currently being processed, including

comments (50-character ﬁeld).

Reﬂects the EDD tape number currently being used

(nn is a 2-digit dump identiﬁer assigned during EDD).

This ﬁeld reﬂects the date and time when DSDI was

executed.

Indicates type of memory currently being dumped

(central memory, extended memory, or a PP number).

If the AP directive is processed, the PP number is

followed by the name of the program currently being

dumped.

Time of day, date, system title line, and system

version name taken from CMR.

comments Up to 36 characters of comments speciﬁed on a

*.ccc...ccc input directive.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-65

Printer Output Listing Examples

Example 1:

Figure 6-7 illustrates the printer output listing produced by the PP directive (dump

PP communication areas).

RA = P P. OURP 01 D S D I 8 2 / 0 1 / 2 0 . 16.22.05. PAGE

CC0DO0O CN 16.52.11. 82/01/06. CDC NETWORK OPERATING SYSTEM. NOS 2

PPOO MTR 1524 2234 MOO OOOO OOOO HTR1 PP01 OSD 0423 0434 WOO OOOO OOOO DSD1

OOOO 0001 OOOO 0003 5570 A C OOOO OOOO OOOO COM OOOO

CP34 0101 1501 OOOO OOOO 0007 A A K A G CP34 3404 1434 0010 W44 0142 1DL1 H 9A7

OOOO OOOO OOOO 0004 0565 OE / ><CH10 OOOO OOOO OOOO OOOO OOOO

OOOO OOOO OOOO oooo oooo P r o g r a m I n E x e c u t i o n / OOOO WOO 1401 0613 54W LAFK"

oooo oooo oooo oooo oooo Control Point AsBlgrtnent, 5534 3457 3737 5733 3357 11.44.00.

oooo oooo oooo oooo oooo Channel Asslgnscnt S543 3550 3334 5033 4157 82/01/06.

oooo oooo oooo oooo oooo W4 0 W W O O W OOW OOW

PP02 RAC 2101 0356 0000 0000 0123 SAC, AS PP03 OOW OWO WOO woo 7430

UTER 1077 OOOO OOOO 0003 3734 H C 4 1 OOW OOW WW W W O OW

CP16 0006 0001 4600 0001 OOOO F A - A CPOO OOOO OWO OOW woo woo

OOOO 0002 0300 OOOO OOOO ec WW 0002 0 235 0022 0001 BB2 R A

OOOO OOOO OOOO OOOO oooo WW WOO OWO OWO OOW

0013 0000 1765 0000 2265 K 0 R W 04 W W 3 31 4 OOW 7325 0 O L l l

0301 6400 WOO OOOO OOOO CA 0724 WOO 2127 1031 1200 GT BWHYJ

0330 OOOO OOOO OOOO 0125 C X A U W W W W 0 0 0 2 1030 1020 BHXKP

PPW OOOO OOOO OOOO OOOO 7350 PP05 1R0 3422 1705 WW OWO WOO 1R0E

OOOO OOOO OOOO oooo oooo

oooo oooo oooo oooo oooo

2043 04W WW

0154 W11 4440

0001 5400

WOO 0400

PSD A<=

As 195 DCPOO CP05

0000 OOOO 0235 0022 0001 B2 R A 40W W W 0 5 0 4 26W WOO 5 E D V

OOOO oooo oooo oooo oooo WW WOO OWO OWO WOO

4002 OOOO 5502 0002 1500 S B B B R 0003 WW 1776 OOW 2777 COW

0301 6400 OOOO OOOO OOOO CA 0301 6400 OWO WOO WOO CA

0330 OOOO OOOO OOOO 0333 C X C O OOW OOW 1401 0620 3735 LAFP4Z

PP06 1SJ 3423 1234 0003 OOOO OOOO 1SJ1 C PP07 1RT 3415 2431 3170 OOOO WW 1HTYY

LDAR 1055 0001 4472 0305 1073 H A9 CEN LDAR 1055 W01 4472 0230 1073 H A9 BXH

CP34 OOOO 0001 4472 0305 OOOO A9 CE CP31 WW 0W1 4472 0230 OWO A9 BX

OOOO 0002 0237 0005 0007 BS4 E 6 0020 OOW WW OOW OOW

OOOO OOOO OOOO WOO OOOO WOO OWO WOO WOO WW

0214 4400 OOOO OOOO OOOO 8L9 3441 4033 3635 5655 111 6 1 6503 2 , I N

0301 6400 OOOO OOOO OOOO CA 57W WOO 2100 3645 0100 Q 3+A

0330 OOOO OOOO OOOO 0333 C X C O 3S47 34W 20W 3345 5415 2*1 P 0*=R

PP10 OOOO OOOO OOOO OOOO 7370

OOOO OOW OOOO OOOO oooo

PP11 QAC

UTER

2101 03S4 OOW

1077 OOW W03

OOW

0003

0216

3734

QAC= BN

H C C 4 1

CPOO oooo oooo oooo oooo oooo

OOOO 0002 0235 0004 0014 BB2 D L

CPI 4 0W4 0002 4476

OOW OWO 0237

0152

W04

OWO

0010

D B9 A)

B4 D H

OOOO oooo oooo oooo oooo WW WOO OWO OWO OWO

4011 0005 5031 0005 5432 51 E/Y E=2 0013 OOW 4126 OOW 4426 K 6 V 9 W

0724 OOOO 2127 1031 1200 6T QWHTJ 0724 OWO 2127 1031 1200 GT QWHYJ

OOOO OOOO 0002 1030 1020 BHXKP OOW OOW 0W2 1030 1020 BHXKP

PP20 PIP 2011 2003 OOOO OOOO 0136

OOOO OOOO OOOO OOOO OOOO

PIPC A3 PP21 CPO

LDAR

0320 0434 OOW

1055 OOW WW

OOW

0225

CPD1 BU

CP03 0311 17D3 OOOO 0002 0331 CIO C B CY CP34 WOO 0W4 6123 0131 WOO D SAY

CH05 OOOO OOOO OOOO OOOO OOOO

1130 6700 6017 1701 0574

0100 6202 0200 6155 0363

0100 6022 3004 1201 3101

3406 1411 3204 0502 1412

IX OOAE

A B B C

A RXDJAYA

1FLIZDEBLJ

OOW 0005 0424

OOW WOO WW

0777 1777 3777

20W 0045 6001

0770 1401 3411

W21

WOO

01W

3W3

3412

0022

OWO

71W

1013

3077

EOT Q ﬁ

G 0 4 A

P ♦ AXCKK

G LA1I1JX

Figure 6-7. Sample PP Directive Output

,*^^.

6-66 NOS Version 2 Analysis Handbook Revision M

Printer Output Listing Examples

Example 2:

Figure 6-8 illustrates the printer output listing produced by the AP directive (dump

analysis of PP and PP memory in line format).

I A • AP.3. DUMP 01 DSDI 82/01/20. 16.22.09. P A G E 3

0000000 FP03 6 .52 11. 62/01/06. CDC NETWORK OPERATING SYSTEM. NOS 2

ANALYSIS OP PP03

PP03 OOOO OOW OOW OOW 743D

OOOO OOW OOW WW OWO

LAST HAIN PROGRAM LOADED -

LAST OVERLAY LOADED - 3ME

CPOO 0000 OOW OOW WW OOW UST MASS STORAGE DRIVER - 6DI

OOW 0W2 0235 W22 W01 BB2 R A

WW OOW OOW WW WW

0004 OOW 3314 OOW «B25 D O L U

0724 OOW 2127 1031 12W GT OVHYJ

OOW WW 0002 1030 1020 BHXKP

RESIDENT DORY POINTS LOW CORE CONSTANTS OPERATING REGISTERS

NAME LOC CALLER NAME LOC ACTUAL EXPECTED Q K

FTN 0163 2205 R A 5 5 4 W 4 BEFORE IDLE 00130 0W76 006000 W6135

DFM 0432 2341 P L 5 6 W 2 1 AF T E R ID L E W1 3 3 00 0 6 4 1 07 7W W7371

EXS 0471 1477 O N 7 0 0 0 0 1 0 W 1

SKS 0506 0355 H N 7 1 0 1 W O I W DFT BUFFER 1 OOOOO W 1 7 0 0 0 0 14 0 000010

RDS 0553 0370 I B 7 2 1 0 W 1 0 W DFT BUFFER 2 030W 0 7 41 0 W S W 6 OOOOOO

VDS 0556 1740 T R 7 3 0 W 3 0 W 3

EMS 0561 0415 C P 7 4 6 2 W

I A 7 5 7 2 3 0

O A 7 6 7 2 3 1

M A 7 7 7 2 3 2

R-REC W010W0 OOOOOOOO ••• WARNING •**

PPU KEKORT

OOOO 0216 2210 2175 2616 0W6 W01 4473 0315 BMRHQ VN F A9 CM 0010 2120 qp

0020 W65 0030 0027 40W ~_ -— X V 5 0030 40W 0301 5 C A

0040 0002 5504 0203 W53 3101 W06 B D DC $YA P W50 7*30 4W4 0021 0312 X5D QCJ

0060 3314 —- —- OL 0070 O W I OIW 10W 0W3 62W 7230 723) 7232 A A H C X Y Z

0100 —— ——• —— W02 0343 0411 B C8DI OHO W04 2711 1006 0621 1002 DWI H F F QHB

0120 0704 2000 1701 0576 3010 22W 0177 0336 COP OAE XHR A C3 0130 1457 6010 3010 3374 0513 1425 0327 2000 L. HXHO EKLUCWP

0140 1720 0676 3076 6010 3010 1301 0546 3074 OFT X HXHKAE-X 0150 1625 60)0 2411 3013 3455 30)4 3456 3076 KU HTIXK1 XL1.X

0160 6010 14W 01W 2205 0443 3410 3076 6210 HL A IE081HI H 0170 1427 3210 0644 0312 0711 3011 1277 5100 LUZHF9CJGIXU (

0200 0510 6010 3010 1015 0731 2005 1455 6010 EH HXHKKCYPEL H 0210 3010 3111 0572 2005 1447 6370 0111 2W5 XHYIE PEL* A1PE

0220 1447 2610 6010 1014 0445 1720 0676 1502 L*VH HXLD+OPF Ml 0230 3500 0665 1476 6210 0356 2001 2551-6173 2 F L HC,PAU(

0240 7762 0250 2W0 1704 0576 3075 6050 3050 0471 3077 P ODE X /X/D X

0260 6173 0020 3051 1340 1006 3474 3574 3374 PXCK5HF) 2 0 0270 3350 IW6 0200 0321 0115 W05 2W0 0036 O/HFB CQAM EP 3

0300 3412 20W 1505 3413 1473 0200 0163 0326 UP ME1KL B A CT 0310 3311 1014 3112 6113 2120 I4W 0200 0163 01HLYJ KQPL B A

0320 0100 0474 5400 0302 1063 5400 0277 3076 A D - CBH - B X 0330 6010 3710 0543 3076 6210 3014 0502 3015 H4HE8X HXLEBXM

0340 34IS 5*00 0314 3011 1377 0542 3015 1702 IH- CLXIX E7XK0B 0350 3403 3076 6004 02W 0506 1440 5500 OHO 1CX DB EFL5 AH

0360 4003 3416 5W3 W01 3417 3003 02W 0553 SC1N/C AIOXCB E5 0370 0727 3607 5300 0106 0504 3407 4W3 3406 GW3CS AFED1CSC1F

0400 3016 4403 3017 5403 W01 2000 0506 3503 XN9CX0-C AP E 2C 0410 5715 0W3 0545 0200 0561 OIW 0315 20W .H CE+B E A CHT

0420 2725 0302 3071 2342 0420 340) 1453 6173 VUCBX S7DP1ALS 0430 7762 0100 2341 5400 0457 1701 3402 1063 ' A S6- D.0A1BH

0440 3412 1217 1111 0410 3602 4W2 0575 3602 1JJOIIDH3B5BE 3B 0450 5200 0457 3411 1406 3401 3077 6301 2572 ) 0.11LF1AX AU

0460 1452 02W 0163 3014 0444 2001 2225 6173 DB A XLD9PARU 0470 7764 1477 0200 0321 5000 0471 5415 0W6 L B CQ/ D -H F

0590 0115 0W7 1466 6170 01W OIW 0355 2000 AM CL A A C P 0510 7464 3105 60)0 3014 1003 1606 6170 0102 YE HXLHCHF AB

0520 SOW 0110 1237 0556 5000 0104 52W 0107 / AHJ4E,/ AD) AC 0530 0420 55W 0107 3413 1441 3412 1473 0200 DP AC1KL61JL B

0540 0163 301) 1014 3112 6113 0551 0340 A XIHLTJ KE(C5 0550 0100 1056 OIW 0370 0314 0100 1740 0334 A H,A C CU 05C1

0560 01W 0415 14W 3413 4471 02W 0773 0370 A DHL IK9 B C C 0570 54W 0606 4771 0705 3107 S3W 0624 0403 - FF* CEYCS FTDC

0600 02W 0663 1404 02W 0746 7106 2116 0532 B P LDB C- FQNEZ 0610 5000 0553 0335 54W 0640 1063 S4W 0633 / ESC2- F5H - FO

0620 4771 0705 3107 2300 0342 0403 0200 0663 » GETGS C7DCB F 0630 4 0 7 1 0 4 0 3 2 0 0 0 1 1 3 5 02W 0746 7306 5 D C P I2S C- F

0640 1131 0515 5C00 0556 6606 0644 7546 3410 IYEH/ E. FF9 -IH 0650 1412 02O0 0746 1401 7106 W13 0525 3013 UB G-U F KECXK

0660 0524 0110 0602 3076 6010 3010 0574 ETAH PBX HXHE 0670 3077 6204 1455 02W 0163 3714 4471 3107 X DL B A 4L9 YG

Figure 6-8. Sample AP Directive Output

0$ms.

Revision M Express Deadstart Dump Interpreter (DSDI) 6-67

Printer Output Listing Examples

Example 3:

Figure 6-9 illustrates the printer output listing produced by the CP directive (dump

active control points). The default list options (XTAF) are used to dump the control

point. This example consists of six pages. Also, notice that the columns cross page

boundaries; that is, the left column is read continuously, from the top of the second

page to the middle of the fourth page. The sequence then continues at the top of the

right column on the second page.

y*SW|L

RA « CP,2. DUMP 00 OSOI 83/06/02. 22.36.17. PA6E

0OQ0000 CH 22.15.04. 83/06/02. CDC NETWORK OPERATING SYSTEM. NOS 2

0400 - CONTROL POINT 02

CP02 EXCHANGE PACKAGE

P 33357 AO 22577 BO 0 (A0)«OOW OOW OOW 0002 2575 BU (80>*WW OWO WW WW OWO

RA 120300 A1 1 Bl 3 (AD'OOW OOW WOO WOO WOO <B1>°3444 4400 OOW WW OOW 199

FL 46600 A2 26250 82 31053 <A2)*00W OWO WW OWO 0024 T (82>«04W 0320 25W OWO OWO D CPU

EH 70070000 A3 14424 B3 0 (AS)=0000 OOW OOW 0004 3521 028 (B3>*00W OOW WW WW OOW

ME 0 A4 33070 B4 43 <A4)>0WO WW WW 0003 3244 CZ9 (84)*OW0 WW OWO 0000 OWO

FLE 0 AS 33072 B5 31367 <A5)*0OW OOW WW 0W3 3273 CZ (B5)>04W 0320 51W OWO OWO D CP<

RA 400 A6 1 B6 33646 CA6)-W00 OWO WW OWO WOO <86)*S110 0336 3001 W03 3274 (HC3XA «

0 A7 33070 B7 43515 <A7)»0DW WW WW W03 3244 CZ9 <87>"00W 0004 0435 0204 3715 DD2BD4H

XO 7777 7777 7700 OOW WW <X0>«WW WW WOO WOO WW

XI WW OOW OOW WW OWO CX1)«0OW OOW WW OOW OWO

X 2 W W W W W W W W W 2 4 T ( X 2 ) * W W W W W W O W O O W O

X3 OWO OOW OOW 0004 3521 D2R (X3)*10W 0150 OOW WW OOW H A/

X4 WOO WOO WW W03 3243 C7.8 (X4><04W 0321 6630 2205 0301 D ca XRECA

X S O O W W W W W 0 0 0 3 3 2 7 3 C Z ( X 5 > > 0 0 M O O W W W W W W W

X 6 2 2 0 3 1 4 W W W W W W W R C L < X 6 ) » W W W W W W W W W W

X7 OOW WW OOOO W03 3244 CZ9 (X7)=04W 0320 1130 0530 0503 D CPIXEXEC

PARARETER SUHRART

USER NAHE NETOPS SENSE SWITCHES W

P R I M A R Y F I L E C C L R 1 O O W W X Llct Option

ORIGIN TYPE SYOT CCL R2 DWOW

TIRE USED WWWW22 CCL R3 OWWO

TIRE REMAINING 1014223327 CCL EF 00

SUBSYSTEM ID 7770 CPU PRIORITY 0077

CPU STATUS X SERVICE CLASS x««« SSSC

RESSA6E 1 - NP/SN102 22.05.09.LLINK LW3103,EN,AL=S,H-N,S

R E S S A G E 2 - S R E Q U E S T * K * D I S P L A Y

CURRENT COWARD - NIP<NIN«199,ISTP»YES,FSTP»YES,RC«5W)

LAST DAYFILE RESSA6E - 22.05.09.NP/SN102 22.0S.O9.LLINK LW3103,EN,RL*3,M-N,SH102 /1D3.SN102 /003

SPECIAL ENTRY POINTS

EJT ENTRY

02SW0 1601 1555 0137 7777 0002 NAM A4 B

025001 OOW OOOO 40W 0002 7776 5 B

025W2 OWO WOO WW WW WOO

0 2 5 W 3 0 7 W W W W W W W W 3 6 G 3

Figure 6-9. Sample CP Directive Output

(Continued)

6-68 NOS Version 2 Analysis Handbook Revision M

Printer Output Listing Examples

(Continued)

RA = CP,2. Ximf 00 DSDI 83/06/02. 22.36.17. PAGE 2

0 0 0 0 0 0 0 C M 2 2 IS.04. 83/06/02. CDC NETWORK OPERATING SYSTEM NOS 2

Beginning of T List Option

CONTROL POINT A R E A /

CPU STATUS STSW 000420 5 /A RESERVED RFCW W047S OOOOOOOO

CPU SU8-CP ACTIVE CONTROL STATEMENT ADDRESS WOWO

NUIBER FPS ASSIGNED 01 DEMAND FILE RANDOM INDEX OOOOOO

ERROR FLAGS OOOO RESERVED ALMW 000476 0077

RESERVED MAX MAGNETIC TAPFS

S U S P E N D O N R O L L O U T F L A 6 0 MAX REMOVABLE PACKS

ROLLOUT CONTROL 00 MAX DEFER BATCH JOBS

RESERVED OOW RESERVED

RECALL CONTROL W01 MAX TINE LIMIT 77

CPU PRIORITY/RECALL JCIW 000421 0 1 7 6 JO MAX SRU LIMIT 77

RESERVED WO MAX FIELD LENGTH 77

CPU SELECTION MAX ECS FIELD LENGTH 77

SUBSYSTEM ID 7770 MAX LINES PRINTED 77

JOS CONTROL/*0IS* FLAGS 0 MAX CARDS PUNCHED 77

USER PRIVACY, PRESERVE E C S 0 RESERVED ACLW 000477 OOW

•NS* LINKAGE 001221 DAYFILE MESSAGES COUNT 7777

CPU RECALL CRITERION CRCW 0004 22 000 3000 0033713 0264 22 C C 4 K8 R CONTROL STATEMENTS CN1 7777

JOB TERMINATION OPTION EOJW 000423 0 RESERVED 77

QUEUE FILE DISPOSITION MS PRUS COUNT 777777

RESERVED OOOOO ACCOUNT ACCESS WORD AACW 000500 OOOCOQOOQOO0O3777777 C

RSVD/1 STATEMENT FLAG M A P III M U LT. MPKW W0S01 0 0 0 0 0 0 2 P

CHAR6E REG OP/CCL LEVEL OOOO RESERVED WWW

JOB ORIGIN TYPE J O T W O O W HAP III ACCUMULATOR MPAW 0000000

OPER ASSIGNED ED.UIP O A E V O O O O AUC1/2/3 ACCUMULATORS AU1W 000502 OOOOOOOOOOOOOOOOOOOO

NEGATIVE FL/IOOB FLSW 0 00424 W 04 JCD AUC4/S ACCUMULATORS A U 2W 00050 3 0 000000 0 000000

•SET* RECOVERY NFL OOW RESERVED WOWO

RA/IOOB 0000120J LENGTH BUFFER 0 ICAW W0S04 OOOO

FL/1008 0446 ADDRESS BUFFER 0 OOOOOO

ZERO 000425 OOOO SDD LENGTH BUFFER 1 OOOO

RA/IOCB - NFL/1006 00W1177 ADDRESS BUFFER 1 oooooo

40008 ♦ NFL/1009 4004 SPEC ENTRY POINT FLAG SEPW W0S05

FL/10OB 0466 RSVD/LOR«/CLB* EN PI FLA6S

ZERO ECSW 000426 OOOO BU ARG>/DNP'/SDRs EN PT FLAGS

RESERVED OOW SSJ«/VAL«/SS*» EN PT FLAGS

ECS RA/10WB 00000225 RESERVED oooo

ECS FL/10008 OOW RESTART/SUPPRESS OHP* FLA6S

PUNCH MODE/OVERRIDE SNSW 000427 0 DM* FILE FLAGS

RESERVED WWOX OM P* FL/1 00 (0 = ENT IRE FL ) oooo

RESERVED-JNSTALLATION WW S5J= PARAMETER 6LK ADDRESS oooooo

PP PAUSE FLAGS OOW SYSTEM PROC CALL WORD SPCW W0506 oooooooooooooooooooo

SENSE SWITCHES 00 CCL - EFF JCDW 000507 0 0 A

RESERVED CCL - RI G oooooo

MESSAGE 1 AREA HS1W 000430 16205025163433355555 NP/SN102 CCL - DATA 0001

000431 5S3S3557334057S34457 2 2 . 0 5 . 0 9 . RESERVED oooooooo

000432 14141116130055143333 LLIHK LW CCL - EF JCRW 000510 00

000433 36341336S60SI6562214 3103,EN,RL CCL - R3 oooooo

000434 54365610461656230000 «3,H-N,S CCL - R2 oooooo

MESSAGE 2 AREA NS2W 000435 53220521250523245547 SRE8UEST • CCL - R1 oooooo

000436 13475504112320140131 * • D I S PL AY SEC NEN/K DIS FUGS DBAW X0511 0 A S B A S ' A S -

000437 OOOOOOWWOOOOOOOOOO ECS AND CD CMF1 KOOE FLAGS

INSTALLATION WORD 0 INOW 000440 OOOOOOOOOOOOOOOOOOOO INPUT BUFFER ADDRESS 014043

INSTALLATION WORD 1 INlW 000441 WOOWOOGOOOOOOCWOO RIGHT SCREEN BFR ADD 014054

INSTALLATION WORD 2 IN2W 000442 OOWOWOOOOOWOOOOW LEFT SCREEN BFR ADD 014054

INSTALLATION WORD 3 IN3W 000443 OOOOOOOOOOOOOOOOOOOO HAP OPTIONS/LIB FLAG LB1W 000512 00

Figure 6-9. Sample CP Directive Output

(Continued)

Revision M Express Deadstart Dump Interpreter (DSDI) 6-69

Printer Output Listing Examples

(Continued)

C P, 2 . 83/06/02. 22.36.17.

O W O W O C H 2 2 15.04. 83/06/02 . CDC NET WORK OP ERATING SYSTE

INSTALLATION WORD 4 IN4M 000444 OOOOOOOOOOOOOOOOOOOO

INSTALLATION WORD 5 I N S W 0 0 0 4 4 5 O W W W W W W W W W O

INSTALLATION WORD 6 IN6W 000446 OOOOOOOOOOOOOOOOOOOO

INSTALLATION WORD 7 IN7W 000447 OOOOOOOOOOOOOOOOOOOO

L I M I T F U G S A C T W 0 0 0 4 5 0 0 0 N ( W O R

OVERFLOW FUGS S R U W O O W

SRU ACCUMULATOR/OVERFLOW 00001651271722

CP ACCUM (NANOUNITS/4) C P T W 000 4 5 1 0 OOO W O 1 O 7 4O5 3 5 5 6 W 0 A 6SS

(0 ACCUMULATORS IOAW 000452 O0314540O0OO0OOO0416 Y»S DN

M13 x HI ♦ H3 H P 1 W 0 0 0 4 5 3 0 0 W 3 6 3 3 B

H14 ■ JTI * H4 000036

RESERVED

ADDER ACCUMULATOR ADAW 0000002

M l • 1 0 0 0 H P 2 N W 0 4 5 4 0 2 3 4 2 0 S I P 0 /

H12 • M1 • R2 001750

RESERVED OOOOOOOO

DISABLE SRU ACCUM, CPM RP3W W0455 0000025700 B. 7X

ION 0000004230

SRU ACCOUNT BLOCK LRT S TLW 00045 6 77777 7

COMPUTED SRU JOB STEP LIMIT 77777777777777

RESERVED S R J W 0 0 0 4 5 7 O W O D C D

SRU JOB STEP LIMIT 777777

SRU ACCUM AT JOB STEP S T A R T 0 0 0 0 0 4 0 3 0 4

RESERVED CPJW 000460 WOO N

CP TIME JOB STEP LIMIT 777777

CP ACCUM AT JOB STEP START 0000000016

CHARGE/ACCOUNTING FLGS FPFW 000461 00

RESERVED

SRU VALIDATION LIMIT 7777

FNT ORDINAL OF PROFILE F I L E O O O O

LEVEL-3 BLOCK TRACK OOOO

LEVEL-3 BLOCK SECTOR OOOO

RESERVED 000462 OOOOOOOOOOOOOOOOOOOO

MAX FL FOR JOB STEP F L C W 0 0 0 4 6 3 3 7 6 5 4 F 4

LAST CARD FL (NFL) 06W

MAXIMUM FL FOR ENTIRE J O B 3 7 6 5

RESERVED OOOOOOOO

JOB STEP RAX ECS FL E L C W 0 0 0 4 6 4 0 2 0 0 B 8

LAST CARD ECS FL (NFL) OOOO

JOB MAX ECS FL 0200

RESERVED OOOOOOOO

E0U1P ASSIGNED COUNT E A C W 0 0 0 4 6 5 0 0 0 1 A

RESERVED• F L I W 0 0 0

SCHEDULER STATUS

NE6ATIVE FL FOR ROLLIN OOOO

ECS FL REQUEST OOOO

CM FL REQUEST OOOO

TXOT SUBSYSTEM TXSW 000466 00

RESERVED

TERMINAL NUMBER T T N W O O W

TERMINAL INT ADDR T I A W O O O O O O

OUTPUT POINTER T I O N O O O O O O

RESERVED PFCW 000467 OOOOOOOOOOOO H

EST ORD OF FAMILY DEVICE 0010

LIMIT FOR SIZE OF OAF

LIMIT FOR NUMBER OF P

LIMIT CUNM SIZE 1APF

NOS 2

REDUCE/LOADER FLAGS

RESERVED FOR LOADER

INTERACTIVE DEBUG CONTROL

GLOBAL LIB INDICATORS

SECOND LIB/GLOBAL IND LB2W 0W513

F I R S T LIB / G L OBA L I N D L B3W 0 W 5 14

FNT ADDR LAST FIL EXEC EOCW 0W51S

ECS FL FOR DMP> CALL PPDW

FL FOR DMP« CALL

DUMP WORD COUNT

FWA OF DUMP

RESERVED 000516

OUTSTANDING CONNECT COUNT

ROLLOUT ALLOWABLE SSCW 000517

CONNECTION/WAIT RESPONSES

COMPUTED CP TIRE LIMIT CPLW 000520

LIST OF FILES INDEX LOFW 000521

RESERVED

LIST OF FILES ADDRESS

RESERVED

A P P L A C C E S S L E V E L A A L W

CM RESIDENCE TIME LMT TSCN

RESERVED

INIT TIRE SLICE OCCUR FUG

CPU TIME SLICE LIMIT

ADMIN/D1AG/USER PW EXP JSCW 000523

PF PW EXP/LOVER JAL/FAL

WRITE DOWN/UNLABELED TAPES

RESERVED

JOB ACCESS LEVEL

USER ACCESS LEV VALIDATION

JOB ACCESS LEVEL LIMIT

USER ACCESS CATEGORY SET

DEFAULT PACK NAME PKNW 0W524

DEFAULT PACK TYPE

RESERVED 000525

RECALL CRITERION RCCW 0W526

RECALL CRITERION 000527

RECALL CRITERION 0W530

RECALL CRITERION 000531

RECALL CRITERION 000532

RECAU CRITERION 0WS33

RECALL CRITERION 000534

RECAU CRITERION 000535

RECALL CRITERION 000536

RECAU CRITERION 000537

RECAU CRITERION 000540

RECAU CRITERION 000541

RECAU CRITERION 000542

RECAU CRITERION 000543

RECALL RESUESTS IR-S RECW 000544

RECAU RESUESTS IR-S 00054S

RECAU REQUESTS IR-S 0W546

RECAU REQUESTS IR-S 000547

RECALL REQUESTS IR-S 000550

RECAU REQUESTS IR-S 0W551

RECALL REQUESTS IR-S 000552

OOW

OWWWO

WWWWDWOOOOOOOW

OOOOOOOOOOOOOOOOOOOO

oooo oooo

oooo

oooooo

woooooowowwo

0021

OOW

0000000000000000

S7777777777777000000 4

OOW

023655

000522 00400323

000

oooowooow

oooooooooooow

oooooo

oooooooooooooooooooo

00000OOWO0OO00W002

00000000000000000003

00000000000000000004

ODOOO00OO0OOOOOOOW5

00000000000000000006

00OOO0O0OOOOOOWW07

D0OOOOO0OOOW0OO0O1O

0D00000000W000O0O11

00000000000000000012

00000000000000000013

00000000000000000014

03000000000000000015

WO0W0OO0O0W0W016

ooooooooowwwooooa

oooooooooooooooooooo

ooooooowwooooooow

oooooooooooooooooooo

Figure 6-9. Sample CP Directive Output

(Continued)

6-70 NOS Version 2 Analysis Handbook Revision M

Printer Output Listing Examples

r(Continued)

0000000 CM 22.15.04. 83/06/02,

DUMP 00

CDC NETWORK OPERATING SYSTEM 83/06/02. 22.36.17.

LIMIT FOR SIZE OF IAPF 7

USER NAME UIDW 000470 16052417202300

CHARGE FUG, USER INDEX 377772

NO EXIT FLA6 EECW 0W471 2

RESERVED 000

REPRIEVE DATA 0077

TERMINAL INPUT POINTER TINW WWOO

REPRIEVE DATA 024010

EJT OROINAL OF JOB TFSW 000472 0017

PRIMARY FILE FNT OFFSET OOOO

RESERVED DO

ROLLOUT TIME TERN 000

EVENT DESCRIPTOR OOOOOCO

RESERVED CSPW 000473 00

LOGOUT EPILOG, .EPILOG REQ 0

PR CHARGE/USER/INHIBIT OEC 2

EOR FUG/CS COUNT OOO0OO11

NEXT STATEMENT INDEX 0134

LIMIT INDEX

INPUT/SKIP FLAGS

EST OROINAL

FIRST TRACK

CURRENT TRACK

CURRENT SECTOR

OVERLAP WORD COUNT

CSSW 000474 0 0065

I RECALL REQUESTS IR-S

| RECAU REQUESTS IR-S

j RECALL REQUESTS IR-S

I l RECALL REQUESTS IR-S

! RECAU REQUESTS IR-S

J RECAU REQUESTS IR-S

I RECA U REQU ESTS I R-S

| RECAU REQUESTS MB-S

j RECAU REQUESTS FB-S

! RECALL REQUESTS H8-S

• RECAU REQUESTS MB-S

I RECAU RESUESTS MB-S

| RECAU REQUESTS HB-S

j RECAU RESUESTS MB-S

j RECAU REQUESTS HB-S

J RECAU REQUESTS MB-S

I RECAU RESUESTS MB-S

I RECAU REQUESTS KB-S

j RECAU REQUESTS MB-S

• RECAU REQUESTS MB-S

j RECAU REQUESTS MB-S

| T Un Option rc«ua«t

I right colian oa «h««t

000553

0W554

000555

000556

000557

000560

000561

REPW 0W562

W0563

000564

0W565

0WS66

000567

000570

W0S71

000572

000573

0WS74

0W575

000576

000577

OOOOOOOOOOOOOOOOOOOO

WWWOOOWWOOWOW

OOOOOOOOOOOOOOOOOOOO

OOOOOOWWWWOWOOO

OOOWOOOOWOOQOWOW

OWWWOWOOWOOWDO

OOOOOOOOOOOOOOOOOOOO

OOOOWOOOOWOOWOWO

OOOOOOOOOOOOOOOOOOOO

WOWWOOOWWOOOOW

woooooowoawooww

oooooooooooooooooooo

owwwowowoowwo

oooooooooooooooooooo

DAYFILE POINTERS AND BUFFER

— — l - J ♦

• - A L i s t O p t i o n \ y

00120275 0011 7777 OOW WOO 0046

00120276 MOO OOOO 0114 OOW 0104

00120277 0007 4200 42W W10 Q003 G7 7 H C

W117777

00120003

00120007

00120013

00120017

00120023

00120027

00120033

00120037

00120043

W12W47

00120053

W120057

0012W63

00120067

00120073

00120077

00120103

WI 20107

55353557334057334157

17160055553636565514

55353557334057334157

0323551617O4O5OO5S33

55353557334057334457

36343336560516562214

WWWWOWOWWWW

14141116130055143333

16343335S5SSS0333336

.23163433355555565504

55353557333757373657

17010411160755511511

SS3SS5S7333757374157

55353557333757404157

17010411160755511603

55353557333757404257

55353557334057334157

S55S56010356343336W

55232520052226112311

16205023163433355555

05260514005540443756

16205023163433355555

33335655202205261117

162050231634333S5SSS

54365610461656231634

55353557334057334457

36343336560516562214

OOWOOOOWOOWWWOO

25152055162033333444

16235055555555555555

O355555552575500OO00

23163433355555565514

1623S0SS5S55S555SSSS

02555555525755000000

23163433355555565514

16205023163433355555

OOOOOOOOOOOOOOOOOOOO

171655O7O11116OS04OO

55353557334057334157

552601221101162400S5

55353SS7334057334157

25235516235516170405

55353557334057334457

3335SSS5S03333365623

1620502316343335S5S5

54365610461656231634

16235055555555555555

44550317152014052405

55353557333757373657

55333557333757374157

17010411160755510211

55353557333757404157

55353557333757404257

17010455031715201405

SS3S3557334057334157

55353557334057334157

OOOOOOOOOOOOOOUOOOOO

55030320552605222311

40443700000000000000

55202205261117252355

00553333330000000000

14141116130OS5143333

16343335S5S550343336

55353557334057334457

33355555503433365623

55353557333757373657

04 570000000000000000

23163433355555565514

16235055555555555555

071415SS52S75S0OMM

23163433355555565514

16235055555555555555

240S04S700000WOOWO

16202500552316343335

16205023163433355555

22.05.06. SUPERVISION GAINED

22.05.06.NP/SN102 22.05.06. CCP VERSI

ON 33, LEVEL 594, VARIANT 594

22.05.06.HP/SN102 22.05.06. PREVIOUS

CS NODE 000, PREVIOUS NS NODE WO

22.05.09.NP/SN102 22.0S.09.LLINK LOO

31O3,EN,RL'3,H-N,SN102 /W3,SN102 /103

22.05.09.NP/SN102 22.05.09.

U1NK LW3103,EN,RL'3,H-N,SN102 /103,S

N102 /003 NS/ 22.04.43.

SN102 , DUMP NP00199 COMPLETED.

22.04.43.NS/ 22.04.43.SN102 , L

OADING (MIC ). 22.04.46.NS/

22.04.46.SN102 , LOADING (BIGLM ).

22.04.56.HS/ 22.04.56.SN102 , L

OADING (NCB ). 22.04.57.NS/

22.04.57.SN102 , LOAD COMPLETED.

22.O5.O6.NP/SN102 22.05.06.NPU SN102

,AC,103 22.05.06.NP/SN102

DAYFILE LINES IN BUFFER

NS/ 22.04.43.SN102 , DUMP NP00199 COMPLETED.

22.04.43.NS/ 22.04.43.SN102 , LOADING (NIC ).

22.04.46.NS/ 22.04.46.SN102 , LOADING (B1GLN >.

Figure 6-9. Sample CP Directive Output

(Continued)

Revision M Express Deadstart Dump Interpreter (DSDI) 6-71

Printer Output Listing Examples

■^^s

(Continued)

ooooow CP,2.

CR 22.15.04. 83/06/02.

DUMP W

CDC NETWORK OPERATING SYSTER NOS 2

83/06/02. 22.36.18.

22.04.56.NS/

22.04.57.HS/

22.05.06.NP/SN102

22.05.09.HP/SN102

Z2.05.09.NP/SN1O2

ATTACKED F

0304 INPUT I

22.04.56.SN102 , LOADING (NCB X.

22.04.57.SN102 , LOAD COMPLETED.

22.05.06.NPU SN102 ,AC,103

22.05.06. SUPERVISION GAINED

22.05.06. CCP VERSION 33, LEVEL 594, VARIANT 594

22.05.06. PREVIOUS CS NODE OW, PREVIWS NS NODE 000

22.05.09. LLINK LW3103,EN,RL»3,H-N,SN102 /W3,SN102 /103

22.05.09.LLINK L003103,EN,RL»3,H-N,SN102 /103,SN102 /003

F Ll»t Option

FNT - 1116 2025 24W W01 17W INPUT AO

FST - 0006 4175 4175 0002 0W5 F6 6 B E

TRACK CHAIN -

4175 0002

FN T - 1 601 1 5W W W W W 1 3W N A R R

FST - 0006 4176 4176 0004 0W5 F6 6 D E

FUT WOO OOW WW WOO WW

EST 77W 6210 0013 1377 0421 127? H KK DOJ

FUT OWO OWO WW WW WOO

EST 77W 6210 0013 1377 0421 1277 H KK DQJ

TRACK CHAIN -

4176 0W5

0315 OUTFIL LO FNT - 1725 2406

FST - 0W7 4205

TRACK CHAIN -

4205 0003

1114 WW 15W OUTFIL M

4205 0W1 0105 G7E7E AAE

FUT

EST

OOW WOO WOO OOW woo

7702 6210 W13 1377 0421 1302 H KK DBKB

0320 ZZZZZCO LI FNT - 3232 3232

FST - 0006 4175

TRACK CHAIN -

4175 W02

3203 3301 1007 ZZZZZCOAHG

4175 0W1 W05 F6 6 A E

FUT

EST

W W W W W O O W W W O O

77W 6210 W13 1377 0421 1277 H KK DSJ

0323 ZZZZZC2 LO FNT - 3232 3232

FST - W06 4177

TRACK CHAIN -

4177 0002

3203 3SW 1507 ZZZZZC2 MG

4177 0W2 0307 F6 6 8CG

FUT

EST

O O W W W W W W W W O O

77W 6210 0013 1377 0421 1277 H KK DGJ

0326 OUTPUT PR FNT - 172S 2420

FST - W10 6742

TRACK CHAIN -

6742 0W1

2524 OOW 1200 OUT PUT J

6742 0W1 0005 H 7 7 A E

FUT

EST

W O O O O W 0 7 0 0 O O W W O O G

7704 4210 3226 1377 0412 130S 7HZVK DJKE

0331 NRF1 LO FHT - 1622 0634

FST - 0W7 4210

TRACK CHAIN -

4210 W03

W W W W 1 5 W N R F 1 H

4210 0002 0705 G7H7H BGE

FUT

EST

W O O W O O W W W O O W W

7702 6210 W13 1377 0421 1302 H XK 0«KB

0334 NRF2 LO FNT - 1622 0635 WW WW 1500 NRF 2 H

FST - 0006 42W 4200 0W1 0705 F7 7 AGE

TRACK CHAIN -

42W 0003

FUT

EST

O O W W W W W W W W O O

77W 6210 W13 1377 0421 1277 H KK DSJ

0337 LIST PR FNT - 1411 2324 WOO WOO 1200 LIST J

FST - W10 7052 7052 0001 WD5 H ) ) A E

TRACK CHAIN -

7052 W01

FUT

EST

O O O O O O W 0 7 W W W W W G

7704 4210 3226 1377 0412 1305 7HZVK DJKE

0342 ZZZZZPP PR FNT - 3232 3232 3220 2000 1200 ZZZZZPP J FUT W W W W 0 7 0 0 O O W W O O G

RA ■

OOOOOW

CP,2.

CM 22 .15.04. 83/06/02.

DUMP OO

CDC NETWORK OPERATING SYSTER

DSDI 83/06/02. 22.36.18.

NOS 2

P A 6 E 6

FST - W10 7053

TRACK CHAIN -

7053 0W1

7 0 5 3 W 0 1 O W S H S 1 A E EST 7704 4210 3226 1377 0412 1305 7HZVK DJKE

0345 ZZZZORB PM FNT - 3232 3232

FST - 0010 7054

TRACK CHAIN -

7054 0W1

0415 02W 1200 ZZZZDMB J

7 0 5 4 0 W 1 W 0 5 H = ■ A E

FUT

EST

W W W O O 0 7 W W W W O O G

7704 4210 3226 1377 0412 1305 7HZVK DJKE

0350 ZZZZZDN LO FNT - 3232 3232

FST - 0007 4216

TRACK CHAIN -

4216 W22

3204 1600 1500 ZZZZZDN H

4216 0022 0703 G7N7N RGC

FUT

EST

WW WOO WW WW WOO

7702 6210 0013 1377 0421 1302 H KK DOKB

Figure 6-9. Sample CP Directive Output

6-72 NOS Version 2 Analysis Handbook Revision M

Printer Output Listing Examples

Example 4:

Figure 6-10 illustrates the printer output listing produced by the MST directive

(dump mass storage/track reservation table). The MST is listed in two columns. The

left column is read from top to bottom, perhaps across page boundaries, and continues

at the top of the right column. The track reservation table is listed in single column

following the MST.

R A - K S T . 1 3 . D U M P 2 2 DSDI 86/05/06. 10.01.2b. PAGE 1

O0OO0W CH 09.52.49. 86/05/06. CDC NETWORK OPERATING SYST

EQUIPMENT 013 - MASS STORAGE TABLE

t i l N O S 2

UHIT RESERVE COUNT DULL 014204

KUHBER OF TRACKS IDOL 014170 31S0 Y/ FZ6+SA OU

tr INTERLOCK / COUNTS 0000 CUKUL TOT UNIT RES COUNTS 00

LENCTH OF TRT 0632 PF INTERLOCK / COUNTS OOOO

FIRST AVAIL TRACK PTR *U5 RESERVED 00

HUM AVAILABLE TRACKS 2301 MACHINE INDEX - 1

CTI/DS PILB/CT ISX OVF ACCL 014171 0 j-ha CKPT FLG/SYSTEH TABLE TRACK OOOO

RESERVED 000 FAMILY IDLE STAT/ACT COUNT WOO

DA ECS CHAIN FIRST TRACK 0000 LOCAL STATUS FLACS STLL 014205 OOOOOO 0 6 A

DIRECT ACCESS FILE CRT 0012 ERROR STATUS 00

FIRST TRACK IQFT *6I5 MACHINE ID 334)

REDEFINITION STATUSES 01 CURRENT USER COUNT DAP OOOO

ALL KF UKLOADE0/ESROR IDLE 0 NEXT EST ORDINAL IK CHAIN 000

RESERVED 0 LOCAL STATUS

FLAC/KST LIKK OSV ADDR SDCL 014172 00010120 AAP AY W SEDEF IN PROC/HULL EQ DDLL 01420b

HST/TRT UPDATE COUNT 0001316127 RESERVED

XF INDEX/CPUMIR INTERLOCKS 00 ORIGINAL NO. OF UNITS - 1

FIRST TRACK LAPF ALCL 014173 4042 575 58 5 CURRENT NO. OF UNITS - I

LABEL TRACK 4000 EQUIPMENT UNIT LIST 0000000000000001

FIRST TRACK PERMITS 4043 LOCAL 1NSTAL AREA ISLL 014207 oooooooooooooooooooo

NUHBER. CATALOG TRACKS 0040 RESERVED TRLL 014210 00000000000000 BH4

FIRST TtACK DAT OOOO FWA OF TRT 021637

FAMILY OR PACK NAHE FFCL 014174 1515060000000b KHF A VER1F FAILURE THRESHOLD 0)4211 0001 ACMA- / A

DEVICE NUHBER 01 RESTRICT ACTIVITY THRESHOLD 03)5

DEVICE AL LIMIT LOWES/UPPER 00 LOU SPACE THRESHOLD 0)46

REL UHIT KULTIUNIT DEV 0 RECOVERED ERROR THRESHOLD 0050

MM UNIT KULTIUNIT DEV 0 UNRECOVERED ERROR THRESHOLD 000)

USER NUM PRIVATE PACK FUGL 014175 00000000000000 VERIFIFICATION FAILURE CUT 0)4212 OOOO

C O N T L B L T R K / D E V R E S M A S K S 5 7 7 7 7 7 RESERVED OOOOOOOO

FLAGS AND DAT INDEX MDCL 014176 1001 HAC8DI ACS RECOVERED ERROR COUNT OOOO

FT-HT FLAG •/« SECTOR LIMIT 0343 UNRECOVERED ERROR COUNT oooo

RESERVED 041100 RESERVED 014213 oooooooo

DRIVER INDEX ol ERROR DATE/TIKE oooooooooooo

SECTOR LIMIT 0343 RESERVED FOR CDC HCLL 014214 oooooooo

TRACK TO BE FLAWED HVCL 014177 OOOO MST WRITE FLAG oooo

RESERVED OOOOOOOOOOOOO PP INTERLOCK BIT MASK 00000400

RESERVED UHIT rLAGS 000 RESERVED 014215 OOOOOOOOOOOOOOOOOOOO

CLOBAL LHSIAL AREA ISGL 014200 OOOOOOOOOOOOOOOOOOOO 014216 oooooooooooooooooooo

I2GL 014201 OOOOOOOOOOOOOOOOOOOO 014217 oooooooooooooooooooo

RES/DEVKE STATS "^ 0,4M2 ° 5 0D

SUSPECT/REST*. ACTIvm °

OLTSTAKDIKG REQITST5 00

UNIT INTERLOCKS °000

CURRENT POSITION *0°0

NEXT JEST POSITION WOO

CHANNEL 2 04

CHANNEL 1 04

ALLOCATION FLAGS DILL 014203 OOOO 9 E

RESERVED POR CDC 0030

•PIT* ORD OF FIRST UNIT OOOO

DEVICE FLACS

MEHORK TYPE - 3 BIT VALUE

CPU TYPE - 3 SIT VALUE

PP PATH TYPE - I BIT VALUE "„

RESERVED 00

ALGORITHM INDEX

TRACK RESERVATION TABIX

021637 40000 4001 4002 4003 4W4 4017 1 1111 SA5B5C5D50

021640 *0004 4005 4006 4007 4010 0217 1 1111 5E5P5C5HB0

021641 -KW10 4011 4012 4013 4014 W17 1111 5I5J5K5L 0

021642 «0014 4013 4016 4017 4020 0217 1 1111 5HSN5O5PB0

021643 tO020 4021 4022 4023 U.1 2 4 0 11 7 1 — 1111 5 Q S R 5 S 5 TA O

Track Link S t at u* B i t *

Byte Ordinal

Figure 6-10. Sample MST Directive Output

Revision M Express Deadstart Dump Interpreter (DSDI) 6-73

Printer Output Listing Examples

Example 5:

Figure 6-11 illustrates the printer output listing produced by the C, D, and E

memory dump directives (instruction parcel, byte, and word format, respectively). The

same portions of central memory are dumped in each format. Auto page eject has

been disabled using the EJOFF directive to allow listing the output from all three

memory dump directives on one page.

C,50,110. C - FORMAT OUKP.

16.52.11. 82/01/06.

OUHP 01

CSC NETWORK OPERATING SYSTEM.

OOOWSO WWO 00OW OWOO OOOOO

WW052 OOOW OOWO WWO 07240 S

OOOW53 30766 01030 10057 43001 X HXHE XA

0000054 34131 44234 12147 302W 1KL71JL B

W0WS5 01633 01401 0W34 17761 A XLA C6

OOW0S6 OOWO WOW WWO 07164

0WWS7 77775 12230 W320 OWW (RX Z

WOW60 W751 1W00 WOW WOOO I

W00061

00W062

W00063

0000064

OOW065

0000066

77770

OOOW

OOOOO

OQOOO

W042

WWO

WWO WWO

OWW 01065

OOOW WOOO

OOOW OOWO

71100 WOOO

WOW OWOO

ooooo

34137

WOW

51426

WOW

OOOW

82/01/20. 16.22.27.

WW072 74647 56475 12W0 WOOO

OOW073 00033 36401 W007 30005

W00074 W033 56404 W035 00025

0000075 W03S 56406 W057 50002

W00076 WOOO OOOOO OOWO WOW

0WW77 OW07 754W 004W 07200

W00100 OOWO OOWO OOOW OWOO

CO A

C2 0

00WO70

0000071

OOOW

74500

033W 24040

W122 250W

40000

07421

0 TOO

/ ARU 0 W00107 W054 13700 01065 34117 E64 AFS60

CH 0,50,110,

OOOCOSO WOO WOO WW OOOO

0 - FORMAT OUKP.

OOW

OOOW52

00W0S3

W00054

OW0055

000WS6

OW0057

0000060

WW050

WW054

0000060

00W064

000W70

0000074

W00100

WW104

3076

3413

0163

OOW

7777

W75

6010

1442

3014

WOO

5122

11 W

WOO OOOO

3010 0574

3412 1473

0100 0341

WOO OOOO

3W0 3200

WOO WOO

E,50,110.

7 2 4 0 5

3001 X HXHE XA

0200 1KL71JL 6

7761 A XLA C6

7164

W W ( R X I

W W I

0000061 7777 OWO WOO WOO WW

OOW062 WW WW W01 0653 4137

00W063 OOW WW WOO WW WOO

0000064 WW OOW WW 0005 1426

W00065 W04 2711 WOO OWO WOO

O00W66 OOW OOW OOW OOW OOW

0000070 WW 0033 0024 0404 OOW

WW071 7450 W01 2225 WOO 7421

0000072 7464 7564 7512 WOO OWO J

AFS64 W00073 W03 3364 OIW 0073 0005 CO A E

0000074 0003 3564 0400 0350 0025 C2 0 C/ U

ELV OOW075 0W3 5564 06W 0575 0002 C F E B

WI W00076 OWO OOW OOOO OWO OOOO

0WW77 OOOO 7754 OOW 40W 72W * 5

0000100 WOO WOO OOOO WOO WOO

0 T0t>

ARU 8 W00107 0005 4137 0001 0653 4117 E64 AFS60

FORMAT OUHP.

OOOWOOOOWOOOWWW OOOOOWWWWOOWWO W0OO0WW0W0007240 30766010301005743001

341314423412147302W 01633014010W3417761 0W0WOOOOW0O007164 777751223W0320WW0 1KL71JL B A XLA C

W7511WWW0W000W 7777OWO0WWWWW0 00000000000106534137 OOWOOOOOOWOOWOWO I

000W0W0WW0051426 WO42711OW0W00OOW OOOOOOOWOOWOOOOWO OOOOOOOOOOOOOOOOOOOO ELV 0W1

W00W330024O4O4O0W 74500W122250W07421 746475647512WW00W 000333640100W730005 0 TOO / ARU

OW33564040003SOW25 0003SS6406000S7SOW2 OOWOWWOOOOOOOWW 0000775400W4W072W C2 0 C/ U C Ft

WWOWWOWOOOQWOO wooowwwwowoow owoowwwwwooooo wwowooowoooowoo

000WWW0WW00O0W 00OOOOWO0W0000O0W OOOOOOOWWOOWOOOW 00054137000106534117

HXHE XA

(RX 1

Figure 6-11. Sample C Directive Output

6-74 NOS Version 2 Analysis Handbook Revision M

Printer Output Listing Examples

Example 6:

Figure 6-12 illustrates the printer output listing of the system ﬁle name table

produced by the FNT CMR dump directive. This table is printed in the same format

as that produced by the D memory dump option (refer to example 5).

RA =

wwwo FNT.

CN 12.31.31. 83/06/12. COC OUKP 00 DSDI

NETWORK OPERATING SYSTEM NOS 2 83/06/12. 12.33.11. PAGE 1

SYSTEM FILENAME TABLE

0W247O6

W024707

2331 2324 0515 OOW 1W0 SYSTEM

W06 4005 OOW WW OOW FSE 00024723 OOW WW OWO WW WOO

00024724 OOW OOW WW OOW W10 W024740 OWO WOO WW WW W16

H 0W24741 OOW OOW OOW OOW WW

0W24710 2223 3004 4137 OWO 13W RSX064 0W2472S OOW WW WW WW WW W024742 OOW OOW OOW OOW W17

0W24711

W024712

0010 4463 OOW OOW OOW H9 W024726 OOW WW OOW WW 0011 I 00024743 WW WW OOW OOW WW

2223 3026 4137 OOW 13W R5XV64 OW24727 OOW WW WW OWO OWO 0W24744 OOW WOO OOW WW 0020

0W24713 0010 4201 OOW OOW OOW H7A 00024730 OOOO WW OOW OOW W12 J 0W24745 WW WW OOW OOW OOW

00024714 2022 1706 1114 03W 13W PROFILC W024731 WW WOO WW WW WOO 00024746 OOW WOO WW WW W21

0W24715 W10 5410 OOW WW WW H'H 0W24732 OOW OOW OOW WW W13 K W024747 OOW OOW OOW OOW WW

00024716

0W24717

2601 1411 0425 23W 13W VALIDUS 00024733 WOO OOW WW WW OWO 00024750 OWO WW WW WW W22

0010 6504 OOW OOW OOW H 0 0W24734 OWO OOW OOW WW W14 L W024751 OOW OOW OOW OOW OOW

tWixﬁO 2601 1411 0425 23W 13W VALIOUS 00024735 WW WW WW WW WOO

W024721 WOS 4215 OOW OOW OOW E7N 0W24736 WW OOW WW WW 0015

00024722 0000 WW WW WOO W07 00024737 OOOO WW WW WW OOW

Figure 6-12. Sample FNT Directive Output

0$ss

Revision M Express Deadstart Dump Interpreter (DSDI) 6-75

Printer Output Listing Examples

Example 7:

Figure 6-13 illustrates the printer output listing produced by the PRORF directive for

CYBER 180 model 990 and 995 machines.

1 R A • P R O R F .

0000000 CH 09.06.46. 85/07/2*.

ftUNP 00

COC NETWORK OPERATING SYSTEM

8S/08/07. 17.23.56.

RADIAL HCI « 1

IOU CURRENT INSTRUCTION REGISTER

0 0000 0800 0000 2000

1 OOOO 2000 0000 007B

2 0080 0000 0000 0000

3 0000 0000 0000 0078

4 FFFF EOOO FOFF EFOF

5 FFFF F30C FOFF 1603

6 0003 EOOO OOFF CO40

7 SCFf EOOO OQFO EOOO

EXECUTINS WORDS OF SOFT CONTROL MEMORIES

ACU HZ OOOO 2000 oooo OOOO

ACU H3 0040 OOOO 0060 OOOO

ACU M4 OOOO OOOO oooo OOOO

BOP OOOO 0800 oooo OOOO

OOOO OOOO oooo 0800

EPN SCH OOOO 4V00 C001 C200

epn tin OOOO OOOO 0100 F200

IBU CS OOOO OOOO oooo OOOO

OODO OOOO oooo OOOO

INU MAP OOOO OOOO 0030 4AF2

INU IBS FFFF FFFF FFFF FFFF

LSO OBOO OOOO OOOO OOOO

SVA BN ODOO oooo BFFF F019

ERROR INFORMATION TABLE

0 OOOO OOOO OOOO F200

1 OOOO OOOO 0101 FZ10

2 OOOO OOOO 0202 F220

OOOO OOOO OEOE F2E0

OOOO OOOO OFOF F2F0

INSTRUCTION BUFFER STACK

0 C70D 9322 OBOO AAOA

1 F81Q 2BA6 AF32 F2S0

2 »»«2 9012 1000 0004

63 8042 225* 6C60 4224

44C6 EBCA 8E66 CI10

AO OOOO 1004 OOOO 67»0

A1 OOOO FFFF 8000 OOOD

A2 OOOO FFFF 8000 OOOO

OFFF 03F2 4E41 5500

07FF 03F2 4E41 5300

HISTORT FILE

0 X-REG 0030 OOOO OOOO OOOO

0 A-REC 1003 OOOO OOSE

0 P, N 0002 4640 9C00

1 X-RE4 OOFt COOO 01 AS 48B9

1 A-REC 10FF OOOF ECOO

1 P, N 0002 4644 4B00

X-REO OOOO OOOO OOOO OOOO

A-RES 1B03 OODO OOOO

r, H 00 02 46 00 8 000

Figure 6-13. Sample PRORF Directive Output

6-76 NOS Version 2 Analysis Handbook Revision M

Install Command

NOS allows you to deadstart from a disk rather than from a tape. To deadstart from a

disk, you must ﬁrst use the INSTALL command to install a deadstart ﬁle that you

specify to a deadstart disk. The disk that you select must be available on a single-unit

device such as an 844 or 885-11/12,1 and must have the Common Test and

Initialization (CTI) software. (For more information about CTI, refer to the CYBER

Initialization Package User's Handbook.) If the deadstart disk is a shared device in a

multimainframe environment, a full INITIALIZE must have been previously done. The

calling job must be of system origin or you must be validated for system origin

privileges and the system must be in debug mode. In addition, in a secured system,

INSTALL is allowed only from jobs with security administrator privileges.

You must re-execute the INSTALL command if a full INITIALIZE becomes necessary,

because PFDUMP and PFLOAD do not dump or load the system deadstart ﬁle (SDF).

The format of the command is:

INSTALL.ﬁlename,EQest.

Parameter Description

ﬁlename Disk ﬁle (assigned to the job) to be installed as an SDF. Default is

SYSTEM, which must be assigned to the job. The name SDF is a

reserved ﬁle name and cannot be speciﬁed for ﬁlename.

INSTALL stops reading the system deadstart ﬁle at the ﬁrst EOF;

therefore, the new running system does not include any changes made

through SYSEDIT after a previous deadstart.

If the deadstart ﬁle to be installed is on tape, it must ﬁrst be copied

to disk before INSTALL can be used.

NOTE

est

The deadstart ﬁle to be installed must contain a valid OPLD directory.

Note that when you use LIBEDIT to update a deadstart ﬁle and write

the new deadstart ﬁle directly to tape, the OPLD record on the tape

will not contain valid disk addresses. Therefore, if you want the

deadstart ﬁle on tape and you intend to later use the INSTALL

command on the deadstart ﬁle, you should have LIBEDIT write the

new deadstart ﬁle to disk, and then copy the ﬁle to tape using the

COPY command (refer to the NOS 2 Reference Set, Volume 3).

EST ordinal of the deadstart disk on which ﬁle ﬁlename is to be

installed.

1. The system deadstart ﬁle cannot be installed to an 819, 885-42, 887, or 9853 disk.

Revision M Install Command 7-1

?*%

f^'

K-Display Utilities 8

FLAW K Display 3.3

INITIALIZE K Display 8-8

Machine Recovery (MREC) Utility K Display 8-16

MREC Procedures 8-17

Using MREC with the K Display .. 8-17

Using MREC without the K Display 8-22

MREC Unit and Controller Reservations 8-23

Mass Storage Extended Subsystem (MSE) K Display 8-25

Network Access Method (NAM) K Display 8-27

Display Control Characters 8-29

NAM K-Display Operation 8-30

NAM Mode Maintenance Commands 8-32

DB Command 8-32

DE Command 8-33

DU Command 8-33

FL Command 8-34

LB Command 8-34

LE Command 8-35

LR Command ... 8-35

RS Command 8-36

ST Command , 8-37

Queue File Transfer Facility (QTF) K Display f 8-41

QTF Selection Classes . 8-41

File Selection Process . 8-41

Visualizing the File Selection Process 8-42

Queued File Size Index 8-44

Queued File Size and Transfer Times 8-44

QTF K-Display Usage 8-45

QTF Operation Under RHF 8-45

QTF Operation Under NAM 8-46

QTF K Display Commands 8-47

CLASS Command 8-48

DISABLE Command 8-49

ENABLE Command 8-49

HELP Command 8-50

IDLE Command 8-50

INCLUDE Command 8-51

SCHED Command 8-52

STATUS Command 8-53

QTF Physical Identiﬁer K Display 8-55

QTF Selection Class K Display 8-57

QTF Transfer K Display 8-59

STOP Command 8-60

REDEFINE K Display 8-61

Remote Batch Facility (RBF) K Display 8-72

Remote Host Facility (RHF) K Display . , 8-75

RHF Initiation >. 8-76

Operator Interface : 8-76

RHF Commands Under K display , 8-78

Application Table Display — 8-79

Network Path Table Display ................ 8-81

RHF Commands Available Under Application or Network Path Display ...... 8-84

RHF Termination 8-84

SCOPE 2 Station Facility (SSF) K Displays ...... , 8-85

Operator Interface ....— 8-85

Station Login 8-85

Enabling and Disabling File Transfers 8-85

Station Recovery .-.. 8-86

S t a t i o n D i s c o n n e c t i o n a n d L o g o u t . . . . . — . . . . . . . . . . . . . . ; . . 8 - 8 6

File Transfer Limit Commands ,,. 8-87

Staged File Transfer Commands 8-87

Spooled File Transfer Commands 8-8&

Transaction Facility (TAF) K Displays ... .... 8-89

TAF Initialization K Display ,-,. 8-89

Restarting TAF K Display 8-93

TAF Status K Display 8-93

TAF K-Display Commands . 8-96

TAF/CRM Status K Dis p l ays and Comman d s . . ........... . . ...... 8-100

K-Display Utilities

This section documents the K displays listed next, along with the utilities used to

present them. Other K-display utilities described elsewhere in this manual include the

permanent ﬁle utilities, queued ﬁle utilities, and the status/control register simulator.

K-display utilities described in other manuals include those for the CDCNET network

(refer to the CDCNET network manuals) and the printer support utility (refer to the

NOS Version 2 Operations Handbook).

K Display

FLAW

INITIALIZE

Machine Recovery (MREC)

Mass Storage Extended Subsystem

(MSE)

Network Access Method (NAM)

Queue File Transfer Facility (QTF)

REDEFINE

Remote Batch Facility (RBF)

Remote Host Facility (RHF)

SCOPE 2 Station Facility (SSF)

Transaction Facility (TAF)

Description

Disk ﬂaw mapping.

Disk initialization.

Multimainframe device recovery.

K display of the MSE.

Operator interface to NAM.

K display of the QTF application.

Online reconﬁguration display.

K display of the RBF Subsystem.

K display of the RHF Subsystem.

K display of the SSF Subsystem.

K display of the TAF Subsystem.

Revision M K-Display Utilities 8-1

K-Display Utilities

/*^k

By using the K display, a job can place information on the console screen and receive

information from the keyboard. The information is passed to the job by DSD. Normally, a*^%

these displays are used for utility programs. '

The job ﬁrst issues a request message on the B,0 display, asking you to bring up the

K display.

When this happens, enter:

K.jsn.

jsn Description

jsn Job sequence name of the requesting job.

Once the display is assigned to a job, you can enter data by typing K. followed by

data. The data is transferred to a speciﬁed area of the job's ﬁeld length when you

terminate the entry. If more than 50 characters are entered as data, the message

LINE TOO LONG. /^^

appears on the screen. DSD does not accept the entry until the data string is

shortened.

K displays are job oriented. The job sequence name associated with each K display

appears at the top of the screen next to the display designator and name.

All parameter entries must be preﬁxed by K period (K.). However, after pressing CR

or NEXT for the ﬁrst parameter entry, everything but the K. is erased. This allows

another parameter to be entered without entering K. ﬁrst. All examples in this /SS%.

section show K. although you may not have to enter it. If you have to enter a DSD

command during parameter entry, backspace to erase the K., enter the command, and

then continue parameter entry by entering K. and the parameter.

8-2 NOS Version 2 Analysis Handbook Revision M

FLAW K Display

The FLAW utility reserves (ﬂaws) tracks on any mass storage device during normal

system operation. Each entry identiﬁes an area of mass storage that is unusable

(ﬂawed area) and prevents the system from accessing it. Since 881 and 883 disk packs

normally contain ﬂaw information in the utility sector, the FLAW utility should be

used on an 881 or 883 only to specify additional areas not currently in the utility

sector. Obtain ﬂaw addresses from the customer engineer.

Flawing tracks on mass storage devices is accomplished using the FLAW K displays

(ﬁgures 8-1 and 8-2).

K . F L A W . A A A M

CURRENT SLF, CLF. SPF, AND CPF ENTRIES

EQ = 0 EST ORDINAL OF DEVICE.

NUM VALUES ENTERED TRT OCTAL VALUES • = DUPLICATE

ENTER EST ORDINAL.

Figure 8-1. FLAW K Display (Left Screen)

Revision M K-Display Utilities 8-3

FLAW K Display

FLAW. AAAM

MASS STORAGE DEVICE FLAWING

SLF, CLF, SPF AND CPF ENTRIES

SLF = NNNN SET LOGICAL TRACK FLAW.

CLF = NNNN CLEAR LOGICAL TRACK FLAW.

SPF = XX SET PHYSICAL TRACK FLAW.

CPF = XX CLEAR PHYSICAL TRACK FLAW.

WHERE

NNNN = VALID LOGICAL TRACK NUMBER.

XX = COO0.T00.SO0 (CYLINDER,TRACK.SECTOR.)

= A0000000 (ADDR/1OB, FOR DE/DP ONLY.)

NUMERIC VALUES ARE CONSIDERED OCTAL UNLESS THE

DIGITS 8 OR 9 APPEAR IN THE NUMBER, OR A *D* IS

APPENDED TO THE NUMBER.

SEE LEFT SCREEN FOR CURRENT ENTRIES,

UP TO 14 ENTRIES ALLOWED

Figure 8-2. FLAW K Display (Right Screen)

All console entry is performed under DSD control. The sequence is:

1. Call the FLAW utility by entering:

X.FLAW.

2. Bring the K display to the left console screen by entering:

K.jsn.

jsn Description

jsn Job sequence name of the job requesting the K display (the

message REQUEST *K* DISPLAY appears on the B,0 display).

3. You can bring the ﬂawing instructions up on the right screen by entering KK.

4. Specify the mass storage device on which tracks are to be ﬂawed. Enter:

K.EQ=est.

est Description

est EST ordinal of the mass storage device.

^^*\

8-4 NOS Version 2 Analysis Handbook Revision M

0f^S

FLAW K Display

5. Enter ﬂaws. A maximum of 14 ﬂaw entries is allowed for each call to the FLAW

utility. In addition, there are four types of ﬂaw entries that may be speciﬁed. The

general format for the K display entry is:

K.xtk=ta.

Variable Description

xtk Speciﬁes one of the following types of ﬂaw entries.

xtk Description

SPF Sets the track reservation table (TRT) entry for the

speciﬁed physical block (track) address(es) in

extended memory or disk to indicate that the block

is unavailable for use.

CPF Clears that TRT ﬂaw entry for the speciﬁed physical

block address(es) in extended memory or disk to

indicate that the block is available for use.

SLF Sets the TRT ﬂaw entry for the speciﬁed logical

track to indicate that the track is unavailable for

use.

CLF Clears the TRT ﬂaw entry for the speciﬁed logical

track to indicate that the track is available for use.

ta Speciﬁes the track address to be reserved. (Refer to table 8-1.)

ta can be one of the following:

ta Description

trk Logical track address for a disk or extended

memory. (Use with SLF and CLF.) The variable trk

can be any octal number in the speciﬁed range for

the particular type of disk drive or extended

memory.

Aaddr Physical block (track) address for extended memory.

(Use with SPF or CPF.) addr is the extended

memory address divided by 108.

Aaddr- Range of physical block addresses for extended

Aaddr memory. (Use with SPF or CPF.)

Ccyl, Physical track address for the disk. (Use with SPF

Ttrk, or CPF.) The variables cyl, trk, and sec can be any

Ssec octal number in the speciﬁed range for the

particular type of disk drive.

6. Initiate ﬂawing of the speciﬁed device by entering:

K.GO.

Revision M K-Display Utilities 8-5

FLAW K Display

The FLAW utility provides two messages in the system dayﬁle that indicate the results

of the ﬂawing operation. The ﬁrst message is: ^^\

nn TRACKS FLAWED.

nn Description

nn Octal number of tracks that were successfully ﬂawed.

The second message appears only if some of the ﬂaws speciﬁed were not processed.

This occurs when the track speciﬁed for ﬂawing is already reserved by the system (but

not as a ﬂawed track). In this case, the following message also appears in the system

dayﬁle.

nn FLAWS NOT PROCESSED,list.

Variable Description

nn Octal number of ﬂaws not processed.

list List of the logical tracks that were not ﬂawed.

The entries described here are similar to those entered in the APRDECK (refer to

section 3, Deadstart Decks) for ﬂawing a device at deadstart time. However, the ﬂaw

entries speciﬁed using the FLAW utility or DSD command INITIALIZE are not

recovered if the device is initialized at deadstart time. Only the ﬂaw entries speciﬁed

in the APRDECK will be recovered. If a device is initialized during normal system

operation (INITIALIZE command), all ﬂaws speciﬁed in the devices TRT, including ^

those entered using the FLAW utility or INITIALIZE command, will be recovered

providing the device has a good label at the time of initialization. If the label is bad, '**%.

or cannot be recognized, all current ﬂaws are cleared. The left console screen

(ﬁgure 8-3) shows all ﬂaw entries made through the FLAW utility and INITIALIZE

command. It lists the ﬂaw entry and its logical track equivalent. If the same logical

track is referenced by more than one ﬂaw entry, an asterisk appears to the right of

those entries, so that you are aware that only the last entry takes effect.

8-6 NOS Version 2 Analysis Handbook Revision M

FLAW K Display

00&S

K . F L A W A A A M

CURRENT SLF, CLF, SPF. AND CPF ENTRIES

EQ = 14 EST ORDINAL OF DEVICE.

NUM VALUES ENTERED TRT OCTAL VALUES * = DUPLICATE

1SLF=4000. 4000 *

2SLF=4000B. 4000 *

3SLF=4001. 4001

SLF=4001.

Figure 8-3. FLAW Utility K Display (Left Screen)

Table 8-1. Flawing Information for Disks and Extended Memory

Logical

Track Cylinders/ Tracks/ Sectors/

Device Type MNE Range1 Device1 Cylinder1 Track1

819, single density DV 4000-5465 633 12 24

819, double density DW 4000-7153 1466 12 24

834 DD 4000-7135 1457 12 40

836 DG 4000-6565 1273 30 57

844-21, half track DI 4000-7137 630 22 30

844-21, full track DK 4000-7137 630 23 30

844-41/44, half track DJ 4000-7147 1464 23 30

844-41/44, full track DL 4000-7147 1464 23 30

885-11/12, half track DM 4000-7221 1511 50 40

885-11/12, full track DQ 4000-7221 1511 50 40

885-42, full track DB 4000-7221 1511 12 40

887, 4K sector DF 4000-7343 1562 46

887, 16K sector DH 4000-7343 1562 13

895 DC 4000-7351 1565 17

9853, 2K sector DN 4000-7726 2601 23 25

Extended memory DE/DP 4000-7620

1. Numbers are in octal.

Revision M K-Display Utilities 8-7

INITIALIZE K Display

The INITIALIZE command can be used to reconﬁgure certain removable devices (844-21

and 844-41/44) to suit your needs. For example, if you currently have two single-unit

844 packs (DK-ls), both packs can be initialized and linked together to form a

multispindle device (DK-2). However, this can be done only if the devices to be linked

meet the following requirements:

• Same device type.

• Same channels.

• Same share status (shared or nonshared).

• Removable.

• Not currently in use.

The INITIALIZE command must be entered to set the initialize status for each device

to be chained. Current multispindle devices can also be initialized providing all packs

that form the device are mounted in a logical order as determined by the unit numbers

list on the E,C display. It is only necessary to enter the INITIALIZE command for the

ﬁrst unit of a current multispindle device. The format of the INITIALIZE command is

described in section 5, DSD Commands.

NOTE

Examine the FAMC and DAFC ﬁelds in the family status display (E,F) before

entering the INITIALIZE command. The user count for the device must be zero before ^s%

this command is valid. /

8-8 NOS Version 2 Analysis Handbook Revision M

0^S

yims

INITIALIZE K Display

The following steps describe the procedures to initialize and (if necessary) ﬂaw tracks

on a mass storage device.

1. Enter the INITIALIZE command for the device(s) to be initialized. Examine the

B,0 display for the following message:

REQUEST *K* DISPLAY.

Note the job sequence name of the job displaying the message is CMS.

2. Activate the K display for that job by entering:

K.CMS.

The INITIALIZE K display (ﬁgure 8-4) appears on the left console screen. All

parameters required to initialize and (if necessary) ﬂaw the speciﬁed device are

entered through the K display.

ypPk\

K. INITIALIZE. CMS

MASS STORAGE DEVICE INITIALIZATION PAGE 1 OF 2

INITIALIZE EQUIPMENT

014

INITIALIZE REQUEST *TOTAL*

OPTION DESCRIPTION

EQ = 014 EST ORDINAL OF DEVICE (5-777).

FM = 0 FAMILY NAME OR PACK NAME IF TY=X (1-7 CHAR).

DN = 0 DEVICE NUMBER (1-77).

UN = 0 PRIVATE AUXILIARY DEVICE USER NAME (1-7 CHAR).

TY o 0 ACCESS TYPE (F OR X ).

LA = 0 LOWER ACCESS LEVEL (1-7 CHAR).

UA = 0 UPPER ACCESS LEVEL (1-7 CHAR).

DM = 0 DEVICE MASK (0-377).

SM = 0 SECONDARY MASK (0-377).

NC = 0 CATALOG TRACKS (0-200).

NP = 0 NUMBER OF PACKS (1-7).

ENTER OLD FAMILY NAME.

Figure 8-4. INITIALIZE Command K Display

The K display lists the options used to initialize a device (under the OPTION

column as shown in ﬁgure 8-4). Refer to table 8-2 for a description of each option.

Refer to table 8-3 for additional information. The EST ordinal of the device to be

initialized is listed under the INITIALIZE EQUIPMENT heading (see ﬁgure 8-4).

Revision M K-Display Utilities 8-9

INITIALIZE K Display

Table 8-2. Device Initialization Options

Option1 Description )

EQ EST ordinal of the device to be initialized. For multispindle devices, this

ordinal must be the ﬁrst of the consecutive units in the multispindle chain.

FM The 1- to 7-character family name. Speciﬁes the permanent ﬁle family in

which the initialized device is to be included. All devices must have a

family name or pack name. The name 0 (single character zero) is reserved

and cannot be used. This option cannot be used to change the family name

of the link device in a multimainframe environment. If you specify TY=X,

this option speciﬁes a 1- to 7-character packname to be associated with an

auxiliary device. To clear an existing entry, enter FM=NULL.

DN The 2-digit (octal) logical device number (from 1 to 77s) that uniquely

identiﬁes the device in its permanent ﬁle family. This option cannot be

entered if you specify TY=X.

UN The 1- to 7-character user name. This option is speciﬁed only when ")

initializing an auxiliary device (TY=X). If speciﬁed, the device is

considered to be a private auxiliary device. Only the user name speciﬁed

will be allowed to create ﬁles on the device (use SAVE, REPLACE, or

DEFINE commands). To clear an existing entry, enter UN = NULL.

TY=F Initialized device may contain direct and indirect access permanent ﬁles.

However, if you specify DM=0, only direct access ﬁles can reside on the

device. If you specify SM=0 and DM=0, the device can contain only

special system permanent ﬁles. Indirect access ﬁles can reside only on a .A*ms

master device (that is, DM=£0). )

TY=X Initialized device is an auxiliary device. This is a mass storage device that

is not part of a permanent ﬁle family. An auxiliary device is a

supplementary permanent ﬁle storage device that may be privately owned

(UN option speciﬁed) or can be shared by many users (UN not speciﬁed).

Auxiliary devices can contain direct or indirect access permanent ﬁles.

LA Lower limit for the access level of the device. This establishes the lowest

access level of ﬁles that can be stored on the device (must not be less than .-*ms

the lower limit for the access level of the device as given in its EST '

entry).

UA Upper limit for the access level of the device. This establishes the highest

access level of ﬁles that can be stored on the device (must not be greater

than the upper limit for the access level of the device as given in its EST

entry).

DM The 3-digit (octal) device mask (from 0 to 377s). This option is required

whenever a permanent ﬁle master device is being initialized. It deﬁnes

which users will have this device as their master device. This option

cannot be entered if you specify TY=X.

1. Device initialization options may be changed only if the total initialization level

(AL) is speciﬁed in the INITIALIZE command.

(Continued) _^

8-10 NOS Version 2 Analysis Handbook Revision M

/uSSS*^

A0&\

INITIALIZE K Display

Table 8-2. Device Initialization Options (Continued)

Option2 Description

SM The 3-digit (octal) secondary device mask (from 0 to 377s) used to control

the residence of direct access ﬁles. This option cannot be entered if you

specify TY=X.

NC Octal number of catalog tracks (from 0 to 200s; must be a power of 2).

This option is used only if the number of catalog tracks speciﬁed as the

system default for the device type is not satisfactory.

NP Number of physical units to be included in a multispindle device. The

default value is 1. Each unit to be included in the multispindle chain must

be deﬁned as removable in the current disk status display (E,M).

2. Device initialization options may be changed only if the total initialization level

(AL) is speciﬁed in the INITIALIZE command.

Table 8-3. Track Flawing Options

Option1 Description2

SPF Sets the track reservation table (TRT) entry for the speciﬁed physical

block (track) address(es) in extended memory or disk to indicate that the

block is unavailable for use.

CPF Clears the TRT ﬂaw entry for the speciﬁed physical address(es) in

extended memory or disk to indicate that the block is available for use.

SLF Sets the TRT ﬂaw entry for the speciﬁed logical track to indicate that

the track is unavailable for use.

CLF Clears the TRT ﬂaw entry for the speciﬁed logical track to indicate that

the track is available for use.

1. Flawing of 881 and 883 disk packs is automatic; only ﬂaws additional to the

current ﬂaw information in the utility sector should be entered.

2. Refer to the APRDECK description in section 3, Deadstart Decks, for entry

formats.

3. Enter the INITIALIZE command for each additional device to be initialized. This

can also be done before activating the K display. In either case, only the ﬁrst

device speciﬁed will be listed (by EST ordinal) in the K display. Thus, to update

the K display to show additional devices, enter the following command:

K.RERUN.

If more than one device is listed, they are initialized one at a time as they

appear in the list from left to right. Multispindle devices (more than one EST

ordinal) are considered one device.

Revision M K-Display Utilities 8-11

INITIALIZE K Display

The system has already checked the label on each mass storage device. If the

label was found to be good, it is necessary to enter device initialization options

(requested in messages that appear in the K display) to identify the device. This

is to ensure that the device selected is indeed the correct device to be initialized.

The messages appear automatically and are displayed until the correct options are

entered. If an incorrect option is entered, it is ignored. Refer to table 8-2 for a

description of each option.

Examine the family status display (E,F) to determine the current option values.

The following messages may be displayed and the appropriate response should be

entered. If none of these messages appears, the device label was not recognized or

was found to be bad. In this event, proceed to step 5.

Message Description

ENTER OLD DEVICE

NUMBER

This message appears if the device to be initialized is a

permanent ﬁle family device. Enter the following

response:

Aeims

ENTER OLD FAMILY

NAME

ENTER OLD PACK

NAME

ENTER OLD USER

NAME

K.DN=dev i cenumber.

This message appears only if more than one family of

permanent ﬁle devices is currently active in the

system. Enter the following response:

K.FM=familyname.

This message appears only if the device to be

initialized is an auxiliary device. Enter the following

response:

K.PN=packname.

This message appears only if the auxiliary device to be

initialized is a private auxiliary device (associated with

a speciﬁc user name). Enter the following response:

K.UN=username.

The user name is written to the account dayﬁle when

the device is mounted.

If you discover that the wrong device was speciﬁed in

the INITIALIZE command, you can clear the initialize

status for that device by entering:

K.CLEAR.

The device to be processed by the clear entry must be

a valid device. That is, the device cannot have a device

number that conﬂicts with another device in its family

name or a pack name that duplicates one already in

the system. Its mask bits must meet standard

requirements. The leftmost device in the list of devices

to be initialized is cleared: One of the preceding

messages will then be displayed for the next device to

be initialized (if any) providing the label on that device

is good.

8-12 NOS Version 2 Analysis Handbook Revision M

INITIALIZE K Display

5. When the following message appears on the K display, enter the options that

specify the new characteristics to be associated with the device when it is

initialized (refer to tables 8-2 and 8-3).

ENTER PARAMETERS

The new options can be entered as a string or one at a time. For example:

K.optioni=valuei,...,optionn=valuen.

K.optioni=valuei.

K.option2=value2.

K.optionn=valuen.

If ﬂaw entries are to be speciﬁed (refer to table 8-3), they must be entered singly,

as shown in the last example. If the label on the device being initialized was

good, all current ﬂaws on that device are normally recovered. However, if the

label was not recognized or was bad, the ﬂaw entries cannot be recovered and

must be entered (if necessary) using this mechanism. A maximum of 14 ﬂaw

entries are permitted. In addition to the SLF, CLF, SPF, and CPF entries, the

ﬂaw information recorded in the utility sector on an 881, 883, or 885 disk pack is

read during initialization of 844 equipment, and the appropriate areas are

reserved by the system automatically.

If the NP option is speciﬁed (NP > 1), the device is to be initialized as a

multispindle device. In this case, the number of packs speciﬁed by NP indicates

the number of spindles to be linked. This is the next n number of devices waiting

to be initialized. Each device must be deﬁned as removable and mounted on

consecutive physical unit numbers. To determine if a device is deﬁned as

removable, examine the disk status display (E,M). If the units are conﬁgured

correctly, the label on each unit is checked. If any label is not recognized or is

bad, that unit is free for initialization and chaining. However, if the label is good,

the message

ENTER IDENTITY OF EQest

appears in the K display (est is the EST ordinal of the device). Enter one of the

following responses.

K.DN=devi cenumber.

K.FM=fami 1yname,DN=devicenumber.

K.PN=packname.

K.PN=packname,UN=username.

This is a precautionary measure to ensure that the devices speciﬁed are the

correct devices to be chained.

6. After all the necessary options have been entered for a speciﬁc device, enter the

following command to proceed with the initialization.

K.GO.

If more devices are waiting to be initialized, repeat steps 4 through 6 of this

procedure for each device.

Revision M K-Display Utilities 8-13

INITIALIZE K Display

Example 1:

Suppose you want to create and initialize a multispindle device by linking spindles

from two separate units. Assume your EQPDECK has the following entries:

EQ40=DL,ST=OFF,UN=4,CH=27/30.

EQ41=DL,ST=OFF,UN=5,CH=27/30.

and you want to initialize multispindle device EQ40, which consists of units 4 and 5.

Enter the following commands:

ON,40.

ON,41.

INITIALIZE.AL,40,41.

MOUNT,40.

MOUNT,41.

K.CMS.

Enter device initialization options as described in steps 4 and 5.

K.NP=2.

When the message ENTER IDENTITY OF EQest appears in the K display, respond as

described in step 5.

K.GO.

8-14 NOS Version 2 Analysis Handbook Revision M

INITIALIZE K Display

0ms

Example 2:

Assume that you want to initialize a multispindle device and your EQPDECK already

deﬁnes the device as a multispindle device:

EQ40=DL,ST=OFF,UN=4/5,CH=27/30.

In this case, enter the following commands:

ON.40.

INITIALIZE,AL,40.

MOUNT,40.

K.CMS.

Enter device initialization options as described in steps 4 and 5.

K.GO.

Example 3:

Suppose you want to initialize a single-spindle device and your EQPDECK deﬁnes the

device as a multispindle device:

E04ODL,ST=OFF,UN=4/5,CH=27/30.

Mount the pack you want to initialize on unit 4 and enter the following commands:

ON,40.

INITIALIZE.AL.40.

MOUNT,40.

K.CMS.

Enter device initialization options as described in steps 4 and 5.

K.PN=1.

K.GO.

0ms

Revision M K-Display Utilities 8-15

Machine Recovery (MREC) Utility K Display

When a machine that has been in a multimainframe conﬁguration has a malfunction

that requires a level 0 deadstart, before you can begin the deadstart, the machine

recovery utility program (MREC) must be run on every machine that shares devices

with the machine that is not working properly. MREC releases local mass storage

space on the shared devices and clears interlocks set before the malfunction occurred.

For example, assume that machines A and B share disk unit 12 and machines B and

C share disk unit 13 as shown in ﬁgure 8-5.

MACHINE MACHINE MACHINE

A B C

DISK DISK

UNIT UNIT

12 13

Figure 8-5. Machine Conﬁgurations

If machine A must be deadstarted using a level 0 deadstart, MREC must be run on

machine B to recover disk unit 12. Machine C need not be involved since it is not

aware of the existence of disk unit 12. However, if machine B must be deadstarted

using a level 0 deadstart, MREC must be run on machine A to recover disk unit 12

and on machine C to recover disk unit 13.

NOTE

Once MREC has been run for an inoperative machine, any level of deadstart on the

machine other than 0 is not possible.

8-16 NOS Version 2 Analysis Handbook Revision M

/#^V

MREC Procedures

The MREC utility can be used through keyboard entry to a K display or by direct

keyboard entry under DSD or DIS control without using the K display.

Using MREC with the K Display

The following procedure describes K-display usage for MREC operations.

1. Call the MREC utility by entering this DSD command:1

X.MREC.

2. Examine the DSD system status display (B,0). When MREC is scheduled to a

control point, it is indicated on the B,0 display. The message

REQUEST *K* DISPLAY

appears in the message ﬁeld for that control point.

3. Activate the K display for that control point by entering:

K.jsn.

jsn Description

jsn MREC job sequence name.

The K display for MREC (ﬁgure 8-6) appears on the left console screen.

The MREC left screen K display lists all of the devices that are shared by the

machine on which MREC is being run. The machine identiﬁer (MID) of this

machine is also given. Information describing the shared devices is given in the

following format.

eq type un dn fm/pn mids sharing device

Header Description

eq EST ordinal of the shared device,

type Device type,

un Unit number of the device,

dn Device number,

fm/pn Family name/pack name.

mids sharing Machine identiﬁers of other machines that are currently

device accessing the device. If there is an * by the machine

identiﬁer, the machine is determined to be down.

1. Under DIS control, the command MREC calls the MREC utility.

Revision M K-Display Utilities 8-17

Using MREC with the K Display

K. MREC. jsn

hh.mm.ss . yy/mm/dd. CDC NETWORK OPERATING SYSTEM

LVLO LVL7 MID=AA NOS version

MACHINE RECOVERY UTILITY PAGE 1 OF 1

EQ TYPE UN DN FM/PN MIDS SHARING DEVICE

10 DJ PACKV2 73 72 71 70

11 DJ

12 DJ

13 DJ 30 MAINTV2 73 72 71 70

14 DL 40 SYS606 73 72 71 70

15 DI SYST06 73 72 71 70

16 DI

17 DI R4IAE 73 72 71 70

20 DJ

ID OF DOWNED MACHINE = 72

EQ(S) TO RECOVER = ALL

y^^lt

Figure 8-6. MREC K Display (Left Screen)

8-18 NOS Version 2 Analysis Handbook Revision M

Using MREC with the K Display

4. The MREC right screen K display contains the parameters and commands that

may be selected. Refer to table 8-4 (MREC parameters) and table 8-5 (MREC

commands) for more information. To activate the MREC right screen K display

(ﬁgures 8-7 and 8-8), enter:

KK.

5. Enter the MREC parameters in the following format:

K.parami=valuei,param2=value2 paranv1=valuen.

The parameters entered (and error messages, if any) are displayed on the lower

portion of the MREC left screen K display.

6. If you want to reset the parameters to their default values or refresh the device

descriptions on the MREC left screen K display, enter:

K.RERUN.

The parameters can then be reentered.

7. Enter the desired parameters and initiate MREC processing by entering:

K.GO.

When processing is complete, the message

PROCESSING COMPLETE

appears at the bottom of the left screen.

8. After all MREC operations are complete, end the MREC utility by entering:

K.STOP.

Revision M K-Display Utilities 8-19

Using MREC with the K Display

Table 8-4. MREC Parameters

Parameter Description

ID=id

EQ=est

OP = x

The 1- or 2-character machine identiﬁer of the inoperative machine that

is to be processed. This parameter must be entered before processing can

take place.

EST ordinals of the devices to be processed. Only devices shared between

the machine that is down and the machine on which MREC is running

are processed. The form of the entry is:

EQ=esti,est2,....estn.

If you omit this parameter, EQ=ALL is assumed. In this case, all

devices shared between this machine and the inoperative machine are

processed.

You can enter this parameter only when using the K display (that is, it

cannot be used if the MREC utility is called by a command) and then,

use it only if a unit or controller cannot be accessed by MREC due to

physical hardware reservations. Its use is invalid if a unit reservation is

not in effect.

x D e s c r i p t i o n

R Directs the MREC utility to release all unit reservations (using the

GRENADE function; refer to section 2, Deadstart) for 844-41/44

equipment.

I Directs the MREC utility to ignore certain functions on the

equipment for which the reservation message was issued. Functions

that do not require the unit to be accessed are still performed.

Table 8-5. MREC Commands

Command Description

GO. Directs the MREC utility to proceed with processing the entered

parameters.

RERUN. Reinitializes the device descriptions and parameters on the K display and

reruns the MREC utility.

STOP. Terminates the MREC utility and ends the K-display interaction.

+ Pages the left screen K display forward to the next screen.

Pages the left screen K display backward to the ﬁrst screen.

) Pages the right screen K display forward to the next screen.

( Pages the right screen K display backward to the previous screen.

■«*^ii\

8-20 NOS Version 2 Analysis Handbook Revision M

/$m\

0ms

Using MREC with the K Display

MREC. jsn

MREC

PARAMETER DESCRIPTIONS

PAGE 1 OF 2

MID OF DOWNED MACHINE TO PROCESS.

EST ORDINALS OF DEVICES SHARED BETWEEN THIS MACHINE

AND DOWNED MACHINE TO PROCESS. ENTER AS

EQ=XXX,XXX XXX. THE DEFAULT IS EQ=ALL.

MAY BE ENTERED IF 844 RESERVE SITUATIONS OCCUR.

THE RESPONSE APPLIES ONLY TO THE CURRENT SITUATION.

OP=I IGNORE CURRENT DEVICE.

OP=R CLEAR ALL UNIT RESERVES ON THAT CONTROLLER

K DISPLAY COMMANDS

RERUN

STOP

INITIATE PROCESSING OF DEVICES SPECIFIED.

REINITIALIZE K-DISPLAY AND RERUN PROGRAM.

TERMINATE PROGRAM.

PAGE LEFT DISPLAY FORWARD.

PAGE LEFT DISPLAY BACKWARD.

PAGE RIGHT DISPLAY FORWARD.

PAGE RIGHT DISPLAY BACKWARD.

Figure 8-7. MREC K Display (Right Screen) (Page 1 of 2)

MREC

DESCRIPTION OF TABLE ENTRIES.

PAGE 2 OF 2

FM/PN

MIDS SHARING DEVICE

* (BESIDE MID)

EST ORDINAL OF EQUIPMENT.

UNIT NUMBER.

DEVICE NUMBER.

FAMILY OR PACK NAME.

MACHINE ID-S OF OTHER MACHINES

CURRENTLY SHARING THE DEVICE.

THE MACHINE IS DOWN. IT CANNOT BE

DETERMINED IF MACHINES RUNNING IN

SHARED RMS MODE ARE DOWN.

Figure 8-8. MREC K Display (Right Screen) (Page 2 of 2)

Revision M K-Display Utilities 8-21

Using MREC without the K Display

You can also call the MREC utility and specify the appropriate parameters without

using the K display. You can do this by making a single keyboard entry (under DSD

or DIS control) in the following format:

X.MREC(param1=valuei,param2=value2 paramn=valuen)

The parameters (parami = valuei) are described in table 8-4. The ID = id parameter must

be entered.

When the MREC command is entered with parameters, the K display is not activated

and processing occurs automatically. If an error occurs using this procedure, the

message

REQUEST *K* DISPLAY

appears on the DSD B,0 display. Activate the K display and continue as described

under Using MREC With the K Display earlier in this section. ^^

8-22 NOS Version 2 Analysis Handbook Revision M

0ims

/0ms

MREC Unit and Controller Reservations

When attempting to access a device, MREC may ﬁnd the controller access or unit

reserved by another machine. When this occurs, the following message appears on the

left screen K display.

EQest,CHCC, CONTROLLER RESERVED.

EQest,UNUU, UNIT RESERVED.

Variable Description

e s t E S T o r d i n a l o f t h e d e v i c e .

cc Channel number.

uu Physical unit number (from 0 to 778).

Assuming the inoperative machine is the machine holding the reservation, you can

clear the reservation or direct MREC to clear it by using the following procedures:

To clear a controller reservation, initiate deadstart on the machine that is down.

To clear a unit reservation, perform one of the following procedures:

• On a device that is connected to a 7155 controller, initiate deadstart on the

machine that is down.

• On a device that is not connected to a 7155 controller, toggle the OFF LINE/ON

LINE switch on the back of the drive to OFF LINE and then back to ON LINE.

• If none of the preceding procedures can be performed, select the OP = R parameter

to clear an 844 device reservation.

Revision M K-Display Utilities 8-23

MREC Unit and Controller Reservations

NOTE

Do not select the OP=R parameter unless the other reservation clearing procedures

cannot be performed. It is recommended that all machines in a multimainframe

environment be put in IDLE status or put in STEP mode when you select the OP=R

parameter.

Once the correct action has been taken, enter

K.GO.

to continue processing. If the reservation is still not cleared, the appropriate reserved

message reappears on the left screen K display. Repeat one of the reservation clearing

procedures described above or enter:

K.OP=I.

This directs the MREC utility to ignore certain functions on the device. Processing may

then continue.

If a device or controller in an independent shared device multimainframe environment

is reserved by a down machine, the previously mentioned messages appear on the

system 'status display (B,0) instead of on the K display. Use the procedures already

described to clear the reservation.

8-24 NOS Version 2 Analysis Handbook Revision M

00ms

Mass Storage Extended Subsystem (MSE) K Display

MSE uses the K display to present messages that require your action. You can use the

MSE K display to reply to these messages and to open and close staging/destaging

operations. You can also use the MSE K display to display a speciﬁc storage module's

ON/OFF status, the number of DRDs within the storage module for staging/destaging,

and other information including the job sequence name, ﬁle name, user index, and

PRU count for the ﬁles being processed. In addition, you can set the maximum number

of DRDs to be used for staging/destaging operations within a speciﬁed storage module.

When MSE enters messages into the K display, a request for the K display appears

on the B,0 display. Enter K,MSE to activate the display for MSE. After you have

responded to all K-display messages, the request for the K display terminates.

The MSE K display provides space for two messages with up to three lines per

message. If you request information about a speciﬁc storage module, the message

area is replaced by the information concerning the storage module. Figures 8-9 and

8-10 show the message and storage module formats of the MSE K display.

Valid MSE commands are:

Command Description

K.m.GO.

K.STAGE.

K.DESTAGE.

K.SMi.

K.SMi,DRD=ST=s.

K.SMi,DRD=DS=d.

Enters a GO response to the message at message ordinal m.

Toggles staging operations between open and close. Staging is

normally open (permitted).

Toggles destaging operations between open and close. Destaging

is normally open (permitted).

Displays information for storage module i (i is an SM identiﬁer

A through H). The display shows SM ON/OFF/MAINTENANCE

status, the number of DRDs for staging/destaging, and DRD

activity including the job sequence name, ﬁle name, user index,

and PRU count for the ﬁles being processed.

The display also shows if a DRD is being used for staging or

destaging, a DRD is under control of a utility, or the state of a

DRD is maintenance mode, off, or idle.

Sets the maximum number of DRDs to be used for staging

within storage module i (i is an SM identiﬁer A through H). s

is an integer 0 through 2; an entry of 0 turns off staging for

SMi.

Sets the maximum number of DRDs to be used for destaging

within storage module i (i is an SM identiﬁer A through H). d

is an integer 0 through 2; an entry of 0 turns off destaging for

SMi.

0^S.

MSE clears a message when an acceptable action is taken. Usually this action is

entering a GO response to the message at message ordinal m. However, for actions

such as emptying the exit tray on a storage module, the message is cleared when the

hardware status indicates the action has occurred and you enter the GO response.

Refer to appendix B of the NOS Version 2 Operations Handbook for the appropriate

action for each message.

Revision M K-Display Utilities 8-25

Mass Storage Extended Subsystem (MSE) K Display

- MSE SUBSYSTEM -

- OPEN STAGE R T R Q * 5 OPEN DESTAGE -

1 . - # — —-Message ordinal

2. *Number of

stage requests

Malting to

become active

VALID COMMANDS within SSEXEC

K.X.GO. CLEARS ORDINAL *X*

K.STAGE OR K.DESTAGE TOGGLES OPEN/CLOSE

K.SMI SELECTS SM • I* INFORMATION

K.SMI.DRD°ST"X SETS MAX »X* STAGE DRDS

K.SMI,DRD=DS=Y SETS MAX •¥• DESTAGE DRDS

A^S

Figure 8-9. MSE K Display (Message Format)

- MSE SUBSYSTEM -

- OPEN STAGE R T R Q = 0 OPEN DESTAGE -

SMI STATUS ON

X MAX DRDS FOR STAGING

Y MAX DRDS FOR DESTAGING

JSN FILE UI PRU-S

DRDO STAGE - WXYZ FILENAM USRIDX PRUCNT

DRD1 DESTAGE - ABCD FILENAM USRIDX PRUCNT

VALID COMMANDS

K.X.GO. CLEARS ORDINAL *X*

K.STAGE OR K.DESTAGE TOGGLES OPEN/CLOSE

K.SMI SELECTS SM *I* INFORMATION

K.SMI,DRD=ST=X SETS MAX *X* STAGE DRDS

K.SMI,DRD=DS=Y SETS MAX *Y* DESTAGE DRDS

Figure 8-10. MSE K Display (Storage Module Format)

8-26 NOS Version 2 Analysis Handbook Revision M

Network Access Method (NAM) K Display

The NAM K display provides a common, centralized interface by which network

applications (CS, NS, NVF, or NETOU) can communicate with the host operator (HOP).

Figure 8-11 shows the format of the NAM K display. Table 8-6 explains the ﬁelds

that appear on the NAM K display.

message

data area

Alert Line ALERTS appl app2 app3 app4 app5

HOP Message Line HOP hop message 1i ne

Response Line appname operator entry repeat message

Figure 8-11. NAM K-Display Format

Table 8-6. NAM K-Display Fields

Field Description

data area

message

The portion of the screen that receives the output or diagnostic

message as a result of a command that you entered. This portion of

the NAM K display is operated as a scrollable paged device. Each

line of data enters the display at the bottom of the data area and

forces the previous lines to shift up by one line. The previous top

line is lost. However, whenever you turn the page-wait status on

and more than a screen full of information is written to the data

area, the prompt MORE DATA., is displayed on the bottom line of

the data area. The display remains ﬁxed until you enter the +

character to display the next page of information (refer to Display

Control Characters later in this section).

The system prompt READY., indicates that you can make

additional keyboard entries. The system prompt MORE DATA.,

indicates that you can enter + to see more data.

appl...app5 The alert line is a list of applications requesting your attention,

hop message line This ﬁeld contains a message received from another host,

appname

operator entry

repeat message

The name of the application (CS, NS, NAM, NVF, or NETOU) with

which you are currently interacting.

This ﬁeld contains up to 40 characters of the last command that

you entered. Commands of more than 40 characters are truncated.

Whenever a command cannot be accepted because the system was

not done processing a previous command, you get the system

prompt REPEAT., in this ﬁeld. Repeat the command.

Revision M K-Display Utilities 8-27

Network Access Method (NAM) K Display

NAM supports both the right and left screen K displays. The right K display provides

help information. "**%.

The NAM K display is available at all times during NAM execution, the message

REQUEST K DISPLAY appears on the B,0 display when the K display is not

assigned to NAM and some application has requested operator intervention. You must

assign the K display to NAM to interact with NAM or an application.

You assign the K display to NAM by entering:

K.NAM.

Figure 8-12 shows the NAM K display as it appears when you initially assign it.

READY..

ALERTS

NAM

Figure 8-12. NAM K Display

The NAM K display has two operating modes:

• NAM mode.

• Application mode.

Initially, the display is in NAM mode. The appearance of NAM indicates that you are

interacting directly with NAM.

8-28 NOS Version 2 Analysis Handbook Revision M

Display Control Characters

You can use the following four characters to control the NAM K display. They are

valid in both NAM mode and application mode. Each of the following characters must

be entered in the ﬁrst character position after K.

Character Description

When in application mode, the * character causes the display to revert

to NAM mode. The last command is aborted and any information

generated by the application, after the asterisk is entered, is discarded

by NAM. The page-wait status is automatically turned on. The *

character is ignored if the display is already in NAM mode.

When in application mode, the / character aborts the last command

without altering the assignment of the NAM K display. The / character

always results in a READY., prompt.

When in NAM mode, the / character is ignored.

When in application mode, the + character turns the page-wait status

on, if the current page-wait status is off. If the current page-wait status

is on, then the command K.+ displays the next available page of

information. The page-wait status is always on in NAM mode. The

page-wait status is initialized to off each time you enter application

mode. Therefore, whenever you select an application, the page-wait

status automatically changes from on to off and you can set it as

desired.

The - character turns off the page-wait status. The - character is

ignored if entered in application mode while the page-wait status is off.

Any attempt to turn the page-wait status off while in NAM mode is

ignored without sending any diagnostic message.

Revision M K-Display Utilities 8-29

NAM K-Display Operation

Initially, the display is in NAM mode. The application mode is invoked by a NAM

mode command that indicates the application with which you wish to interact. The

application mode is indicated by the appearance of the name of that application in the

appname portion of the response line (refer to ﬁgure 8-10). In application mode, all

entries except the asterisk are passed by NAM to a previously designated application.

An asterisk character returns the display to NAM mode.

Entering any command (processed either by NAM or by an application) immediately

causes the command to appear in the operator entry ﬁeld. When processing of the

command completes, the prompt READY., appears in the bottom line of the data

area. If you key something in, other than one of the display control characters *, /,

+ , or - before the prompt READY., appears, the character is displayed in your entry

ﬁeld but the command is not acted upon and the message REPEAT., appears to the

right of your entry.

Table 8-7 describes the commands for assigning the K display to one of the

applications CS, NS, NVF, or NETOU, and other frequently used commands. If

appname or the remote host number are invalid or the application is not currently

active, you will receive a diagnostic message. These commands are available in NAM

mode only.

Table 8-7. NAM Mode Commands

Command Description

K.AP = appname

K.AP

K.HELP

K.IG=appname

Assigns the NAM K display to the speciﬁed application. The

application name appears in the appname portion of the

response line (refer to ﬁgure 8-10). appname is one of the

supervisory programs CS, NS, NVF, or NETOU. The

command satisﬁes any alert request posted by the application.

When you type the application name, that application name

is removed from the alert line, the page-wait status is turned

off, and the last page of the application's recent history buffer

appears in the data area of the display (refer to Recent

History Command in section 15, Network Operations.

Assigns the NAM K display to the application whose name

appears in the leftmost position of the alert line. The

page-wait status is turned off and the last page of the

application's recent history buffer appears in the data area of

the display.

Displays information about the NAM mode commands on the

right K display.

Causes NAM to ignore all alert requests from the speciﬁed

application. If the application had an alert request pending,

its name is removed from the alert list. NAM informs the

application that its alert request was acknowledged and

ignored. The rest of the display remains unaltered. This

command is cancelled automatically when you enter the

AP = appname command. The assignment of the NAM K

display is not altered by the IG=appname command.

(Continued)

.A^m\

8-30 NOS Version 2 Analysis Handbook Revision M

NAM K-Display Operation

Table 8-7. NAM Mode Commands (Continued)

Command

K.IG

K.END

K.SEND,rrr,string.

Description

Causes NAM to ignore all alert requests from the application

whose name appears in the leftmost position of the alert line.

NAM informs the application that its alert request was

acknowledged and ignored. The rest of the display remains

unaltered. This command is cancelled automatically when you

enter the AP=appname command.

Clears the NAM K display. It causes the data area of the

display to go blank but leaves the alert line and response

line unaltered. The END command is used to stop the

generation of the status display (ﬁgure 8-12). Otherwise, if

the K display is dropped, NAM automatically ends after 60

to 90 seconds.

Causes NAM to send the message text if a logical link exists

between the speciﬁed remote host and this host. This

command applies only to hosts that are connected using 255x

NPUs, since no logical link exists between hosts that are

connected through either a public data network or a

CDCNET network. Use the NAM ST command to determine

which hosts are connected.

Parameter Description

rrr Host node number of the remote host (from 1

to 255).

0*^S

string Message text of no more than 40 characters.

The following message is displayed on the speciﬁed remote

host's NAM K display after the alert line.

hh.mm.ss. FROM sss string

Variable Description

hh.mm.ss

sss

string

Time the message is received.

Host node number of the sending host (from 1

to 255).

Message text of no more than 40 characters.

K.ST Causes the NAM status display to appear on the K-display

screen.

Revision M K-Display Utilities 8-31

NAM Mode Maintenance Commands

While in NAM mode, the HOP can enter the following network maintenance

commands. The commands are sent to NAM or to applications to determine the status

of the network and aid in the debugging of network problems. The HOP may select

NAM or a single network supervisory application or NAM and all network supervisory

applications using the following commands:

Command Description

DB2 Activates the online debugging code.

DE2 Deactivates the online debugging code.

DU2 Causes NAM to dump its ﬁeld length and/or request an application to

dump its ﬁeld length.

FL Changes NAM's maximum ﬁeld length.

LB2 Begins the logging of network trafﬁc in the debug log ﬁle.

LE2 Ends the logging of network trafﬁc in the debug log ﬁle.

LR Releases the debug log ﬁle.

RS Dumps the statistics data to a permanent ﬁle, resets the statistics

counters to 0 (zero), and continues gathering statistics.

ST Causes the NAM status display to appear on the K-display screen (refer

to ﬁgure 8-13).

DB Command

The debug begin (DB) command causes NAM to turn on its online debugging code

and/or request an application to turn on its online debugging code.

The command format is:

DB=mode.

m o d e D e s c r i p t i o n

ALL NAM turns on its online debugging code and requests all applications to

turn on their online debugging code.

appname NAM requests the speciﬁed application to turn on its online debugging

code. The speciﬁed application must be accessing the network or you will

receive a diagnostic message.

NAM NAM turns on its online debugging code.

2. These commands are supported by the network supervisory applications CS, NS, and NVF. They are

ignored by IAF, TAF, and many other network applications.

8-32 NOS Version 2 Analysis Handbook Revision M

/*'^%k

DE Command

The debug end (DE) command causes NAM to turn off its online debugging code and/or

request an application to turn off its online debugging code.

The command format is:

DE=mode.

m o d e D e s c r i p t i o n

ALL NAM turns off its online debugging code and requests all applications to

turn off their online debugging code.

appname NAM requests the speciﬁed application to turn off its online debugging

code. The speciﬁed application must be accessing the network or you will

receive a diagnostic message.

NAM NAM turns off its online debugging code.

DU Command

The dump (DU) command causes NAM to dump its ﬁeld length and/or request an

application to dump its ﬁeld length.

The command format is:

DU=mode.

m o d e D e s c r i p t i o n

ALL NAM dumps its ﬁeld length and requests all applications to dump their

ﬁeld length.

appname NAM requests the speciﬁed application to dump its ﬁeld length. The

speciﬁed application must be accessing the network or you will receive a

diagnostic message.

NAM NAM dumps its ﬁeld length.

Revision M K-Display Utilities 8-33

FL Command

The ﬁeld length (FL) command changes NAM's maximum ﬁeld length to the speciﬁed

value. If the ﬁeld length speciﬁed is less than NAM's current maximum ﬁeld length or

NAM's running ﬁeld length plus 10008, you will receive a diagnostic message.

If NAM reaches its maximum ﬁeld length, use the FL command to increase NAM's

maximum ﬁeld length. If the K display is in application mode when NAM reaches its

maximum ﬁeld length, NAM automatically switches the K display to the NAM mode.

Until you enter the FL command, all other NAM commands are rejected and NAM

stops servicing the network. Frequent need for this command indicates that the initial

ﬁeld length for NAM is inadequate and should be increased by changing the MAXFL

parameter of the NIP command described in the NOS Version 2 Installation

Handbook.

The command format is:

FL=ﬂ.

Parameter Description ^

ﬂ Speciﬁed maximum ﬁeld length in octal, ﬂ must not exceed 3600008. If

it does, NAM issues a diagnostic message to the K display.

LB Command

The log begin (LB) command causes NAM to begin logging network trafﬁc in its debug

log ﬁle and/or request an application to begin logging network trafﬁc in its debug log

ﬁle (refer to the NAM Version 1 Host Application Programming Reference Manual for

information on creating the application's log ﬁle).

The command format is:

LB-mode.

mode Description ^

ALL NAM begins logging network trafﬁc in its debug log ﬁle and requests all

applications to begin logging network trafﬁc in their debug log ﬁles.

appname NAM requests the speciﬁed application to begin logging network trafﬁc in

its debug log ﬁle. The speciﬁed application must be accessing the network

or you will receive a diagnostic message.

NAM NAM begins logging network trafﬁc in its debug log ﬁle. NAM must be

installed with the network trace option or this command is ignored. Refer

to the NOS Version 2 Installation Handbook for more information.

To release the debug log ﬁle, enter the log release (LR) command.

y*^^?\

<**^

''**%.

8-34 NOS Version 2 Analysis Handbook Revision M

LE Command

The log end (LE) command causes NAM to end logging network trafﬁc in its debug log

ﬁle and/or request an application to end logging network trafﬁc in its debug log ﬁle.

The command format is:

LE=mode.

mode Description

ALL NjAJVI ends logging network trafﬁc in its debug log ﬁle and requests all

applications to end logging network trafﬁc in their debug log ﬁles.

appname NAM requests the speciﬁed application to end logging network trafﬁc in

its debug log ﬁle. The speciﬁed application must be accessing the network

or you will receive a diagnostic message.

NAM NAM ends logging network trafﬁc in its debug log ﬁle.

To release the debug log ﬁle, enter the log release (LR) command.

LR Command

The log release (LR) command causes NAM to release its debug log ﬁle (if one exists)

and/or request an application to release its debug log ﬁle. When the local debug log

ﬁle is released, its contents are copied to a permanent ﬁle. Logging continues on a new

local debug log ﬁle.

The command format is:

LR=mode.

m o d e D e s c r i p t i o n

ALL NAM releases its debug log ﬁle and/or requests all applications to release

their debug log ﬁles.

appname NAM requests the speciﬁed application to release its log debug ﬁle. The

speciﬁed application must be accessing the network or you will receive a

diagnostic message.

NAM NAM releases its debug log ﬁle.

Revision M K-Display Utilities 8-35

RS Command

The reset statistics (RS) command causes NAM to dump its statistics to a permanent

ﬁle and restart gathering statistics and/or request an application to dump its statistics

to a permanent ﬁle and restart gathering statistics (refer to the NAM Version 1 Host

Application Programming Reference Manual for descriptions of the network statistics).

You can use this command to help analyze NAM and/or application performance.

The command format is:

RS=mode.

mode Description

ALL NAM dumps its statistics to a permanent ﬁle, resets the statistics

counters to 0 (zero), and continues gathering statistics and/or requests all

applications to dump their statistics to permanent ﬁles, reset their

statistics counters to 0 (zero), and continue gathering statistics.

a p p n a m e N A M r e q u e s t s t h e sp e c i ﬁ e d a p p l i c a t i o n t o d u m p i t s s t a t i s t i c s t o a ]

permanent ﬁle, reset its statistics counters to 0 (zero), and continue

gathering statistics. The speciﬁed application must be accessing the

network or you will receive a diagnostic message.

NAM NAM dumps its statistics to a permanent ﬁle, resets the statistics

counters to 0 (zero), and continues gathering statistics.

8-36 NOS Version 2 Analysis Handbook Revision M

ST Command

The status (ST) command causes the NAM status display (ﬁgure 8-13) to appear on the

K-display screen.

The command format is:

S T.

The NAM status display provides status information regarding all applications,

couplers, and logical links currently active in the host. This display is refreshed

periodically to display the current conditions of the applications and logical links.

The ﬁrst line of the NAM status display contains the network invocation number

(NIN), host regulation level (REG LVL), the number of applications accessing the

network, and the maximum ﬁeld length of NAM.

0^ms NIN = 014 REG LVL = NO OF APPLS = 5 MAXFL = 100000

APP JSN STATUS N C N A C NSM NDM TIME UP

IAF IAF oooooo 30 07.54.35

RBF RBF oooooo 08.10.00

TAF TAF oooooo 10 08.25.46

TVF AABA 001000 08.04.00

NVF AAAM oooooo 09.09.45

EST EM HN NSM NHM NLM IVTSTAT PRUST NPUREJ

054 01 0000 46125

LOG- LINK HN T N H S T NCN AC NHDQ NLDQ TIME UP

01 0 1 0 S 4 4 08.01.30

Figure 8-13. NAM Status Display

Revision M K-Display Utilities 8-37

ST Command

The NAM status display shows the application status entries in the following format:

app

jsn

status

jsn s t a t u s 1 ncn ac nsm ndm timeup

Header Description

Name of the application. The application name may be one of the following

CDC-written applications or a site-written application.

app Description

CS Communications Supervisor.

IAF Interactive Facility.

I T F I n t e r a c t i v e Tr a n s f e r Fa c i l i t y.

MCS Message Control System.

N J F N e t w or k J o b E n tr y F a c i l i t y.

NETFS Network File Server.

NETLS Network Log Server.

NLTERM Network Log Terminator.

NETOU Network Operator Utility.

NS Network Supervisor.

NVF Network Validation Facility.

PLATO Plato-NAM Interface.

PSU Printer Support Utility.

PTF Permanent File Transfer Facility.

PTFS Permanent File Transfer Facility Server.

QTF Queued File Transfer Facility.

QTFS Queued File Transfer Facility Server.

R B F Re m o t e Ba t c h F a c i l i t y.

TAF Transaction Facility.

TCF Terminal Cluster Facility.

TLF Tie Line Facility.

TVF Terminal Veriﬁcation Facility.

VEIAF NOS/VE Interactive Facility.

Job sequence name of the executing job table entry.

Status of the application in octal. Each of the 18 bits (numbered left to

right) represents the following condition:

/J^tK

Bit Number Description

4-5

7-14

15-17

Force ﬂag.

Wait ﬂag.

Rollout ﬂag.

ON ﬂag.

Swap ﬂag.

IN ﬂag.

Reserved.

NVF response ﬂags.

8-38 NOS Version 2 Analysis Handbook Revision M

ST Command

^ m s

0§ms.

Header Description

ncn

nsm

ndm

Ignore alert ﬂag. If set to Y, NAM ignores an alert request from the

application. If set to N, NAM accepts an alert request from the application.

This ﬁeld is blank if the application is not allowed to use the NAM K

display.

Number of current connections to the application.

Reserved.

Number of asynchronous supervisory messages currently queued for the

application.

Number of data messages and synchronous supervisory messages queued for

the application.

timeup Time, in. the format hh.m.ss, that the application became active.

The NAM status display shows the coupler status for all NPU and MDI/MTI coupler

entries in the EST currently being serviced by NAM. These entries have the following

format:

nsmest em hn

Header Description

nhm nlm Ivtstat prust npurej

est Equipment status table ordinal of the front end NPU or MDI/MTI.

em Equipment mnemonic NP for the NPU or ND for the MDI/MTI.

hn Host node number of the coupler.

nsm Number of asynchronous supervisory messages currently queued for the

coupler.

nhm Reserved.

nlm Reserved.

ivtstat Number of characters transferred downline on interactive connections in

the last 30 seconds.

prust Number of characters transferred downline on PRU connections in the last

30 seconds.

npurej Number of times data was rejected by the NPU.

Revision M K-Display Utilities 8-39

ST Command

The NAM status display shows the logical link entries in the following format,

h n t n h n s t n c n a c n h d q n l d q t i m e u p

Header Description

hn Node number of the NPU or MDI/MTI coupler.

tn Terminal node number of the NPU or MDI/MTI on host-to-network logical

links, or host node number of the remote host on host-to-host logical links.

h Logical link regulation level as reported by the host.

n Regulation level as reported by the NPU or MDI/MTI. n can have the

following values:

n Description

0 Logical link is down or disabled for data.

1 Only asynchronous supervisory messages are allowed.

2 Only asynchronous supervisory messages and high priority connections

are allowed.

3 All network connections are allowed.

s Supervision indicator. If CS is using this logical link to send supervisory

messages, the ﬁeld is set to S; otherwise, it is blank.

t Host-to-host logical link indicator. If this logical link is between two hosts, ")

the ﬁeld is set to H; otherwise, it is blank.

ncn Number of current connections on the logical link.

a c R e s e r v e d .

nhdq Reserved.

nldq Reserved.

timeup Time, in the format hh.mm.ss., that the logical link became active.

8-40 NOS Version 2 Analysis Handbook Revision M

Queue File Transfer Facility (QTF) K Display

The QTF utility provides a multistreaming capability for transferring queued ﬁles

between mainframes using connections through RHF or NAM. QTF allows you to

specify which ﬁles are to be transferred. You do this by deﬁning a set of selection

classes. You can deﬁne or modify values for selection classes using the CLASS

command described later in this subsection. You can specify a range of ﬁle sizes and

the maximum number of simultaneous connections allowed. You can also specify a

number of reserved connections that cannot be used by lower-priority selection classes.

QTF Selection Classes

Selection classes are designated by a single letter in the range from A to L. Selection

class A is the highest priority, and L is the lowest priority.

QTF uses selection classes to ﬁnd ﬁles to assign to the PIDs (physical identiﬁers) for

remote hosts. Associated with each PID is one or more LIDs (logical identiﬁers). The

/^N system PID/LID table is created by system program CLDT from the LID conﬁguration

ﬁle LIDCMid (refer to LID Conﬁguration File in section 10) and is viewed and

modiﬁed using the L-display utility LIDOU (refer to LIDOU L Display in section 9).

QTF reads the system PID/LID table to determine which PIDs have connections

through RHF or NAM and which LIDs are associated with each PID.

File Selection Process

Each copy of QTF has a ﬁxed number of connection slots available. The goal of the

QTF ﬁle. scheduling algorithm is to ﬁnd ﬁles for as many of these connection slots as

possible within the constraints of the selection class deﬁnitions. On each pass through

its ﬁle selection process, QTF proceeds through all deﬁned selection classes from the

highest priority A to the lowest priority L. For each selection class, QTF attempts to

ﬁnd one ﬁle for each available PID that does not already have a ﬁle assigned. The ﬁle

selection process continues until one of the following occurs:

• All connection slots are in use, or

• All PID/selection class combinations have been checked and

- Are in use, or

- Are unavailable (disabled or the maximum number of connections per selection

class has been reached), or

- No ﬁles were found.

• The sum of the reserved connections for higher-priority selection classes is equal

to or exceeds the number of free connection slots.

Revision M K-Display Utilities 8-41

Visualizing the File Selection Process

A ^ ^ S

Visualizing tbe File Selection Process

To visualize the ﬁle selection process, imagine that selection classes and PIDs form a

two-dimensional matrix of jacks on a telephone switchboard with PIDs assigned across

the columns and selection classes assigned down the rows. Above each jack is a light.

When a light is ﬂashing, a queued ﬁle is available for the corresponding PID/selection

class combination. The plug cables represent the connection slots. A jack can have only

one plug inserted at a time. Inserting a plug into a jack establishes a connection and

causes the light to stop ﬂashing and a ﬁle transfer to begin. A plug is removed from a

jack after one or more ﬁles have been transferred or a connection error occurs.

Figure 8-14 shows the connection slot array (plugs) and the PID/selection class matrix

(switchboard).

Connecti on

S l o t s

1 2 3 4

XXXX

SeLecti on

Class

PID

• s a PID

A o 0 0 0

B0 0 0 0

C0 0 0 0

L0 0 o0

Figure 8-14. Connection Slot Array and PID/Selection Class Matrix

On each ﬁle selection cycle, QTF looks for a free plug cable (an unused connection

slot) and, starting at the upper left corner of the PID/selection class switchboard,

searches across each row in turn looking for a ﬂashing light (queued ﬁle matching the

PID/selection class combination). When the plug is inserted into the jack, that

connection is established. If another plug cable is free, the search continues from the

current jack until QTF runs out of plugs or the bottom right corner of the switchboard

is reached.

8-42 NOS Version 2 Analysis Handbook Revision M

Visualizing the File Selection Process

For each selection class row for which a reserved number of connections is deﬁned,

QTF checks the number of plugs inserted in that row when the rightmost jack for the

row is reached. If the number of plugs inserted is less than the number of reserved

connections, QTF places enough plugs in unused jacks in that row until the number of

reserved jacks is ﬁlled, only one free plug remains, or the row is ﬁlled; whichever

occurs ﬁrst.

For each selection class row for which a maximum number of connections is deﬁned

and the maximum is reached, QTF continues searching at the leftmost jack of the

next row down.

You can minimize the number of different operating system calls QTF must make to

cover all of the switchboard jacks. Do this by either minimizing the number of

selection classes (number of rows) or by deﬁning overlapping selection criteria for

selection classes. For example, if selection classes A, B, and C all have the same ﬁle

size index range, and QTF does not ﬁnd a ﬁle for a PID in selection class row A,

QTF can consider selection class rows B and C checked as well. If selection class D

has a different, nonoverlapping ﬁle size index range, QTF must call the operating

system to check the jacks in selection class D regardless of the results of the checks

for selection classes A, B, and C (assuming that the ﬁle selection process has not yet

terminated).

Revision M K-Display Utilities 8-43

Queued File Size Index

QTF ﬁle size selection criteria are speciﬁed in terms of a queued ﬁle size index. The

NOSTEXT symbols FSI1 through FSI6, deﬁned in NOS common deck PPCOM (refer to

the NOS Version 2 Installation Handbook), allow you to completely and uniquely

partition the range of queued ﬁle sizes into seven subranges. Each subrange is

identiﬁed by an integer value in the range from 1 to 7, known as a ﬁle size index.

The FSIn values deﬁne end points for each ﬁle size subrange in units of PRUs (640

6-bit characters).

FSI File size subrange

1 0 ^ size < FSI1

2 FSI1 ^ size < FSI2

3 FSI2 ^ size < FSI3

4 FSI3 ^ size < FSI4

5 FSI4 as size < FSI5

6 FSI5 ^ size < FSI6

7 F S I 6 ^ s i z e

Each NOS queued ﬁle is assigned a ﬁle size index value based on its size. This value

is displayed on the DSD Q displays (refer to the NOS Version 2 Operations Handbook).

You can examine the NOS common deck PPCOM to determine the default values for

the FSIn symbols. However, to modify the current values, you need to rebuild

NOSTEXT, the NOS operating system, and QTF under RHP. For each FSIn symbol,

the value of FSIn must be less than the value of FSIn+1.

The ﬁle size ranges are displayed on the QTF help display (ﬁgure 8-15). A-m^

Queued File Size and Transfer Times

Given a typical set of ﬁle size index values, the following shows the approximate ﬁle

transfer time in minutes for various line speeds. The time is based on typical line

utilization and only one transfer active through the line. NAM line speed is shown in

minutes.

FUe Size 1200 9600 19.2k 56k RHF/

FSI (PRUs) bps bps bps bps LCN

1319 21 1.5 <1.0 <1.0 <0.5

2511 34 4.2 2.5 <1.0 <0.5

31023 68 8.5 4.2 1.5 <0.5

42047 136 17.0 8.5 3.0 <0.5

54095 272 34.0 17.0 6.0 <0.5

68191 544 68.0 34.0 12.0 <0.5

999999 >544 >68.0 >34.0 >12.0 >0.5

8-44 NOS Version 2 Analysis Handbook Revision M

0$ms

QTF K-Display Usage

The following procedures describe DSD K-display usage for QTF operations. QTF is

used by both the RHF and NAM subsystems.

QTF Operation Under RHF

Normally, QTF is deﬁned as an autostart application in the RHF conﬁguration ﬁle

(refer to RHF Conﬁguration Files in section 10). That means a QTF job starts

automatically when the RHF subsystem is initiated. The RHF application table K

display shows how many copies of QTF are deﬁned and how many copies are currently

active.

QTF terminates when the K.IDLE command terminates the RHF subsystem. To

manually terminate a copy of QTF, bring up the RHF application table display by

entering the following commands:

K.RHF.

K.APPL.

When the display appears, enter:

DISABLE,ord.

where ord is the ordinal of an active QTF in the RHF application table display. The

message QTF, NETWORK IDLEDOWN IN PROGRESS should appear in the QTF job

dayﬁle. QTF completes any ﬁles in progress and then ends. If QTF does not terminate

within a reasonable amount of time, use the DSD DROP command to terminate QTF.

If a fault condition occurs, QTF normally restarts itself. To manually restart QTF,

bring up the RHF application table display by entering the following commands:

K.RHF.

K.APPL.

When the display appears, enter:

ENABLE,ord.

where ord is the QTF ordinal in the RHF application table display.

You can determine the jsn of QTF by using the DSD system status display (B,0) and

the RHF application table display. When QTF is active at a control point, it is

indicated on the B,0 display.

Activate the K display for that QTF by entering:

K.jsn.

where jsn is the job sequence name for QTF. The QTF transfer display appears on the

left screen.

Revision M K-Display Utilities 8-45

QTF Operation Under NAM

Normally, QTF is started by NAMI during NAM subsystem initiation. The NAM status

K display shows how many copies of QTF are currently active.

QTF terminates automatically when the NAM subsystem is terminated. To manually

terminate a copy of QTF, use the NVF IDLE or DISABLE commands (refer to NVF

Control Commands in section 15) to either idle QTF or immediately terminate QTF.

Enter the following commands:

K.NAM.

K . * .

K.AP=NVF.

K.IDLE,AP=QTF. or K.DISABLE,AP<=QTF.

The message QTF, NETWORK IDLEDOWN IN PROGRESS or QTF, NETWORK

SHUTDOWN should appear in the QTF job dayﬁle. If the IDLE,AP=QTF command is

used, QTF completes any ﬁles in progress and then ends. If QTF does not terminate

within a reasonable amount of time, use the DISABLE,AP=QTF command or DSD

DROP command to terminate QTF.

If a fault condition occurs, QTF normally restarts itself. To manually restart QTF,

enter the following command (refer to the NAMI command in section 16):

X.NAMI(RS=QT)

You can determine the jsn of QTF by using the NAM status K display. Bring up the

NAM status K display by entering the following commands:

K.NAM.

K . * .

K.ST.

Activate the K display for that QTF by entering the following command:

K.jsn.

where jsn is the job sequence name for QTF. The QTF transfer display appears on the

left screen.

,-/^^.

.A^m^

8-46 NOS Version 2 Analysis Handbook Revision M

QTF K Display Commands

QTF commands have the same general format whether they are entered through the K

display or read from an input ﬁle using the INCLUDE command.

• Each QTF command must be on one line and only the first 80 columns of a

command line are signiﬁcant.

• A command consists of a command verb followed by 0 to 10 parameters.

• A command verb and parameters are separated by one or more spaces or a

comma.

• A command is terminated by a single period.

• Most command verbs and parameters can be abbreviated. The exceptions are IDLE

and STOP.

QTF accepts the following K display commands:

Command Description

CLASS Deﬁne or modify QTF selection class values.

DISABLE Change the status of QTF elements from enabled to disabled.

ENABLE Change the status of QTF elements from disabled to enabled.

HELP Displays information about the available QTF commands.

IDLE Begins the idle-down process of QTF.

INCLUDE Causes QTF commands to be read from an input ﬁle.

SCHED Modify QTF scheduling parameters.

STATUS Displays information about the status of ﬁle transfers in progress, PIDs,

or QTF selection classes.

STOP Causes an immediate termination of QTF.

+ Page forward to the next screen of a multipage display.

Page backward to the previous screen of a multipage display.

Comment line (anything following a period is ignored).

/$pn

Revision M K-Display Utilities 8-47

CLASS Command

The CLASS command allows you to deﬁne or modify QTF selection class values.

The command format is:

K.CLASS,SC=selclass,FSI=fs1,MAXIMUM=max,RESERVE=res.

Parameter Description

selclass Selection class name; specify one character in the range from A to L.

Selection class A is the highest priority class and L is the lowest

priority class. This parameter is required.

File size index; specify the ﬁle size index or range of indexes using

one of the following formats:

fsi

Format Description

fsi File size index in the range from 1 to 7.

fsil..fsiu , File size index range, where fsil ^ fsiu.

* An * speciﬁes all ﬁle sizes. It is equivalent to entering

1..7. If FSI is omitted, the default is FSI=*.

max Maximum connections; specify the maximum number of simultaneous

connections for ﬁles in this selection class in the range from 0 to the

maximum available connections (speciﬁed by the MAXCONS parameter

in the QTF SCHED command) or specify *. If 0 or * is speciﬁed, all

connections can be used for ﬁles in this selection class. However, only

one connection per PID per selection class is permitted. Default

is 0 or *.

res Reserved connections; specify the number of connections reserved for

ﬁles in this selection class in the range from 0 to the maximum

available connections (speciﬁed by the MAXCONS parameter in the

QTF SCHED command). If zero is speciﬁed, no connections are

reserved for this selection class. For each selection class, the number

of available connections is reduced by the sum of reserved connections

for all higher-priority selection classes. To prevent a possible

overcommitment of connections, QTF allows only mxcons-1 connections

to be reserved. Default is zero.

The following default selection class is deﬁned at QTF initiation:

CLASS,SC=A,FSI»•,MAXIMUM**,RESERVE=0.

which allows one ﬁle transfer to any one remote host with no restrictions on ﬁle size

and no reserved connections. To modify the default selection class or deﬁne other

classes, use additional CLASS commands.

z«*^^.

8-48 NOS Version 2 Analysis Handbook Revision M

DISABLE Command

The DISABLE command changes the status of the selected QTF elements from enabled

to disabled.

The command format is:

K.DISABLE,element.

Parameter Description

element One of the following:

PID=pid

SC = selclass

SC = selclass,PID=pid

Option Description

pid Physical identiﬁer of a given mainframe. Specify PID = * to

disable all PIDs.

selclass Selection class in the range from A to L. Specify SC = * to

disable all selection classes.

ENABLE Command

The ENABLE command changes the status of the selected QTF elements from disabled

to enabled.

The command format is:

K.ENABLE,element.

Parameter Description

element One of the following:

PID=pid

SC = selclass

SC = selclass,PID=pid

Option Description

pid Physical identiﬁer of a given mainframe. Specify PID=* to

enable all PIDs.

selclass Selection class in the range from A to L. Specify SC = * to

enable all selection classes.

Revision M K-Display Utilities 8-49

HELP Command

The HELP command displays the available QTF K display commands, ﬁle size ranges

in decimal PRUs, and the parameters for the QTF SCHED command.

The command format is:

K.HELP.

The HELP K display (ﬁgure 8-15) appears on the right screen.

HELP PAGE 1 OF 1

COMMANDS FSI SIZE 1RANGE

CLASS,SC=A,FSI=N..N, MAXIMUMS, RESERVED. 0. 319

DISABLE,ELEMENT. 320. 511

ENABLE.ELEMENT. 512. 1023

IDLE. 1024. . 20 47

INCLUDE,FILE=PFN. 2048. . 40 95

SCHED,PARAMETERS. 4096. . 8191

STATUS,TYPE. (TYPE=PID,SC,OR TRANSFER) 8192. .999999

*STOP.

+ OR PARAMETERS = (FILE SIZE RANGES

SELSECS=N IN DECIMAL PRUS)

ELEMENT = ONE OF- MAXCONS=N

SC=X OR * TIMEOUT=N

PID=XXX OR * DISABLED

SC=X,PID=XXX RETRY=N

X=A,B,C L. LDTRFSH»N

Figure 8-15. QTF HELP K Display (Right Screen)

IDLE Command

The IDLE command causes QTF to idle down and terminate. The message QTF,

NETWORK IDLEDOWN IN PROGRESS appears in the QTF job dayﬁle and

IDLEDOWN appears on the B,0 display. QTF completes any ﬁle transfers in progress

and then ends.

The command format is:

K.IDLE.

QTF does not accept an abbreviation for this command.

To restart QTF, enter the following command:

X.NAMI(RS=QT)

•^585\

8-50 NOS Version 2 Analysis Handbook Revision M

INCLUDE Command

The INCLUDE command causes QTF to rewind the speciﬁed input ﬁle and begin

processing QTF commands from that ﬁle.

The command format is:

K.INCLUDE,FILE=inﬁle.

Parameter Description

inﬁle File name from which QTF commands are to be processed. QTF

searches for a local ﬁle with name inﬁle. If the local ﬁle is not found,

QTF attempts to GET or ATTACH permanent ﬁle inﬁle from user

name SYSTEMX. If QTF performs a GET or ATTACH, it returns the

ﬁle after processing all of the commands. This parameter is required.

A ﬁle of QTF commands processed by using the INCLUDE command can itself contain

another INCLUDE command. Upon executing the second INCLUDE command, QTF

switches to the new ﬁle and does not read anything following the INCLUDE command

in the old ﬁle.

Revision M K-Display Utilities 8-51

SCHED Command

The SCHED command allows you to modify QTF scheduling parameters.

The command format is:

K.SCHED,SELSECS=se1 sees,MAXCONS=mxcons,TIME0UT=tmom1 ns,

DISABLE=d1sm1ns,RETRY=maxrtry,LDTRFSH=1dtm1ns.

Parameter Description

selsecs File selection recall period in the range from 1 to 4095 seconds;

speciﬁes the minimum amount of time between attempts to acquire

new queued ﬁles. Whenever QTF is unable to ﬁll all connection slots

with ﬁles, it waits selsecs seconds before again attempting to acquire

ﬁles for unused slots. Default is 20 seconds.

mxcons Maximum simultaneous connection slots (to all hosts) in the range

from 1 to 4. Default is 4 connections.

tmomins Timeout period in the range from 1 to 60 minutes; speciﬁes the

maximum amount of time QTF waits for a reply from a remote host or

access method. For the RHF variant of QTF, tmomins also speciﬁes

how long QTF waits for NAD code conversion resources when

applicable. Default is 2 minutes.

dismins Disable period in the range from 1 to 60 minutes; speciﬁes the amount

of time for which QTF will not attempt transfers to a PID after an

error occurs. You can use the ENABLE command to clear the error

disabled condition. Default is 2 minutes.

maxrtry Maximum retries in the range from 1 to 50 retries; speciﬁes the

maximum number of immediate retries that QTF attempts to complete

a ﬁle transfer before error disabling the PID. When QTF cannot

complete a connection request (the connection is rejected by the local

or remote access method or QTFS), QTF disables the PID without any

immediate retrys. Once the PID is disabled, QTF returns the ﬁle to

the local queue and does not attempt to acquire any ﬁles for LIDs

associated with the PID until either dismins minutes have elapsed or

you use the ENABLE,PID command to enable the PID. Default is 2

retries.

ldtmins PID/LID table refresh period in the range from 1 to 60 minutes;

speciﬁes how often QTF updates its internal tables from the system

PID/LID table. Default is 2 minutes.

.-^%.

8-52 NOS Version 2 Analysis Handbook Revision M

STATUS Command

The STATUS command displays the status of selected QTF elements or ﬁle transfers.

The command format is:

K.STATUS,display.

display Description

PID Selects the QTF physical identiﬁer K display which shows the status

of PIDs, selection classes for each PID, and the number of queued ﬁles

transferred.

SC Selects the QTF selection class K display which shows the selection

class deﬁnitions and the number of ﬁles transferred.

TRANSFER Selects the QTF transfer K display which shows transfers in progress

and the number of ﬁles already transferred.

Each QTF status display consists of a two-line display header and a variable number of

display items. The lower portion of the screen contains a QTF command and message

area. Figure 8-16 shows the general format for the QTF status displays. The display

ﬁelds are described following the ﬁgure.

STATUS,type nnnnnn FILES SENT xxx PAGE n OF m

display header

display item 1

display item 2

display item k

COMMAND- < last cmd (1-50) >

Idledown < last cmd (51-80) >

< o p r m s g >

Figure 8-16. General Format of QTF Status K Display

Revision M K-Display Utilities 8-53

STATUS Command

Field Description

type Display type; PID, SC, or TRANSFER.

nnnnnn Number of queued ﬁles transferred.

xxx Access method; RHF or NAM.

n Current page number.

m Maximum page number.

last cmd The last QTF command processed. If the command is incorrect, an asterisk

(*) appears in the line above the command to show the approximate

position of the character or parameter in error. If the error is in the

second part of the command (positions 51 to 80), a plus sign (+) appears

instead of an asterisk.

idledown This ﬁeld contains the message *IDLEDOWN* if QTF is in idle-down

mode; otherwise, this ﬁeld is blank.

opr msg This ﬁeld contains an operator message about the last QTF command

processed. Refer to the NOS Version 2 Operations Handbook for message

descriptions.

,,>=^S\

8-54 NOS Version 2 Analysis Handbook Revision M

STATUS Command

QTF Physical Identiﬁer K Display

Figure 8-17 shows a sample QTF STATUS.PID K display.

STATUS,PID nnnnnn FILES SENT RHF PAGE 1 OF 1

PID STATUS SC DISABLED SC IN-USE

M01 ENABLED

M02 DISABLED

M03 NO LIDS i\———————\_ —————

M04 NO PATH

M05 REJECTING

Figure 8-17. QTF STATUS.PID K Display

Each entry in the QTF physical identiﬁer display appears in the following format:

pid status sc disabled sc in-use

Header Description

pid

status

Physical identiﬁer of the remote host speciﬁed in the NOS system LID

table.

Status of the PID; status can be one of the following:

status Description

DISABLED

ENABLED

NO LIDS

NO PATH

The PID was disabled by the QTF DISjABLE

command. Use the QTF ENABLE command to change

the PID status to ENABLED.

The PID is available for ﬁle selection for enabled

selection classes.

No enabled LIDs are associated with this PID in the

system LID table or the PID is disabled. Use the

LIDOU L-display utility to display and modify the

system LID table.

The system LID table does not deﬁne a path in

service to this PID through the access method (RHF

or NAM) that is available to this copy of QTF. Use

the LIDOU L-display utility to determine whether an

RHF or NAM path to the PID is deﬁned and its

status. If the path status is OUT OF SERVICE, use

the RHF network path table display or the NAM

status display to display the status of the network

elements linking the two hosts.

Revision M K-Display Utilities 8-55

STATUS Command

Header Description

status (Continued)

status Description

REJECTING

sc disabled

sc m-use

QTF has temporarily disabled ﬁle transfers to the PID

because QTF was unable to establish a connection to

the PID or was unable to successfully transfer a ﬁle

after the maximum number of retries. QTF returns the

PID to ENABLED status after the speciﬁed disable

period has elapsed. (Refer to the QTF SCHED command

for descriptions of the RETRY and DISABLE

parameters.) You can use the QTF ENABLE command

to immediately return the PID to ENABLED status. If

the PID returns to REJECTING status, examine the

QTF dayﬁle to determine whether a connection reject

or a particular ﬁle is causing the problem.

Currently, for RHF paths and for some types of NAM

paths, the path status in the system LID table does not

reﬂect the status of the remote PID or access method

but rather the status of the local host network elements

used to access the PID. Thus, a remote host may be

out-of-service (due to scheduled maintenance, for

example), yet the local system LID table shows the

path status as IN SERVICE. Each time the disable

period elapses, QTF attempts to send ﬁles to the PID.

If the access method rejects the connection, QTF

returns the PID to REJECTING status.

Selection classes that are unavailable for this PID. They were disabled

by the QTF DISABLE command. Use the QTF ENABLE command to

enable a particular PID and selection class combination.

Selection classes currently in use for this PID. The QTF transfer display

shows the ordinals for connection slots and the selection class associated

with each slot.

8-56 NOS Version 2 Analysis Handbook Revision M

STATUS Command

QTF Selection Class K Display

Figure 8-18 shows a sample QTF STATUS,SC K display.

STATUS,SC nnnnnn FILES SENT NAM PAGE 1 OF 1

SC STATUS FSI RANGE MAXIMUM RESERVED IN-USE AVAILABLE

A ENABLED .. 6 1 2

B ENABLED .. 6 1 2

C DISABLED .. 7

D UNDEFINED

E UNDEFINED

F UNDEFINED

G UNDEFINED

H UNDEFINED

I UNDEFINED

J UNDEFINED

K UNDEFINED

L UNDEFINED

Figure 8-18. QTF STATUS,SC K Display

Each entry in the QTF selection class display appears in the following format:

sc status fsi range maximum reserved in-use available

Header Description

status

fsi range

maximum

reserved

Selection class.

Status of the selection class; status can be one of the following:

status Description

DISABLED The selection class was disabled by the QTF

DISABLE command. Use the QTF ENABLE command

to change the selection class status to ENABLED.

ENABLED The selection class is available for ﬁle selection for

enabled PIDs.

UNDEFINED The selection class has not been deﬁned. Use the QTF

CLASS command to deﬁne or modify a selection class.

File size index range. This ﬁeld shows the inclusive range of ﬁle size

index values that are allowed for this selection class.

Maximum number of connection slots allowed for ﬁles in this selection

class. This field contains dashes (—) if no maximum is specified.

Number of connection slots reserved for ﬁles in this selection class.

This field contains dashes (—) if no connection slots are reserved.

Revision M K-Display Utilities 8-57

STATUS Command

Header

m-use

Description

Number of connection slots that have been assigned ﬁles in this

selection class. This field contains dashes (—) if no connection slots are

in use.

available Number of free connection slots available for assigning ﬁles in this

selection class. This field contains dashes (—) if no connection slots are

available.

^■a^v

""^\

8-58 NOS Version 2 Analysis Handbook Revision M

STATUS Command

QTF Transfer K Display

Figure 8-19 shows a sample QTF STATUS,TRANSFER K display.

STATUS,TRANSFER nnnnnn FILES SENT RHF PAGE 1 OF 1

ACN SC FILE LID/PID DC DD PRUS STARTED STATUS

1 A ABCD SV6 M03 PR C8 1234 11.03.45 *START XFR

CONNECTION ESTABLISHED TO M03.

2 A AAED MFF MFF IN US 22 11.10.00 WAIT REM

ACQUIRED DC=IN, ST=MFF, D0=MA2.

00ms

Figure 8-19. QTF STATUS/TRANSFER K Display

Each entry in the QTF transfer display appears in the following format:

acn sc ﬁle lid/pid dc dd prus started status1 id/pid dc

message

Header Description

acn

ﬁle

lid

pid

Application connection number; this is the ordinal of the connection slot.

Selection class associated with this connection slot.

Job sequence name associated with the queued ﬁle assigned to this

connection slot.

Logical identiﬁer of the host associated with this connection slot.

Physical identiﬁer of the remote host to which this slot is connected or is

being connected.

Disposition code of the queued ﬁle. Refer to the ROUTE command in the

NOS Version 2 Reference Set, Volume 3 for a list of disposition codes

and their meanings.

Data declaration. This ﬁeld deﬁnes the format of the ﬁle data during

transfer. Values of C6 and C8 indicate that the ﬁle is to be sent in

code-conversion mode. Values of UH, US, and UU indicate that a binary

transfer is to be used.

Revision M K-Display Utilities 8-59

STOP Command

Header Description

prus

started

status

message

Size of the ﬁle in physical record units (PRUs) in decimal. One PRU

contains the equivalent of 640 6-bit characters.

This ﬁeld shows the time (hh.mm.ss) at which the connection to the remote

host was initiated. The time is updated when the connection* is established

and at the beginning of each ﬁle transfer sequence.

Status of the connection or ﬁle transfer; status can be one of the following:

status Description

ACQ FILE File has been acquired.

CONNECT Issuing connect request.

END CONN Connection terminating.

FILE XFR File transfer in progress.

IDLE Connection established.

QUEUED File transfer successful.

START XFR Initiating ﬁle transfer.

STOP XFR File transfer completing.

WAIT ACQ Waiting for acquire queued ﬁle response.

WAIT REM Waiting for response from remote host or access method.

This ﬁeld contains the last dayﬁle message issued by QTF for this

connection slot. Refer to the NOS Version 2 Operations Handbook for

message descriptions.

STOP Command

The STOP command causes QTF to terminate immediately. The message QTF,

NETWORK SHUTDOWN appears in the QTF job dayﬁle. QTF abandons any ﬁle

transfers in progress, returns ﬁles to the system I/O queues, and ends.

The command format is:

K.STOP.

QTF does not accept an abbreviation for this command.

8-60 NOS Version 2 Analysis Handbook Revision M

REDEFINE K Display

Use the REDEFINE utility to reconﬁgure 844 and 885-11/123 disk drives online and

thereby logically eliminate a unit that is malfunctioning without performing a level 0

initial deadstart. A single unit or multiunit device4 that fails can be replaced with an

unused unit. Unused units must be available or made available across channels or on

the same channel as the failing unit by physically moving the disk pack from the

failing unit to the replacement unit. Table 8-8 gives the equipment requirements for

reconﬁguration.

Table 8-8. Equipment Requirements for Reconﬁguration

Operation

To Be Special

Performed Current Equipment Replacement Equipment Considerations

Must be unloaded, Must be in Must be

removable EST unloaded,

removable

Add or Yes No Yes Equipment must

Return a have less than

Unit eight units.

Delete a Unit Must be removable;

if not unloaded, the

unit must be in the

EST

NA NA Equipment must

have at least

one unit.

Replace a No Not necessary Yes, if in The system does

Unit EST not perform

label veriﬁcation

on packs that

are unloaded

before the

failing unit is

replaced. If the

unloaded pack is

not moved from

the failing unit

or the wrong

pack is moved,

the system

issues an error

message when

the replacement

unit is accessed.

Recable a No NA No None.

Unit

3. When reconﬁguring an 885-11/12 disk drive, a customer engineer must be present.

4. When reconﬁguring a multiunit device, all units of the device must be on the same controller(s).

Revision M K-Display Utilities 8-61

REDEFINE K Display

NOTE

■'*^%k

There are special cases when you cannot perform an online reconﬁguration. In these

cases, you must perform a level 0 deadstart and deﬁne the failing device to another

drive. The special cases are:

• Reconfiguring an 885-11/12 disk drive that is the only system device defined.

• Reconfiguring an 844 or 885-11/12 disk drive that is the only temporary device

deﬁned.

• Reconfiguring a device designated as an independent shared device.

• Reconfiguring an 819 or 885-42 disk drive.

The reconﬁguration sequence follows:

1. Request the reconﬁguration of the disk device deﬁned by EST ordinal est by

entering:

REDEFINE,est.

The.message REQUEST *K* DISPLAY appears at the appropriate control point on

the system status display (B,0). You can reenter the REDEFINE command as

many times as there are devices to be reconﬁgured. Multispindle devices that

include two or more units are considered to be one device.

If an INITIALIZE command is being processed, the REDEFINE entry is not

processed until the initialization is complete.

2. Bring the initial REDEFINE K display (see ﬁgure 8-20) to the left console screen

by entering:

K.jsn.

jsn Description

jsn Job sequence name of the requesting job.

To bring up the right console screen K display, enter: /*^\

KK.

As shown in the left screen of ﬁgure 8-20, the current EST description of the

device being reconﬁgured appears under the heading CURRENT EQUIPMENT

CONFIGURATION. The EST description of the device as changes are made

appears under the heading REQUESTED EQUIPMENT CONFIGURATION. Any

devices listed under IDLED EQUIPMENTS are devices that have been previously

selected by your REDEFINE command but have yet to be processed in the

reconﬁguration run.

The left screen K display may not list all relevant devices. If more than one

device is listed, they are processed one at a time as they appear in the list with

one exception: all shared devices are processed prior to nonshared devices.

8-62 NOS Version 2 Analysis Handbook Revision M

REDEFINE K Display

You are guided through the reconﬁguration process in two ways. First, by the list

of commands and parameters shown on the right screen K display of ﬁgure 8-21,

and second, by the system responses and error messages.

yd^ptoSV,

CMS

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

LVLO - LVL7 MID=AA NOS version

MASS STORAGE FAILURE RECOVERY

CURRENT EQUIPMENT CONFIGURATION

011 DJ-1 CH24.01 UL=01

REQUESTED EQUIPMENT CONFIGURATION

011 DJ-1 CH24.01 UL=01

CHANGED UNITS

IDLED EQUIPMENTS

010 DJ-1 CH01.24 UL=00

ST = I

ST « I

Figure 8-20. REDEFINE K Display (Left Screen)

K DISPLAY COMMANDS

CLEAR CLEAR IDLE AND SUSPEND ON CURRENT EQUIPMENT.

END CLEAR ALL DEVICE IDLES AND END *CONFIG*.

GO INITIATE PROCESSING OF ENTERED CONFIGURATION.

RERUN RESTART *CONFIG* UTILITY PROCESSING.

RESET RESET CURRENT EQUIPMENT TO DEFAULT PARAMETERS.

SUSPEND SUSPEND SYSTEM OPERATION ON CURRENT EQUIPMENT.

CH=C1,C2 C1 AND OPTIONALLY C2-ARE NEW ACCESS CHANNELS.

EQ=EEE SET EST ORDINAL EEE TO BE PROCESSED.

UL=U1,U2,... SET UNIT LIST AS SPECIFIED.

UR=UU UNIT UU IS TO BE RECABLED WITH A NEW DRIVE.

Figure 8-21. REDEFINE K Display (Right Screen)

Revision M K-Display Utilities 8-63

REDEFINE K Display

3. Enter all of the valid parameters (refer to table 8-9) that deﬁne the

reconﬁguration characteristics for the speciﬁed device. Parameters are entered ^^v

singly. Parameters are processed left to right; no terminator is necessary. ?

If you enter an incorrect parameter (for example, the wrong channel number), the

error can be corrected by entering the correct parameter.

4. Enter

K.GO

to initiate processing of the parameters when all parameters and commands (refer

to table 8-10) have been entered for the speciﬁc device. Enter this command after

each set of parameters to signal the system to go ahead with the reconﬁguration

you have deﬁned. If more devices remain to be reconﬁgured, repeat steps 3 and 4.

5. Enter

K.END

to end a reconﬁguration run when there are no more devices to be processed.

8-64 NOS Version 2 Analysis Handbook Revision M

REDEFINE K Display

Table 8-9. Reconﬁguration Parameters

Parameter Description

0$ms.

CH=ci,C2 Speciﬁes the channel numbers (ci and, optionally, C2) to be used

under the new device deﬁnition.

Use this parameter when the entire device is to be redeﬁned to

alternate channel number(s). You can also use it to add or delete

channels from an equipment deﬁnition. New channels can be

added by specifying new channel numbers for a deﬁned

equipment. Channels can be deleted by specifying CH = . when

deleting a unit. Acceptable values for ci and C2 are from 0 to 13s

for systems having 10 or less PPs; from 0 to 13s and from 20s to

338 for systems having more than 10 PPs. Leading zeros can be

omitted.

EQ=est Speciﬁes the equipment with EST ordinal est is to be processed.

Enter this parameter when the processing of equipments is order

dependent; for example, when a device must ﬁrst be made

available before it can replace a failing device. The system

ignores all parameters not processed before you enter the EQ = est

parameter. Parameters entered after EQ=est refer to the

speciﬁed equipment until you enter another EQ = est or a

command that causes the next equipment in the list to be

selected for processing.

UL=ui,u2,...,Un Speciﬁes the unit list for the new conﬁguration.

When a unit is to be replaced, added, or deleted, the entire unit

conﬁguration must be entered with this parameter. The

equipment must be unloaded (multimainframe mode) or otherwise

have an unavailable status if a unit is to be added or deleted.

Any number of units can be changed. By specifying UL = . the

current unit conﬁguration is deleted.

NOTE

If the unit number speciﬁed in the UL= parameter represents a

unit that is not deﬁned in the EST, the unit number is accepted

without validation. Ensure that the unit number entered

represents a valid device.

UR=un Speciﬁes that the device with unit number un is to be recabled.

This parameter is used when a device is to be physically replaced

by a new device with the same unit number. More than one

device can be speciﬁed for recabling at the same time.

jtg^S.

Revision M K-Display Utilities 8-65

REDEFINE K Display

Table 8-10. Reconﬁguration Commands

Command Description

ABORT Discontinues processing of the current command. It is entered only in

response to a detected error condition and is used to initiate error

recovery procedures. This command cannot be followed immediately by

an END command.

CLEAR Clears the suspend and redeﬁnition request status for the current

equipment. Label and read/write veriﬁcation of the device is performed.

This command should be entered only when no further processing is

desired for the current equipment.

END Terminates the reconﬁguration processing. The K-display parameters are

set to the default values as control is returned. This command cannot be

entered immediately after an ABORT command.

GO Initiates the processing of speciﬁed reconﬁguration parameters previously

entered.

IGNORE Informs the system on which the command was entered to ignore

processing on this device (multimainframe mode only). This command

should be entered during an add or delete unit reconﬁguration run on

the machine(s) within the multimainframe complex which, for control

reasons, cannot add to or delete from the speciﬁed equipment. Also, you

can use this command to ignore an error message pertaining to a

marginally unacceptable servo timing check, which is performed

automatically when reconﬁguring an 885-11/12 disk drive.

NEXT Enter this command in response to an 885-11/12 disk drive servo timing

check that meets requirements. This command causes processing to

continue with the next device or the next step of processing. This

command is valid only when reconﬁguring 885-11/12 disk drives.

RECHECK Retries a veriﬁcation/diagnostic process that previously gave an error.

Only the commands RECHECK, ABORT, and IGNORE are accepted by

the system when an error message is issued.

RERUN Sets the parameters to default values and updates the list of equipment

to be reconﬁgured.

RESET Resets the parameters to default values.

SUSPEND Causes system processing on the speciﬁed device to be suspended

indeﬁnitely while the device is in a not ready state. Only diagnostic

access to the device is allowed. All other jobs accessing the device will

be unable to continue until the device is returned to a ready state. More

than one equipment can be suspended at the same time.

8-66 NOS Version 2 Analysis Handbook Revision M

REDEFINE K Display

You can stop the reconﬁguration procedure by entering either the K.ABORT command

/#^ or the K.CLEAR command (depending on which command is posted in the list of valid

*• commands on the K display). In either case, processing advances to the next device in

the list of devices under IDLED EQUIPMENTS (see ﬁgure 8-20).

If the device speciﬁed is a shared device in a multimainframe environment and

reconﬁguration is not desirable on one or more of the mainframes, use the following

procedure:

1. Enter at the console of each mainframe for which reconﬁguration of the shared

device is not desired.

REDEFINE,est.

est Description

est EST ordinal of the shared device.

*s^ 2. Enter

K.jsn.

jsn Description

jsn Job sequence name of the job requesting the REDEFINE K

display.

3. Enter

K.IGNORE

and processing on the shared device in the list is ignored by that mainframe. The

machine must wait until the shared device is done with its processing.

4. Enter

K.END

to end the reconﬁguration procedure.

Enter either the RERUN or RESET command to clear the IGNORE command.

Figure 8-22 shows the output for a sample reconﬁguration run.

ORD TYPE CHANNELS UNITS STATUS

06 DJ-1 CH26.32 UL=06 ST =

07 DJ-1 CH26.32 UL=07 ST =

10 DJ-1 CH32 UL=01 ST =

11 DJ-1 CH13 UL=02 ST =

12 DI-1 CH26 UL=03 ST =

13 DI-1 CH32 UL=11 ST =

14 DI-1 CH26,32 UL=04 ST = I

/^N Figure 8-22. Reconﬁguration Run Output

Revision M K-Display Utilities 8-67

REDEFINE K Display

Following are examples of the reconﬁguration of disk devices and how their status

changes in the REDEFINE K display. /*=%

Example 1:

This example illustrates returning a unit to the system.

Assume the unit of est ordinal 7, the system's spare disk drive, was used to replace a

failing disk drive. Now that the defective unit has been repaired, it is to be returned

as the unit of the spare disk drive.

Disk conﬁguration before the REDEFINE:

EST Equipment Channel Unit

Ordinal Type Number(s) Number Status

7 D J - 0 0 0 - I - -

Enter the following commands:

Command Description

REDEFINE,7. 7 is the EST ordinal of the shared device.

K,jsn. jsn is the job sequence name of the job requesting the REDEFINE K

display.

K.CH = 32,26 32 and 26 are the channel numbers to be used by est ordinal 7.

K.UL=7 7 is the unit number of the device being returned.

K.GO Initiates the processing of all the parameters and commands entered

for EST ordinal 7.

The system responds with the message:

EQ 7 REDEFINITION COMPLETE.

To end the reconﬁguration, enter:

K.END

Disk conﬁguration after the REDEFINE:

EST Equipment Channel Unit

Ordinal Type Number(s) Number Status

7 D J - 1 3 2 , 2 6 7

8-68 NOS Version 2 Analysis Handbook Revision M

0$ms.

REDEFINE K Display

Example 2:

This example illustrates reconﬁguring a failing unit.

Assume EST ordinal 10 and EST ordinal 11 are deﬁned in the EST display.

Equipment 10 is a spare unit currently not being used. Equipment 11 has a pack

mounted and is the failing device. Before the failing device can be reconﬁgured, the

spare unit must be removed from the EST. The following stipulations apply when

deleting a unit from the EST:

• If the EST ordinal is defined in a single mainframe environment or if it is not

shared in a multimainframe environment, then having device unavailable status

(U status in the E,M display) for that EST ordinal is sufﬁcient.

• If the EST ordinal is shared in a linked shared device multimainframe

environment, then that EST ordinal must have device unavailable status and be

globally unloaded (U and N status in the E,M display) before a reconﬁguration

can be performed.

Disk conﬁguration before the REDEFINE:

EST Equipment Channel Unit

Ordinal Type Number(s) Number Status

1 0 D J - 1 3 2 1 I —

1 1 D J - 1 1 3 2

To perform the reconﬁguration, enter the following commands:

Command Description

REDEFINE,10. 10 is the EST ordinal of the spare device.

REDEFINE,11. 11 is the EST ordinal of the failing device.

KJsn. jsn is the job sequence name of the job requesting the REDEFINE K

display.

K.CH = . Deletes the channel number(s) assigned to EST ordinal 10 from the

EST.

K.UL=. Deletes the unit number of EST ordinal 10 from the EST.

K.GO Initiates the processing of all the parameters and commands entered

for EST ordinal 10.

The system responds with:

EQ 10 REDEFINITION COMPLETE.

Enter the reconﬁguration commands for the failing device as follows:

Command Description

K.CH = 32 32 is the channel number that was assigned to EST ordinal 10.

K.UL = 1 1 is the unit number that was assigned to EST ordinal 10.

K.GO Initiates the processing of all the parameters and commands entered

for EST ordinal 11.

Revision M K-Display Utilities 8-69

REDEFINE K Display

The system responds with:

SPIN DOWN UNIT 02.

02 is the unit number of EST ordinal 11.

After the unit is spun down, the system responds with:

MOVE PACK FROM UNIT 02 TO UNIT 01 AND SPIN UP.

After the pack is moved and the new unit is spun up, the system responds with:

EQ 11 REDEFINITION COMPLETE.

To end the reconﬁguration, enter:

K.END

Disk conﬁguration after the REDEFINE:

EST Equipment Channel Unit

Ordinal Type Number(s) Number Status

1 0 D J - 0 0 0

1 1 D J - 1 3 2 1

Example 3:

This example illustrates reconﬁguring devices across channels.

Units may be reconﬁgured across channels with the REDEFINE command. Assume )

that EST ordinal 12 is the failing unit (or possibly the failing channel) and is on

channel 26. EST ordinal 13 is the spare unit on channel 32. The following commands

illustrate a reconﬁguration across channels.

Disk conﬁguration before the REDEFINE:

EST Equipment Channel Unit

Ordinal Type Number(s) Number Status

1 2 D I - 1 2 6 3

1 3 D I - 1 3 2 1 1

Enter the following commands:

Command Description

REDEFINE,12. 12 is the EST ordinal of the failing device.

REDEFINE,13. 13 is the EST ordinal of the spare device.

K,jsn. jsn is the job sequence name of the job requesting the REDEFINE K

display.

K.EQ=13 Selects the spare unit to be processed ﬁrst.

K. SUSPEND Suspends the spare unit to clear it from the EST.

8-70 NOS Version 2 Analysis Handbook Revision M

/g^v

J^^S.

REDEFINE K Display

The system responds with:

SYSTEM USAGE OF EQ 13 SUSPENDED.

To continue with the reconﬁguration on EST ordinal 12, enter:

K.EQ=12 Selects the failing unit for processing.

K.CH = 32 Assigns the old channel number of EST ordinal 13 to EST ordinal 12.

K.UL=11 Assigns the old unit number of EST ordinal 13 to EST ordinal 12.

K.GO Initiates the processing of all the parameters and commands entered

for EST ordinal 12.

The system responds with:

EQ 12 REDEFINITION COMPLETE.

Disk conﬁguration at this point in example 3.

EST Equipment Channel Unit

Ordinal Type Number(s) Number Status

12 DI-1 32 11 I—

13 DI-1 32 11 IS-

All SUSPEND status ﬂags must be cleared before a reconﬁguration run can be ended.

The system therefore automatically selects EST ordinal 13 again for the next EST

ordinal to be processed.

To continue with the reconﬁguration, enter:

Command Description

K.CH = 26 Assigns the old channel number of EST ordinal 12 to EST

ordinal 13.

K.UL=3 Assigns the old unit number of EST ordinal 12 to EST ordinal 13.

K.GO Initiates the processing of all the parameters and commands entered

for EST ordinal 13.

The system responds with:

EQ 13 REDEFINITION COMPLETE.

To end the reconﬁguration, enter:

K.END

Disk conﬁguration after the REDEFINE:

EST Equipment Channel Unit

Ordinal Type Number(s) Number Status

1 2 D I - 1 3 2 1 1

1 3 D I - 1 2 6 3

Revision M K-Display Utilities 8-71

Remote Batch Facility (RBF) K Display

Whenever RBF is in operation, user connections and activity are shown in the RBF K

display (refer to ﬁgure 8-23).

RBF Version PSR Level of Number of Terminals Number of Batch

Number RBF Software Active (Decimal) Devices Active (Decimal)

REMOTE BATCH TERMINAL STATUS.

RBF VER 1.2- 525 TERMINALS = 1 DEVICES - 2

TERM. DEV. TCLASS. USR/JSN. FAMILY/.

NAME. TYP. /STATE. FILESIZE. ERRORS.

LOET1 CO. 200UT. AADC NZDGTYR.

CR1. STOP.

L P 1 . G O . A J Z A 4 0

FORMS

CODE

Figure 8-23. RBF K Display

Select the RBF K display with this DSD command:

K.RBF.

Data on the RBF display is updated periodically to reﬂect the current status of batch

terminals connected to RBF.

Terminals are identiﬁed by termname, a unique name deﬁned by the site. Terminals

are displayed in alphabetical order. There are multiple lines on the K display per

terminal; one line for the terminal console, and one line for each batch device (card

reader, line printer, card punch, or plotter).

If there are more terminals and devices active than can be displayed on one screen,

the message

MORE LINES FOLLOW.

appears at the lower left corner of the display. You can display additional screens by

entering:

K.+

The K.+ command advances the display page-by-page and end-around from the last

page to the ﬁrst. All screens are displayed in a forward direction. Use of the console

input K.- does not move the screen back to the previous display.

Entries in the display have the following format:

term dev tclass usr/jsn family/

name typ /state ﬁlesize errors

forms

code

8-72 NOS Version 2 Analysis Handbook Revision M

Remote Batch Facility (RBF) K Display

Header

term name

dev typ

tclass/state

Description

Name of the terminal. A unique name, termname, assigned by the

installation during network startup. This ﬁeld is used for console

devices only.

Device type. The device type code is one of the following: •

Code Description

CPi

CRi

LPi

PLi

Console device.

Card punch.

Card reader.

Line printer.

Plotter.

The ordinal number, i, of the device can range from 1 to 7.

If this ﬁeld is in the same line as the termname, the contents of the

ﬁeld is the terminal class mnemonic (refer to the Remote Batch

Facility Reference Manual). If this ﬁeld occurs in a line not

containing a termname, the device status code is one of the following:

Code Description

ABRT File in transmission is to be discarded.

CONN Device is connected (initial state).

END Device stops transmission at EOI.

ENDA File being aborted; device will stop at EOI.

ENDC End connection.

ENDI Device will stop at EOI; idle down requested.

GO Device is ready for input.

NULL Console is not connected, but RBF devices are. This code

appears only when the device type is a console (device type

code CO).

PREC Preconnect status (device connection not complete).

STOP Device is not ready for transmission of data.

STPA Device is stopped; current ﬁle to be aborted.

STPE Device is stopped.

STPI Device is stopped due to idle down request.

Revision M K-Display Utilities 8-73

Remote Batch Facility (RBF) K Display

Header Description

usr/jsn

family/ﬁlesize

errors

forms code

If a ﬁle transfer is in progress, jsn is displayed in this ﬁeld. If no ﬁle

is being transferred, then the user name appears in this ﬁeld.

If this ﬁeld is in the same line as the termname, the content of this

ﬁeld is the family associated with the username. The user index and

family combination determine the terminal identiﬁer used by the

system for routing jobs. Otherwise, the contents of this ﬁeld is the

size of the output ﬁle in PRUs.

Contents of this ﬁeld, if present, is one of the following error

messages:

Message Description

DISK Data has been lost due to an unrecoverable disk

ERROR failure.

DISK FULL A disk-full indication was received while RBF was

attempting to write to disk.

NOT The device the user speciﬁed is not available (for

READY example, the line printer is out of paper or the card

reader has a card jam).

QUEUE The system input queue has reached its limit of jobs

FULL waiting to begin execution.

Forms code for output devices as speciﬁed by the user and deﬁned by

the site.

8-74 NOS Version 2 Analysis Handbook Revision M

Remote Host Facility (RHF) K Display

/fpSN

Remote Host Facility (RHF) K Display

The Remote Host Facility (RHF) links NOS to a loosely coupled network (LCN)

providing transfer of permanent ﬁles, queued ﬁles, and maintenance facilities for the

LCN hardware. Each system in an LCN conﬁguration is connected to one or more LCN

trunks by network access devices (NADs). Several types of NADs are available,

allowing the connection of various types of computer systems to an LCN. Each system

has an RHF that provides some or all of the following capabilities.

RHF runs at a control point and contains an operator interface package, control

tables, network application code, and the PP routines that drive the RHF network

hardware.

The following applications are also available as part of RHF:

Application Description

PTF, PTFS Permanent File Transfer Facility (PTF) and Permanent File Transfer

Facility Servicer (PTFS). PTF and PTFS provide users access to remote

permanent ﬁles. A local user activates PTF with the MFLINK

command (refer to the NOS Version 2 Reference Set, Volume 3). When

a remote user enters the MFLINK command, RHF activates a PTFS

application on the local host to service the remote request.

MHF Maintenance Host Facility (MHF). MHF automatically loads

controlware of local NADs during RHF initiation and performs

maintenance logging. If a local NAD fails while RHF is active, MHF

automatically dumps the NAD's memory to a permanent ﬁle and

reloads the controlware, for a description of LNAD statement

parameters that control automatic NAD dumping and loading (refer to

section 10, LID/RHF Conﬁguration Files).

MHF also periodically checks all local NADs and any remote NADs

enabled for logging (refer to the network path table display described

later in this section) and copies new NAD error log entries to the

binary maintenance log (BML).

MHF starts automatically when RHF is initiated and remains active

until RHF terminates.

QTF, QTFS Queue File Transfer Facility (QTF) and Queue File Transfer Facility

Servicer (QTFS). QTF and QTFS allow you to transfer input and

output ﬁles to a remote system. When RHF is initiated, the system

automatically activates QTF. When a remote host QTF application has

a ﬁle to transfer, RHF initiates QTFS on the local host to service the

remote request.

The following subsections describe the initiation, operation, control, and termination of

the RHF.

Revision M K-Display Utilities 8-75

RHF Initiation

Before network operation can begin, you must initiate RHF using the DSD RHFffff

command (refer to section 5, DSD Commands).

When RHF is initiated, it starts all enabled applications deﬁned by the network

conﬁguration as autostart applications (refer to RHF Conﬁguration Files in

section 10). These applications are typically MHF and QTF.

Application MHF loads controlware into all local NADs that are deﬁned in RHF's

conﬁguration with AUTOLOAD=YES and have an EST status of ON.

Operator Interface

The operator interface for RHF consists of using the following displays:

Display Description

Application table Lists all active applications. '*s^\

Network path table Shows how the remote hosts are connected to the local host.

An example of each display is shown later in this section. To understand how to use

the information provided in the displays it is helpful to understand the sample LCN

network shown in ﬁgure 8-24.

8-76 NOS Version 2 Analysis Handbook Revision M

Operator Interface

MFA

NOS

EST 31

EST 32

NAD

TCU

1 I 2

NAD

TCU

0 I 1 I 2 | 3

Oil I 2 I 3

TCU

NAD

ND-3F

■©•

EST 33

NAD

TCU

0|1 I 2 I 3

0 I 1 | 2 | 3

TCU

NAD

ND-4A

t-*

EST 34

NAD

TCU

0 | 1 | 2 | 3

TCU

NAD

ND-G2

MFB

LIDs: BBB. XYZ

NOS

0 I 1 | 2 I 3

TCU

NAD

ND-F1

MFC MFD

LID: CCC LIDs: EEE, BBB. DDD

NOS NOS

Figure 8-24. Sample LCN

MFA is the local NOS host and MFB, MFC, and MFD are the remote NOS hosts. MFA

has four EST ordinals set up during installation (EST 31, 32, 33, and 34) to be used

with the three remote hosts. Each NAD can connect to a maximum of four channels

that can be used to communicate with NADs. These trunk control units (TCUs) are

represented by the numbers 0, 1, 2, and 3 in the diagram (refer to A in ﬁgure 8-24).

The horizontal lines between the NADs of the local mainframe (MFA) and the NADs

of the remote mainframes (MFB, MFC, and MFD) depict the connections between the

NADs. For example, the top horizontal line shows the connections between the

channels starting at TCU 0 of the local host and TCU 3 of the remote host (refer to

B in ﬁgure 8-24).

Also during the installation process, each remote NAD is given a remote NAD

address to uniquely identify that particular NAD. For example, the remote host MFB

has two NADs associated with it. Their remote NAD addresses are 3F and 4A (refer

to C in ﬁgure 8-24).

Revision M K-Display Utilities 8-77

RHF Commands Under K display

Use the RHF K display to communicate with RHF. Bring up this display with this

DSD command:

K.RHF.

Two K displays are available under RHF, the application table display (see ﬁgure 8-25)

and the network path table display (see ﬁgure 8-26). When you ﬁrst enter K,RHF, the

application table display appears. The next time and subsequent times you enter

K,RHF, the display that appears will be either the application table or the network

path table, whichever was displayed last.

The following commands are available under the RHF K display:

Command Function

APPL

IDLE

PATH

ord,ND=rna,AC = mac,

DD=dd,RT=rteb,

LT=lteb,LOG=status

Displays the application table (see ﬁgure 8-25).

Begins the idle-down process of RHF and all its associated

applications.

Displays the network path table (see ﬁgure 8-26).

Modiﬁes entries associated with the network path table

display ordinal ord.

Pages the current display forward.

Pages the current display backward.

.jj££f?$£iV

8-78 NOS Version 2 Analysis Handbook Revision M

Application Table Display

The application table display (see ﬁgure 8-25) lists all active applications. In this

ﬁgure, USRAP represents an application written by the site.

MAX ACTIVE MAX

ORD APPL ENABLED COPIES COPIES CONNECTS

1QTF YES

2QTFS YES

3PTF YES

4PTFS YES

5USRAP YES

6MHF YES

ACTIVE APPLICATIONS (NETON PERFORMED)

JOB JOB

NAME JOBORD CONNECTS NAME JOBORD CONNECTS

ABCA 26 AARM 22

AAQT 31 ACAL 33

ACXQ 24 ACXY 27

Figure 8-25. Application Table Display

/ ^ ^ V

Revision M K-Display Utilities 8-79

Application Table Display

Each entry in the display appears in the following format:

ord appl enabled maxcoples actlvecopies maxconnects

jobname jobord connects

Header Description

ord Application table display ordinal.

appl Name of the application.

enabled Speciﬁes whether the application communicates with RHF.

maxcopies Maximum number of copies of the application that can

simultaneously communicate with RHF.

activecopies Number of copies of the application that are currently

communicating with RHF.

maxconnects Maximum number of network connections allowed by each copy of

the application.

jobname Job sequence name of the application.

jobord Executing job table ordinal of the application.

connects Number of network connections currently used by this copy of the

application.

8-80 NOS Version 2 Analysis Handbook Revision M

Network Path Table Display

The network path table display, illustrated in ﬁgure 8-26, shows how the remote hosts

are connected to the local host. In this ﬁgure, A 1 means that a TCU is enabled. A 0

means that a TCU is not enabled. The leftmost digit of these entries corresponds with

TCU 0, the next with TCU 1, and so forth. For example, if a NAD has TCUs 1 and 2

enabled, the corresponding entry should be 0110. Refer to ﬁgure 8-24.

LOCAL REMOTE REMOTE

TRUNK TRUNK NAD ACCESS CON LOG

ORD PID EST CH ENABLED ENABLES ENABLES ADDR DEST CODE COUNT ERRS

1 M F B 031 YES 0110 0110 3F F0F0 NO

2032 YES 0001 1000 4A F0F0 NO

3033 10 NO 0001 0100 3F F0F0 NO

4033 10 YES 0100 1000 4A F0F0 NO

5 M F C 031 NO 0110 0110 62 F0F0 NO

6032 NO 0001 1000 62 F0F0 NO

7033 10 YES 0101 1010 62 F0F0 NO

8034 11 YES 1000 0001 62 F0F0 NO

9 M F D 031 YES 0010 0010 F1 F0F0 NO

10 033 10 YES 0001 0010 F1 F0F0 NO

Figure 8-26. Network Path Table Display

Revision M K-Display Utilities 8-81

Network Path Table Display

Each entry in the display appears in the following format:

ord pid est ch

Header

ord

pid

est

enabled

local trunk

enables

remote

trunk

enables

remote nad

addr

dest

access code

con count

log errs

enabled local remote remote

t r u n k t r u n k n a d

enables enables addr

dest access con log

code count errs

Description

Network path table display ordinal.

Physical identiﬁer of the remote mainframe for which the path is

deﬁned.

EST ordinal of the NAD that deﬁnes the local end of the path.

Channel number of the NAD that deﬁnes the local end of the path.

Speciﬁes whether RHF uses the path for starting new connections. For

example, if a customer engineer wants to run diagnostics on a trunk

connecting two NADs (local and remote), you would disable the

appropriate path. When the connection count on that path falls to zero,

the customer engineer can run concurrent diagnostics on that trunk

without disturbing the operation of RHF.

Bit pattern specifying which TCUs to use on the local NAD for

communications with the remote NAD.

Bit pattern specifying which TCUs the remote NAD uses in returning

communications to the local NAD.

Hardware address (hexadecimal) of the remote NAD.

Destination device address (hexadecimal).

Access code of the remote NAD; the software access code is the two

leftmost characters and the hardware access code is the two rightmost

characters.

Number of connections currently using this path.

Speciﬁes whether the MHF application logs trunk errors (detected by

the remote NAD) into the binary maintenance log (BML).

8-82 NOS Version 2 Analysis Handbook Revision M

Network Path Table Display

When the network path table is displayed, you can enter the following command to

make changes.

ord,ND=rna,AC=rnac,DD=dd.RT=rteb,LT=lteb,LOG=status.

Parameter Description

ord Network path table display ordinal.

rna New remote NAD address in hexadecimal.

rnac New remote NAD access code in hexadecimal.

dd New remote NAD destination device address in hexadecimal.

rteb New remote trunk enable bit pattern (nonzero 4-bit binary number

with the leftmost bit representing TCU 0).

lteb New local trunk enable bit pattern (nonzero 4-bit binary number with

the leftmost bit representing TCU 0). The same number of bits must

be set for both RT and LT or the command will be rejected.

status Status of error logging for the speciﬁed remote NAD. Enter YES to

have MHF periodically copy the trunk errors occurring on the speciﬁed

remote NAD to the binary maintenance log (BML). MHF does not copy

trunk errors when you enter NO.

ord is the only required parameter and it must come ﬁrst. All other parameters are

optional and can appear in any order, but at least one, besides ord, must be present.

NOTE

Any change to rna, dd, or status for a given path results in an equivalent change for

all paths using that remote NAD. This is because rna, dd, and status are associated

with the remote NAD rather than the path to the remote NAD.

When RHF ﬁnds a faulty local NAD, it turns off the NAD's EST entry, activates the

MHF application, and issues the following job dayﬁle message:

RHF, NAD ON EST est HAS BEEN TURNED OFF

MHF dumps the NAD's memory, reloads the NAD's controlware, and turns on the

NAD's EST entry.

When RHF ﬁnds a path conﬁgured to use a faulty or nonexistent local NAD trunk

control unit (TCU), it disables the path and notiﬁes you with the following message

on the B,0 display:

RHF,BAD TCU ON PATH XXX, PATH TURNED OFF.

Use the GO,RHF command to acknowledge the message.

Revision M K-Display Utilities 8-83

RHF Commands Available Under Application or Network Path Display

When the application table or network path table is displayed, you can enter the

following commands to enable or disable the application or network path.

ENABLE,ord.

DISABLE,ord.

Parameter Description

ord Ordinal of the application or network path on the current display.

For example, if the application table display is up, EN ABLE,ord enables the application

speciﬁed by the display ordinal ord.

If an application is being enabled (whether currently enabled or disabled) and that

application is deﬁned as an autostart application, RHF initiates a copy of that .^^s

application. This feature may be used, for example, if the MHF application is

normally disabled and you want to start logging NAD errors. It may also be used if

the copy of the QTF application has been dropped accidentally and a new copy of the

QTF application must be started.

RHF Termination

You can terminate an RHF operation using the IDLE command.

The command format is:

IDLE.

When you enter the IDLE command the system waits for applications to stop and the

message IDLE-DOWN IN PROGRESS appears on the system status display. When

activity stops, the system drops RHF and the message RHF ENDED appears. The

IDLE command allows a gradual shutdown of RHF activities. No new connections are

allowed.

8-84 NOS Version 2 Analysis Handbook Revision M

SCOPE 2 Station Facility (SSF) K Displays

The following commands control operation of the SSF Subsystem. Initiate the SSF

Subsystem before issuing any commands. SSF can be initiated automatically when the

system is brought up, either manually by entering an SSFffff command, or by entering

the ENABLE,SSF command.

Operator Interface

The SSF uses the K display for operator communications; therefore, the K display must

be assigned to SSF before any other SSF Subsystem commands can be entered. Use the

following command to assign the K display to SSF.

K.SSF.

Station Login

After SSF is assigned to a control point, SSF searches the equipment status table

(EST) for a SCOPE 2 communication coupler. If SSF ﬁnds an available coupler, it asks

the SCOPE 2 mainframe to establish communications. If no coupler is found, a message

appears at the console indicating that SSF is ready to log in to a SCOPE 2 mainframe.

To log in, ﬁrst enter an ON command (refer to section 5, DSD Commands, for a

description of the ON command), then enter the following command:

K.LOGIN.

Enabling and Disabling File Transfers

File transfers between NOS and SCOPE 2 must be enabled before actual ﬁle transfers

can begin; however, this occurs automatically when the SSF Subsystem is logged in.

You can disable ﬁle transfers by entering the following command:

K.OFFSTAT.X.

Parameter Description

x Speciﬁes the SCOPE 2 mainframe PID.

When you enter a K.OFFSTAT command, ﬁle transfers in progress are completed, but

no new ﬁle transfers are initiated. This command does not affect command, display, or

message processing.

File transfers can be reenabled using the following command:

K.ONSTAT.x.

Parameter Description

x Speciﬁes the SCOPE 2 mainframe PID.

5. Refer to appendix C for more information on the SSF Subsystem.

Revision M K-Display Utilities 8-85

Station Recovery

If NOS fails, follow the SSF initiation procedure as described in section 5, DSD

Commands, to recover the station. SSF does not require reinitiation if SCOPE 2 fails

or if there is an FLPP failure that results in a reload of the FLPP.

Station Disconnection and Logout

Station disconnection refers to the termination of communications between SSF and

SCOPE 2 as a result of a hardware or software error or as a result of an operator

command. When the disconnection occurs, all pending requests to the SCOPE 2

mainframe (such as screen refreshes, messages, and acknowledgments) are dropped.

Before entering an operator command to disconnect SCOPE 2, you should perform the

following steps:

1. Idle down the ﬁle transfers by entering the K.OFFSTAT command. The station is

idle when all staging SPOT jobs are ﬁnished and the spooling SPOT job has been

swapped out.

2. If the system is in restricted or privileged mode, enter the CLROP command to

relinquish operator control of SCOPE 2. Refer to the SCOPE 2.1 Operator's Guide

for a description of the CLROP command.

To disconnect the SCOPE 2 mainframe, you can either drop the station control point by

issuing a STOP or IDLE Subsystem command as described in section 5, DSD

Commands, or you can log out of a speciﬁc SCOPE 2 mainframe by using the following

command:

K.LOGOUT.

K.LOGOUT,p1d.

Parameter Description

pid PID of the mainframe you want to disconnect (required only in a

multimainframe environment).

8-86 NOS Version 2 Analysis Handbook Revision M

File Transfer Limit Commands

The following commands allow you to set limits on the number of ﬁle transfers that

can be in progress concurrently for various types of ﬁle operations. The system default

value for each of the following commands is 2; however, the default values can be

changed, as an installation option, when SSF is installed. Refer to the NOS Version 2

Installation Handbook for more information on the default values for ﬁle transfer limit

commands.

Staged File Transfer Commands

The following commands control the number of concurrent SPOT jobs performing staged

ﬁle operations:

Command Description

K.PURGE,pid,n.

K.READTP,pid,n.

K.WRITETP,pid,n.

K.GETPF,pid,n.

K.SAVEPF,pid,n.

Speciﬁes the maximum number of concurrent SPOT jobs allowed

to purge ﬁles from a NOS permanent ﬁle device, pid is the PID of

the SCOPE 2 mainframe, n is the maximum number of purge

operations (0 ^ n^ 7).

Speciﬁes the maximum number of concurrent SPOT jobs allowed

to read tape ﬁles from NOS to SCOPE 2. pid is the PID of the

SCOPE 2 mainframe, n is the maximum number of read

operations (0 ^ n ^ 7).

Speciﬁes the maximum number of concurrent SPOT jobs allowed

to write tape ﬁles from SCOPE 2 to NOS. pid is the PID of the

SCOPE 2 mainframe, n is the maximum number of write

operations (0 ^ n ^ 7).

Speciﬁes the maximum number of concurrent SPOT jobs allowed

to read permanent ﬁles from NOS to SCOPE 2. pid is the PID of

the SCOPE 2 mainframe, n is the maximum number of permanent

ﬁle read operations (0 ^ n ^ 7).

Speciﬁes the maximum number of concurrent SPOT jobs allowed

to write ﬁles from SCOPE 2 to a NOS permanent ﬁle device, pid

is the PID of the SCOPE 2 mainframe, n is the maximum number

of permanent ﬁle write operations (0 ^ n ^ 7).

V^Sfﬂfv

Revision M K-Display Utilities 8-87

Spooled File Transfer Commands

The following commands control the number of I/O ﬁles that the spooling SPOT job

xSTA can transfer concurrently. Separate limits can be deﬁned for both input and

output ﬁles; however, the system limits the combined total of input and output ﬁles to

four concurrent transfers.

Command Description

K.INPUT,pid,n. Speciﬁes the maximum number of concurrent input ﬁle transfers

from NOS terminals to SCOPE 2. pid is the PID of the SCOPE 2

mainframe, n is the maximum number of input ﬁles (0 ^ n ^ 4).

K.OUTPUT,pid,n. Speciﬁes the maximum number of concurrent output ﬁle transfers

from SCOPE 2 to NOS terminals, pid is the PID of the SCOPE 2

mainframe, n is the maximum number of output ﬁles

(0 ^ n ^ 4).

■^

8-88 NOS Version 2 Analysis Handbook Revision M

Transaction Facility (TAF) K Displays

The following commands control operation of the TAF Subsystem. Initiate the TAF

Subsystem by using the TAFffff command before issuing these commands.

TAF Initialization K Display

When the transaction executive is brought to a control point, the message

REQUEST *K* DISPLAY appears at the control point if a DISPLAY,ON command is

speciﬁed in the TAF conﬁguration ﬁle. Respond with the entry:

K . TA F.

The TAF initialization K display appears on the system console as shown in

ﬁgure 8-27.

TAF

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

TAF INITIALIZATION OPTIONS

OPTION DESCRIPTION

SCP = 31 SUB CONTROL POINTS (2 - 31).

CMB = 40 COMMUNICATION BLOCKS (19 - 40).

REC = NO RECOVERY MODE (YES OR NO).

MFL = 376500B MAXIMUM FIELD LENGTH (40000B - 376500B).

ECS = OK EXTENDED MEMORY FIELD LENGTH (OK - 400K).

TLF = TASKLIB TASK LIBRARY FILE (1-7 CHARACTERS).

INITIALIZE CRF RECOVERY FILES (NONE,ALL OR

1-7 DIGITS).

INT = CRM,NONE INITIALIZE CRM RECOVERY FILES (NONE OR ALL)

ERO = CRF,NO ERROR OVERRIDE (YES OR NO).

BFL = 70000B CMM BASE FL (20000B - 200000B).

EFL = OB CMM EXPANDABLE FL (0 - 100000B).

TFL = 30000B CMM TARGET FL (10000B - 100000B).

INT = CRF,NONE

Figure 8-27. TAF Initialization K Display (Left Screen)

Revision M K-Display Utilities 8-89

TAF Initialization K Display

Any of the following initialization commands can then be entered. If no values are to

be changed, enter the command:

K.END

Values are decimal unless otherwise indicated.

Command Description

K.BFL=n Changes the starting and minimum ﬁeld length allocated by TAF to

CRM (200008 ^ n ^ 200000s). Default is 70000s.

K.CMB=n Changes the maximum number of communication blocks allowed to

the TAF Subsystem (19 ^ n ^ 40). Default is 40.

K.ECS=n Sets the extended memory ﬁeld length to be used by the

transaction executive; n is octal thousands of words. Default is 0.

K.EFL=n Changes the maximum additional central memory ﬁeld length made

available to CRM for buffers and capsules (0 ^ n ^ 100000s).

Default is 0.

K.END Ends input of the transaction executive initialization parameters.

Initialization is completed when the TAF status K display appears

(ﬁgure 8-28).

K.ERO=CRF,op Speciﬁes whether to override certain I/O and logic errors when

processing the communication recovery ﬁles (CRF).

op Description

NO Aborts if I/O or logic errors are encountered while

processing the communication recovery ﬁles. This is the

default setting.

YES I/O or logic errors encountered on a run unit header

record result in the loss of that run unit with no

indication to the terminal user. The loss is noted on the

recovery report.

I/O or logic errors encountered on a message record

within a run unit result in a loss of that run unit. A

status ﬁeld is set in the run unit header, allowing TAF to

inform the terminal user of the run unit loss.

I/O or logic errors encountered on the CRF header record

result in an unconditional abort of the TAF Subsystem.

A ^ S

8-90 NOS Version 2 Analysis Handbook Revision M

TAF Initialization K Display

Command Description

K.GO

K.INT=typ,ﬁleid

Ends input of the transaction executive initialization parameters.

Initialization is complete when the TAF status K display appears

(ﬁgure 8-28).

Speciﬁes which communication recovery ﬁles (CRF) are to be

initialized. This is the only way to initialize a CRF. Files speciﬁed

on a RECOVER command in the TAF conﬁguration ﬁle and

speciﬁed in this command are initialized. Files speciﬁed on a

RECOVER command but not speciﬁed in this command are used for

recovery.

This command also speciﬁes whether CYBER record manager

(CRM) data base recovery ﬁles are initialized or recovered. For

CRM recovery ﬁles, this command is valid regardless of TAF

assembly parameters.

typ

CRF

CRM

ﬁleid

Description

Communication recovery ﬁles. This parameter is valid

only if the installation parameter IPTAR equals 1.

CYBER Record Manager after-image and before-image

recovery ﬁles.

Description

n A digit from 1 to 7 that deﬁnes a CRF to be initialized.

The digit corresponds to the ID parameter on the

RECOVER command in the TAF conﬁguration ﬁle. This

parameter is not valid for CRM recovery ﬁles.

ALL If typ is CRF, all communication recovery ﬁles deﬁned by

RECOVER commands in the TAF conﬁguration ﬁle are

initialized. If typ is CRM, all CRM recovery ﬁles are

initialized.

This parameter must be used with caution when typ is

CRM since the CRM update history currently on the

after-image recovery ﬁles is lost.

NONE If typ is CRF, communication recovery ﬁles are

initialized; all communication recovery ﬁles speciﬁed in

RECOVER commands in the TAF conﬁguration ﬁle are

recovered. If typ is CRM, all CRM data bases are

recovered based on information in the existing recovery

ﬁles. No CRM recovery ﬁles are initialized. This ﬁleid is

the default for both typ=CRF and typ=CRM.

Revision M K-Display Utilities 8-91

TAF Initialization K Display

Command

K.MFL=n

K.REC=a

K.SCP=n

K.STOP

K.TFL=n

Description

K.TLF=ﬁlename

Sets the maximum ﬁeld length to be used by the transaction

executive (400008 ^ n ^ 376500s). Default is 376500s.

Speciﬁes whether to set the recovery bit in the user area of each

terminal status table entry (YES or NO). If YES, the user recovery

bit is set. If NO, the value of the user recovery bit is not changed

from what it was before the command was issued. Default is NO.

Changes the number of subcontrol points (2 ^ n < 31). Default

is 31.

Aborts the TAF Subsystem initialization unconditionally.

Changes the value used as the upper bound for TARGET. This is

the amount of memory CRM uses for data and index blocks

(100008 ^ n ^ 1000008). Default is 300008. For more information,

refer to the CYBER Record Manager Advanced Access Methods

Version 2 Reference Manual.

Changes the name of the system task library ﬁle (any valid ﬁle

name). Default is TASKLIB.

8-92 NOS Version 2 Analysis Handbook Revision M

0ms

Restarting TAF K Display

On a level 3 recovery deadstart, or if the TAF Subsystem aborts, the TAF procedure

ﬁle will automatically restart TAF by transferring control to the TAF automatic

recovery program, if sense switch 4 is set. The TAF automatic recovery program

recovers the central memory pointers and variables deﬁned during TAF initialization.

If a DISPLAY,ON command is in the TAF conﬁguration ﬁle, the automatic recovery

program brings up the K display on the left console screen. This display is identical

to the TAF initialization K display, except that values speciﬁed by the TAF

initialization commands replace any default values that were in the TAF initialization

K display.

TAF Status K Display

When the TAF Subsystem is executing, the TAF status K display (ﬁgure 8-28)

indicates the:

• Latest transaction sequence number.

• Number of words of unused memory.

• Maximum field length.

• Global task dump limit.

• Subsystem default values for memory dump parameters.

The subsystem default values are used to control memory dumps when parameters are

not included in the CMDUMP or DSDUMP command. Any of these default values can

be changed by specifying the corresponding parameter in the K.DSDUMP command.

Revision M K-Display Utilities 8-93

TAF Status K Display

hh.mm.ss. yy/mm/dd.

TAF

CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

TAF STATUS

TRANSACTION SEQUENCE NUMBER 5

UNUSED FIELD LENGTH 7700

MAXIMUM FIELD LENGTH 376000

GLOBAL TASK DUMP LIMIT 0

CMDUMP/DSDUMP DEFAULT VALUES

OPTION DESCRIPTION

FW = 0 FWA OF TASK MEMORY TO BE DUMPED.

LW = 100000 LWA OF TASK MEMORY TO BE DUMPED.

EP = 1 IF EP=1, DUMP EXCHANGE PACKAGE.

IF EP»0, DO NOT DUMP EXCHANGE PACKAGE

OQ = BC OUTPUT QUEUE.

QD « 0 QUEUE DESTINATION.

Figure 8-28. TAF Status K Display (Left Screen)

/■SS8|v

>*fﬁ%v

8-94 NOS Version 2 Analysis Handbook Revision M

TAF Status K Display

The default parameter values for the CMDUMP and DSDUMP commands are given on

the TAF status K display shown in ﬁgure 8-28.

Parameter Description

EP Exchange package:

EP Description

0 Do not dump the exchange package.

1 Dump the exchange package.

FW First word address of the task memory to be dumped.

LW Last word address of the task memory to be dumped.

OQ6 Output queue:

O Q D e s c r i p t i o n

BC Local batch.

RB Remote batch.

PF Permanent ﬁle.

QD6 Queue destination:

Printer identiﬁer (if OQ=BC).

User name (if OQ=RB).

Permanent ﬁle name (if OQ=PF).

6. The functioning of this parameter is described in the TAF Version 1 Reference Manual.

Revision M K-Display Utilities 8-95

TAF K-Display Commands

When the transaction executive is at its control point, the following commands can be

entered from the system console or submitted from tasks using the KPOINT request.

Any task can issue the K.DUMP command. Only tasks that reside on the system task

library can issue the other K-display commands. (Refer to the TAF Version 1 Reference

Manual for additional information on the KPOINT request and the system task

library.)

Command

K.ASSIGN.est

K.ASSIGN,est,db,n

K.DEBUG

K.DROP,n

Description

K.DSDUMP,FW=addr,

LW=addr,EP=pkg,

OQ=outq,QD = qdest,

DB=db

Assigns a magnetic tape unit to be used for a journal ﬁle,

est is the EST ordinal of the tape unit. The ﬁrst form of the

command makes unit est available for the transaction

executive to assign to the next tape journal ﬁle that

encounters end of reel. Two tape units may be preassigned.

If a tape has not been preassigned in this manner, an

end-of-reel on a journal ﬁle causes subsequent entries for

that ﬁle to be placed on disk.

The second form of the command forces journal ﬁle n (n = l,

2, or 3) for data base db, deﬁned as a tape ﬁle, from disk to

tape. The transaction executive copies the data from the

disk journal ﬁle to tape est and places all subsequent

entries for that ﬁle on the tape. This command is necessary

after the transaction executive initialization to assign tape

units to the tape journal ﬁles or after an end-of-reel on a

tape journal ﬁle when no tape had been preassigned to the

transaction executive. All data residing on the disk for the

tape journal ﬁle must ﬁt on the tape assigned by this

command or the transaction executive will unload the tape

and issue the message *UNABLE TO USE TAPE*.

Turns on the application interface program (AIP) debug

option, which logs all messages on trace ﬁle ZZZZZDN. Use

this command only when TAF is installed with the DEBUG

option.

Drops an executing task at subcontrol point n.

Allows you to modify the standard system default

parameters controlling memory dumps. The command does

not directly cause a dump. Rather, it sets default values to

be used when a subsequent CMDUMP request is received or

when abort conditions occur. Refer to TAF Status K Display,

earlier in this section, for explanations of the parameters.

z<^\

8-96 NOS Version 2 Analysis Handbook Revision M

TAF K-Display Commands

Command

K.DUMP,fwa,lwa

J$3zm^\

0nms

Description

Dumps all or part of the ﬁeld length of the Transaction

Facility from the ﬁrst word address (fwa) to the last word

address (lwa) of the area to be dumped. The default value

for fwa is 0 and for lwa is 377777s. The default base is

octal. If no parameters are speciﬁed, the entire ﬁeld length

is dumped. The output is routed to a printer that has an ID

of 0.

Unlike other K-display commands, the K.DUMP command

can be issued from any task. Other K-display commands can

be issued by tasks only if they are on the system task

library (refer to the TAF Version 1 Reference Manual).

Since secure information may be contained in a dump of the

Transaction Facility, the following safeguards have been set

up to protect dumped information; however, the installation

must take the ultimate responsibility for the protection of

dumped information.

• The global task dump limit (GTDL) can be set by the

K.DUMPLIM command to limit the number of times the

K.DUMP command can be issued from tasks. The initial

value of the GTDL is 0 (zero), so the K.DUMP command

is disabled from use by a task by default.

• For all dumps of the Transaction Facility, whether you

initiated it or a task did, a one-page header precedes the

dump. This header page indicates the output is secure

and should be given only to the TAF central site systems

analyst.

• When the Transaction Facility is dumped, the message

TAF FIELD LENGTH DUMP RELEASED is issued to

the system dayﬁle, the Transaction Facility dayﬁle, and

line one of the control point.

Revision M K-Display Utilities 8-97

TAF K-Display Commands

Command Description

K.DUMPLIM,n

K.IDLE

K.JEND,db,n

K.MAXFL,n

K.MESSAGE,TN=b.

message

Sets GTDL to value n (0 ^ n ^ 9999999). If n is not

speciﬁed, the GTDL is set to 0.

The GTDL is the number of times the K.DUMP command

can be issued from tasks. This value is displayed on the

TAF status K display shown in ﬁgure 8-28. The initial value

of the GTDL is 0. When the GTDL is 0, no dumps of the

Transaction Facility can occur from tasks. Thus, the

K.DUMP command is disabled from tasks by default.

To enable the K.DUMP command for tasks, issue the

K.DUMPLIM command to set the GTDL to a value greater

than zero. Each time a task issues a K.DUMP command,

the GTDL is decreased by one until it equals zero. When

the ﬁrst K.DUMP command is issued from a task with the

GTDL equal to zero, the message

GLOBAL TASK DUMP LIMIT EXHAUSTED

is issued to the Transaction Facility dayﬁle, the system

dayﬁle, and line one of the control point. Also, the message

DUMPS.LOST

is displayed on the K display in place of the value of the

GTDL. This message remains until the value of GTDL is set

to a value greater than or equal to zero. The K.DUMPLIM

command should be used with care in system tasks, since

this might allow unauthorized users to alter the GTDL.

Idles down the transaction control point. Once idle down has

been initiated, no new transactions will be permitted but

currently executing transactions will be allowed to ﬁnish.

Forces end-of-reel processing (writes an EOI and rewinds the

ﬁle) on tape journal ﬁle n of data base db. If n is not a

tape journal ﬁle, the command is ignored.

Alters the transaction executive maximum ﬁeld length. The

transaction executive does not attempt to obtain more than

n words of storage. This command is rejected if the value

for n is more than 376500s or less than the ﬁeld length

currently required for TAF.

Directs the transaction executive to send message to a

terminal speciﬁed by terminal name b.

^®^Si

>^^^V

8-98 NOS Version 2 Analysis Handbook Revision M

TAF K-Display Commands

Command

K.NODEBUG

K.OFFTASK,a,db

K.ONTASK.a.db

K.ROLLTIM,nnnnnn

K.SWITCH

Description

K.TBCON,nn

K.TST,TN = a,DB=db,

UU = nnnn,UL=mmmm,

NN = b

Turns off the application interface program (AIP) debug

option, which logs all messages on trace ﬁle ZZZZZDN. Use

this command only when TAF is installed with the DEBUG

option.

Disables the use of task a, where a is the task name in the

data base db task library directory (dbTASKL). The data

base name db is not speciﬁed for tasks in the system task

library.

Reverses the effect of a previous OFFTASK command for the

speciﬁed task a in the data base db task library directory

(dbTASKL). The data base name db is not speciﬁed for tasks

in the system task library (TASKLIB).

Changes the amount of time that TAF will retain its ﬁeld

length between communication input messages; nnnnnn is

speciﬁed in units of milliseconds. Refer to the installation

parameter ITRTL in the NOS Version 2 Installation

Handbook.

Causes the console K display to change to a display listing

all allowable TAF K display commands. When K.SWITCH is

entered a second time, the previous K display returns. This

command activates task KDIS and forces TAF to remain

rolled in.

Changes the number of TAF/CRM batch concurrency users;

nn is less than or equal to the value speciﬁed on the

TBCON command in TAF's conﬁguration ﬁle (TCF).

Changes entries in the terminal status table for terminal a.

The following entries can be changed: data base name db,

user area upper 12 bits (nnnn), user area lower 12 bits

(mmmm), and new terminal name b. The changes do not

affect the network and simulation ﬁles. Do not use this

command if the terminal is logged in.

Revision M K-Display Utilities 8-99

TAF/CRM Status K Displays and Commands

You can use the K display to monitor the status of CRM, CRM data bases, or CRM

data base permanent ﬁles while TAF/CRM is running if CRMTASK is present on the

system task library. To get the CRMTASK K display enter:

K.DIS,CRMTASK.

After the K display is assigned to the task, the display in ﬁgure 8-29 is brought to the

left screen.

TAF

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

CRMTASK COMMANDS

CRMSTAT SEE ALL CRM DATA BASES.

CRMSTAT.DB SEE CRM DATA BASE DB.

CRMSTAT,DBPFN SEE CRM DATA BASE FILE DBPFN.

DBDOWN,DB DOWN DATA BASE DB.

DBDOWN,DBPFN DOWN DATA BASE FILE DBPFN.

DBUP.DB UP DATA BA SE D B.

DBUP,DBPFN UP DATA BASE FILE DBPFN.

END END CRMTASK PROCESSING.

MENU SEE THIS DISPLAY.

+ PAGE LEFT SCREEN FORWARD.

PAGE LEFT SCREEN BACKWARD.

Figure 8-29. CRMTASK Commands K Display (Left Screen)

■^Sfev

8-100 NOS Version 2 Analysis Handbook Revision M

TAF/CRM Status K Displays and Commands

0m^-..

The following commands are available under CRMTASK.

Command Description

K.CRMSTAT

K.CRMSTAT,db

K.CRMSTAT,dbﬁle

K.DBDOWN,db

K.DBDOWN,dbﬁle

K.DBUP,db

K.DBUP,dbﬁle

K.END

K.MENU

K.+

K-

Selects the CRM status K display (ﬁgure 8-30). This display

shows the status of all CRM data bases.

Selects the CRM data base status K display (ﬁgure 8-31). This

display shows the status of the speciﬁc CRM data base with

identiﬁer db. It also shows the ﬁle name and status of the

before-image ﬁles.

Selects the CRM ﬁle status K display (ﬁgure 8-32). This display

shows the status of the speciﬁc permanent ﬁle in a CRM data

base with ﬁle name dbﬁle.

Makes a speciﬁc CRM data base with identiﬁer db unavailable

for processing. When this command executes, the CRMTASK

command directory appears on the K display.

Makes a speciﬁc permanent ﬁle in a CRM data base with ﬁle

name dbﬁle unavailable for processing. When this command

executes, the CRMTASK command directory appears on the K

display.

Makes a speciﬁc CRM data base with identiﬁer db available for

processing. When this command executes, the CRMTASK

command directory appears on the K display.

Makes a speciﬁc permanent ﬁle in a CRM data base with ﬁle

name dbﬁle available for processing. When this command

executes, the CRMTASK command directory appears on the K

display.

Ends CRMTASK K-display processing.

Returns to the CRMTASK commands K display (ﬁgure 8-29).

Pages the left screen forward to the next screen.

Pages the left screen backward to the ﬁrst screen.

Revision M K-Display Utilities 8-101

TAF/CRM Status K Displays and Commands

TAF

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID'=AA

CRM

NOS version

STATUS

TRANSACTIONS IN INPUT QUEUE ss

ACTIVE TRANSACTIONS =

TRANSACTIONS IN OUTPUT QUEUE

DATA DATA DATA

BASE STATUS BASE STATUS BASE STATUS

TT UP BA UP BB UP

BC UP BD UP BE UP

BF IDLE BG UP BH UP

BI UP BJ IDLE CA UP

CB UP CC UP CD UP

CE DOWN CF UP CG DOWN

CH UP CI DOWN CJ UP

./<a^8k

Figure 8-30. CRM Status K Display (Left Screen)

K. TAF

hh.mm.ss. yy/mm/dd CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

DATA BASE NAME = AB AFTER IMAGE FILE = ZZABA01

DATA BASE STATUS = UP PRU-S REMAINING = 20

ocrunt XRiMut riuto oihiuo

FILE STATUS F I L E S TAT U S F I L E STATUS

ZZABB01 UP ZZABB02 UP ZZABB03 UP

ZZABB04 DOWN ZZABB05 UP ZZABB06 DOWN

ZZABB07 UP

FILE STATUS F I L E S TAT U S F I L E STATUS

ABPFN01 UP ABPFN02 UP ABPFN03 IDLE

ABPFN04 UP ABPFN05 DOWN ABPFN06 UP

ABPFN07 UP ABPFN08 UP ABPFN09 UP

ABPFN10 DOWN ABPFN11 UP ABPFN12 UP

/^SftjSk

Figure 8-31. CRM Data Base Status K Display (Left Screen)

8-102 NOS Version 2 Analysis Handbook Revision M

TAF/CRM Status K Displays and Commands

TAF

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

CRM FILE STATUS

FILE NAME = TTRNFDM

FILE STATUS = UP

RECOVERABLE = YES

PACK NAME = PACKNAM

DEVICE TYPE = DEFAULT

ATTACH MODE = M

SIZE OF PRIMARY KEY = 30

NUMBER OF ALTERNATE KEYS = 6

ACTIVE USERS = 4

ACTIVE LOCKS = 8

Figure 8-32. CRM File Status K Display (Left Screen)

Revision M K-Display Utilities 8-103

jr**'*

L-Display Utilities 9

FOTD L Display 9-2

LIDOU L Display 9-3

QDSPLAY L Display 9-6

SCTD L Display . 9-10

SDSPLAY L Display 9-12

SDSPLAY Commands 9-16

Changing System Attributes . 9-17

L.keyword=value Commands 9-17

SUBSYST L Display 9-20

**%.

* " %

yams.

L-Display Utilities

This section documents the following L displays and the utilities used to present them.

Utility Description

FOTD Displays family ordinal table (FOT).

LIDOU Displays logical identiﬁer table (LID).

QDSPLAY Displays the contents of a ﬁle in the queued ﬁle table (QFT).

SCTD Displays the contents of the service class control table, which shows the

valid service classes for each origin type.

SDSPLAY Displays system values used to control job ﬂow.

SUBSYST Displays subsystem information.

The L display is like the K display in that you must start the program by entering

the name of the utility. When the L display is ready for use, DSD assigns it to the left

screen automatically. Only one L-display program is active at a time. The program

need not occupy a control point or memory while you are looking at the display.

Using the L display, you can run any utility program you have created. The L

display is an interface between your program and DSD. This interface allows the site

analyst to write CPU programs that show displays on the system console.

All entries must be preﬁxed by L period (L.). However, when pressing CR or NEXT

after the ﬁrst entry, everything but the L. is erased. This allows another command to

be entered without entering L. ﬁrst. All examples in this section show L., although

you may not have to type it. If it becomes necessary to enter a DSD command during

parameter entry, simply backspace to erase the L., enter the command, and then

continue by typing L. and the entry.

After you call a speciﬁc utility, enter the ﬁrst command in the following format:

L.commandst ring.

Parameter Description

commandstring Any input (command, data, or parameter) that is deﬁned by the

job as valid input.

Execute your own L-display utility by entering the following command:

LDISopt.

Parameter Description

opt One to three characters deﬁning your utility.

Revision M L-Display Utilities 9-1

FOTD L Display

The FOTD L-display utility displays all the family names known to the system and the

corresponding family ordinal. To initiate the FOTD utility enter one of the following

commands:

FOTD,L=out ﬁle,LO=opt1on.

FOTD.outﬂle,option.

Parameter Description

outﬁle Output ﬁle name. This parameter is valid only if a list option is

speciﬁed. The default outﬁle is ﬁle OUTPUT.

option List option. Enter one or more of the following:

option Description

D Formats the data for the DSD L display. This is the

default list option if the parameters outﬁle and option

are not speciﬁed.

L Formats the data for a line printer.

If you do not specify outﬁle and option, the data is written to the L-display buffer once

and the utility ends.

Figure 9-1 shows the FOTD L display.

/*&%.

L. FOTD

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID-AA NOS version

FAMILY ORDINAL TABLE DISPLAY

DEFAULT FAMILY = FAMILY1 AVAILABLE ENTRIES = 4B

ORD. FAMILY ORD. FAMILY ORD. FAMILY ORD. FAMILY

1. FAMILY1 2. FAMILY2 3. FAMILY3 4. FAMILY4

5. FAMILY5 6. FAMILY6 7. FAMILY7 10. FAMILY8

11. FAMILY9 12. FAMIL10 13. FAMIL11 14. FAMIL12

Figure 9-1. FOTD L Display

The FOTD L display contains the default family name and the octal number of FOT

entries that remain available for assignment. The display also shows each family name

and its corresponding family ordinal.

9-2 NOS Version 2 Analysis Handbook Revision M

LIDOU L Display

*^s LIDOU L Display

The LIDOU utility displays the LID table conﬁguration and allows you to add or delete

LIDs, LID attributes, and PID attributes in the LID table conﬁguration. This modiﬁed

conﬁguration is retained across a level 1, 2, or 3 recovery deadstart. A level 0

deadstart causes a return to the conﬁguration deﬁned by the LID conﬁguration ﬁle.

To initiate the LIDOU utility, enter one of the following commands:

LIDOU.

LIDOU,I=1nf1le,L=outﬁle.

The ﬁrst command format can be entered only from the console under DSD control;

also, all input and output is processed using the LIDOU L display. Figure 9-2 shows

the initial LIDOU L display (LID display).

NOTE

The information shown in ﬁgure 9-2 corresponds to the sample LCN shown in

section 8.

From this display you can select the HELP display shown in ﬁgures 9-3 and 9-4. You

can toggle between these two L displays as described in the HELP display.

The second command format can be entered under DIS control or by a system origin

batch job. In either case, the I and L parameters must be speciﬁed.

Parameter

I=inﬁle

L=outﬁle

Description

LIDOU reads input directives from ﬁle inﬁle. Valid input

directives are the same as the commands shown in the HELP L

display (ﬁgure 9-3) without the L. preﬁx and without the BACK,

HELP, +, or - commands.

LIDOU writes output to ﬁle outﬁle.

Revision M L-Display Utilities 9-3

LIDOU L Display

L. LIDOU

hh.mm ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

PID LID ENABLED ATTR MFTYPE NETWORK STATUS

MFB YES H-V- NOS RHF IN SERVICE

MBF YES

BBB YES s—

XYZ YES s—

MFC YES -L-- NOS RHF IN SERVICE

MFC YES

CCC YES S—N

MFD YES -LVN NOS SSF IN SERVICE

MFD YES

DDD NO s—

EEE NO

BBB YES

«_•-

Figure 9-2. LIDOU L Display (LID)

L. LIDOU

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID* =AA NOS version

LID OPERATOR UTILITY HELP DISPLAY PAGE 1 OF 2

COMMAND DESCRIPTION

L.END TERMINATE LIDOU

L.BACK FORMAT LID DISPLAY

L.HELP FORMAT HELP DISPLAY

L.+ PAGES THE L DISPLAY FORWARD

L.- PAGES THE L DISPLAY BACKWARD

L.OUT ROUTES A LISTING OF THE LID TABLE

L.PID,PA=XXZ SET ALL APPLICABLE PID ATTIBUTES

L.PID,NT=NNN,NA=Z ENABLE/DISABLE NETWORK STATUS

L.PID,LD=LID,LA=YYYZ ADD OR SET ATTRIBUTES FOR LID

L.PID,LD=LID,DELETE DELETE LID FROM LID TABLE

L.PID,MF=MAINFRM SET MAINFRAME DESCRIPTION

X CAN BE V OR N. Y CAN BE S, B, V OR N. Z CAN BE E OR D.

ALL COMMANDS CAN BE ENTERED FROM THE HELP OR LID DISPLAY

ENTER L.+ FOR MORE HELP (COMPLETE ATTRIBUTE DESCRIPTIONS)

Figure 9-3. LIDOU L Display (HELP) (Page 1 of 2)

9-4 NOS Version 2 Analysis Handbook Revision M

LIDOU L Display

L. LIDOU

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

LID OPERATOR UTILITY HELP DISPLAY

ATTRIBUTE DESCRIPTIONS

S - STORE AND FORWARD

B - LOOPBACK

V - VALIDATE

D - DISABLE

E - ENABLE

N - NOLIST

PARAMETER DESCRIPTIONS

PID IS THE THREE CHARACTER PHYSICAL IDENTIFIER

LID IS AN ASSOCIATED THREE CHARACTER LOGICAL IDENTIFIER

X CAN BE V OR N. Y CAN BE S, B, V OR N. Z CAN BE E OR D.

MAINFRM IS A (ONE TO SEVEN CHARACTER) MAINFRAME DESCRIPTOR

NNN IS THE NETWORK (-RHF*, *SSF*. *NAM*. OR *ALL*)

ENTER L.- TO SEE THE AVAILABLE COMMANDS

PAGE 2 OF 2

LIDS FOR LINKED PIDS ONLY.

LIDS FOR HOST PIDS ONLY.

NOTE THAT S AND B CANNOT

BE SPECIFIED TOGETHER AND

ALSO THAT E AND D CANNOT

BE SPECIFIED TOGETHER.

Figure 9-4. LIDOU L Display (HELP) (Page 2 of 2)

Although not shown in the HELP L display, some or all PID characteristics can be

changed with one command by including all of the appropriate parameters on that

command. For example, all PID characteristics for a given PID can be changed with

the following command:

L.pid,PA=xxz,MF=ma1nfrm,NT=nnn,NA=z.

Revision M L-Display Utilities 9-5

QDSPLAY L Display

The DSD utility QDSPLAY displays the content of a queued ﬁle listed in the queued

ﬁle table (QFT). ASCII8 ﬁles are automatically translated into display code before they

are displayed.

NOTE

On a secured system, the console must be in security unlock status to accept this

command (refer to the UNLOCK,username,password command in section 5).

To bring the QDSPLAY L display to the left console screen, enter the following

command:

QDSPLAY,jsn.

Parameter Description

jsn Job sequence name of the queued ﬁle you want to examine.

Figures 9-5 and 9-6 show the initial QDSPLAY L display.

/*^^S

9-6 NOS Version 2 Analysis Handbook Revision M

QDSPLAY L Display

.. QDSPLAY

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

QDSPLAY COMMANDS

COMMAND

PAGE 1 OF 2

DESCRIPTION

BACK RETURN FROM HELP/SS TO PREVIOUS FILE DISPLAY.

BOI POSITION FILE TO BOI AND DISPLAY DATA.

DAY POSITION PRINT FILE TO DAYFILE AND DISPLAY DATA.

DROP DROP FILE FROM QUEUE AND TERMINATE.

END RETURN FILE TO QUEUE AND TERMINATE.

HELP LIST L DISPLAY COMMANDS FOR QDSPLAY UTILITY.

LINE SET FILE DISPLAY MODE TO LINE FORMAT.

OCTAL SET FILE DISPLAY MODE TO OCTAL FORMAT.

SS DISPLAY IN OCTAL FROM BEGINNING OF SYSTEM SECTOR

+ PAGE DISPLAY FORWARD (CIRCULAR FOR SS).

PAGE DISPLAY BACKWARD.

Figure 9-5. Initial QDSPLAY L Display (Page 1 of 2)

QDSPLAY

hh.mm.ss. yy/mm/dd

QDSPLAY COMMANDS

CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

PAGE 2 OF 2

COMMAND

SL.N

SP.N

SR.N

SR,*

BR.N

DESCRIPTION

SKIP 1 LINE FORWARD IN FILE AND DISPLAY DATA.

SKIP N LINES FORWARD IN FILE AND DISPLAY DATA.

SKIP 1 PAGE FORWARD IN FILE AND DISPLAY DATA.

SKIP N PAGES FORWARD IN FILE AND DISPLAY DATA.

SKIP 1 RECORD FORWARD IN FILE AND DISPLAY DATA.

SKIP N RECORDS FORWARD IN FILE AND DISPLAY DATA.

SKIP TO EOI IN FILE.

SKIP 1 RECORD BACKWARD IN FILE AND DISPLAY DATA.

SKIP N RECORDS BACKWARD IN FILE AND DISPLAY DATA.

Figure 9-6. Initial QDSPLAY L Display (Page 2 of 2)

Revision M L-Display Utilities 9-7

QDSPLAY L Display

Valid QDSPLAY commands are:

Command Description

L.BACK Returns from the L.SS or L.HELP command displays and displays the

contents of the ﬁle at the current position. This command is ignored if

you are already displaying the contents of the ﬁle.

L.BOI Positions the ﬁle to the beginning-of-information (BOI) and then displays

the data from that point.

L.BR,r Skips backward r records in the ﬁle and displays the data from the

beginning of that record. If r is not speciﬁed, the default is one record.

The record count, r, is decimal unless a B postradix is speciﬁed to make

the number octal. The maximum skip count is 3777778. A BOI terminates

the skip regardless of the number of records speciﬁed or skipped. This

command is not allowed when the HELP display or system sector is being

displayed.

L.DAY Positions the print ﬁle to the beginning of the dayﬁle record and displays

the data from that point.

L.DROP Drops the ﬁle from the queue and terminates the QDSPLAY utility.

L.END Returns the ﬁle to the queue and terminates the QDSPLAY utility.

L.HELP Lists the L display commands for the QDSPLAY utility.

L.LINE Sets the ﬁle display mode to line format (not allowed when displaying the

system sector).

L.OCTAL Sets the ﬁle display mode to octal format.

L.SL,n Skips n lines forward in the ﬁle and displays the data from the beginning

of that line. If n is not speciﬁed, the default is one line. The line count,

n, is decimal unless a B postradix is speciﬁed to make the number octal.

The maximum skip count allowed is 377777s. A line can be up to 14

words. An EOR, EOF, or EOI ends the skip regardless of the number of

lines speciﬁepl or skipped. This command is allowed only when displaying

data in line format and is not allowed when the HELP display or system

sector is being displayed.

L.SP,p Skips forward p pages in the ﬁle and displays the data from the

beginning of that page. If p is not speciﬁed, the default is one page. The

page count, p, is decimal unless a B postradix is speciﬁed to make the

number octal. The maximum skip count allowed is 377777s. A page is

based on a carriage control 1 as the ﬁrst character in a line. An EOR,

EOF, or EOI ends the skip regardless of the number of pages speciﬁed or

skipped. The command is allowed only when displaying data in line

format and is not allowed when the HELP display or system sector is

being displayed.

a*^$k

9-8 NOS Version 2 Analysis Handbook Revision M

QDSPLAY L Display

Command Description

f^ L.SR,r Skips forward r records in the ﬁle and displays the data from the

beginning of that record. If r is not speciﬁed, the default is one record. If

an asterisk (*) is speciﬁed for the record count, r, a skip forward to the

EOI occurs. The record count, r, is decimal unless a B postradix is

speciﬁed making the number octal. The maximum skip count allowed is

3777778. An EOI ends the skip regardless of the number of records

speciﬁed or skipped. This command is not allowed when the HELP

display or system sector is being displayed.

L.SS Displays the data from the beginning of the system sector in octal display

code format.

L.+ Pages the QDSPLAY, HELP, or SS display forward.

L.- Pages the HELP display backward.

Revision M L-Display Utilities 9-9

SCTD L Display

The service class control table L-display (SCTD) utility displays the contents of the

service class control table (SCT). To begin the SCTD utility, enter one of the following

commands:

SCTD,L=out ﬁle,LO=opt1on.

SCTD,outﬁle,option.

Parameter Description

outﬁle Output ﬁle name. This parameter is valid only if a list option is

speciﬁed. The default outﬁle is ﬁle OUTPUT.

option List option. Enter one or more of the following:

option Description

D Formats the data for the DSD L display. This is the

default list option if the parameters outﬁle and option

are not speciﬁed.

L Formats the data for a line printer.

Figures 9-7 and 9-8 show the SCTD L display. This display shows the various origin

types along with the service classes valid for each origin type.

At the bottom of ﬁgure 9-8, the display shows how the service classes are prioritized.

The priority levels are PO through P7.

When entering the DSD CLASS or PCLASS command, use this utility to see the

service classes for each origin type or the priority level. Refer to the DSD CLASS or

PCLASS command in section 5 for further information.

Valid SCTD L display commands are:

Command Description

L.+ Pages the display forward.

L.- Pages the display backward.

L.END Terminates the SCTD utility.

,«^^H

-^•^V

9-10 NOS Version 2 Analysis Handbook Revision M

SCTD L Display

L. SCTD

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

SERVICE CLASS CONTROL TABLE DISPLAY PAGE 1 OF 2

SYSTEM ORIGIN SERVICE CLASSES

S Y B C R B T S D I N S S S M A C T 1 0

1 1 1 2 1 3

LOCAL BATCH ORIGIN SERVICE CLASSES

PRIORITY ORDERED SERVICE CLASSES

P O P 1 P 2 P 3 P 4 P 5 P 6 P 7

B C 1 0 1 1 1 2 1 3

Figure 9-7. SCTD L Display (Page 1 of 2)

L. SCTD

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

SERVICE CLASS CONTROL TABLE DISPLAY PAGE 2 OF 2

REMOTE BATCH ORIGIN TYPE SERVICE CLASSES

INTERACTIVE ORIGIN SERVICE CLASSES

T S D I

PRIORITY ORDERED SERVICE CLASSES

P O P 1 P 2 P 3 P 4 P 5 P 6 P 7

B C 1 0 1 1 1 2 1 3

Z*^ Figure 9-8. SCTD L Display (Page 2 of 2)

Revision M L-Display Utilities 9-11

SDSPLAY L Display

The SDSPLAY utility displays system values used to control job ﬂow. This utility

allows you to alter the queue priorities and weighting factors associated with the input,

output, and executing queues for each service class. Also, you can alter the service

limits associated with each service class. System values that you change using the

SDSPLAY utility are valid only until the next deadstart.

To initiate the SDSPLAY utility, enter one of the following commands:

SDSPLAY,L=out ﬁle,L0=opt1on.

SDSPLAY,out f11e,opt 1on.

Parameter Description

outﬁle

option

Output ﬁle name. This parameter is valid only if a list option of

L is speciﬁed. The default outﬁle is ﬁle OUTPUT.

List option. Enter one of the following:

option Description

D Formats the data for the DSD L display. This is the

default list option if the parameters outﬁle and option

are not speciﬁed.

L Formats the data for a line printer.

Figures 9-9 through 9-11 show the SDSPLAY-HELP display. This display consists of

three screens. The ﬁrst screen (ﬁgure 9-9) shows the display commands used to control

the utility. The other two screens (ﬁgures 9-10 and 9-11) show the service class

deﬁnitions.

L. SDSPLAY - HELP.

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

SDSPLAY COMMANDS PAGE 1 OF 3

COMMAND DESCRIPTION

CLASS=XX DISPLAY THE DATA FOR SERVICE CLASS XX.

END END THE PROCESSING OF SDSPLAY.

GO SET ALL ENTERED VALUES INTO THE SYSTEM.

HELP SHOW L DISPLAY COMMANDS FOR SDSPLAY.

RESET IGNORE ALL VALUES ENTERED SINCE THE

LAST GO.

+PAGE THE DISPLAY FORWARD.

—PAGE THE DISPLAY BACKWARD.

Figure 9-9. SDSPLAY-HELP Display (Page 1 of 3)

9-12 NOS Version 2 Analysis Handbook Revision M

SDSPLAY L Display

L. SDSPLAY - HELP

SERVICE CLASS DEFINITIONS PAGE 2 OF 3

SY SYSTEM SERVICE CLASS.

BC BATCH SERVICE CLASS.

RB REMOTE BATCH SERVICE CLASS.

TS INTERACTIVE SERVICE CLASS.

DI DETACHED INTERACTIVE SERVICE CLASS.

NS NETWORK SUPERVISOR SERVICE CLASS.

SS SUBSYSTEM SERVICE CLASS.

MA MAINTENANCE SERVICE CLASS.

CT COMMUNICATION TASK SERVICE CLASS.

Figure 9-10. SDSPLAY-HELP Display (Page 2 of 3)

0ms

L. SDSPLAY - HELP

SERVICE CLASS DEFINITIONS PAGE 3 Of 3

10 INSTALLATION CLASS 0.

11 INSTALLATION CLASS 1.

12 INSTALLATION CLASS 2.

13 INSTALLATION CLASS 3.

THE FOLLOWING SERVICE CLASSES MAY BE DEFINED AT YOUR SITE.

14 INSTALLATION CLASS 4.

15 INSTALLATION CLASS 5.

16 INSTALLATION CLASS 6.

17 INSTALLATION CLASS 7.

18 INSTALLATION CLASS 8.

19 INSTALLATION CLASS 9.

II INSTALLATION CLASS I.

IJ INSTALLATION CLASS J.

IK INSTALLATION CLASS K.

IL INSTALLATION CLASS L.

Figure 9-11. SDSPLAY-HELP Display (Page 3 of 3)

0ms.

Revision M L-Display Utilities 9-13

SDSPLAY L Display

Figures 9-12 through 9-14 show the SDSPLAY-CLASS display. This display consists of

three screens. The ﬁrst screen shows queue priorities and weighting factors associated r*m\.

with the speciﬁed service class. The other two screens show service limits associated 1

with the speciﬁed service class. For additional information concerning queue priorities

and service limits, refer to the QUEUE and SERVICE commands in section 5, DSD

Commands.

NOTE

The current number of jobs for the speciﬁed service class is displayed at the bottom

of the second screen. However, this value may not reﬂect the actual count of jobs

running under the displayed service class since the displayed value is updated only at

the initial processing of the SDSPLAY utility or after issuing an L.RESET or L.GO

command.

L. SDSPLAY - CLASS,

hh.mm.ss. yy/mm/dd. CDC NETWORK OPERATING SYSTEM

MID=AA NOS version

ATTRIBUTES FOR SERVICE CLASS SY PAGE 1 OF 3

OPTION DESCRIPTION

INPUT QUEUE LOWER PRIORITY.

INPUT QUEUE UPPER PRIORITY.

OUTPUT QUEUE LOWER PRIORITY.

OUTPUT QUEUE UPPER PRIORITY.

EXECUTION QUEUE LOWER PRIORITY.

EXECUTION QUEUE UPPER PRIORITY.

EXECUTION QUEUE INITIAL LOWER PRIORITY.

EXECUTION QUEUE INITIAL PRIORITY.

INITIAL TERMINAL PRIORITY AFTER I/O.

INPUT QUEUE WEIGHTING FACTOR.

EXECUTION QUEUE WEIGHTING FACTOR.

OUTPUT QUEUE WEIGHTING FACTOR.

(ALL PRIORITY VALUES ARE 0-7777B).

(WEIGHTING FACTORS MUST BE A POWER OF 2 FROM 1 - 4000B).

INLP 7770B

INUP 7776B

OTLP 7000B

OTUP 7776B

EXLP 2000B

EXUP 7000B

EXIL 4000B

EXIP 7000B

EXTP 4004B

INWF ss 1B

EXWF 1B

OTWF 1B

Figure 9-12. SDSPLAY-CLASS Display (Page 1 of 3)

9-14 NOS Version 2 Analysis Handbook Revision M

SDSPLAY L Display

0R0"S

L. SDSPLAY - CLASS.

ATTRIBUTES FOR SERVICE CLASS SY PAGE 2 OF 3

OPTION

CB = 3030B

DESCRIPTION

CT 100B

CM 20B

NJ 7777B

CP 6770B

FL 3777B

AM 77777B

SD 20B

EC 3777B

EM 7777B

GILB+G/UB CPU PRIORITY BOUNDS (2-77B).

LB = LOWER BOUND, UB = UPPER BOUND

CP TIME SLICE (1-7777B MILLISEC0NDS*64).

CM TIME SLICE (0-7777B SECONDS).

MAXIMUM NUMBER OF JOBS (0-7777B).

CONTROL POINT SLICE PRIORITY (0-7777B).

MAXIMUM CM FL/100B, ANY JOB (0-3765B).

MAXIMUM CM FL/100B, ALL JOBS (0-77777B)

CPU JOB SWITCH DELAY (1-7777B MSECS).

MAXIMUM EM FL/1000B, ANY JOB (0-7777B).

MAXIMUM EM LF/1000B, ALL JOBS (0-7777B)

THE CURRENT JOB COUNT FOR THIS SERVICE CLASS IS 2B.

Figure 9-13. SDSPLAY-CLASS Display (Page 2 of 3)

L. SDSPLAY - CLASS.

ATTRIBUTES F O R S E R V I C E C L A S S S Y P A G E 3 O F 3

INDEX VALUE DESCRIPTION

DS = 0 (UNLIMITED) MAXIMUM DIRECT ACCESS FILE

SIZE (0-7).

FC = 0 (UNLIMITED) MAXIMUM NUMBER PERMANENT

FILES (0-7).

CS = 0 (UNLIMITED) MAXIMUM PRUS, ALL INDIRECT

FILES (0-7).

FS = 0 (UNLIMITED) MAXIMUM PRUS, ONE INDIRECT

FILE (0-7).

OPTION DESCRIPTION

DT = DI SERVICE CLASS ON DETACH (NOT SS OR DS).

TD = 341B SUSPENSION TIMEOUT DELAY (0-7777B

SECONDS" MOB).

NOTE - EXTF>, DT AND TD ARE USED ONLY FOR INTERACTIVE JOBS.

Figure 9-14. SDSPLAY-CLASS Display (Page 3 of 3)

Revision M L-Display Utilities 9-15

SDSPLAY Commands

You can enter the following SDSPLAY commands from either the HELP or CLASS

displays and from any screen.

Command Description

L.CLASS=xx

L.END

L.HELP *

L.GO

L.RESET

L.+

L.-

Selects the CLASS display and displays the system attributes for the

service class speciﬁed by xx, where xx is any service class deﬁned on

the HELP display.

Terminates the SDSPLAY utility. If you made changes to system

attributes, the changes will not be saved unless you issue a L.GO

command before issuing the L.END command.

Selects the HELP display and displays the ﬁrst screen.

Saves any changes you made to system attributes using the

L.keyword=value commands by entering the new values into the

system tables. The current display remains on the screen.

Clears any changes you made to system attributes using the

L.keyword=value commands since you issued the last L.GO command.

The current display remains on the screen. Following a L.RESET

command, the CLASS displays will display values from the system

tables.

Pages the HELP or CLASS display forward to the next screen. When

the last screen is being displayed, the next L.+ command advances the

display to the ﬁrst screen. You cannot page from one display (CLASS

or HELP) to the other display; nor can you page from one service class

to another service class within the CLASS displays.

Pages the HELP or CLASS display backward to the previous screen.

When the ﬁrst screen is being displayed, this command has no effect on

the display.

"">

9-16 NOS Version 2 Analysis Handbook Revision M

Changing System Attributes

y#flfl$£Vv

Changing System Attributes

To alter system attributes using the SDSPLAY utility, you select the CLASS display

and specify the service class using the L.CLASS command. Next, you make changes

using L.keyword = value commands. The values you enter apply only to the service

class being displayed. You can enter multiple keyword=value pairs separated by

commas in one command line. The only limit is the L display input line length

restriction.

When you change system values, the updated values are displayed immediately if the

corresponding screen is being displayed. If your changes affect a different screen, the

updated values will be displayed when you page to that screen. Your changes will not

be entered into the system tables until you enter a L.GO command. You can cancel

your changes and reset the system values by entering a L.RESET command, but it

must be entered before you enter the L.GO command.

You can change system values for a different service class by entering another

L.CLASS command followed by more L.keyword=value commands.

Lukeyword = value Commands

You can enter the following L.keyword = value commands from a CLASS display.

Although you can enter numeric values in either octal or decimal, most of the values

are displayed in octal. Octal is the input default; you specify decimal values with a

radix of D following the number. Also, numbers containing an 8 or 9 are interpreted as

decimal.

The keywords are listed in the same order as they appear in the CLASS display.

Keyword Description

INLP Priority assigned to a ﬁle entering the input queue. The value of INLP

can range from 0 to 7777s, but must be less than the value of INUP.

INUP Highest priority a ﬁle can reach in the input queue; aging stops when

this priority is reached. The value of INUP can range from 0 to 7777s,

but must be greater than the value of INLP.

OTLP Priority assigned to a ﬁle entering the output queue. The value of OTLP

can range from 0 to 7777s, but must be less than the value of OTUP.

OTUP Highest priority a ﬁle can reach in the output queue; aging stops when

this priority is reached. The value of OTUP can range from 0 to 77778,

but must be greater than the value of OTLP.

EXLP Lowest priority for a job in the execution queue. The value of EXLP can

range from 0 to 7777s, but must be less than the value of EXUP.

Revision M L-Display Utilities 9-17

L.keyword=value Commands

| Keyword Description

| EXUP

J EXIL

j EXIP

| EXTP

| INWF

| EXWF

j OTWF

| A M

I CB

1 CM

I CP

j CS

I CT

I DS

I DT

1 EC

I EM

Priority assigned to a job when initially assigned to a control point.

However, online interactive jobs with terminal I/O available are scheduled

at EXTP priority. The value of EXUP can range from 0 to 7777s, but

must be greater than the value of EXLP.

Priority a job receives when it initially exhausts its CM or CPU time

slice. Thereafter, whenever the time slice is exhausted the job's priority

will be set to the value of EXLP. The value of EXIL can range from 0 to

77778, but must be in the range of values for EXLP and EXUP.

Initial scheduling priority for an executing job. The value of EXIP can

range from 0 to 7777s, but must be in the range of values for EXLP and

EXUP.

Terminal job priority assigned to a terminal job that has terminal input or

output available. The value of EXTP can range from 0 to 7777s, but must

be in the range of values for EXLP and EXUP.

Weighting factor for input queue priority calculations. The value of INWF

must be a power of 2 in the range 1 to 4000s.

Weighting factor for execution queue priority calculations. The value of

EXWF must be a power of 2 in the range 1 to 4000s.

Weighting factor for output queue priority calculations. The value of

OTWF must be a power of 2 in the range 1 to 4000s.

Maximum ﬁeld length divided by 100s for all jobs of the speciﬁed service

class. Refer to the SERVICE command in section 3.

Upper and lower bound CPU priorities. Refer to the SERVICE command

in section 3.

Central memory time slice in seconds. Refer to the SERVICE command in

section 3.

Control point slice priority. Refer to the SERVICE command in section 3.

Cumulative size in PRUs allowed for all indirect access permanent ﬁles.

Refer to the SERVICE command in section 3.

Control point time slice (seconds). Refer to the SERVICE command in

section 3.

Size in PRUs allowed for individual direct access permanent ﬁles. Refer to

the SERVICE command in section 3.

Service class to which a detached job is assigned. Refer to the SERVICE

command in section 3.

Maximum user-accessible extended 'memory ﬁeld length in words divided

by UEBS for any job of the speciﬁed service class. Refer to the SERVICE

command in section 3.

Maximum extended memory length in words divided by UEBS for all jobs

of the speciﬁed service class. Refer to the SERVICE command in section 3.

/***%.

9-18 NOS Version 2 Analysis Handbook Revision M

L.keyword=value Commands

Keyword Description

FC Number of permanent ﬁles allowed. Refer to the SERVICE command in

section 3.

FL Maximum ﬁeld length divided by' 100s for any job of the speciﬁed service

class. Refer to the SERVICE command in section 3.

FS Size in PRUs allowed for individual indirect access permanent ﬁles. Refer

to the SERVICE command in section 3.

NJ Maximum number of jobs. Refer to the SERVICE command in section 3.

SD CPU job switch delay (milliseconds). Refer to the SERVICE command in

section 3.

TD Suspension timeout delay. Refer to the SERVICE command in section 3.

Revision M L-Display Utilities 9-19

SUBSYST L Display

The SUBSYST L-display utility displays information about all the subsystems supported

by NOS. To begin the SUBSYST utility enter one of the following commands:

SUBSYST,L=out f11e,LO=opt1on.

SUBSYST,outﬁle,opt ion.

Parameter Description

outﬁle Output ﬁle name. This parameter is valid only if a list option is

speciﬁed. The default outﬁle is ﬁle OUTPUT.

option List option. Enter one of the following:

option Description

D Formats the data for the DSD L display. This is the

default list option if the parameters outﬁle and option

are not speciﬁed.

L Formats the data for a line printer.

If you do not specify outﬁle and option, the data is written to the L display.

Refer to the NOS Version 2 Operations Handbook for further information regarding

the SUBSYST L display.

9-20 NOS Version 2 Analysis Handbook Revision M

LID/RHF Conﬁguration Files 10

LID Conﬁguration File 10-1

Creating the LIDCMid File 10-1

NPID Statement 10-2

NLID Statement . ." 10-3

Equipment Conﬁguration Example 10-4

Specifying the Size of the LID Table 10-6

Building or Rebuilding the LID Table ....: 10-6

Listing the LID Table 10-7

Changing LIDS Using the LIDOU Display 10-7

Specifying Requirements for Running NOS/VE in Dual-State Mode 10-8

RHF Conﬁguration Files 10-9

RCFGEN Utility 10-9

RCFGEN Input 10-11

Statement Order 10-12

Network Conﬁguration Statements 10-13

LNDR Statement 10-13

APPL Statement 10-13

NPID Statement 10-15

PATH Statement , 10-15

LNAD Statement 10-16

RNAD Statement 10-18

DEBUG Statement 10-18

CHARGE Statement , 10-19

LCN Examples ... 10-20

QTF Conﬁguration Requirements 10-23

RHF Conﬁguration File Directives 10-23

NAM Conﬁguration File Directives 10-24

QTF Procedure File ... v. 10-25

Procedure File Example 10-26

jrffﬁP&V

LID/RHF Conﬁguration Files 10

This section contains information about LID conﬁguration ﬁles, RHF conﬁguration ﬁles,

and QTF conﬁguration requirements.

LID Conﬁguration File

You create a LID conﬁguration ﬁle to deﬁne the physical and logical machines

available for your system. After a level 0 deadstart, the system program CLDT

searches for the LID conﬁguration ﬁle on user index 377777s. It will use the

information in this ﬁle to generate a logical identiﬁer table (LDT) in central memory.

This section describes how to create and maintain the LID conﬁguration ﬁle

(LIDCMid). Speciﬁcally, how to:

• Create the LIDCMid file using NPID and NLID configuration statements.

• Specify the size of the LID table using the CMRDECK LDT entry.

• Build or rebuild the LID table using the X.CLDT command.

• List the current LID table using the LISTLID command.

• Change LIDs and LID attributes using the LIDOU display.

• Specify the requirements for running NOS/VE in dual-state mode.

Creating the LIDCMid File

The ﬁrst line of the LIDCMid ﬁle should be a 1- to 7-character name for the LID

conﬁguration ﬁle. Use this format for naming the ﬁle:

LIDCMid

Value Description

id 2-character machine identiﬁer for the mainframe.

You can place comment lines in the ﬁle by putting an asterisk in column 1 or by

putting the word COMMENT beginning in column one.

Example:

LIDCM64

•LIDCMID FILE FOR MAINFRAME 64.

You should deﬁne each PID (physical identiﬁer) and all the LIDs (logical identiﬁers)

associated with the PID by using NPID and NLID statements.

Revision M LID/RHF Conﬁguration Piles 10-1

NPID Statement

Use one NPID statement to deﬁne each physical mainframe available to your system.

Use this format for the NPID statement:

NPID,PID=p1d,ENABLED=statUS,MFTYPE=type,AT=at,NETDIS=SUb.

Parameter Description

PID=pid

ENABLED=status

MFTYPE=type

AT=at

NETDIS=sub

Physical identiﬁer; pid is the 3-character physical identiﬁer for

the mainframe. For NOS, pid should be Mid, where id is the

2-character machine identiﬁer deﬁned by the MID=id

CMRDECK entry. Required.

Speciﬁes whether the mainframe identiﬁed by PID is available.

You can enter YES or NO. Default is YES.

Mainframe type; type is any 1- to 7-character string specifying

the mainframe type or mainframe identiﬁer (for example,

NOSBE, NOS, CY200, NOSMF2, CLSH1). Required.

Speciﬁes the attribute associated with the mainframe identiﬁed

by PID. You can specify multiple attributes by using a slant

separator (AT=VALID/NLIST) or by repeating the parameter

(AT=VALID.AT=NLIST).

y<^KT^k

at Description

VALID

NVALID

NLIST

Speciﬁes that USER command prevalidation is

required. This parameter causes the local host to

verify that a user is authorized. This prevalidation

is performed in addition to the normal user

validation performed at the remote host.

Speciﬁes that USER command prevalidation is not

required.

Speciﬁes that this PID is not available to the end

user through the LISTLID command.

The default attributes are: USER command prevalidation is not

required and PID is listable using the LISTLID command.

Speciﬁes which network accesses to the mainframe identiﬁed by

PID are disabled. By default, all network accesses are enabled.

You should not specify NETDIS for the host PID. Any

combination of network accesses may be used (for example,

NETDIS=RHF/SSF).

sub Description

NAM

RHF

SSF

Network Access Method.

Remote Host Facility.

SCOPE Station Facility.

10-2 NOS Version 2 Analysis Handbook Revision M

NLID Statement

j^pKsy

NLID Statement

Use NLID statements after each NPID statement to deﬁne the relationship between

each PID and its associated LIDs and the attributes of the LIDs. Use this format for

the NLID statement:

NLID,LID=1i d,ENABLED=Stat US,AT=at

Parameter Description

LID=lid

ENABLED=status

AT=at

00$S

Logical identiﬁer; lid is the 3-character logical identiﬁer for the

mainframe identiﬁed by the last NPID statement. One lid entry

must be the same as the last pid. Required.

Speciﬁes whether the mainframe identiﬁed by LID is available.

You can enter YES or NO. Default is YES.

Speciﬁes the attribute associated with the mainframe identiﬁed

by LID. You can specify multiple attributes by using a slant

separator (AT=STOREF/VALID) or by repeating the parameter

(AT=STOREF,AT=VALID).

LOOPB

STOREF

VALID

NVALID

NLIST

Description

Loopback capability for RHF testing. This testing

can be performed on one mainframe with one

NAD. The attribute of LOOPB is valid only for

LIDs deﬁned for the host (local) mainframe.

You should not specify the LOOPB attribute if the

STOREF attribute is speciﬁed.

Store and forward capability indicating that the

speciﬁed mainframe will act as part of the

network in store and forward mode. Data will

pass through the machine and into the network.

You should not specify the STOREF attribute if

the LOOPB attribute is speciﬁed.

Speciﬁes that USER command prevalidation is

required and can be speciﬁed only if the STOREF

attribute is speciﬁed. This parameter causes the

mainframe deﬁned as being in store and forward

mode (AT=STOREF) to verify that a user is

authorized. This prevalidation is performed in

addition to the normal user validation performed

at the remote host that processes the job.

Speciﬁes that USER command prevalidation is not

required and can be speciﬁed only if the STOREF

attribute is speciﬁed.

Speciﬁes that this LID is not available to the end

user through the LISTLID command. LIDs with

this attribute might be reserved for onsite

maintenance purposes.

Revision M LID/RHF Conﬁguration Files 10-3

Equipment Conﬁguration Example

Figure 10-1 shows an equipment conﬁguration consisting of six mainframes. Figure

10-2 shows the LIDCMid ﬁles for the three NOS mainframes in the conﬁguration:

LIDCM64, LIDCM42, and LIDCM05.

SCOPE 2

PID-MF2

LID-SC2

LID-MF2

7683

C120

PID«120

LID-120

6683

2S5X NPU NETWORK

CDCNET NETWORK

NOS

MID-64

PID-M64

LID-M64

LID" AAA

LID-AAB

NOS

MID-42

PID-M42

LID=M42

LID=BBB

LID=BBC

NAD NAD

NOS

MID°05

PID-M05

LID«M05

LID»DDD

NAD NAD NOS/BE

PID=NBE

LID=NBE

LID=CCC

>*SS^v

Figure 10-1. Equipment Conﬁguration

10-4 NOS Version 2 Analysis Handbook Revision M

Equipment Conﬁguration Example

LIDCM64

•LIDCM64 LIDCMID FILE FOR MAINFRAME 64. NOTE THAT LID=LBK IS

• DEFINED AS A LOOPBACK LID AND THAT THE NAM NETWORK IS

DISABLED TO NOS PID=M42.

NPID,PID=M64,MFTYPE=NOSHOST,AT=VALID.

NLID.L1D=M64.

NLID,LID=AAA.

NLID.LID=AAB.

NLID.LID=LBK,AT=LOOPB.

NPID.PID=M42.MFTYPE=NOS42,NETDIS=NAM.

NLID.LID=M42.

NLID,LID=BBB.

NLID.LID=BBC.

NLID.LID=NBE,AT=STOREF/NLIST.

NLID,LID=CCC,AT=STOREF.

NPID,PID=MF2,MFTYPE=SC0PE2,AT=NVAL1D.

NLID,LID=MF2.

NLID,LID=SC2.

NPID.PID=120,MFTYPE=C120,AT=NVALID.

NLID,LID=120.

NPID,PID=M05,MFTYPE=NOS05.

NLID.LID-M05.

NL1D.LID=DDD.

NLID.LID=NBE,AT=STOREF.

NLID.LID=CCC,AT=STOREF.

LIDCM42

•LIDCM42 LIDCMID FILE FOR MAINFRAME 42. NOTE THAT LID=NBE IS

DEFINED AS A STORE AND FORWARD LID FOR THE NOS/BE MAINFRAME.

NPID.PID=M42,MFTYPE=NOSHOST.

NLID,LID=M42.

NLID.LID=BBB.

NLID.LID=BBC.

NLID,LID=M64,AT=STOREF.

NLID.LID=CCC.AT=STOREF.

NLID,LID=NBE.AT=STOREF.

NLID,LID=LBK.AT=LOOPB.

NPID,PID=M64,MFTYPE=N0S64,NETDIS=NAM.

NLID.LID=M64.

NLID,LID=AAA.

NLID.LID=AAB.

NPID.PID=NBE,MFTYPE=NOSBE,AT=NVALID.

NLID,LID=NBE.

NLID,LID=CCC.

NPID,PID=120,MFTYPE=C120.AT=NVALID.

NLID.LID=120.

LIDCM05

•LIDCM05 LIDCMID FILE FOR MAINFRAME 05. AGAIN, NOTE THAT

LID=NBE IS DEFINED AS A STORE AND FORWARD LID FOR THE

NOS/BE MAINFRAME.

NPID,PID=M05,MFTYPE=M05.

NLID.LID=M05.

NLID.LID=LBK,AT=LOOPB.

NLID,LID=M64,AT=STOREF.

NLID,LID=CCC.AT=STOREF.

NLID,LID=NBE,AT=STOREF.

NPID,P1D=NBE,MFTYPE=N0SBE,AT=NVALID.

NLID,LID=CCC.

NLID.LID=NBE.

NPID.PID=M64,MFTYPE=N0S64.

NLID.LID=M64.

Figure 10-2. LIDCMid Files for Example Equipment Conﬁguration

Revision M LID/RHF Conﬁguration Files 10-5

Specifying the Size of the LID Table

You should specify the size of the LID table (LDT) in central memory with this

CMRDECK entry:

LDT=nnnn.

Value Description

nnnn Number of central memory words allocated to the LDT

(default=1, minimum=0, maximum=11008).

You can calculate the value for nnnn by using this formula:

(3+lid)*p1d

Value Description

lid Maximum number of LIDs allowed per PID.

p i d To t a l n u m b e r o f P I D s i n a l l n e t w o r k s .

Building or Rebuilding the LID Table

When you deadstart the system, the system program CDLT automatically looks for the

LIDCMid ﬁle under user index 377777s. It will use the ﬁle to automatically create the

LID table in central memory.

If the system cannot ﬁnd the ﬁle, or if there are errors in the ﬁle, an error message ^^

will be displayed in the dayﬁle. You can then create or correct your LIDCMid. ﬁle ^)

and execute the X.CLDT command to build or rebuild the LID table in central

memory.

To build or rebuild the LID table in central memory, follow these steps:

1. Create or correct your LIDCMid ﬁle and ensure that the ﬁle is stored as an

indirect access ﬁle on user index 377777s.

2. Ensure that all subsystems that access the LID table (NAM, RHF, and SSF) are

not active.

3. Enter this command:

X.CLDT

The system will build or rebuild the LID table in central memory.

10-6 NOS Version 2 Analysis Handbook Revision M

Listing the LID Table

You can use the LISTLID command to list the conﬁguration of the LID table. The

LISTLID command can be executed through an interactive or batch environment. Use

this format for the command:

LISTLID,LID=lid.PID=p1d,L=l1st

Parameter Description

LID = lid

PID or PID=pid

L=list

List the attributes of PIDs in which LID lid exists. LID lid

must be enabled and it must be a listable LID. If the LID

keyword is speciﬁed, a LID value lid must also be speciﬁed. If

the LID keyword is omitted, the default is to list all listable

LIDs.

List the attributes of a given LID (LID=lid speciﬁed) or all

LIDs (LID omitted) under a given PID (PID=pid speciﬁed) or

all PIDs (PID speciﬁed). If the PID keyword is omitted, the

default is to list all listable LIDs speciﬁed by the LID

parameter.

Speciﬁes the local ﬁle to which all listable output is to be

written. If the L keyword is omitted, the default is ﬁle

OUTPUT.

Changing LIDS Using the LIDOU Display

You can change LIDs by using the LIDOU display. Refer to section 9, L-Display

Utilities, for information about the LIDOU display.

00S

Revision M UD/RHF Conﬁguration Files 10-7

Specifying Requirements for Running NOS/VE in Dual-State Mode

You must create the appropriate conﬁguration ﬁles before users can route batch jobs

from the NOS side of a dual-state system to the NOS/VE side.

You must deﬁne the LIDs for NOS/VE batch jobs in the LIDCMid ﬁle. You can

deﬁne a maximum of 30 NOS/VE LIDs.

Create the indirect access ﬁle LIDVEid (id is a 2-character machine identiﬁer for the

mainframe) on user index 377777s. The LIDVEid ﬁle is used by the dual-state IRHF

job. This ﬁle contains the LIDs to be used for NOS/VE batch jobs, in the following

format:

nvelld 1

nvelid 2

nvelld 30

Each LID must be on a separate line. ■'^\

In the LIDCMid ﬁle used by CLDT, you should use an NPID statement to deﬁne a

PID for the NOS/VE side of your dual-state system and include NLID statements for

each of the LIDs in the LIDVEid ﬁle. The PID value used is not signiﬁcant but must

be unique. In a QTF network environment, you should also deﬁne the NOS/VE LIDs

as store and forward (AT=STOREF) under the NOS host PID to allow storing and

forwarding of batch jobs from other hosts through the NOS side of a dual-state

system to the NOS/VE side.

10-8 NOS Version 2 Analysis Handbook Revision M

RHF Conﬁguration Files

You must deﬁne the RHF conﬁgurations including all NADs, applications, and physical

identiﬁers (PIDs) to be used by or accessible to RHF. You may deﬁne multiple

conﬁgurations. Each RHF in the network must have its own deﬁnition of the

conﬁguration; each deﬁnition will be different from the deﬁnition of the conﬁguration

for other RHFs in the network. Use the RCFGEN utility to deﬁne these conﬁgurations.

RCFGEN Utility

The RCFGEN utility reads network conﬁguration statements to create the RHF

conﬁguration record, which RHF uses to deﬁne the network and determine proper

access.

Each conﬁguration record has a name of one to seven letters or digits, starting with

a letter. The default name set by RCFGEN when creating a record (and used by RHF

when searching for a record) is RCFpid, where pid is the three-character physical

^^ identiﬁer of the host. You can use the RN=recname parameter on the RCFGEN and

RHF commands to specify a different conﬁguration record name.

When RHF is initiated, it searches for the speciﬁed conﬁguration record ﬁrst on local

ﬁle RCFILE, then on permanent ﬁle RCFpid (user name SYSTEMX, direct or indirect

access), and ﬁnally on the system library.

By default, RHF searches for conﬁguration record name RCFpid. If you provide a

subsystem procedure ﬁle for RHF, you can use the command RHF.RN = recname to

specify a different name. Refer to the NOS Version 2 Installation Handbook for

[^ information about creating a procedure ﬁle for RHF.

Revision M LID/RHF Conﬁguration Files 10-9

RCFGEN Utility

The following command calls the RCFGEN utility.

RCFGEN,1=i nput,L=1i st,O=out put,RN=recname,LO=opt i on.comment s

The parameters are order independent and optional; if omitted, the defaults are used.

P a r a m e t e r D e s c r i p t i o n

I=input

L=list

0=output

RN = recname

L0=option

comments

Speciﬁes the local ﬁle from which the network conﬁguration

statements are to be read. The default is ﬁle INPUT. If only the

keyword I is speciﬁed, ﬁle COMPILE is used.

Speciﬁes the local ﬁle to which all listable output is to be

written. The default is ﬁle OUTPUT.

Speciﬁes the local ﬁle to which the conﬁguration tables are to

be written. The default is ﬁle RCFILE.

Speciﬁes the name given to the conﬁguration record being

generated. The default name, RCFpid, is used if RN is omitted,

if only the keyword RN is speciﬁed, or if recname is 0.

Speciﬁes the list options used when generating listable output. If

LO is omitted or option is 0, only the network conﬁguration

statements and diagnostics are listed. If only the keyword LO is

speciﬁed, macro deﬁnitions and table-generating deﬁnitions are

also listed. LO=ALL speciﬁes all list options. (The non-default

list options may help analysis of RCFGEN or RHF problems.)

Speciﬁes an optional 1- to 70-character string that is placed in

the preﬁx table of the conﬁguration record being generated.

(During initiation, RHF displays the string in the job dayﬁle.)

10-10 NOS Version 2 Analysis Handbook Revision M

RCFGEN Input

The input to RCFGEN consists of network conﬁguration statements. The syntax of

these statements conforms to COMPASS statements. Conﬁguration statements (except

LNAD and RNAD statements) must not start before character position 3.

Use the following network conﬁguration statements.

S t a t e m e n t D e s c r i p t i o n

LNDR

APPL

NPID

PATH

LNAD

RNAD

DEBUG

CHARGE

Deﬁnes the maximum number of local NAD drivers (NDRs)

allowed to execute at one time.

Deﬁnes application programs that are allowed to access RHF.

Deﬁnes the physical identiﬁer of a remote mainframe.

Deﬁnes the paths to a remote mainframe through the LCN

network.

Deﬁnes information necessary to address local NADs.

Deﬁnes the addressing information necessary to access a

remote NAD.

Deﬁnes debug parameters.

Deﬁnes the charge that is transferred to a user control point

for each RHF call.

Revision M LID/RHF Conﬁguration Files 10-11

Statement Order

Network conﬁguration statements must be speciﬁed in the following order to properly

deﬁne a network.

1. LNDR statement (if necessary).

2. APPL statement to deﬁne application programs.

3. Sets of NPID and PATH statements that deﬁne all portions of the network. The

PATH statement must be associated with a given physical mainframe (NPID

statement). The following structure is required of these statements when deﬁning

a network:

a. NPID statement.

b. All paths (PATH statement) associated with the preceding NPID.

4. LNAD statements to deﬁne local NADs.

5. RNAD statements to deﬁne hardware addressing of remote NADs.

6. DEBUG and CHARGE statements (order independent).

At least one of each conﬁguration statement is required in the conﬁguration ﬁle with

the exceptions of the LNDR, DEBUG, and CHARGE statements. Defaults are speciﬁed

in the individual statement descriptions.

/*^%

10-12 NOS Version 2 Analysis Handbook Revision M

Network Conﬁguration Statements

The network conﬁguration statements are described in the order that they must be

speciﬁed to deﬁne a network.

LNDR Statement

The maximum number of NAD drivers is the maximum number of PPs that may

contain a NAD driver at one time, regardless of the number of NADs and the number

of drivers allowed per NAD.

To deﬁne the maximum NDRs allowed for all NADs, enter:

LNDR MAXNDRS=nn

P a r a m e t e r D e s c r i p t i o n

MAXNDRS=nn Maximum number of PPs that may contain NDRs at any one

time. Default is 4. The maximum value is the number of PPs

available for NAD drivers, nn must not be less than the

largest value speciﬁed for MAXNDRS on any LNAD statement.

0^s

APPL Statement

Each application deﬁnition uses additional RHF ﬁeld length. This additional ﬁeld

length is equal to mxcopys*(5 + 3*mxcons) central memory words (mxcopys and mxcons

are deﬁned below). To deﬁne an RHF application, enter:

APPL NAME=name,ENABLED=statUS,MXC0NS=mxcons,MXCOPYS=mxcopys,SVR=status,

ASTART=stat us,SYS0RG=st at us

Parameter

NAME=name

ENABLED=status

MXCONS=mxcons

MXCOPYS=mxcopys

Description

Application name; name is 1- to 7-alphanumeric characters,

where the ﬁrst character must be alphabetic. Required.

Availability of the application when RHF is initiated,

status may be YES or NO. Default is YES.

Maximum simultaneous connections that are allowed for

this application. The maximum value for mxcons is 127.

Default is 1.

Maximum number of simultaneously active copies of this

application that are allowed. The maximum value for

mxcopys is 127. Default is 1.

Revision M LID/RHF Conﬁguration Files 10-13

APPL Statement

Parameter

/ * ^ ! \

Description

ASTART=status

SYSORG=status

SVR=status Servicer program status; speciﬁes if the application is a

servicer, that is, an application program that is started

automatically by RHF upon request of an application

program on another host, status may be YES or NO.

Default is NO.

Application startup status; speciﬁes whether the application

is started when RHF is initiated or when the operator

enables the application, status may be YES or NO. If NO is

speciﬁed, the application is started by the user, or for a

servicer application, by a request from a remote application.

Default is NO.

System origin status; speciﬁes whether the application must

have privileged application status to gain access to RHF.

status may be YES or NO. If status is YES, the application

must have an SSJ= entry point and be loaded from the

system library. If status is NO, the application need not

reside on a system library; however, the user must have

CUCP (system control point) validation. Default is YES.

In determining the number of allowed connections and copies of an application, note

that each NAD has a maximum of 127 active connections. This number is restricted to

35 during NAD controlware loading but may be increased by modifying the appropriate

NAD controlware load parameters in LOADBC.

When deﬁning the APPL statements you must follow certain restrictions for

system-supplied applications QTF, QTFS, PTF, PTFS, and ITF. The maximum

simultaneous connections should be set to one (default) for QTFS, PTF, and PTFS.

The maximum number of simultaneously active copies should be set to one (default)

for QTF and ITF. The maximum simultaneous connections for QTF should be set to

four unless installation parameters ACNMAXC for QTF and MAXFILEXFR for FIP

are changed. The SVR=YES parameter must be speciﬁed for QTFS and PTFS and

must not be speciﬁed for QTF, PTF, or ITF. ASTART=YES should be speciﬁed for

QTF.

For ITF, the maximum simultaneous connections should be equal to the value of the

PI parameter on the ITF command in the JOBITF record on the NAMI startup

master ﬁle. The released default is 2.

The MHF application is deﬁned by default in RHF's conﬁguration ﬁle; no separate

deﬁnition is needed. If you choose to deﬁne MHF, you must specify ASTART=YES to

allow initial loading of controlware into local NADs.

■^^k

10-14 NOS Version 2 Analysis Handbook Revision M

NPID Statement

yms

NPID Statement

Each remote mainframe deﬁnition requires three words in RHF's ﬁeld length. To deﬁne

a remote mainframe, enter:

NPID PID=pid,ENABLED=statUS,MFTYPE=type

Parameter Description

PID=pid Physical identiﬁer of the remote mainframe; pid must be a

unique 3-character physical identiﬁer. Required.

ENABLED = status

MFTYPE=type

Availability of the mainframe identiﬁed by PID. status

may be YES or NO. Default is YES.

Mainframe type; type is any 1- to 7-character string

specifying the mainframe type or mainframe identiﬁer. For

example, you could use NOS1, NOS2, NOS/BE, or CY200.

Required.

PATH Statement

Every path deﬁned for a remote mainframe requires two words in RHF's ﬁeld length.

To deﬁne a path for a remote mainframe, enter:

PATH ENABLED=status,LT=tttt,RT=rrrr,RNAD=raddr,LNAD=laddr,AC=aaaa

Parameter Description

ENABLED = status

LT=tttt

RT = rrrr

Availability of the path when RHF is initialized, status

may be YES or NO. Default is YES.

Local trunk control units (TCUs) enabled; tttt is a 4-digit

nonzero binary number indicating the network trunk

connections for the local NAD. Required.

Remote TCUs enabled; rrrr is a 4-digit nonzero binary

number indicating the network trunk connections for the

remote NAD. Required.

Symbolic address of the remote NAD entry for this path

referenced in the RNAD statement. Required.

RNAD=raddr

LNAD=laddr

AC = aaaa

At least one PATH statement is required for each PID deﬁned

Symbolic address of the local NAD entry for this path

referenced in the LNAD statement. Required.

Access code; aaaa is a 4-digit hexadecimal access code for

the remote NAD. Default is 0.

jA&$&S

Revision M LID/RHF Conﬁguration Files 10-15

LNAD Statement

Each local NAD deﬁnition requires ﬁve words in RHF's ﬁeld length. To deﬁne a local

NAD, enter:

laddr LNAD CH=ch,MAXNDRS=n,DEDICATE=status,CMPATHS=nn,CMBUFFS=nn,

AUTODUMP=statUS,AUTOLOAD=StatUS,TRACE=t rw,LOG=statUS

Parameter Description

laddr

CH=ch

MAXNDRS=n

DEDICATE = status

CMPATHS=nn

CMBUFFS=nn

AUTODUMP=status

AUTOLOAD = status

Symbolic address referenced in a preceding PATH

statement. This parameter is required and must begin in

character position 1 or 2.

Channel number; ch is a 2-digit octal number of the

channel to which the NAD is connected.

Maximum number of NAD drivers that may be assigned at

one time to this NAD. n can be from 1 to 3. Default is 1.

Dedicated channel status; speciﬁes whether the driver will

always hold the NAD channel reservation between

consecutive blocks of one I/O request, status may be YES

or NO. Default is YES. YES should be speciﬁed unless

some non-CDC driver requires high-performance access to

the NAD channel.

Maximum number of convert mode paths. You can specify

a number from 0 to 63. The default is 0. If nn is 0, then

code conversion is done in the CPU. Refer to table 10-1

for additional information.

Maximum number of convert mode buffers. You can specify

a number from 0 to 63. The default is 0. Refer to table

10-1 for additional information.

Automatic dump status; speciﬁes whether RHF will

automatically dump the NAD's memory if the NAD fails,

status may be YES or NO. Default is NO. (YES is

intended for fault analysis along with

TRACE = YES/FULL).

Automatic load status; speciﬁes whether RHF will

automatically load the NAD's controlware during

initialization and reload the NAD if it fails, status may be

YES or NO. Default is YES.

10-16 NOS Version 2 Analysis Handbook Revision M

LNAD Statement

Parameter Description

TRACE=trw

00&S

LOG=status

NAD trace word; controls the value of trace word in the

NAD microcode initialization parameters used by MHF when

loading the NAD (refer to the 380-170 Network Access

Device Reference Manual for a complete description), trw

may be NO (no trace), YES (standard trace buffers), or

FULL (maximum trace buffers). Default is NO.

Because TRACE decreases NAD performance, TRACE should

be speciﬁed only for fault analysis. AUTODUMP=YES

should be speciﬁed along with TRACE = YES or FULL to

capture the NAD memory dump. TRACE = YES or FULL

produces a trace word value of 2954 or 529F hexadecimal,

respectively.

Error logging status; speciﬁes if local NAD trunk errors are

to be transmitted by MHF to the mainframe's binary

maintenance log. status may be YES or NO. Default is YES.

The software relies on the NAD to determine the actual number of paths and buffers

reserved for code conversion. The algorithm used by the NAD is:

Paths reserved = the smaller of (paths requested) or (buffers reserved * 2/3).

Buffers reserved = the smaller of (buffers requested) or (total buffers - 2).

The code conversion parameters should be adjusted according to the average number of

concurrently active connections doing code conversion and the average size of the ﬁles

being converted. Table 10-1 gives some suggested values for the code conversion

parameters. In this table, the term average connection implies four conversion

connections. The term average ﬁle size is a ﬁle approximately 500 PRUs long.

Table 10-1. Suggested Code Conversion Parameters

Connections 96K-Byte NAD 128K-Byte NAD

Less than average connections, average ﬁle size 2 paths

3 buffers

Less than average connections, large ﬁle sizes 2 paths

5 buffers

Average connections, average ﬁle sizes 4 paths

5 Duffers

Greater than average connections, average ﬁle 6 paths

sizes 9 buffers

Greater than average connections, large ﬁle 5 paths

sizes 9 buffers

2 paths

3 buffers

2 paths

5 Duffers

6 paths

8 buffers

7 paths

11 buffers

6 paths

13 buffers

Revision M LID/RHF Conﬁguration Files 10-17

RNAD Statement

Every remote NAD deﬁned requires four words in RHF's ﬁeld length. To deﬁne a

remote NAD, enter:

raddr RNAD ND=nn,DD=d,LOG=status

Parameter Description

,>e'^%\

raddr

ND=nn

DD=d

LOG=status

Symbolic address referenced in a preceding PATH

statement. This parameter is required and must begin in

character position 1 or 2.

Remote NAD address; nn is a 2-digit hexadecimal address

of the remote NAD. Default is 0.

Remote NAD exit port; d is a 1-digit hexadecimal address

of the exit port of the remote NAD. Default is 0.

Error logging status; speciﬁes if remote NAD trunk errors

are to be recorded by MHF in the mainframe's binary

maintenance log. status may be YES or NO. Default is

NO. Do not specify LOG=YES if the deﬁned NAD is a

local NAD (as in the case of a loopback path). Instead, use

the LOG parameter on the LNAD statement to control

error logging.

DEBUG Statement

The DEBUG statement controls the manner in which RHF uses queue entries.

DEBUG TRACE=statUS

Parameter

TRACE = status

Description

RHF trace status; speciﬁes whether the RHF trace is on

or off. status may be YES or NO. Default is NO. NO

speciﬁes that RHF trace is off and queue entries freed by

RHF are placed at the top of the empty queue and reused

immediately. RHF uses a slightly smaller amount of

processing time when the RHF trace is off. Default is NO.

YES speciﬁes that RHF trace is on and a queue entry is

reused only after all queue entries ahead of it have been

used. Also, when the RHF trace is on, you can use it for

analysis of an RHF dump and resolution of the associated

RHF problem.

T ^ S

10-18 NOS Version 2 Analysis Handbook Revision M

J<nlif^^\

CHARGE Statement

The CHARGE statement speciﬁes the amount of system resources a user control point

is charged for an RHF call. RHF distinguishes two different types of calls; those that

require a large amount of processing time and those that require a small amount of

processing time.

CHARGE TYPE=type,CPA=cpa,CPB=cpb,IO=io,CMFL=cm,PP=pp

P a r a m e t e r D e s c r i p t i o n

TYPE=type Type of call for which the charge is being speciﬁed.

Type Description

Requires small amount of RHF processing time.

Requires large amount of RHF processing time.

CPA=cpa

CPB = cpb

IO=io

CMFL=cm

PP=pp

Time to be charged for central processor cpa (decimal

milliseconds). Default is 2 milliseconds for a type 1 call; 10

milliseconds for a type 2 call.

Time to be charged for central processor cpb (decimal

milliseconds). Default is 0.

Input/output time to be charged (decimal milliseconds). Default

is 0.

Central memory ﬁeld length to be charged, cm is the ﬁeld

length in octal divided by 100s. Default is 10s.

PP time to be charged (decimal milliseconds). Default is 0.

A CHARGE statement is not required. You may, however, enter two CHARGE

statements: one for type 1 calls, a second for type 2 calls.

Revision M LID/RHF Conﬁguration Files 10-19

LCN Examples

Figure 10-3 illustrates a sample LCN network and is the basis of the network

conﬁguration statements shown in ﬁgures 10-4 and 10-5. Figure 10-4 shows the

network conﬁguration statements for the RHF on mainframe MFA. Figure 10-5 shows

the network conﬁguration statements for the RHF on mainframe MFD.

NAD

ND-01

TCU

0 11 12 13

B-

0 I 1 12 13

TCU

NAD

ND-3F

CH2

MFA

LIDs: MFA. AAA

NOS

CH6

CH7

NAD

ND-02

TCU

0 1 1 1 2 1 3

O i l 1 2 1 3

TCU

NAD

ND-4A

CH7

MFB MFC

LIDs: MFB, XYZ LIDs: MFC, BBB, CCC

NOS NOS

Maintenance Host

CH 11

CH20

NAD

ND-03

TCU

O| 1 [2 j 3

#-*

0 I 1 I 2 I 3

TCU

NAD

ND-62

CH3

NAD

ND-04

TCU

oLLLill

Oil 1 2 I 3

TCU

NAD

ND=11

CH4

MFD

LIDs: MFD, EEE, BBB,

DDD

NOS

Figure 10-3. Sample LCN Network

10-20 NOS Version 2 Analysis Handbook Revision M

LCN Examples

LNDR MAXNDRS=3

APPL NAME=QTF,MXC0NS=4,ASTART=YES

APPL NAME =QTFS,MXC0NS=1,MXCOPYS=8,ASTART=NO,SVR=YES

APPL NAME=PTF,MXCONS=1,MXCOPYS=8,ASTART=NO

APPL NAME=PTFS,MXC0PYS=8,SVR=YES

APPL NAME=USRAP,MXCONS=6,ENABLED=NO

APPL NAME=ITF,MXC0NS=2,MXCOPYS=1,ASTART=NO,SVR=NO

NPID PID=MFB,ENABLED=YES,MFTYPE=NOS2

PATH ENABLED=YES,LT=0110,RT=0110,RNAD=RN2,LNAD=LN1,AC=FOFO

PATH ENABLED=YES,LT=0001,RT=1000,RNAD=RN1,LNAD=LN2,AC=F0F0

PATH ENABLED=NO,LT=0001,RT=0100,RNAD=RN2,LNAD=LN3,AC=F0F0

PATH ENABLED=YES,LT=0100,RT=1000,RNAD=RN1,LNAD=LN3,AC=F0F0

NPID PID=MFC.MFTYPE=N0S2

PATH ENABLED=NO,LT=0110,RT=0110,RNAD=RN3,LNAD=LN1,AC=F0F0

PATH ENABLED=NO,LT=0001,RT=1000,RNAD=RN3,LNAD=LN2,AC=F0F0

PATH LT=0101,RT=1010,RNAD=RN3,LNAD=LN3,AC=F0F0

PATH LT=1000,RT=0001,RNAD=RN3,LNAD=LN4,AC=F0F0

NPID PID=MFD,MFTYPE=N0S2

PATH LT=0010,RT=0010,RNAD=RN4,LNAD=LN1,AC=F0F0

PATH LT=0001,RT=0010,RNAD=RN4,LNAD=LN3,AC=F0F0

LN1 LNAD CH=6

LN2 LNAD CH=7

LN3 LNAD CH=20

LN4 LNAD CH=11

RN1 RNAD DD=0,ND=4A,L0G=YES

RN2 RNAD DD=0,ND=3F,L0G=YES

RN3 RNAD DD=0,ND=62,L0G=YES

RN4 RNAD DD=0,ND=11,L0G=YES

Figure 10-4. Network Conﬁguration Statements for the RHF on Mainframe MFA

Revision M LID/RHF Conﬁguration Files 10-21

LCN Examples

APPL NAME=QTF,ASTART=YES,MXC0NS=4

APPL NAME=QTFS,MXC0PYS=4,SVR=YES

APPL NAME=PTF,MXCOPYS»6

APPL NAME=PTFS,MXC0PYS=6,SVR-YES

APPL NAME=ITF,MXC0NS=2,MXCOPYS=1,ASTART=NO,SVR=NO

NPID PID=MFA,MFTYPE=NOS2

PATH RNAD=RN01,LNAD=LN1,LT=0010,RT=0010,AC=F0F0

PATH RNAD=RN03, LNAD=LN 1, LT<=0010, RT=0001, AC=F0F0

NPID PID=MFB,MFTYPE=N0S2

PATH RNAD=RN3F,LNAD=LN1,LT=0010,RT=0100,AC=F0F0

NPID PID=MFC,MFTYPE=NOS2

PATH RNAD=RN62,LNAD=LN1,LT=0010,RT=0010,AC=F0F0

LN1 LNAD CH=4

RN01 RNAD ND=01,LOG=NO

RN03 RNAD ND=03,LOG=NO

RN3F RNAD ND=3F,L0G=N0

RN62 RNAD ND=62,L0G=N0

/^^^k

Figure 10-5. Network Conﬁguration Statements for the RHF on Mainframe MFD

|,^®K

>^^8\

10-22 NOS Version 2 Analysis Handbook Revision M

0$*S

008S,

0^S.

QTF Conﬁguration Requirements

You must create the appropriate conﬁguration ﬁles before you can perform queued ﬁle

transfers using RHF or NAM.

RHF Conﬁguration File Directives

For each copy of QTF that you want RHF to start automatically on a NOS host, you

must include the following network conﬁguration statement in the RCFGEN input ﬁle

for that host:

APPL NAME=QTF.MXC0NS=4,ASTART=YES

Refer to the APPL statement under RHF Conﬁguration Files earlier in this section.

Each copy of QTF will establish at most four simultaneous connections. For most

RHF conﬁgurations, a single copy of QTF should be sufﬁcient.

To allow a NOS host to receive queued ﬁles, you must include the following network

conﬁguration statement in the RCFGEN input ﬁle for that host:

APPL NAME=QTFS,SVR=YES,MXCOPYS=n

where n is the maximum number of QTFS servers you want to have active. Each copy

of QTFS services one connection.

Refer to the appropriate reference manuals for other RHF implementations for

information on how to conﬁgure non-NOS QTF and QTFS applications and

connections.

Revision M LID/RHF Conﬁguration Files 10-23

NAM Conﬁguration File Directives

In the NDLP input ﬁle, you must include QTFS OUTCALL statements for each PID to

which you want to transfer queued ﬁles and one APPL statement for QTF:

QTF: APPL,PRIV,PRU,NETXFR,MXCOPY=n.

where n is the maximum number of QTF initiators you want to have active.

Parameter records in the NAMI job record ﬁle NAMSTRT specify which applications

NAM automatically starts. Each default parameter record produced when QTF is

installed includes one JOB(JOBQTF,QT) directive. To automatically start more than

one copy of QTF, you can add additional job records to the NAMSTRT ﬁle and must

add additional JOB directives to each parameter record desired. Refer to the NAM5 -

Network Access Methods Version 1 subsection of the Special Product Installation

Information section of the NOS Version 2 Installation Handbook for more information

on NAMI job records.

Each copy of QTF will establish at most four simultaneous connections for ﬁle

transfers. For many NAM conﬁgurations, a single copy of QTF should be sufﬁcient.

You might consider using more than one copy of QTF if you expect a large number

of queued ﬁle transfers over slow communication lines.

To allow a NOS host to receive queued ﬁles, you must include in the NDLP input

ﬁle for that host one or more QTFS INCALL statements and one APPL statement for

QTFS:

QTFS: APPL,RS,PRIV,PRU,NETXFR,MXC0PY=n.

where n is the maximum number of QTFS servers you want to have active. Each copy

of QTFS services one connection. Refer to the Network Access Method Network

Deﬁnition Language Reference Manual for more information on APPL, INCALL, and

OUTCALL statements.

The default NAMI parameter records include a JOB(JOBQTFS,QS) directive. The

JOBQTFS job record creates or modiﬁes permanent ﬁle ZZQTFS, which contains the

job ﬁle submitted by NAM to start a new copy of QTFS when necessary.

10-24 NOS Version 2 Analysis Handbook Revision M

QTF Procedure File

The QTF procedure ﬁle contains a set of default initialization commands for each of

the two QTF variants, RHF and NAM. You can edit the commands in the procedure

and capture the modiﬁed procedure in a new system ﬁle without rebuilding QTF, or

you can include UPDATE modiﬁcations in the USER ﬁle for the RHP installation

procedure. The QTF procedure is deck QTFPROC on the RHPlpsrin program library

ﬁle (refer to the NOS Version 2 Installation Handbook).

For the NAM variant, another option is to modify the JOBQTF record (on ﬁle

NAMSTRT), which is used to start QTF at NAM initiation.

Format of the call to the QTF procedure ﬁle is:

QTF,variant,I=inﬂie.

Parameter Description

variant

inﬁle

RHF or NAM, identiﬁes which QTF variant to initiate. The

default is RHF.

File name from which initialization commands are to be

processed. Default is to use the commands for the variant

included in the QTF procedure ﬁle. QTF reads the

initialization commands until end-of-record. Each command on

the ﬁle is a separate line (equivalent to a K-display entry

without the K. preﬁx). Specifying I = inﬁle is equivalent to

entering the command K.INCLUDE,FILE = inﬁle.

00^S

/SS*V

Revision M LID/RHF Conﬁguration Files 10-25

Procedure File Example

Figure 10-6 shows some typical sets of initialization commands that you can include in

the QTF procedure ﬁle. The example assumes you started with the default QTF

procedure ﬁle.

A^m$\

,PROC,QTF*I, ...

DATA,XXDEFI

IF,$NAM$.EQ.$NO$.L1.

* DEFAULT DIRECTIVES FOR NOS RHF VARIANT

SCHED,MAXCONS=4.MAXIMUM OF FOUR CONNECTIONS.

CLASS,SC=A. DEFINE THREE SELECTION CLASSES

CLASS,SC=B. WITH IDENTICAL CRITERIA,

CLASS,SC=C. ALLOWS UP TO 3 CONNECTIONS PER PID.

.ELSE.L1.

^ •

.* DEFAULT DIRECTIVES FOR NOS NAM VARIANT

•

SCHED,MAXC0NS=4.

CLASS.SC=A,FSI=1..4.

CLASS,SC=B,FSI=1..6.

CLASS,SC=C,FSI=1..6.

CLASS, SC=D, FSI =7, MAXIMUMS. AT MOST ONE LARGE FILE.

DISABLE,SC=D,PID=M03. NEVER ALLOW LARGEST FILES TO M03.

.IF,(TIME.GT.0700).AND.(TIME.LT.1700),PRIME.

INCLUDE,F=PRIME. MUST BE LAST DIRECTIVE

.ELSE,PRIME.

INCLUDE,F«=OFF. MUST BE LAST DIRECTIVE

.ENDIF,PRIME.

.ENDIF,Ll.

.DATA,PRIME

. PRIME-HOURS PARAMETERS

. ENTER "K.INCLUDE,F=PRIME." TO CHANGE NAM PARAMETERS

DISABLE,SC=D. NO LARGE FILES DURING PRIME TIME.

ENABLE,SC=B,PID=M03. PERMIT MEDIUM FILES TO M03.

ENABLE,SC=C,PID=M03.

.DATA,OFF

. OFF-HOURS PARAMETERS

. ENTER "K.INCLUDE,F=OFF." TO CHANGE NAM PARAMETERS

ENABLE,SC=D. ALLOW LARGE FILES EXCEPT TO M03.

DISABLE,SC=B,PID=M03. DISALLOW MEDIUM FILES TO M03.

DISABLE,SC=C,PID=M03.

A^ms

■^^s

Figure 10-6. Procedure File Example

10-26 NOS Version 2 Analysis Handbook Revision M

Mass Storage Extended Subsystem (MSE) 11

Introduction 11-1

MSE Utilities 11-2

Utility Description Terminology .'. 11-3

Cartridge 11-3

S M M a p . . 1 1 - 3

7990 Catalog , 11-5

7990 Files 11-11

SSEXEC 11-12

Subfamily .... 11-12

Group .. 11-12

Volume 11-12

MSE Utilities 11-13

SSDEF 11-13

SSLABEL 11-14

SSLABEL Directives 11-15

OP=AS — Add SMs 11-16

OP=RS — Remove SMs 11-16

OP=AB — Add Cubicles :.. 11-16

OP=RB —. Remove Cubicles 11-16

OP-AM — Add Cartridges . . 11-17

OP=RM — Remove Cartridges 11-17

OP=RC — Restore Lost Cartridges 11-17

OP=FX — Fix Cartridge Labels 11-18

OP=IB — Control Cartridge Allocation 11-18

OP=FC — Free Cartridge Files 11-18

Parameters for SSLABEL Directives 11-19

SSLABEL Update Sequence .. 11-22

Restrictions to SSLABEL 11-23

SSLABEL Example 11-23

SSMOVE ■ . . . 1 1 - 2 4

SSMOVE Directives 11-27

Speciﬁc File (SF) Directives 11-27

Value Speciﬁer Directives 11-28

Excluding Destage/Release Processing 11-32

Selection Algorithms . 11-32

Destage and Release Algorithms ... 11-33

Decision Algorithm Hierarchy 11-34

Disk Space/Dump Tape Management 11-35

Restrictions to SSMOVE 11-35

Output Example 11-36

SSVAL 11-40

Release Processing .. 11-42

Problem Reporting 11-43

Error Detection and Classiﬁcation 11-43

SM Map Analysis 11-43

7990 Catalog Analysis 11-44

PFC Analysis 11-45

Release Processing and Problem Fixing 11-46

Validation Report 11-47

E r r o r T y p e s 1 a n d 2 1 1 - 4 7

E r r o r T y p e s 3 a n d 4 1 1 - 4 7

E r r o r T y p e s 5 , 6 , a n d 7 1 1 - 4 7

Intersections ,... ■. . . . - — • 1 1 - 4 8

Validation Report Example 11-48

Typical SSVAL Runs 11-49

Restrictions to SSVAL ..... .... H-50

SSUSE • H-51

Basic Usage Report ................ . 11-53

Optional Report A 11-56

Optional Report B , .. 11-57

Optional Report C ......■ . H - 5 8

Optional Report D ... 11-59

SSDEBUG H-60

SSDEBUG Directives 11-61

OP=RS.— Read AUs 11-61

OP=RF — Read File , 11-61

O P - R P — R e l e a s e F r o z e n C h a i n S p a c e , . . . . . . . 11 - 6 1

OP=RL — Remove Cartridge Entry From 7990 Catalog 1.1-61

OP=RC — Remove Cartridge Entry From SM Map 11-61

OP-CF — Change Flag in SM Map or 7990 Catalog 11-61

Parameters for the SSDEBUG Directives ... 11-62

Using SSDEBUG ............. 11-65

Restrictions to SSDEBUG ,. 11-66

y,z Relationship with SM Map Ordinal 11-66

Format of Disk Files Read From 7990 by SSDEBUG 11-67

7990 Stripe Header and Trailer 11-69

7990 Block Header and Trailer 11-70

SSBLD .. H-71

SSBLD Statements .... 11-71

Node/Path Linkage Statements 11-71

SM/CU Linkage Statements 11-73

SSBLD Example ., 11-73

7990 Equipment Conﬁguration Constraints 11-74

SSALTER 11-75

Console Input 11-75

MSE Operational Procedures .... 11-78

Initialization 11-78

Disk Space Management — 11-79

7990 Space Management 11-80

Backup and Recovery — 11-82

File Dumping ■...- ... 11-82

Full Dumps ,, 11-82

Incremental Dumps .. 11-83

Dump Tape Management 11-83

SM Map/7990 Catalog Backup 11-84

File Reloading ....... ............. .., 11-85

Error Conditions and Corrective Actions 11-86

C l e a r i n g P e r m a n e n t F i l e E r r o r F l a g s . . . . . . . . . . 1 1 - 8 6

Permanent File Recovery ...... 11-86

Cartridge Restoration and Reuse 11-87

Removal of Faulty or Missing Cartridges 11-88

7990 Catalog/SM Map/PFC Problems 11.-89

7990 Catalog/SM Map Mismatch . ..... 11-89

7990 Catalog Chain Problems 11-90

PFC/7990 Catalog Mismatch 11-90

Mass Storage Extended Subsystem (MSE) 11

Introduction

The Mass Storage Extended Subsystem (MSE) is the product consisting of the 7990

hardware, the channel interface, the diagnostics, and the operational software.

The 7990 hardware is a large-capacity online mass storage device, which is a

cost-effective extension to the disk ﬁle storage system and an alternative to

conventional magnetic tape storage. Storing ﬁles on the 7990 retains the security, data

integrity, and online access capabilities provided by disk and reduces the operational

and data integrity problems caused by storing, retrieving, and mounting tape volumes.

You can use both the 7990 and magnetic tapes to protect ﬁles from hardware and

system failures.

The 7990 hardware is composed of the following components:

Component Description

Cartridge

7991 Storage

Module (SM)

7990 Control

Unit (CU)

A plastic housing that encloses magnetic tape on which data is

stored under program control. This cartridge is compatible with an

IBM 3850 cartridge.

The hardware unit that houses up to 312 usable data cartridges.

The SM also consists of the cartridge accessor unit, which picks

cartridges from and puts cartridges in their cubicles, and one or

two data recording drives (DRD), which read data from and write

data to the cartridges. The cubicles are assigned coordinate

locations that are identiﬁed by the ordered pair (y,z), where the z

axis is horizontal and the y axis is vertical.

The hardware unit that is the controller for up to four 7991 SMs.

The CU acts like a tape controller for up to four IBM channel

interfaces and includes one or two data recording controllers

(DRCs). The MSE software supports up to two IBM channel

interfaces per CYBER mainframe.

Revision M Mass Storage Extended Subsystem (MSE) 11-1

MSE Utilities

The basic function of MSE is to store data on the 7990 and move it to disk upon

request for access by an authorized user. Control of ﬁle movement between a disk and

a 7990 is transparent to the applications programmer; however, there are operational

changes and additions that site personnel should note. These include modiﬁcations to

operational procedures in the areas of permanent ﬁle backup, permanent ﬁle recovery,

and disk space management. In addition, new MSE utilities are introduced. Detailed

information about the call and operation of each MSE utility, listed next, is included in

this section.

Utility

SSDEF

SSLABEL

SSMOVE

SSVAL

SSUSE

SSDEBUG

SSBLD

SSALTER

Description

Creates the system ﬁles (SM maps and 7990 catalogs) necessary for

MSE processing.

Manages allocation of cubicles and assignment of cartridges in

the SM.

Controls the destaging of ﬁles (creating 7990 images) and the

release of disk space.

Controls the release of 7990 space and analyzes the SM maps, 7990

catalogs, and PFC entries to identify and ﬂag discrepancies within

these three components.

Provides reports on the assignment and availability of cartridges

and cubicles within an SM.

Corrects error conditions detected by SSVAL and recovers data from

7990 cartridges.

Builds the unit device table (UDT) for SSEXEC. The UDT describes

the 7990 hardware conﬁguration.

Provides dynamic modiﬁcation of the 7990 hardware conﬁguration.

11-2 NOS Version 2 Analysis Handbook Revision M

0ms

Utility Description Terminology

Review the following terms; they are deﬁned as used in the descriptions of the MSE

utilities later in this section.

Cartridge

A cartridge is the 7990 storage component consisting of magnetic tape. The 7990

hardware always writes or reads 6652 8-bit byte records, called stripes, to or from the

magnetic tape. Each cartridge has 27,087 stripes. The ﬁrst 20 stripes contain a

manufacturer's label. The second 20 stripes contain a 7990 cartridge label written by

the SSLABEL utility. The remaining 27,047 stripes are organized into software-deﬁned

allocation units (AUs) of 14 stripes each for a total of 1931 AUs per cartridge.

An allocation unit is the smallest allocatable portion of a cartridge. One or more AUs

may be allocated to contain data for a single ﬁle. The allocated AUs for each ﬁle are

organized into volumes (deﬁnition follows), which are chained together in the 7990

catalog to identify the sequence of AUs that must be accessed in order to read a ﬁle.

A head-of-chain (HOC) ﬂag identiﬁes the ﬁrst volume in the chain, a link ﬁeld

identiﬁes the next volume in the chain, and an end-of-chain (EOC) ﬂag identiﬁes the

last volume in the chain.

The addition, removal, and reassignment of cartridges are managed by the MSE

utilities. The cartridge labels, and also the 7990 catalogs and SM maps (deﬁnitions

following), contain information concerning the location and content of the cartridges.

The descriptions of the MSE utilities later in this section contain further information

on cartridge management.

SM Map

An SM map is a direct access permanent ﬁle that contains information indicating how

cubicles in an SM are assigned to a family and identifying the cartridges that reside in

the SM. There is one SM map for each SM in the conﬁguration. The permanent ﬁle

name of the SM map is SMMAPi, where i is the SM identiﬁer (a letter from A to H);

its user index is 377760s; and its family is the default family on the mainframe on

which SSEXEC executes (refer to the deﬁnition of SSEXEC later in this section).

An SM map contains an entry for each possible coordinate pair (y,z) that identiﬁes a

cubicle in the SM, from (0,0), which is bottom left, to (21,15), which is upper right.

Certain coordinate pairs are reserved for customer engineering use, system use, and

diagnostic purposes.

Revision M Mass Storage Extended Subsystem (MSE) 11-3

SM Map

Each entry in an SM map has this format:

59 53 47 41

code ord familyname

6'///A sub

family id unused

cm csn

Field Description

code

ord

familyname

subfamily

Number from 0 to 6 indicating how the cubicle is assigned.

c o d e D e s c r i p t i o n

Unassigned.

Reserved for customer engineer.

Reserved for system use.

Reserved for a different SM map.

Assigned to the cartridge scratch pool.

Assigned to a subfamily.

No cubicle exists at these coordinates.

Ordinal for this cubicle in the 7990 catalog of the subfamily to

which the cubicle is assigned. This ordinal is referred to as the

FCT ordinal (refer to the deﬁnition of 7990 catalog, next). This

ﬁeld is meaningful only if code = 5.

The 7-character name, in 6-bit display code, of the family to which

the cubicle is assigned. This ﬁeld is meaningful only if code = 5.

Linkage error ﬂag that is set by the SSVAL utility when an SM

map entry of a cubicle assigned to a family has no corresponding

entry in the 7990 catalog.

Number from 0 to 7 identifying the subfamily to which the cubicle

is assigned. This ﬁeld is meaningful only if code = 5.

Letter from A to H identifying the SM.

Cartridge manufacturer code.

cm Description

A-

B-

Cartridge manufacturer is IBM.

Reserved for future cartridge manufacturer.

csn The 8-character cartridge serial number of the cartridge assigned to

the cubicle. If no cartridge is assigned, this ﬁeld contains spaces.

1. The pool is an area of the SM that stores scratch cartridges managed by the SSLABEL utility. . ^ ^ \

11-4 NOS Version 2 Analysis Handbook Revision M

y < ^ \

7990 Catalog

The zero entry in the SM map is the map header entry. In this entry the code ﬁeld is

6, the leftmost 6 bits of the second word contain the SM identiﬁer, and the remaining

bits contain the permanent ﬁle name of the SM map.

The SM map is updated whenever the SSLABEL, SSVAL, or SSDEBUG utility causes

a change in cubicle or cartridge assignment. It is recommended that the SM map be

backed up after every update to avoid problems such as:

• Mismatches between SM map and 7990 catalog entries

• Lost SM maps because of a disk failure or other problem

• Attempts to access cartridges that are no longer available

• Attempts to store cartridges in cubicles that are no longer empty

Thus, you should make a copy of the SM map on tape or another device or family in

order to retain the latest version of the SM map. If a device containing SM maps is

reloaded, the latest version of the SM maps should be recovered from the backup copy

after the reload is completed. After recovering the SM maps, you should run the

SSVAL utility to check that the entries in the SM maps and 7990 catalogs match. If

there are inconsistencies, corrective action should be taken as described under Error

Conditions and Corrective Actions later in this section.

7990 Catalog

A 7990 catalog is a disk-resident direct access permanent ﬁle that contains information

describing which allocation units of each cartridge assigned to a particular subfamily

are allocated to 7990 ﬁles and which AUs are available for allocation. There is one

7990 catalog for each subfamily of a family that can have 7990-resident ﬁles, and it

resides on the master device for the subfamily. The permanent ﬁle name of the 7990

catalog ﬁle is SFMCATi and its user index is 37776is, where i is the subfamily

identiﬁer (a number from 0 to 7). For example, ﬁle SFMCAT3 and user index 377763s

identify the 7990 catalog for subfamily 3.

A 7990 catalog is partitioned into subcatalogs, one subcatalog for each SM used by

the subfamily. The maximum number of subcatalogs in a 7990 catalog is eight (the

maximum number of SMs in a conﬁguration). Each subcatalog consists of two parts,

the ﬁle and cartridge table (FCT) and the allocation summary table (AST). The FCT

has an entry for each cartridge assigned to the subfamily from the given SM. The

maximum number of FCT entries in a subcatalog is 312 (the maximum number of

cartridges in an SM). The AST contains information used by the allocation algorithm

to select the cartridges on which a ﬁle will reside.

Revision M Mass Storage Extended Subsystem (MSE) 11-5

7990 Catalog

The ﬁrst part of the 7990 catalog contains a header and at most eight subcatalog

entries. The header identiﬁes the family and subfamily of the 7990 catalog. Each

subcatalog entry contains the length and location of its FCT and AST, the SM

identiﬁer, unallocated AUs available for small and large ﬁles, and the date and time of

the last SSVAL run that resulted in the release of 7990 space assigned to orphan ﬁles.

The format of the header is:

59 17 11 2 0

familyname sub

family W////A id

unused

Field Description

familyname The 7-character name, in 6-bit display code, of the family for this

7990 catalog.

subfamily Number from 0 to 7 identifying the subfamily for this 7990 catalog,

id Number identifying this ﬁle as a 7990 catalog ﬁle, id is set to 1.

The format of the subcatalog entry is:

59 53 47 41 135 26 17 11

id length FCT loc AST loc unused

unused date time

AUs small AUs large AUs group unused

Field Description

SM identiﬁer for the subcatalog (a letter from A to H).

Number of FCT (and AST) entries in the subcatalog.

Location (beginning PRU number) of the FCT.

Location (beginning PRU number) of the AST.

Date and time of the last releasing of orphan ﬁles on the SM

identiﬁed by id.

Total number of free AUs available for small ﬁles on all cartridges

in this SM.

Total number of free AUs available for large ﬁles on all cartridges

in this SM.

Maximum number of free AUs available within one group in

this SM.

length

FCT loc

AST loc

date time

AUs small

AUs large

AUs group

s*^S,

11-6 NOS Version 2 Analysis Handbook Revision M

7990 Catalog

Each cubicle from the given SM assigned to the subfamily has an entry in the FCT of

the subcatalog. This entry contains the coordinates (y,z) of the assigned cubicle. If a

cartridge has been assigned to the cubicle, the FCT entry also contains the csn of the

cartridge, usage information, status ﬂags, and information about each of the AUs of the

assigned cartridge. The format of each FCT entry is:

59 55 53 51 47 41 35 32 29 17 11 2 0

csn cm

yzﬁrst AU large ﬁrst AU small ﬂags ord

ocU ocl2 OCl.|

ﬁrst

stripe stripe pru max AUs cdp oclu

unused

unused ﬂawed AU

read stripe count write stripe count

soft read error soft write error

hard read error demarked stripe load count load error

cartridge link cartridge link

•

cartridge link cartridge link

Field Description

csn

ﬁrst AU large

ﬁrst AU small

The 8-character cartridge serial number, in 6-bit display code, of

the cartridge assigned to this cubicle. If no cartridge is assigned,

this ﬁeld contains spaces.

Cartridge manufacturer code.

cm Description

A - C a r t r i d g e ma n u f a c t u r e r is IB M .

B- Reserved for future cartridge manufacturer.

y coordinate of this cubicle.

z coordinate of this cubicle.

First free AU available for large ﬁles in this cartridge.

First free AU available for small ﬁles in this cartridge.

Revision M Mass Storage Extended Subsystem (MSE) 11-7

7990 Catalog

Field Description

ﬂags One of the following ﬂags:

Bit Description

17 Inhibit allocation ﬂag indicating that space from the

cartridge assigned to this cubicle is not to be allocated

to a ﬁle. This ﬂag is set or cleared by a directive to

the SSLABEL utility.

16 Lost cartridge ﬂag indicating that the cartridge assigned

to this cubicle was not there the last time SSEXEC

tried to pick it. This ﬂag can be cleared by a directive

to the SSLABEL utility.

15 Excessive write parity error ﬂag set by SSEXEC when

an error threshold has been passed. Space from this

cartridge is not to be allocated. This ﬂag can be cleared /tmk^

by a directive to the SSDEBUG utility. )

14 Linkage error ﬂag indicating that the SM map entry for

the location (y,z) is inconsistent with the FCT entry.

This ﬂag is set by the SSVAL utility and cleared by the

SSDEBUG utility.

13 Free cartridge ﬂag. When this ﬂag is set, SSEXEC

inhibits further allocation. Files can be removed by

running the SSVAL utility. This ﬂag is set or cleared

by a directive to the SSLABEL utility.

12 Reserved.

ord Ordinal of the FCT entry in the subcatalog.

ocli Off-cartridge link specifying the ordinal of the FCT entry for the

cubicle containing the next cartridge on which the ﬁle on the

cartridge assigned to this cubicle resides. If a ﬁle does reside on

multiple cartridges, the entire ﬁle must be contained within a ^s^

16-cartridge group. Also, since there are only three off-cartridge j

link ﬁelds, if a cartridge contains several ﬁles, only three can be

contained on other cartridges.

11-8 NOS Version 2 Analysis Handbook Revision M

7990 Catalog

Field

ﬁrst stripe

stripe

pru

max AUs

cdp

oclu

ﬂawed AU

read stripe count

write stripe

count

soft read error

soft write error

hard read error

demarked stripe

load count

load error

Description

First usable stripe following the cartridge label. This ﬁeld is set

to 40.

Number of stripes in each AU. This ﬁeld is set to 14.

Number of PRUs that can be written on each stripe. This ﬁeld is

set to 13.

Maximum number of AUs in a cartridge. This ﬁeld is set to 1931.

Cartridge division point between small and large ﬁles kept on the

same cartridge. The cartridge division point is deﬁned with the

B=n parameter to an SSLABEL directive. The default is B = 600,

specifying that the ﬁrst 600 AUs of a cartridge are available for

small ﬁles.

Off-cartridge link usage ﬁeld. When an ocli ﬁeld is used, a

corresponding bit is set in this ﬁeld. Bit 0 is set for ocli, bit 1 is

set for ock, and bit 2 is set for ocl3.

Number of ﬂawed AUs on this cartridge.

Accumulated number of stripes read from this cartridge.

Accumulated number of stripes written to this cartridge.

Accumulated number of soft read errors detected on this cartridge.

Accumulated number of soft write errors detected on this cartridge.

Accumulated number of hard read errors detected on this cartridge.

Accumulated number of stripes demarked on this cartridge.

Accumulated number of times this cartridge has been loaded for a

read/write operation.

Accumulated number of cartridge load errors detected using this

cartridge.

Revision M Mass Storage Extended Subsystem (MSE) 11-9

7990 Catalog

Field Description

j*S$S

cartridge link There is a 30-bit cartridge link ﬁeld for each AU in the cartridge. j

The bits are divided into the same ﬁelds and represent the same

information for each AU.

Bit(s) Description

29 Free/busy ﬂag indicating whether or not this AU is

currently allocated to a ﬁle.

28 Unused.

27 AU conﬂict ﬂag indicating an allocation conﬂict

involving this AU. This ﬂag is set by the SSVAL utility

or by SSEXEC.

26 Frozen chain ﬂag indicating a problem with this

allocation chain. This ﬂag is set by the SSVAL utility

or by SSEXEC. This AU is not reused until this ﬂag is '^

cleared by a directive to the SSDEBUG utility.

25 Start of fragment ﬂag indicating this AU is the

beginning of a chain fragment.

24 Flawed AU ﬂag indicating this AU is not to be

allocated to a ﬁle.

23 Continuation AU ﬂag indicating whether or not this AU

is the ﬁrst AU of a volume. If this ﬂag is clear, the "'^^

AU is the ﬁrst AU of the volume; if set, the AU is not

the ﬁrst AU of the volume.

22-21 Off-cartridge link ﬁeld indicating that the next volume

of the ﬁle is on another cartridge. The next cartridge is

identiﬁed by one of the ocli ﬁelds. The value of the

off-cartridge link ﬂag (1, 2, or 3) speciﬁes which ocli

ﬁeld to use. The link ﬁeld speciﬁes the number of the

ﬁrst AU on the next volume of the ﬁle. This ﬁeld is ^^

meaningful only if the chain control ﬁeld indicates that ]

this volume is the ﬁrst or middle volume of the ﬁle.

20-19 Chain control ﬁeld indicating whether this volume is

the ﬁrst (1), last (2), only (3), or middle (0) volume of

the ﬁle.

18-12 File length ﬁeld indicating the number of AUs in the

ﬁle following the ﬁrst accessed AU.

11-0 Link ﬁeld indicating the ﬁrst AU on the next volume of

the ﬁle. If the continuation AU ﬂag is clear, this ﬁeld

contains the ordinal of the ﬁrst AU of the volume. If

the continuation AU ﬂag is set, this ﬁeld contains the

ordinal of the ﬁrst AU within the volume.

11-10 NOS Version 2 Analysis Handbook Revision M

j^ms

7990 Files

The 7990 catalog is updated whenever the SSLABEL, SSMOVE, SSVAL, or SSDEBUG

utility causes a change in cartridge or cubicle assignment that affects the subfamily.

Because the 7990 catalog for a subfamily resides on its master device, it will be

backed up by PFDUMP whenever the master device for the subfamily is dumped.

Consequently, when PFLOAD reloads all ﬁles, the 7990 catalogs are automatically

recovered and all 7990-resident ﬁles as indicated in a recovered PFC will also have

entries in the recovered 7990 catalog. Thus, no special operational procedures are

needed to back up a 7990 catalog. It is possible, however, that the 7990 catalog will be

inconsistent with the SM maps or cartridge labels. If such inconsistencies do exist,

corrective action will have to be taken as described under Error Conditions and

Corrective Actions later in this section.

7990 Files

In an MSE environment, permanent ﬁles can be categorized according to whether or

not they reside on 7990. A disk ﬁle is a permanent ﬁle that resides on disk but not on

7990. A 7990 ﬁle is a permanent ﬁle that resides on 7990 and may or may not also

reside on disk, depending on how the disk space is managed (refer to Disk Space

Management later in this section). Depending on backup requirements (BR parameters),

both disk and 7990 ﬁles can also have backup images on tape (refer to the NOS

Version 2 Reference Set, Volume 3).

When a user deﬁnes a direct access ﬁle, initially it is a disk ﬁle. A disk ﬁle becomes

a 7990 ﬁle when it is destaged to 7990; that is, an image of the ﬁle is created on

7990. Files are destaged through use of the SSMOVE utility, which is run periodically

to manage disk space. When SSMOVE is run, ﬁles are destaged to 7990 and/or their

disk space released depending on certain ﬁle characteristics (refer to SSMOVE later

in this section). Thus, after an SSMOVE run a ﬁle can reside on disk, on 7990, or on

both. If the ﬁle does have a 7990 image, the asa ﬁeld in the PFC entry for the ﬁle

indicates the location of the 7990 copy.

When a user attaches a 7990 ﬁle, it is staged to disk from 7990 (that is, a disk

image is created) if the ﬁle is not on disk. If a direct access ﬁle is attached in write

mode, the AFOBS ﬂag is set in the PFC entry for that ﬁle. The current version of

the ﬁle resides on disk only. If an indirect access ﬁle is replaced or appended to, the

asa ﬁeld is cleared in the PFC entry for that ﬁle. Setting the AFOBS ﬂag or clearing

the asa in the PFC ensures that the current version of the ﬁle will be copied to 7990

when that ﬁle is again selected for being released from disk.

If the ﬁle is purged, its disk space, but not its 7990 space, is immediately released.

The SSVAL utility must be run to release 7990 space allocated to purged ﬁles. Thus,

because a purged ﬁle has no PFC entry linking to its 7990 catalog entry, the 7990

image that still exists before SSVAL is run is called an orphan ﬁle. However, a user

can never access an orphan ﬁle.

When a ﬁle is destaged to 7990, control information is written on each stripe to

which the ﬁle data is written. This information is sufﬁcient to identify the ﬁle.

Revision M Mass Storage Extended Subsystem (MSE) 11-11

SSEXEC

SSEXEC is the main processing program that is responsible for controlling MSE

activities. The MSE utilities issue requests to SSEXEC to destage ﬁles from disk to

7990, purge unneeded 7990 ﬁles, label or relabel cartridges, update SM maps and 7990

catalogs, and so forth. In a multimainframe environment, two versions of this program

exist: the mainframe to which the 7990 device is physically connected (the master

mainframe) has a program called SSEXEC, and all other mainframes (the slave

mainframes) have a program called SSSLV. Refer to section 13, Multimainframe

Operations, for more information.

Subfamily

Each permanent ﬁle family consists of eight subfamilies, subfamily 0 through

subfamily 7. The lower 3 bits of the user index identify the subfamily to which a user

belongs. For example, a user whose index ends in 3 (or Oil in bit notation) belongs to

subfamily 3. When the SSDEF utility is run to create MSE system ﬁles, the SM maps

are created on the master device under user index 377760s (subfamily 0) and one 7990

catalog is created under each user index 37776is (subfamily i), i=0,1,2,...,7. When the

SSLABEL utility is used to assign an SM, cubicle, or cartridge to a family, it is

possible to specify assignment to particular subfamilies of the family.

Group

A group is a software-deﬁned structure for allocating cartridges within a subfamily. A

group of cartridges is a number of cartridges (up to 16) in which ﬁles belonging to a

subfamily can reside. Any ﬁle can overﬂow from one cartridge in a group to any other

cartridge in the same group. However, no ﬁle can overﬂow from a cartridge in one

group to any cartridge in another group. Cartridges are placed in or removed from

speciﬁc groups by using the SSLABEL utility. The SSVAL utility compares the 7990

catalogs (for each subfamily) with the same information obtained from the PFC entries.

Volume

A volume is a software-deﬁned structure used to simplify the allocation and access of

large 7990 ﬁles. Up to 128 consecutive AUs in a cartridge that have the same status

(either allocated or hot allocated) are considered to belong to the same volume. These

consecutive AUs are composed of stripes that all have the same volume serial number

(a physical identiﬁer that is written to the tape on the cartridge). The MSE software

ﬁlls cartridges (unallocated volumes) from the beginning to the end with no rewinding.

This action minimizes repositioning of the tape in a cartridge for multivolume ﬁles.

When a ﬁle is staged back to disk, the control information on each stripe is veriﬁed.

If a discrepancy is detected, an error message is issued, the ﬁle stage is aborted, an

error ﬂag is set in the PFC entry to indicate that the 7990 ﬁle could not be accessed,

and the AU conﬂict ﬂag is set in the 7990 catalog entry for the particular

cartridge(s) and AU(s).

<^ES3s

11-12 NOS Version 2 Analysis Handbook Revision M

0^\.

MSE Utilities

MSE utilities are described next. All of these utilities must be run from system origin

jobs. In a multimainframe environment, the SSDEF, SSUSE, and SSBLD utilities can

be run on any mainframe that has access to the family being processed. However, the

remaining utilities must be run on the mainframe on which SSEXEC executes.

SSDEF

SSDEF creates the system ﬁles (SM maps and 7990 catalogs) that are necessary for

MSE processing. If an SM is added to the 7990 hardware conﬁguration, SSDEF is used

to create the SM map for that SM. If a family is to be permitted to have 7990-resident

ﬁles, SSDEF is used to create the eight 7990 catalogs for that family (one catalog for

each subfamily).

NOTE

If the SM parameter is speciﬁed, SSDEF will create an SM map for the speciﬁed SM

(refer to the deﬁnition of SM map earlier in this section). It is recommended that the

SM map be copied on tape or on another device or family immediately after it is

created.

The format of the SSDEF command is:

SSDEF,Pi,P2-

pi Description

FM=familyname Family for which 7990 catalogs are to be created, one catalog for

each subfamily.

FM Same as FM = system default family.

FM omitted No 7990 catalogs are to be created. SM = id or SM must be

speciﬁed.

SM=id SM identiﬁer of the SM for which an SM map is to be created; id

is a letter from A to H.

SM Same as SM=A.

SM omitted No SM map is to be created. FM = familyname or FM must be

speciﬁed.

Example:

SSDEF,SM=B.

SMMAPB, the SM map for SM B, is created, and its entries are as described in the

previous deﬁnition of SM map earlier in this section. However, since SSDEF does not

assign cubicles, the entries for cubicles available for use initially indicate that the

cubicles are unassigned.

Revision M Mass Storage Extended Subsystem (MSE) 11-13

SSLABEL

SSLABEL manages cartridge assignment and cubicle allocation in an SM. The following

functions are performed through use of directives to SSLABEL.

Add an SM to a subfamily (AS directive).

Remove an SM from a subfamily (RS directive).

Add cubicles to a subfamily, the pool, or the reserved area (AB directive).

Remove cubicles from a subfamily, the pool, or the reserved area (RB directive).

Add cartridges to a subfamily or the pool (AM directive).

Remove cartridges from a subfamily or the pool (RM directive).

Restore an abnormally removed cartridge (RC directive).

Repair a cartridge label or overwrite a family label (FX directive).

Inhibit or allow further allocation of ﬁles to a cartridge (IB directive).

Initiate freeing all ﬁles from an existing cartridge (FC directive).

Input to SSLABEL is via a directive ﬁle. SSLABEL reads the appropriate SM maps

and 7990 catalogs to determine how to process each directive and then issues requests

to SSEXEC to read and/or write cartridge labels and to update the SM maps and 7990

catalogs. SSLABEL generates a report detailing the action taken for each input ^^

directive. If the assignment information or cartridge label is not appropriate or conﬂicts j

with data in the SM map or 7990 catalog, the cartridge label information is included

on this report and the cartridge is put into the exit tray. It may be possible to restore

such a cartridge, as described under Cartridge Restoration and Reuse later in this

section.

NOTE

SSLABEL updates the SM map for the speciﬁed SM (refer to the deﬁnition of SM map

earlier in this section). It is recommended that the SM map be copied on tape or on

another device or family immediately after each update. i

11-14 NOS Version 2 Analysis Handbook Revision M

SSLABEL Directives

The format of the SSLABEL command is:

SSLABEL,Pi,P2.

pi Description

I=ﬁlename File containing the directives to SSLABEL.

I S a m e a s I = I N P U T .

I o m i t t e d S a m e a s I = I N P U T.

L=ﬁlename File on which listable output is to be written.

L Same as L=OUTPUT.

L = 0 N o o u t p u t ﬁ l e i s t o b e g e n e r a t e d .

L o m i t t e d Sa m e a s L = O U T P U T.

V Z Directives are contained on the SSLABEL command. The I

parameter is ignored.

Z omitted Directives are contained on the ﬁle speciﬁed by the I parameter.

SSLABEL Directives

The directives to SSLABEL can be speciﬁed on a separate ﬁle (speciﬁed by the I

parameter) or after the SSLABEL command (Z speciﬁed). If on the input ﬁle, each

^ directive must be speciﬁed on a separate line using the OP=directive option.

Parameters for a directive are on the same line, are separated by commas, and end

with a period:

0P=directive,Pi,P2,...,pn.

Example 1:

SSLABEL,I=DIRFILE.

/^ DIRFILE is the directive ﬁle; it contains the following directives:

0P=AM,N=4,PK=D.

OP=RM,CN=44455566,FM,SB=1,PK=F.

Two directives to SSLABEL are speciﬁed. OP=AM adds four cartridges to the pool of

SM A (SM parameter not speciﬁed). OP=RM removes the cartridge whose csn is

44455566 from subfamily 1 of the default family. (Refer to the following descriptions of

directives and parameters.)

If the directives are contained on the command, they follow the command terminator.

The ﬁrst character following the terminator is the separator. Any character that does

not appear in any of the directives can be used as the separator character. Each

directive must be preceded by the separator and terminated by a period.

Revision M Mass Storage Extended Subsystem (MSE) 11-15

SSLABEL Directives

Example 2:

SSLABEL,Z./0P=AM,N=4,PK=D./0P=RM,CN=44455566,FM,SB=1,PK=F.

The slant is used as the separator. This command performs the same functions as those

in the preceding example.

The following directives are available with SSLABEL. Some of these directives cause

cartridges to be physically moved to and from cubicles and the input or exit tray

(refer to OP=AM, OP=RM, OP=RC, and OP=FX in the following paragraphs). The

remaining directives cause only logical operations to occur, updating the MSE system

ﬁles. Descriptions of the parameters for these directives follow the directives.

OP=AS — Add SMs

OP=AS adds an SM to a subfamily. The 7990 catalog for the speciﬁed subfamily is

updated to reﬂect that cartridges and permanent ﬁles for the subfamily can reside on

the speciﬁed SM. This directive, however, does not manipulate cartridges or cubicles.

OP=RS — Remove SMs

OP = RS removes an SM from a subfamily. The 7990 catalog for the speciﬁed subfamily

is updated to reﬂect that cartridges and permanent ﬁles for the subfamily cannot

reside on the speciﬁed SM. Before OP=RS can be speciﬁed, all cubicles in the speciﬁed

SM must have been removed previously from the family (refer to the OP=RB

directive). This directive, however, does not manipulate cartridges or cubicles.

OP=AB — Add Cubicles

OP=AB adds an unassigned cubicle within an SM to a subfamily (PT=F), the pool

(PT=P), or the reserved area of the SM (PT=R). More than one cubicle (N = n) can be

added at a time. Speciﬁc cubicles (YI and Zl parameters) can be added, but they must

be currently unassigned. For PT=R, YI and Zl must be used to add multiple cubicles;

N=n is not valid. The SM map is updated to reﬂect the new assignment of cubicles.

When PT=R is speciﬁed, the cubicle is reserved to a different SM map. Cubicles

reserved for system use, storage module maintenance, or customer engineering use are

reserved at compile time. Cubicles assigned to the pool are not assigned to a family or

subfamily.

OP=RB — Remove Cubicles

OP=RB removes an assigned empty cubicle from a subfamily (PK=F), the pool

(PK=P), or the reserved area of the SM (PK=R). More than one cubicle (N = n) can be

removed at a time. SSLABEL reads the SM map and selects the ﬁrst empty cubicle

assigned to the subfamily, pool, or reserved area to be removed. Speciﬁc cubicles (YI

and Zl parameters) can be removed, but they must be empty. The SM map is updated

to reﬂect that the cubicles are unassigned.

11-16 NOS Version 2 Analysis Handbook Revision M

0ms

SSLABEL Directives

OP=AM — Add Cartridges

OP=AM adds a cartridge to either a speciﬁc subfamily (PT=F) or the pool (PT = P).

SSLABEL selects the ﬁrst empty cubicle assigned to the subfamily or pool as the new

location for the cartridge. More than one cartridge (N=n) can be added at a time or a

speciﬁc cartridge (CN=csn) from the pool can be added to a subfamily. The SM map,

7990 catalog, and cartridge label are updated to reﬂect the new assignment of the

cartridge. Cartridges added to a subfamily may be added to a speciﬁc group using the

GR=n parameter. If cartridges are added to the pool, the GR=n parameter is not

valid. The small and large boundaries for ﬁle size are speciﬁed by the B=n parameter

with the OP=AM option. Default is B = 600, implying that the ﬁrst 600 AUs of the

cartridge are available for small ﬁles. Cartridges added to the pool are not assigned to

a family or subfamily. A customer engineer cartridge can be stored into location (0,0)

or (0,15) by using the CC = loc parameter.

OP=RM — Remove Cartridges

OP=RM either removes an empty cartridge from a subfamily (PK = F) and moves the

cartridge to the pool (PT=P) or it moves any cartridge from the pool (PK=P) and

moves it to the exit tray (PT=D). Any cartridge currently assigned to a subfamily

cannot be removed unless it is empty; that is, all AUs must be unallocated. To remove

a cartridge that is not empty, ﬁrst use the OP = FC directive to free up the cartridge.

If the cartridge speciﬁed by the CN = csn parameter is lost (does not reside in its

assigned cubicle) and if the cartridge is assigned to a family, the LT parameter should

be speciﬁed. This allows the appropriate entries in the SM map and 7990 catalog to be

deleted even though the cartridge is not available to have its label updated. If LT is

not speciﬁed, an error message is issued and SSLABEL aborts. SSLABEL will not

update the SM map and 7990 catalog with the LT parameter unless the lost cartridge

ﬂag is set in the 7990 catalog. SSLABEL will set the lost cartridge ﬂag if SSLABEL is

run to remove a cartridge that is not present.

If the cartridge speciﬁed by the CN=csn parameter is lost and it is in the pool, then

the LT parameter should not be used. Rather, SSDEBUG must be run to remove the

SM map entry and SSLABEL must be run to reassign the cartridge to the pool.

A speciﬁed number (N = n) of cartridges may be removed from a group of a subfamily

by using the GR=n parameter.

OP=RC — Restore Lost Cartridges

OP=RC restores to its proper cubicle a cartridge that was inadvertently removed from

an SM. If restoration is successful, the lost ﬂag in the 7990 catalog is cleared. If data

recorded on the cartridge label does not agree with the information in the 7990 catalog

and the SM map entry for the cubicle to which the cartridge is to be restored, the

cartridge label information is reported and the cartridge is put into the exit tray for

use in further processing of the cartridge.

NOTE

Not more than one lost cartridge can be restored at a time using the OP = RC

directive. Also, since customer engineer cartridges do not exist in the SM map,

OP=RC will not restore a cartridge to a customer engineer cartridge cubicle.

Revision M Mass Storage Extended Subsystem (MSE) 11-17

SSLABEL Directives

OP=FX— Fix Cartridge Labels

OP=FX writes a scratch label on a cartridge identiﬁed by the CN=csn parameter and

adds the cartridge to the pool. This directive is intended for use when a cartridge label

has been destroyed, but the cartridge itself is not physically damaged and can be

reused. It can also be used when a cartridge with a family label is to be assigned to a

different subfamily using the OP=AM directive, but it is not feasible to ﬁrst remove

the cartridge normally using the OP=RM directive. For example, if SSLABEL is run

to add a cartridge to a subfamily, and a system failure occurs before the 7990 catalog

and SM map are updated but after the cartridge is relabeled, then the cartridge label

does not match the corresponding entries in the 7990 catalog and SM map. Hence,

OP=RM cannot be used to remove the cartridge from the subfamily, but OP=FX can

be used to rewrite the cartridge label and then OP=AM can be used to add the

cartridge to a subfamily. However, if a family label is to be overwritten, the

FM=familyname and SB = subfamily parameters must identify the family and subfamily

to which the cartridge was assigned. The SM map and 7990 catalog are updated to

reﬂect the new cartridge label.

OP=RM cannot be used with the LT parameter to remove pool cartridges. SSLABEL

sets the SM map error ﬂag because no 7990 catalog entries exist for pool cartridges.

Use SSDEBUG to remove pool cartridges.

OP=IB — Control Cartridge Allocation

OP=IB sets or clears the inhibit allocation ﬂag in the 7990 catalog entry for the

speciﬁc cartridge (CN=csn). If the ﬂag is set (ON), SSEXEC does not allocate new

7990 ﬁles to this cartridge. If the ﬂag is cleared (OF), allocation of ﬁles to this

cartridge is enabled.

OP=FC — Free Cartridge Files

OP=FC sets or clears the free cartridge ﬂag in the 7990 catalog entry for the speciﬁed

cartridge (CN = csn). If the ﬂag is set (ON), SSEXEC will inhibit further allocation.

Files can be removed by running the SSVAL utility. Refer to the free cartridge

discussion under SSVAL later in this section. If the SM map entry is missing,

SSDEBUG must be used to set the free cartridge ﬂag. Refer to the discussion under

SSDEBUG later in this section.

y^^.

^^\

11-18 NOS Version 2 Analysis Handbook Revision M

Parameters for SSLABEL Directives

The descriptions of the parameters to the SSLABEL directives follow. Not all

parameters are valid with all directives, as indicated.

Pi Description

B=n

B omitted

CC=loc

CM=A-

CM omitted

CN=csn

CN omitted

Number of AUs on this cartridge to be used to store small ﬁles.

The remaining AUs (1931-n) are reserved for large ﬁles. The

distinction between small and large ﬁles is controlled by an

assembly parameter. B=n is valid only with OP=AM.

Same as B=600.

Location into which the cartridge is to be stored.

loc Description

A Store cartridge into location (0,0).

B Store cartridge into location (0,1 5).

Any value of CC other than A or B is not accepted.

Cartridge manufacturer code is A-, indicating IBM.

Same as CM=A-.

Cartridge serial number of the cartridge to be added, removed, or

repaired; not valid if PK=pkloc is speciﬁed. If CN = csn is speciﬁed,

n must be 1 if N = n is speciﬁed.

Cartridge serial number of the cartridge is not speciﬁed.

Same as CN.

FM=familyname Family to which SSLABEL adds or from which it removes a

cartridge or SM. With OP=FX, this parameter speciﬁes the family

to which the cartridge was assigned.

FM Same as FM = system default family.

FM omitted Same as FM = system default family.

GR=n Group to which SSLABEL adds or from which it removes a

cartridge; 1 ^ n ^ 20. With OP=AM, this parameter is ignored if

PT=P is speciﬁed. GR=n is valid only with OP=AM or OP=RM.

GR Not permitted.

GR omitted Default groups are chosen sequentially. SSVAL runs most efﬁciently

if SSLABEL chooses the default groups.

Revision M Mass Storage Extended Subsystem (MSE) 11-19

Parameters for SSLABEL Directives

Pi Description

LT omitted

N=n

N omitted

PK=pkloc

SM map and 7990 catalog entries are to be updated, even though

the cartridge is lost and its label cannot be updated; valid only

with OP = RM.

If the cartridge is lost and OP=RM is speciﬁed, an error message

is issued and SSLABEL aborts.

Number of cartridges or cubicles to be added, removed, or repaired;

1 ^ n ^ 100; not valid if PT=R is speciﬁed. If CN = csn is

speciﬁed, n must be 1.

Same as N = l.

The inhibit allocation ﬂag or free cartridge ﬂag in the 7990 catalog

is to be cleared; valid only with OP=IB or OP=FC.

The inhibit allocation ﬂag or free cartridge ﬂag in the 7990 catalog

is to be set; valid only with OP=IB or OP=FC.

Location from which the cartridge or cubicle is to be picked; not

valid if CN=csn is speciﬁed.

PK omitted

pkloc Description

Cartridge is to be picked from the input tray. PK = D is

valid only with OP=AM, OP=RC, or OP=FX.

Cartridge or cubicle is to be picked from the speciﬁed

family (FM=familyname) and subfamily

(SB = subfamily). PK=F is valid only with OP=RM or

OP=RB.

Cartridge or cubicle is to be picked from the pool.

PK=P is valid only with OP=AM, OP=RM, or

OP=RB. PK=P is not valid if PT=P is speciﬁed.

Cubicle is to be picked from the reserved area of the

SM. PK=R is valid only with OP=RB.

Same as PK=P.

'*a^\

11-20 NOS Version 2 Analysis Handbook Revision M

Parameters for SSLABEL Directives

Pi Description

PT=ptloc

PT omitted

SB = subfamily

SB omitted

SM=id

SM omitted

YI=yi

ZI=zi

YI=yi,ZI = zi

Location into which the cartridge or cubicle is to be put.

ptloc Description

Cartridge is to be put into the exit tray. PT=D is valid

only with OP=RM.

Cartridge or cubicle is to be put into the speciﬁed

family (FM=familyname) and subfamily

(SB=subfamily). PT=F is valid only with OP=AM or

OP=AB.

Cartridge or cubicle is to be put into the pool. PT=P is

valid only with OP=AM, OP=RM, or OP=AB. PT=P

is not valid if PK = P is speciﬁed.

Cubicle is to be put into the reserved area of the SM.

PT=R is valid only with OP=AB.

Same as PT=P.

Subfamily to which SSLABEL adds or from which it removes a

cartridge or SM; 0 ^ sub ^ 7. With OP=FX, this parameter

speciﬁes the subfamily to which the cartridge was assigned.

Same as SB = 0.

SM identiﬁer of the SM to be used by SSLABEL; id is a letter

from A to H.

Same as SM=A.

Row of the SM to be added or removed; 0 ^ yi ^ 21; valid only

with OP=AB or OP = RB.

Column of the SM to be added or removed; 0 ^ zi *£ 15; valid

only with OP=AB or OP=RB.

y and z coordinates of the cubicle to be added or removed;

0 ^ yi ^ 21, 0 ^ zi ^ 5; valid only with OP=AB or OP=RB.

The following coordinates (y,z) are reserved for customer

engineering and/or diagnostic programming purposes: (0,0), (0,6),

(0,15), (11,0), (11,1), (11,15), (15,0), (15,1), (21,6), and (21,15). The

following coordinates are reserved for system use: (0,1), (0,14),

(21,0), and (21,14).

Revision M Mass Storage Extended Subsystem (MSE) 11-21

SSLABEL Update Sequence

pi Description

YI=yi,ZI = zi, Rectangle of cubicles to be added or removed; cubicles with y

YF=y2,ZF=z2 coordinates between yi and y2 and z coordinates between zi and Z2

are included; valid only with OP=AB or OP=RB. At most, 100

cubicles can be included in the rectangle. YF and ZF must both be

speciﬁed, if either is speciﬁed. YF and ZF cannot be speciﬁed

unless both YI and ZI are speciﬁed. If a reserved cubicle is

included in the rectangle, the directive will be accepted even

though the reserved cubicle cannot be added or removed. zi = 6 and

Z2 = 6 are not valid.

YI and ZI With OP=AB, the next available cubicle closest to the top (for

omitted assignment to a family) or the bottom (for assignment to the pool)

is to be selected. With OP=RB, the ﬁrst empty assigned cubicle is

to be selected.

SSLABEL Update Sequence

Each directive to SSLABEL updates SM maps, 7990 catalogs, and cartridge labels,

whichever are appropriate, to reﬂect the changes in cartridge, cubicle, or SM

assignment. Because the 7990 catalog is a disk-resident permanent ﬁle, it will be

backed up on a dump tape whenever PFDUMP dumps the master device for its

particular subfamily. Thus, it is not necessary for the analyst to back up the 7990

catalogs immediately after an SSLABEL run. However, the backup and recovery of SM

maps do require special operational procedures, which should be performed immediately

after an SSLABEL run (refer to the deﬁnition of SM map earlier in this section).

When SSLABEL is run to change the assignment of a cartridge, the update sequence

consists of a series of steps to delete the old assignment information from the MSE

system ﬁles, relabel the cartridge, and add the new assignment information to the MSE

system ﬁles. If an interruption such as a system failure, SSLABEL abort, or SSEXEC

abort prevents SSLABEL from completing the update sequence, the location of the

affected cartridge and the status of the SM maps and 7990 catalogs depend on the

point of interruption, as follows:

• If the cartridge label, 7990 catalog, and SM map do not all match, then the

cartridge is put into the exit tray. The OP=RC directive cannot be used to restore

the cartridge because of the inconsistency. However, OP=FX can be used to

overwrite the cartridge label and add the cartridge to the pool, if the FM and SB

parameters specify the family and the subfamily on the cartridge label.

• If the cartridge label, 7990 catalog, and SM map do match, then the cartridge may

be returned to its original location, the new location, or the exit tray, depending on

the exact point of interruption. If the cartridge is in the input tray, OP=RC can be

used to restore the cartridge to the location indicated on the cartridge label.

Cartridge Restoration and Reuse, later in this section, describes the procedure for

restoring cartridges found in the exit tray.

i^^.

y*J*S^.

11-22 NOS Version 2 Analysis Handbook Revision M

^Px

Restrictions to SSLABEL

The following restrictions apply to the SSLABEL utility.

• SSEXEC must be running when SSLABEL is run.

• Only one copy of SSLABEL can be run at a time.

• SSLABEL, SSVAL, and SSDEBUG cannot be run at the same time.

SSLABEL Example

Figure 11-1 shows the format of an SSLABEL report. The cartridge label information

is included on the report because of a mismatch with the SM map. An error message

is issued for the ﬁrst directive on the command (in this case, the only directive).

SSLABEL REPORT FILE

SSLABEL,Z./0P=RC,PK=D,SM=B.

1 OP=RC,PK=D,SM=B.

1 0P=RC,PK=D,SM=B.

CM=A-

CSN = 66157234

FAMILY = SYSTST

SUBFAMILY = 0

SM = B

Y = 2

Z = 6

*** ERROR 8 DIRECTIVE 1

UNEXPECTED SM, Y, Z, FAMILY OR SUBFAM.***

Figure 11-1. SSLABEL Report File

Revision M Mass Storage Extended Subsystem (MSE) 11-23

SSMOVE

SSMOVE manages disk and 7990 residence. That is, SSMOVE determines which ﬁles

should be left on disk, which ﬁles should be released from disk and moved to 7990,

and which ﬁles should be resident both on disk and on 7990.

The selection process includes algorithms that weigh certain ﬁle characteristics, as

follows:

• Files are selected for destaging to 7990 based on file length, time since the last

update, and the preferred residence speciﬁed by the user.

• Files are selected for release from disk based on time since the last access and the

backup requirement speciﬁed by the user.

SSMOVE reads the PFC entries for a particular family and calculates release and

destage values (refer to Selection Algorithms later in this section) for each ﬁle to

determine its residence. If a ﬁle has both disk and 7990 images but is to reside only

on 7990, SSMOVE releases the disk space for the ﬁle. For ﬁles that do not have a J^\

7990 image, SSMOVE creates entries on the SSMOVE/SSEXEC communication ﬁle,

MVOCOM, to identify the ﬁles to be destaged and to specify whether or not the ﬁle's

disk space is to be released upon completion of the destage. SSEXEC then processes

each destage and destage/release request on MVOCOM.

SSMOVE generates an output report that includes a list of input directives to

SSMOVE, site-deﬁned values used in the destage/release decision-making process, ﬁles

selected for processing, and a summary of the number of ﬁles expected to reside on

each device after destage/release processing.

11-24 NOS Version 2 Analysis Handbook Revision M

SSMOVE

0am\.

The format of the

SSMOVE,pi,p2...

SSMOVE command is:

• .Pn-

Description

DN=device

DN omitted

FM=familyname

FM omitted

I=ﬁlename

1 = 0

I omitted

L=ﬁlename

L = 0

L omitted

LB = n

LB omitted

LO = F

LO = P

LO omitted

Device number of the only disk from which ﬁles are to be

destaged/released.

SSMOVE destages and releases ﬁles from all devices in a speciﬁed

family.

Same as DN.

Family to be used by SSMOVE.

Same as FM = system default family.

File containing the directives to SSMOVE.

Same as I = INPUT.

No input directives ﬁle exists. SSMOVE uses the default

parameters.

Same as I = INPUT.

File on which listable output is to be written.

Same as L=OUTPUT.

No output ﬁle is to be generated.

Same as L=OUTPUT.

Large ﬁle boundary is used when sorting ﬁles for destaging. All

ﬁles smaller than n PRUs are small ﬁles.

Same as LB = 365.

All ﬁles selected for staging, destaging, or releasing are to be

included in the report ﬁle.

Only ﬁles actually processed are listed in the report ﬁle. Refer to

the PX parameter.

Individual ﬁles are not to be listed in the report ﬁle.

Same as LO.

Revision M Mass Storage Extended Subsystem (MSE) 11-25

SSMOVE

NW omitted

PX=xxx

PX omitted

UI=userindex

UI = 0

UI omitted

Description

No wait. SSMOVE will not wait for completion of destage and

release processing by SSEXEC.

SSMOVE will wait for completion of destage and release processing

by SSEXEC.

xxx is a character string that speciﬁes which types of SSMOVE

processing should not be done.

xxx Description

Exclude ﬁle archiving or releasing disk space.

Exclude creating a backup copy by destaging a ﬁle from

disk to the 7990.

Exclude direct access ﬁles.

Exclude freeing a ﬁle from the 7990 by clearing its asa

value from the ﬁle's PFC entry.

Exclude indirect access ﬁles.

Exclude staging a ﬁle from the 7990 to disk.

For example, the PX=ABFS parameter could be used to obtain a

report of the results of an SSMOVE run using a given set of

parameters without actually performing any of the selected actions.

All selected SSMOVE processing is to be done.

Same as PX.

Destage and release processing is restricted to ﬁles having user

index userindex.

All user indexes are processed.

11-26 NOS Version 2 Analysis Handbook Revision M

SSMOVE Directives

The directives to SSMOVE are speciﬁed on a separate ﬁle. Each directive must be

speciﬁed on a separate line. Three types of input directives can be included in an

SSMOVE directive ﬁle. They are: comment directives, speciﬁc ﬁle directives, and value

speciﬁer directives.

Comment directives have an asterisk (*) in the ﬁrst column followed by any message

or comment the user wishes to enter.

The speciﬁc ﬁle and value speciﬁer directives are described in the following

paragraphs:

Speciﬁc File (SF) Directives

SSMOVE uses SF directives to select a given ﬁle or set of ﬁles on which to perform

the speciﬁed processing action. The format of the SF directive is:

SF,FN=ﬁlename,UI=userindex,PO=process.

Parameter Description

FN=ﬁlename The 1 to 7 characters that specify a permanent ﬁle or set of

permanent ﬁles. Asterisks in the ﬁle name are wild card

characters.

UI=userindex User index of the ﬁle speciﬁed by the FN=ﬁlename parameter.

PO=process Processing action to be performed on the ﬁle speciﬁed by the

FN=ﬁlename parameter.

process Description

A A r c h i v e t h e s p e c i ﬁ e d ﬁ l e .

B . Create a backup copy by destaging the speciﬁed ﬁle

from disk to the 7990.

F Free the speciﬁed ﬁle from the 7990 by clearing its asa

value in the ﬁle's PFC entry.

S Stage the speciﬁed ﬁle from the 7990 to disk.

For example, the following SF directive selects all permanent ﬁles with a maximum of

two characters in the ﬁle name beginning with Z from user index 002622s to be staged

to disk.

SF,FN=Z*,UI=2622,PO=S.

The SF directive can be used without specifying the FN parameter to establish default

values for the UI and/or PO parameters. These default values will be used for

subsequent SF, FN=ﬁlename directives that are missing UI and/or PO parameters.

Default values for UI and/or PO parameters can be redeﬁned by using another SF

directive without specifying the FN parameter.

Revision M Mass Storage Extended Subsystem (MSE) 11-27

SSMOVE Directives

Value Speciﬁer Directives

SSMOVE uses the value speciﬁer directives to redeﬁne the values of the weight factors

or thresholds (installation parameters) used in the algorithms that select ﬁles to be

destaged or released. The site analyst uses these value speciﬁers to increase or

decrease the importance of certain ﬁle characteristics used to determine which ﬁles are

to be destaged and/or released. For example, specifying a large MN parameter prohibits

SSMOVE from selecting small ﬁles for destage/release processing.

The general format of a value speciﬁer directive statement is:

directive,qualiﬁer(s).parameter(s).

Variable

directive

qualiﬁer(s)

Description

2-character value speciﬁer directive keyword.

directive Description

Keyword for backup requirement directive.

Keyword for ﬁle requirement directive.

Keyword for preferred residence directive.

Keyword for site management directive.

Keyword for weight adder directive.

Keyword for weight multiplier directive.

One or more qualiﬁer keywords that specify the ﬁle type

(direct/indirect access) and decision type (destage/release) to be

used in the selection process. If all qualiﬁer keywords are

omitted, all four combinations of ﬁle types and decision types

are used in the selection process.

qualiﬁer(s) Description

Keyword for indirect access ﬁles.

Keyword for direct access ﬁles.

Keyword for destage decisions.

Keyword for release decisions.

11-28 NOS Version 2 Analysis Handbook Revision M

SSMOVE Directives

Variable Description

parameter(s) Parameter keyword and a value (positive integer) for the

speciﬁed directive keyword. Multiple parameter=value

combinations can be entered in the same value speciﬁer

directive statement.

parameter(s) Description

With BR

Directive:

N=wf

Y=wf

MD = wf

With FR

Directive:

Weight factor to be used as the backup

requirement value for decisions involving

ﬁles with a BR=N attribute (no backup).2

Weight factor to be used as the backup

requirement value for decisions involving

ﬁles with a BR=Y attribute (tape backup).

Weight factor to be used as the backup

requirement value for decisions involving

ﬁles with a BR=MD attribute (media

dependent, tape, or 7990 backup).

/fﬂlSP*^

MN = min

MX=max

DD=dd

TH = th

Minimum ﬁle size in PRUs.

Maximum ﬁle size in PRUs.

Number of days since the ﬁle was last

accessed or updated.

Destage or release threshold. If a ﬁle's

calculated value is less than the speciﬁed

threshold, the ﬁle is not a candidate for

destage or release processing. TH = 0 allows

all ﬁles to be candidates for destage and

release processing unless they are excluded

by FR directive parameters or by other

run-time parameter values.

/f!$"X 2. The ﬁle owner speciﬁes the backup requirement attribute using the BR parameter on the DEFINE or

CHANGE command (refer to the NOS Version 2 Reference Set, Volume 3).

Revision M Mass Storage Extended Subsystem (MSE) 11-29

SSMOVE Directives

Variable

parameter(s)

Description

(Continued)

parameter(s)

With PR

Directive:

L = wf

D = wf

M=wf

N = w f

With SM

Directive:

MG=mg

SG=sg

Description

Weight factor to be used as the preferred

residence value for decisions involving ﬁles

with a PR=L attribute (locked to disk

preference).

Weight factor to be used as the preferred

residence value for decisions involving ﬁles

with a PR=D attribute (disk preference).3

Weight factor to be used as the preferred

residence value for decisions involving ﬁles

with a PR=M attribute (7990 preference).3

Weight factor to be used as the preferred

residence value for decisions involving ﬁles

with a PR=N attribute (no preference).3

Master goal specifying the percent of master

device disk space not to be exceeded.

Secondary goal specifying the percent of

secondary device disk space not to be

exceeded.

Available for site use.

^S^X

3. The ﬁle owner speciﬁes the preferred residence attribute using the PR parameter on the DEFINE or

CHANGE command (refer to the NOS Version 2 Reference Set, Volume 3).

11-30 NOS Version 2 Analysis Handbook Revision M

SSMOVE Directives

Variable Description

/ i ^ \

parameter(s) (Continued)

parameter(s)

With WA

Directive:

AG=wa

LN = wa

AC = wa

DV=sf

With WM

Directive:

AG=wf

LN = wf

AC = wf

Description

Weight adder to be added to the weighted

ﬁle age.

Weight adder to be added to the weighted

ﬁle length.

Weight adder to be added to the weighted

ﬁle access count.

Scaling factor given to the divisor when

calculating destage and release values.

Weight factor given to the ﬁle age (days

since the ﬁle was last accessed or updated).

Weight factor given to the ﬁle length.

Weight factor given to the ﬁle access count.

For example, the following value speciﬁer directive speciﬁes that the minimum size for

an indirect access ﬁle to be destaged is three disk PRUs.

FR,IA,DS,MN=3.

The following value speciﬁer directive requires that all ﬁles to be destaged or released

must be less than 98765 PRUs long.

FR,MX=98765.

If a directive error is detected on the directive input ﬁle, SSMOVE ignores the

incorrect directive and continues to run in report mode. A message is issued to the job

dayﬁle and no processing is done.

Revision M Mass Storage Extended Subsystem (MSE) 11-31

Excluding Destage/Release Processing

SSMOVE uses FR directives to select candidates for destage and release processing.

After selecting the candidates and making the ﬁnal processing decisions, SSMOVE

checks the PX parameter to determine if any processing should be excluded.

If FR,RL,MN=9999999 is speciﬁed, all ﬁles less than 9999999 PRUs will be given a

negative release value. Any ﬁle less than 9999999 PRUs will not be released unless

the ﬁle has been speciﬁcally selected for releasing with an SF directive.

IF PX=A is speciﬁed, SSMOVE selects ﬁles for processing without considering that

release processing is prohibited. When the ﬁnal processing decisions have been made,

any ﬁles selected for releasing will not be processed. That is, ﬁles selected for

releasing only will not be processed; and ﬁles selected for both destaging and releasing

will not be processed.

If PX=B is speciﬁed, destage processing is excluded just as release processing was in

the previous description.

Destaging cannot be excluded with an FR,DS,MN = 9999999 directive as releasing can.

Even though all ﬁles less than 9999999 PRUs will be given a negative destage value

and not selected for destaging, they will be made candidates for destaging in case any

of them are selected for releasing. When the ﬁnal processing decisions are made, any

ﬁle that is a candidate for destaging and has been selected for releasing will be

destaged and released. However, any destaging candidate not selected for releasing will

not be destaged.

Selection Algorithms

SSMOVE determines which ﬁles to destage and/or release according to the following

algorithms. Files that reside only on 7990 are not considered because they have been

destaged and released previously. For all other ﬁles, SSMOVE checks the ﬁle length

and excludes from further consideration any ﬁle whose length in PRUs is less than the

minimum length threshold (refer back to the MN parameter) or greater than the

maximum length threshold (refer back to the MX parameter).

If the current image of the ﬁle resides on both disk and 7990, SSMOVE uses the

release algorithm to determine whether or not to release disk space; the destage

algorithm is not used. If the current ﬁle resides on disk only, SSMOVE uses the

destage algorithm to determine whether or not to destage the ﬁle. If the ﬁle is to be

destaged to 7990, SSMOVE also determines via the release algorithm whether or not to

release the ﬁle's disk space.

> ^ ^ v

11-32 NOS Version 2 Analysis Handbook Revision M

Selection Algorithms

Destage and Release Algorithms

An analyst uses the FR directive to establish ﬁle requirements on ﬁles that are to be

eligible for destage and/or release processing. These requirements include ﬁle length

(minimum and maximum) and ﬁle age. A ﬁle's age is the number of days since the ﬁle

was last updated (for destaging) or accessed (for releasing). Separate requirements can

be established for the various combinations of ﬁle types (direct/indirect) and decision

types (destage/release).

Files that meet the eligibility requirements are assigned separate destage and release

values. SSMOVE calculates these values for each eligible ﬁle according to the following

equation, which considers various ﬁle attributes from the ﬁle's PFC entry (length, age,

access count, preferred residence, and backup requirements). These ﬁle attributes are

weighted according to site-deﬁned values (refer to Value Speciﬁer Directives earlier in

this section).

value =

(ai+a2* length) •(b1+b2*age)*(res)*(bkup)

(ci+C2*access)»sf

Variable Description

length

age

res

bkup

access

Site-deﬁned length weight adder speciﬁed by the WA directive LN

parameter.

Site-deﬁned length weight factor speciﬁed by the WM directive LN

parameter.

Length of the ﬁle in PRUs.

Site-deﬁned age weight adder speciﬁed by the WA directive AG

parameter.

Site-deﬁned age weight factor speciﬁed by the WM directive AG

parameter.

Number of days since the ﬁle was last updated (for destage) or

accessed (for release).

Site-deﬁned preferred residence weight factor speciﬁed by the PR

directive. The PR directive parameter depends on which preferred

residence attribute the ﬁle owner speciﬁed.

Site-deﬁned backup requirement weight factor speciﬁed by the BR

directive. The BR directive parameter depends on which backup

requirement attribute the ﬁle owner speciﬁed.

Site-deﬁned access count weight adder speciﬁed by the WA directive

AC parameter.

Site-deﬁned access count weight factor speciﬁed by the WM

directive AC parameter.

Number of times the ﬁle was accessed.

Site-deﬁned scaling factor speciﬁed by the WA directive DV

parameter.

Revision M Mass Storage Extended Subsystem (MSE) 11-33

Selection Algorithms

If a ﬁle's destage/release value exceeds a site-deﬁned threshold (FR directive TH

parameter), the ﬁle is selected for destage/release processing. Conversely, if the ﬁle's s^s

destage/release value does not exceed the threshold, the ﬁle is not considered for )

destage/release processing.

Files eligible for releasing are sorted according to their release values. The ﬁles on

each disk with the greatest release values are selected to be released. When enough

ﬁles have been selected and released, the site-deﬁned disk space availability goals will

be achieved.

Decision Algorithm Hierarchy

Several different functions of SSMOVE can lead to apparent contradictory processing

decisions. The decisions are made using the following rules, which are listed in order

of precedence.

1. SSMOVE does not consider any ﬁles for processing if they are excluded by the

PX=I, PX = D, or UI parameters to SSMOVE or if the ﬁle has a PR=L attribute.

2. Files identiﬁed for speciﬁc processing using the SF directive are selected for the

speciﬁed processing regardless of the ﬁle requirements controlled by the FR

directive.

3. If DN = 0, all ﬁles that meet the destage requirements are selected to be destaged.

Otherwise, only ﬁles on the speciﬁed device are destaged.

4. No ﬁle with a BR=Y attribute will be selected to be released (including ﬁles

speciﬁed for processing with an SF directive) if the ﬁle data has not been included

in a full or incremental dump using PFDUMP.

5. If DN = 0, all ﬁles that meet the release requirements are selected to be released, if

needed, to meet the disk space availability goals. Otherwise, only ﬁles on the

speciﬁed device are released.

6. If a ﬁle is selected for a speciﬁc processing action (such as destage) and the

corresponding action is precluded by a PX parameter (in this case, PX=B), then the

selected action is not performed. In all cases, however, the SSMOVE device status

report is based on which ﬁles were selected for processing rather than whether the

PX parameter actually allowed the selected processing to occur.

0^^\

11-34 NOS Version 2 Analysis Handbook Revision M

Disk Space/Dump Tape Management

As more disk-resident ﬁles are created and more 7990-resident ﬁles are staged to disk,

it will be necessary to monitor the availability of disk space. It is recommended that

SSMOVE be used as a periodic disk space management procedure to avoid frequent

disk-full conditions (refer to Disk Space Management later in this section). SSMOVE

can also be used to reduce the amount of data written on dump tapes and thereby

avoid mamtaining large numbers of dump tapes. This is accomplished by destaging to

7990 and/or releasing disk space of ﬁles that need not be on the dump tape (refer to

Dump Tape Management later in this section).

An analyst can specify what percent of disk space is to contain indirect access ﬁles

and what percent is to contain direct access ﬁles. For direct access ﬁles, the analyst

can further specify the percent of master device and/or secondary device disk space.

These percent values are applied to all devices in the family on which SSMOVE is

run or a speciﬁc device can be managed by using the DN parameter to SSMOVE.

The speciﬁed percentages of disk space are considered to be goals in that other

site-deﬁned constraints and ﬁle-busy conditions may make it impossible to achieve

these goals.

Restrictions to SSMOVE

The following restrictions apply to the SSMOVE utility.

• Only one copy of SSMOVE can be run at a time when it is in the process of

destaging ﬁles. A second SSMOVE aborts if the ﬁrst one has not completed.

• SSMOVE and PFDUMP should not be run at the same time.

• SSMOVE cannot be run when SSVAL is running.

Revision M Mass Storage Extended Subsystem (MSE) 11-35

Output Example

The SSMOVE report consists of the following ﬁve sections:

1. A list of directives (ﬁgure 11-2) from the input directive ﬁle plus any diagnostic

messages.

2. A list of run-time parameter values (ﬁgure 11-3) used in the release and destage

decision-making process. These values consist of the site-deﬁned, assembly-time

values as modiﬁed by the value speciﬁer directives.

3. An optional list of the ﬁles (ﬁgure 11-4) selected for processing. This list shows the

ﬁle name, ﬁle type, user index, length, date, access count, processing action, and

destage and release values.

4. A device status report (ﬁgure 11-5) that includes a summary of the number of ﬁles

and total ﬁle length expected to reside on each device as a result of release

processing. Also, a subfamily report that includes a summary (for each subfamily) of

the number of ﬁles and ﬁle length for ﬁles to be destaged, ﬁles that reside only on

7990, and ﬁles that reside on 7990 and disk.

5. A destage abandonment report (ﬁgure 11-6) that is not produced if the NW

parameter is speciﬁed. This report provides a list of the number of ﬁles not

destaged and an optional list of the abandoned ﬁles. It also provides an updated

device status report reﬂecting only the ﬁles that were destaged.

,^*fe\

A^S

11-36 NOS Version 2 Analysis Handbook Revision M

Output Example

/ffty

SSMOVE REPORT.

SSMOVE,I=B1GO.LO* F.FM=MTST.

1 FR,MN=0,DD=0.

2 SM,MG=0.SG=0.

yy/mm/dd. hh.mm.ss. PAGE

Figure 11-2. SSMOVE Report: List of Directives

SSMOVE REPORT.

yy/mm/dd. hh.mm.ss. PAGE

RUN-TIME PARAMETER VALUES

« O E S

DIRECT

TAGE«

INDIRECT

•REL

DIRECT

EASE*

INDIRECT

FILE REQUIREMENTS

F R M N 0

MX 9999999

D D 0

T H 1

9999999

WEIGHT MULTIPLIER

WM AG

100

WEIGHT ADDER

WA AG

1000

PREFERRED RESIDENCE

PR L

100

BACKUP REQUIREMENT

BR N

SITE MANAGEMENT

SM MG

Figure 11-3. SSMOVE Report: Run-time Parameter Values

yms

Revision M Mass Storage Extended Subsystem (MSE) 11-37

Output Example

yy/mm/dd. hri.mm .ss . PAG E 4

SSMOVE REPORT.

NAME TYPE UI LENGTH DATE ACC-CT ACTION{« = NOT DONE PER »PX* OPTION) DES-VAL REL-VAL

DIR301 DIR. 30 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 • 22000

DIR305 DIR. 30 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 22000

DIR302 DIR. 30 1200 84.09.1 1 . 0 DESTAGE AND RELEASE. 6005 22000

DIR303 DIR. 30 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 22000

DIR304 DIR. 30 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 22000

IND3020 IND. 30 619 84.09.1 1 0 DESTAGE AND RELEASE. 3100 16190

IND3019 IND. 30 618 84.09.1 1 0 DESTAGE AND RELEASE. 3095 16180

IND3018 IND. 30 617 84.09,1 1 0 DESTAGE AND RELEASE. 3090 16170

IND3017 IND. 30 616 84.09.1 1 0 DESTAGE AND RELEASE. 3085 16160

IND3016 IND. 30 615 84.09.1 1 0 DESTAGE AND RELEASE. 3080 16150

IND3015 IND. 30 614 84.09.1 1 0 DESTAGE AND RELEASE. 3075 16140

1ND3014 IND. 30 613 84.09.1 1 0 DESTAGE AND RELEASE. 3070 16130

IND3013 IND. 30 612 84.09.1 1 0 DESTAGE AND RELEASE. 3065 16120

IND3012 IND. 30 611 84.09.1 1 0 DESTAGE AND RELEASE. 3060 16110

IND3011 IND. 30 610 84.09.1 1 0 DESTAGE AND RELEASE. 3055 16100

IND3010 IND. 30 609 84.09.1 1 0 DESTAGE AND RELEASE. 3050 16090

IND309 IND. 30 608 84.09.1 1 0 DESTAGE AND RELEASE. 3045 16080

IND308 IND. 30 607 84.09.1 1 0 DESTAGE AND RELEASE. 3040 16070

IND307 IND. 30 606 84.09.1 1 0 DESTAGE AND RELEASE. 3035 16060

IND306 IND. 30 605 84.09.1 1 0 DESTAGE AND RELEASE. 3030 16050

IND305 IND. 30 604 84.09.1 1 0 DESTAGE AND RELEASE. 3025 16040

IND304 IND. 30 603 84.09.1 1 0 DESTAGE AND RELEASE. 3020 16030

1^303 IND. 30 602 84.09.1 1 0 DESTAGE AND RELEASE. 3015 16020

IND302 IND. 30 601 84.09.1 1 0 DESTAGE AND RELEASE. 3010 16010

IND301 IND. 30 600 84.09.1 1 0 DESTAGE AND RELEASE. 3005 16000

DIR313 DIR. 31 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 22000

DIR314 DIR. 31 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 22000

DIR312 DIR. 31 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 22000

DIR315 DIR. 31 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 22000

DIR311 DIR. 31 1200 84.09.1 1 0 DESTAGE AND RELEASE. 6005 22000

IND3120 IND. 31 619 84.09.1 1 0 DESTAGE AND RELEASE. 3100 16190

1ND3119 IND. 31 618 84.09.1 1 0 DESTAGE AND RELEASE. 3095 16180

IND3118 IND. 31 617 84.09.1 1 0 DESTAGE AND RELEASE. 3090 16170

IND3117 IND. 31 616 84.09.1 1 0 DESTAGE AND RELEASE. 3085 16160

IND3116 IND. 31 615 84.09.1 1 0 DESTAGE AND RELEASE. 3080 16150

IND3115 IND. 31 614 84.09.1 1 0 DESTAGE AND RELEASE. 3075 16140

IND3114 IND. 31 613 84.09.1 1 0 DESTAGE AND RELEASE. 3070 16130

IND3113 IND. 31 612 84.09.1 1 0 DESTAGE AND RELEASE. 3065 16120

IND3112 IND. 31 611 84.09.1 1 0 DESTAGE AND RELEASE. 3060 16110

IND3111 IND. 31 610 84.09.1 1 0 DESTAGE AND RELEASE. 3055 16100

Figure 11-4. SSMOVE Report: Optional list of Files

11-38 NOS Version 2 Analysis Handbook Revision M

Output Example

SSKOVE REPORT.

yy/m/dd. hh.ua.ss. PAGE 6

(BEFORE) DEVICE STATUS (AFTER) MR CENTS

EO Bll DT-N TYPE FILES / PRU FILES / PRU EXP. GOAL FLA6.

2 7 D K - 1 I K S . 6 2 3 6 9 5 2 0 0 0 0

2 7 M C - 1 D I R . 2 3 2 5 4 2 * y 1 0 0 8 1 0 • •

•* - DEVICE SPACE GOAL NOT KET

SU8FANILT REPORT

FILES TO DESTAGE FILES ONLY ON 7990 FILES ON 7990

SUB DIRECT INDIRECT DIRECT INDIRECT DIRECT INDIRECT

FAMILY MKSER PSU NWBER PRU NUMBER PRU NWtBER PRU NUMBER PRU NUN9ER PRU

6000

MOO

to 12190 6363 20 12190 I(8 6363 20 12190

n12190 6352 20 12190 3 6352 20 12190

21201 10357 1 2 0 1 ] 10S57 1201

0 0 12362 3 0 1 0 I 12362 3010

0 0 6357 3 0 1 0 : 6357 3010

0 0 6357 3 0 1 0 2 6357 3010

13050 27915 69 56298 27919 69 56298

00 0 0 0 0 0 0 0 0

Figure 11-5. SSMOVE Report: Device Status Report

yRrS

SSKOVE REPORT.

DESTAGE ABANDONMENT REPORT

FILENAME UI

yy/aa/dd. hh.ca.ss. PAGE

1 0 NO SPACE

2 0 NO STORAGE NODULE AVAILABLE

3 0 KO CARTRIDGE OR GROUP AVAILABLE

*0FILE ALREADY DESTAGED

5 0 FILE BUST / PFN PROBLEM

«0CATALOG ACCESS ERROR

7 0 OVERFLOW NOT LEGAL

8 0 GROUP FULL

9001SK READ ERROR

10 CARTRIDGE LOST

11 CLOSED DESTAGE

(BEFORE) DEVICE STATUS (AFTER) PERCENTS

E O D N D T - N T Y P E F I L E S / P R U F I L E S I P R U E X P. G O A L F L A G .

2 7 D K - 1 I N D . 6 2 3 6 9 5 2 0 0 0 0

7 DK-1 DIR. 23 2542* 1008

** - DEVICE SPACE GOAL NOT NET

SUBFAMILY REPORT

FILES NOT DESTAGED FILES ONLY ON 7990 FILES ON 7990

SUB DIRECT INDIRECT DIRECT INDIRECT DIRECT INDIRECT

FAMILY NUMBER PRU NUMBER PRU NUMBER PRU NUMBER PRU NUMBER PRU NUMBER PRU

38 6363

37 6352

37 10357

42 12362

37 6357

19 27915

0 0

20 12190

21201

53010

69 56298

0 0

38 6363

37 6352

37 10357

42 12362

37 6357

19 27919

0 0

20 -12190

20 12190

21201

53010

69 56298

0 0

Figure 11-6. SSMOVE Report: Destage Abandonment

Revision M Mass Storage Extended Subsystem (MSE) 11-39

SSVAL

SSVAL either performs release processing or reports on problems with the current MSE

system ﬁles. When release processing, SSVAL makes 7990 space available that is

presently allocated to ﬁles that are no longer needed, or ﬂags ﬁles to be staged to disk

in order to free up cartridges. When problem reporting, SSVAL reports on irregularities

or discrepancies found in the current 7990 catalogs and PFC entries for the speciﬁed

family and, optionally, in certain SM maps. The function to be performed is determined

by whether or not the RF parameter is speciﬁed, as described next.

NOTE

SSVAL updates the SM map for the speciﬁed SM. (Refer to the deﬁnition of SM map

earlier in this section.) It is recommended that the SM map be copied on tape or on

another device or family immediately after every update of the SM map.

The format of the

SSVAL,pi,P2,...

SSVAL command is:

Description

AM omitted

FM=familyname

FM omitted

FX=n

FX omitted

L=ﬁlename

L=0

L omitted

The SM map for the SM speciﬁed by the SM = id parameter is to

be analyzed in addition to the 7990 catalogs; not valid if

RF=ﬁlename or RF is speciﬁed.

SM maps are not to be analyzed.

Family to be analyzed; not valid if the RF parameter is speciﬁed.

Same as FM = system default family; not valid if the RF

parameter is speciﬁed.

Same as FM = system default family if the RF parameter is not

speciﬁed. The family on the release data ﬁle is used if the RF

parameter is speciﬁed.

Error threshold. If the total error count is greater than n, neither

release processing nor problem ﬁxing is performed.

Same as FX=0.

File on which listable output is to be written.

Same as L= OUTPUT.

No output ﬁle is to be generated.

Same as L=OUTPUT.

'*=%.

11-40 NOS Version 2 Analysis Handbook Revision M

SSVAL

RF=ﬁlename

RF omitted

RL omitted

SB = subfamily

SB omitted

SM = id

SM omitted

ST=n

ST omitted

Description

File that contains the release data ﬁle.

Same as RF=ZZZZRDF.

Current versions of the 7990 catalogs are to be analyzed.

Release processing is to be performed; valid only if the RF

parameter is speciﬁed.

No release processing is to be performed.

Subfamily to be processed. Up to eight subfamilies can be selected

by the numbers 0 through 7. For example, SB = 723 selects

subfamilies 2, 3, and 7.

Same as SB=01234567.

Same as SB = 01234567.

SM identiﬁer of the SM to be used. Up to 8 SMs can be selected

by the letters A through H. For example, SM = ACG selects SM A,

C, and G.

Same as SM=ABCDEFGH.

Scattered ﬁle criterion. Files are indicated as scattered if they are

contained on at least n more cartridges than the minimum number

needed to contain them.

Same as ST=1. That is, ﬁles are scattered if they are contained on

more than the minimum number of cartridges needed to contain

them.

Same as ST=1.

Revision M Mass Storage Extended Subsystem (MSE) 11-41

Release Processing

If RF=ﬁlename or RF is speciﬁed, SSVAL determines which 7990 ﬁles are no longer /

needed and issues a request to SSEXEC to purge these ﬁles so their 7990 space can be

reused. The procedure is for SSVAL to analyze copies of the 7990 catalogs and PFC

entries for the speciﬁed family that are contained on the release data ﬁle (RDF)

speciﬁed by the RF parameter. Those 7990 ﬁles described in a 7990 catalog but not

having a PFC entry (that is, orphans) can be purged. During the analysis (refer to

both 7990 Catalog Analysis and PFC Analysis later in this section), SSVAL keeps track

of the error conditions it discovers, and if the error count is less than or equal to the

threshold speciﬁed by the FX parameter, release processing is performed if RL is

speciﬁed. That is, trouble-free orphans are purged and their 7990 space is made

available for reuse. The current 7990 catalog is updated to reﬂect that these ﬁles no

longer exist. A validation report is issued that lists the errors encountered, the number

of trouble-free orphans, and the amount of released 7990 space. If RL is not speciﬁed,

no release processing is performed but the validation report is issued, which lists the

errors encountered, the number of trouble-free orphans, and the amount of releasable

7990 space. xs?v

The RDF used for this analysis is a ﬁle produced during a previous PFDUMP run

and it contains versions of the 7990 catalogs and PFC entries that were current at

the time of the dump. The site analyst chooses which RDF to use depending on how

long after a ﬁle was purged he/she wants to wait before releasing 7990 space. For

example, the site analyst might run SSVAL every week for release processing

purposes and use the RDF from the previous week's full dump. There are some

restrictions as to which RDFs can be used (refer to 7990 Space Management later in

this section).

Unlike the purging of disk-only ﬁles, where the ﬁle space on disk is immediately 1

made available for reassignment, the MSE software will not immediately free up a

permanent ﬁle's 7990 space following a purge of the associated disk copy of the

permanent ﬁle (assuming both a 7990 and a disk permanent ﬁle image). The reason

for not freeing up the subject 7990 space is that a subsequent PFLOAD, following a

disk failure, could reload the PFC entry for the associated permanent ﬁle. Therefore,

7990 space for purged permanent ﬁles that were resident on the 7990 must not be

made available for reuse until the possibility of such a recovery action cannot occur.

The site analyst determines which dump tapes can and cannot be used for reloading /*^

ﬁles and PFC entries following a disk failure. When a PFDUMP is taken, two ﬁles

are normally produced: the actual dump ﬁle used with PFLOAD to perform a

recovery, and the RDF ﬁle to be used with SSVAL to make 7990 space available after

purging ﬁles.

Using an RDF ﬁle allows the corresponding dump tape to be used (later dump tapes

are also allowable), but earlier dump tapes are not allowed. Because the MSE

software cannot control which dump tapes actually are used in reloading, it is the

site analyst's responsibility to observe the above restriction.

The concurrent execution of PFDUMP with OP=S not speciﬁed and SSVAL with RL

speciﬁed causes SSVAL to run very slowly. PFDUMP acquires and keeps the PF

utility interlock for its duration, and this gives the appearance that SSVAL is hung,

when it is actually just waiting. Therefore, operations personnel should not attempt to

run both of these utilities with the above options at the same time.

If the RF parameter is not speciﬁed, SSVAL will set the free ﬁle ﬂag for all ﬁles }

that totally or partially reside on all cartridges on which the free cartridge ﬂag is

set. -sbv

11-42 NOS Version 2 Analysis Handbook Revision M

Problem Reporting

If RF=ﬁlename or RF is not speciﬁed, SSVAL reports on problems with the current

7990 catalogs and PFC entries for the speciﬁed family. If AM is speciﬁed, problems

with SM maps are also included in the report. SSVAL examines the MSE system ﬁles

and PFC entries and searches for problem chains and fragments, problem asa values,

and SM map/7990 catalog mismatches. The procedures SSVAL uses to detect and

classify inconsistencies and discrepancies are described next, under Error Detection and

Classiﬁcation. SSVAL keeps track of the error conditions, if any, and if the error count

is less than or equal to the threshold speciﬁed by the FX parameter, problem ﬁxing is

performed. That is, SSVAL sets ﬂags in the appropriate entries of the SM map, 7990

catalog, and/or PFC entries to prevent propagation of errors due to the inconsistencies

or discrepancies found and to permit error recovery by the SSDEBUG utility. A count

of the errors is recorded in the job dayﬁle. A validation report is issued, which lists

the errors encountered, the number of trouble-free orphans, and the amount of

releasable 7990 space.

NOTE

For release processing or problem ﬁxing, SSVAL predicts what it expects to happen

(if no problems occur) prior to doing the releasing or ﬁxing. If SSVAL encounters

problems, the reports may not reﬂect the actual status of the cartridges. In this case,

run SSVAL in report mode following the release processing or problem ﬁxing to

verify the actual status of the 7990 catalogs and cartridges.

Error Detection and Classiﬁcation

SSVAL detects and classiﬁes errors according to the following procedures. During the

SM map analysis, SSVAL detects and classiﬁes errors in the SM map. During the 7990

catalog analysis, SSVAL detects errors with chains of AUs; during the PFC analysis,

SSVAL classiﬁes these errors. Whenever an error is encountered, the total error count

is increased by 1. The action taken for each type of error is discussed under Release

Processing and Problem Fixing later in this section.

SM Map Analysis

If the AM option is speciﬁed, SSVAL attempts to locate problems with SM map entries

by comparing the 7990 catalogs and the SM map. For each coordinate pair (y,z) in a

7990 catalog entry, SSVAL locates the corresponding SM map entry. A type 1 error

exists if the code ﬁeld in the SM map entry is not 5 (assigned to a subfamily), or if

the family, subfamily, or csn ﬁelds in the SM map entry do not match those in the

corresponding 7990 catalog entry. SSVAL also scans the SM map for all other entries

assigned to the subfamily and reports as a type 2 error any of these entries that does

not have a corresponding 7990 catalog entry.

Revision M Mass Storage Extended Subsystem (MSE) 11-43

Error Detection and Classiﬁcation

7990 Catalog Analysis

The 7990 catalog analysis locates problems with chains of AUs (refer to the deﬁnition

of cartridge earlier in this section) and identiﬁes on each HOC whether any of the

following problems exist or whether any part of the 7990 ﬁles resides on a cartridge

that is lost or has excessive write parity errors.

SSVAL scans the 7990 catalog for HOC entries that are allocated and follows each

chain until it terminates. Normal termination occurs with an EOC entry. Abnormal

termination occurs when no EOC is found, an AU links to an unallocated AU, or an

AU links to an AU previously found in the chain being followed (looping chain).

During the chain scans, the following types of chains can be encountered; they are

linkage problems that are identiﬁed on the HOC entry.

Chain Description

Intersecting More than one chain links to the same AU.

Scattered ﬁle The number of cartridges used for the ﬁle exceeds the value

speciﬁed by the ST parameter.

SSVAL also locates any AUs that are allocated but were not on any chain being

followed. Such AUs are linked together to form partial chains without an HOC. These

partial chains are called fragments and the ﬁrst AU in a fragment is designated as the

start of fragment. Each fragment chain is followed until it terminates. The abnormal

termination conditions previously listed can also occur with fragments. Intersections can

occur, but a fragment chain that intersects the start of another fragment chain is not

an intersection; rather, one is the tail end of the other. •*^^\

>^*S\

11-44 NOS Version 2 Analysis Handbook Revision M

ymm*\

Error Detection and Classiﬁcation

PFC Analysis

The PFC analysis is performed to classify the errors encountered on chains during the

7990 catalog analysis. For each PFC entry with asa=£0 (the file has a 7990 image and

the asa value identiﬁes the ﬁrst AU in the chain containing the ﬁle), SSVAL classiﬁes

the following errors. Error type 3 exists if the asa value is invalid. Error type 4 exists

for any of the following reasons:

• The AU specified by the asa value is not allocated or is not an HOC entry.

• The chain does not terminate normally.

• The chain intersects with another chain or fragment.

• More than one PFC entry points to the chain.

• The chain includes a cartridge for which the lost or excessive parity error flag is

set.

SSVAL also classiﬁes the following error conditions:

• Error type 5 exists if an orphan chain terminates abnormally or intersects with

other chains or fragments. Trouble-free orphans (chains without a PFC entry

pointing to them and without linkage problems) are not classiﬁed as errors.

• Error type 6 exists if an orphan is a fragment.

• Error type 7 exists if a chain or fragment points to an unallocated AU.

SSVAL generates informational report messages if either of the following conditions is

true.

• The system error flag is set in the PFC.

• The read error flag is set in the PFC.

Revision M Mass Storage Extended Subsystem (MSE) 11-45

Release Processing and Problem Fixing

If the total error count calculated during the analyses described previously exceeds the

value speciﬁed by the FX parameter, then neither release processing nor action to ﬂag

or ﬁx the detected error conditions is performed. The validation report, however, is

produced. Otherwise, action taken depends on the parameter speciﬁed and the type of

errors found, as follows.

If SSVAL was run for release processing purposes (RF=ﬁlename or RF speciﬁed),

SSVAL issues a request to SSEXEC to release trouble-free orphans if RL was

speciﬁed. If RL was not speciﬁed, no release processing is performed. If SSVAL was

run for problem reporting purposes (RF omitted and FX=n, where n is greater than

the number of validation errors), the following action is taken.

• For error type 1, the linkage error flag is set in the 7990 catalog entry.

• For error type 2, the linkage error flag is set in the SM map entry.

• For error type 3, there are three alternatives:

- If the ﬁle also has a disk image, the asa ﬁeld in the PFC entry is cleared.

Thus, the good disk image will not be released and the ﬁle is accessible even

if the 7990 image cannot be retrieved.

- If the ﬁle does not have a disk image, no action is taken.

- However, if the disk image can be reloaded from tape, it is recommended that

the ﬁle be reloaded using PFLOAD with OP=Z so that the asa ﬁeld will be

cleared.

• For error type 4, the action taken is both that taken for error type 3 and that

taken for error type 5, 6, or 7.

• For error type 5, 6, or 7, the frozen flag is set in the 7990 catalog entry for the

initial AU on the problem chain or fragment. This enables the problem

chain/fragment/AU to be made available to the SSDEBUG utility, but prevents

these AUs from being overwritten until then. Thus, the SSDEBUG utility can be

used to inspect or save data from the corresponding AUs or cartridges. If the AUs

are not allocated, the inhibit ﬂag is also set in the 7990 catalog for that cartridge

so that no new ﬁles will overwrite any data that might be on the cartridge.

11-46 NOS Version 2 Analysis Handbook Revision M

jf^v

0ims

Validation Report

The validation report consists of a report heading and a series of report groups for

each subfamily and SM being reported on. The heading identiﬁes the subfamily, the

SM, whether or not there are any problems, and the last purge date and time for the

SM (the last time orphans on this SM were released). There is one report group for

each error detected, and the actual information recorded in a report group depends on

the type of error, as described next. Each report group, however, contains the following

items:

• Error type (a number from 1 to 7).

• Identification (refer to the particular error type, described next).

• Chain information (7990 catalog ordinal, AU number, A or U designation for

allocated or unallocated, H or E designation for HOC or EOC).

• Error description.

After the last report group, the validation report lists the number of trouble-free

orphans, the amount of released or releasable 7990 space, the total number of errors

detected, and whether or not the MSE system ﬁles were updated.

Error Types 1 and 2

Error types 1 and 2 identify mismatches between the SM map and the 7990 catalog. In

the validation report, the identiﬁcation ﬁeld lists the 7990 catalog ordinal, the y and z

coordinates, and the csn of the cartridge in error. The chain ﬁeld is blank because

problem chains are not identiﬁed as either error type 1 or 2. The analyst should run

the SSUSE utility to produce a detailed report of the appropriate SM map and 7990

catalog entries to determine the exact problem.

Error Types 3 and 4

Error types 3 and 4 identify problem chains and problem asa values. In the validation

report, the identiﬁcation ﬁeld lists the permanent ﬁle name and user index of the

affected ﬁle; the dump control date and time (from the PFC entry for the ﬁle) to

identify the backup ﬁle, if any; and the letter N (no) or Y (yes) to indicate whether or

not the ﬁle has a disk image. The chain ﬁeld lists the 7990 catalog ordinal and volume

number for all AUs in the affected chain. An A or U indicates whether each AU is

allocated or unallocated, and an H or E identiﬁes the HOC or EOC. An error

description is printed for each error detected; one chain can have several errors.

Error Types 5, 6, and 7

Error types 5, 6, and 7 identify problem orphans, fragments, and problem unallocated

AUs. In the validation report, the information reported is the same as for error types 3

and 4, except for the identiﬁcation ﬁeld. Instead of the permanent ﬁle identiﬁcation,

the word ORPHAN (error type 5), FRAGMENT (error type 6), or UNALLOCATED

(error type 7) is printed. Error type 7 is an unallocated AU that is pointed to by a

chain or fragment. Each such AU is also reported with the chain for the corresponding

orphan or fragment.

Revision M Mass Storage Extended Subsystem (MSE) 11-47

Validation Report

Intersections

Intersections occur when more than one chain links to the same AU; they are classiﬁed

as either error type 4 or 5. Thus, they are reported as explained previously.

Intersections are also reported in a separate entry consisting of a heading and

additional information identifying the 7990 catalog ordinals and AU numbers of the

intersecting chains. Therefore, when intersections are reported, there are two entries

for the affected subfamily and SM: one lists only the intersections and the other lists

all the errors encountered.

Validation Report Example

Figure 11-7 shows the format of a validation report. SSVAL was run for problem

reporting purposes (RF not speciﬁed), and no errors were detected.

SSVAL - VALIDATION REPORT SSVAL - VER •.*

FAMILY = MTST

SSVAL,FM=MTST,SB =0,SM=A.

L= OUTPUT

RF = 0

AM = 0

SM = A

FM = MTST •

FX = 0

RL = 0

SB = 0

ST = 0

SUBFAMILY =0 SM = A — GOOD ~ LPDT - ******** ********

RELEASABLE 7990 FILES 12

RELEASABLE 7990 AUS = 800

TOTAL VALIDATION ERRORS

CATALOGS NOT MODIFIED

••REPORT COMPLETE**

^S^S

Figure 11-7. SSVAL Validation Report

11-48 NOS Version 2 Analysis Handbook Revision M

0ms

Typical SSVAL Runs

As described previously, SSVAL is run either to make 7990 space available for reuse or

to report on problems with the current MSE system ﬁles and/or PFC entries. The

following examples show typical SSVAL runs that may be used periodically for these

purposes.

Example 1:

The following SSVAL command causes 7990 space to be made available for reuse.

SSVAL,RF=DUMP1,RL.

DUMP1 is the release data ﬁle produced by a previous PFDUMP run from which

SSVAL can identify all the 7990 ﬁles that were orphans at the time of the dump. If no

error conditions are detected in the 7990 catalog and PFC entries contained on ﬁle

DUMP1 (FX=0 by default), the orphans are purged and the 7990 space assigned to

them is released. The last purge date and time ﬁeld in the subcatalog (in the 7990

catalog) for each SM for each subfamily is updated to the time of the SSVAL run if

any ﬁle from that SM is purged. This is to ensure that a subsequent SSVAL run does

not release the same space a second time, as the 7990 space may be reallocated to a

new ﬁle.

If errors were detected, they are listed on the validation report and release processing

is not performed. This is indicated by the CATALOGS NOT MODIFIED message at

the end of the validation report. The analyst can rerun SSVAL with the following

command to ensure that release processing is performed (n is the number of errors

detected during the previous SSVAL run).

SSVAL,RF=DUMP1,RL,FX=n.

After this SSVAL run has completed, an analysis of the current version of the SM

map, 7990 catalog, and PFC entries should be made to determine whether or not the

errors detected in the ﬁrst SSVAL run exist in the current MSE system ﬁles and PFC

entries. This is accomplished using the command:

SSVAL,FM=fam i1yname.AM.

familyname is that on the release data ﬁle, DUMP1.

Example 2:

Upon completion of any device reload that includes recovery of a 7990 catalog and/or

SM map, SSVAL should be run to determine whether any SM map/7990 catalog

mismatches exist. The following call accomplishes this.

SSVAL ,FM=f ami 1 yname, SB=ni,n2 nh,AM.

familyname is that for which recovery was done; ni,n2,...,nhare the affected subfamilies

if just some of the devices of the family were reloaded.

Revision M Mass Storage Extended Subsystem (MSE) 11-49

Restrictions to SSVAL

SSVAL analyzes the SM map, 7990 catalog, and PFC entries for the subfamilies

speciﬁed by the SB parameter and reports any discrepancies or inconsistencies. For ^^

example, if the recovery was not scheduled and SSLABEL was run to add or remove '

cartridges or cubicles from one of the affected subfamilies after the last incremental or

full dump for these subfamilies, then the SM map will reﬂect the results of the

SSLABEL run but the 7990 catalog will not. Such mismatches will be reported on the

validation report.

If the following SSVAL command is then made, the errors detected during the

previous run will be ﬂagged, as described previously under Release Processing and

Problem Fixing.

SSVAL, FM=f am il yname ,SB=ni,n2,... ,nh,AM,FX=n.

familyname and m,n2,...,nhare the same as in the previous SSVAL run; n is the

number of errors detected during the previous run. After investigating the cause of

these errors, the analyst can run the SSDEBUG utility to correct the error conditions

(refer to SSDEBUG later in this section).

If a device reload includes recovery of all SM maps (default family, user

index=3777608), the latest SM maps should be recovered from backup copies. Then

each family that has 7990-resident ﬁles should be analyzed using the following

command to detect SM map/7990 catalog mismatches.

SSVAL, FM=f ami 1 yname, AM.

It is recommended that the previous SSVAL run be made periodically to check whether

any unexpected error conditions exist. As the site analyst becomes more familiar with

MSE processing, these periodic SSVAL runs can be done less frequently. "^)

Restrictions to SSVAL

The following restrictions apply to the SSVAL utility.

• SSVAL cannot be run when SSMOVE is in the process of destaging files.

• Only one copy of SSVAL can be run at a time.

• SSVAL, SSLABEL, and SSDEBUG cannot be run at the same time. ^^

11-50 NOS Version 2 Analysis Handbook Revision M

SSUSE

0$ms

00b\

SSUSE

SSUSE reads data in the 7990 catalogs and SM maps and produces reports on the

availability of space on 7990 cartridges and the allocation of cubicle space within an

SM. The reports may not be completely up to date because the 7990 catalogs and SM

maps can be updated while the reports are being generated.

The types of reports that SSUSE generates are:

Report Description

Basic usage report

Optional report A

Optional report B

Optional report C

Optional report D

Usage report that lists general information about the use of

each SM in a subfamily.

SM map report that lists the contents of an SM as described

in the SM map.

Cartridge summary report that lists general status information

for each cartridge entry in the 7990 catalog. The report

identiﬁes the available AUs and ﬂags set for each cartridge in

the 7990 catalog.

Detailed cartridge report that lists cartridge usage information

for each cartridge entry in the 7990 catalog.

Detailed AU status report that lists AU status information for

each entry in the 7990 catalog plus cartridge usage

information.

The format of the SSUSE command is:

SSUSE,p,,p2 Pn-

pi Description

CM=A-

CM omitted

CN=csn

CN omitted

FM=familyname

FM omitted

L=ﬁlename

Cartridge manufacturer code is A-, indicating IBM.

Same as CM=A-.

Cartridge serial number of a speciﬁc cartridge. The CN = csn

parameter is valid only if the OP=D parameter is speciﬁed.

Cartridge serial number is not speciﬁed.

Same as CN.

Family to be reported on.

Same as FM = system default family.

File on which listable output is to be written.

Same as L=OUTPUT.

Revision M Mass Storage Extended Subsystem (MSE) 11-51

SSUSE

Report

L=0

L omitted

OP=op

OP omitted

SB=subfamily

SB omitted

SM = id

SM omitted

Description

No output ﬁle is to be generated.

Same as L=OUTPUT.

Type of report to be generated. Multiple options can be speciﬁed

(for example, OP=AB).

op Description

Optional report A and basic usage report.

Optional report B and basic usage report.

Optional report C and basic usage report.

Optional report D and basic usage report.

Basic usage report only is to be generated.

Same as OP.

Subfamily to be reported on. Up to eight subfamilies can be

selected by the numbers 0 through 7. For example, SB = 0273

selects subfamilies 0, 2, 3, and 7.

Same as SB = 01234567.

SM identiﬁer of the SM to be used. Up to eight SMs can be

selected by the letters A through H. For example, SM = AGC

selects SM A, C, and G.

Same as SM=A.

» < s § \

11-52 NOS Version 2 Analysis Handbook Revision M

Basic Usage Report

The basic usage report includes a title line, which identiﬁes the family and SM being

reported on. The statistics in this report contain data for each group followed by totals

for SMs and subfamilies. The following items are listed.

• Number of cubicles reserved for each subfamily.

• Number of cartridges in each subfamily.

• Number of unassigned and unflagged AUs for small files.

• Number of unassigned and unflagged AUs for large files.

• Number of flagged cartridges.

• Number of flagged AUs.

• If any empty cubicles are assigned, they are listed first for each subfamily. This

line shows a hyphen (-) as the group designator since groups are assigned only to

cartridges.

Figure 11-8 shows the format of a basic usage report.

Revision M Mass Storage Extended Subsystem (MSE) 11-53

Basic Usage Report

yy/mm/dd. hh.mm .ss. PAGE

SSUSE REPORT FILE

SSUSE BASIC REPORT FAMILY * MTST

CUBE = CUBICLES CARTRIDGE FLAGS AU FLAGS

CART = CARTRIDGES M = MISSING FA = FLAWED AND ALLOCATED

I = INHIBIT ALLOCATION FU = FLAWED AND UNALLOCATED

** = SUBFAMILY TOTAL F = FREE CARTRIDGE SF = START OF FRAGMENT

-= UNASSIGNED GROUP L = OFF CARTRIDGE LINKS AVAILABLE FC ■FROZEN CHAIN

P = EXCESSIVE PARITY ERRORS AC = AU CONFLICT

E = MAP ERROR(AS DETECTED BY SSVAL)

k n i t / n C D P A D T D T P l f t C C c i A ﬁ ﬁ c r i

SUB SM GR CUBE CART

—nvnunout.

A U A U

MFA F U S F F C AC

(SMALL) (LARGE)

0 A

0 A 1167 2662

0 A SX 1167 2662

1 A

1 A 1168 2662

1 A sz 1168 2662

2 A

2 A 1168 2606

2 A ■ s 1168 2606

3 A

3 A 1168 2586

3 A ■ s 1168 2586

4 A

4 A 1168 2662

4 A ss 1168 2662

5 A

5 A 1168 2662

5 A ss 1168 2662

6 A

6 A 5400

6 A ss 5400

7 A

7 A 1200 2662

7 A ** 1200 2662

Figure 11-8. SSUSE Basic Report

(Continued)

A ^ S

11-54 NOS Version 2 Analysis Handbook Revision M

Basic Usage Report

0$m\.

jfﬂ^fﬁ^.^

0m\

(Continued)

SSUSE REPORT FILE yy/mm/dd. hh. mm.ss. PAGE

SSUSE BASIC REPORT SM AND SUBFAMILY TOTALS FAMILY = MTST

SUB SM GR CUBE CART

AVAILABLE

A U A U

(SMALL) (LARGE)

NUMBER CARTRIDGES FLAGGED

M I F L P E

ER AU FL

SFFA FU FC AC

SM TOTALS FOR SPECIFIED SUBFAMILIES

■« A « 3 9 17 13607 18509 15

SUBFAMILY TOTALS FOR SPECIFIED SM-S

0 s s s s 4 1167 2662

1 s s . . 5 1168 2662

2 s s s s 5 1168 2606

3 " « 5 1168 2586

4 s s s s 5 1168 2662

5 " « 5 1168 2662

6 • • « 5 5400

7 s s s s 5 1200 2662

Figure 11-8. SSUSE Basic Report

Revision M Mass Storage Extended Subsystem (MSE) 11-55

Optional Report A

Optional report A includes a title line that identiﬁes the SM being reported on. This

report is made up of 15 pages, each page describing the SM map for speciﬁc y and z

coordinates. Each page describes 22 y coordinates (rows) and 1 z coordinate (column).

Column 6 (z=6) is not reported on because it contains the data recording drives. The

information listed in optional report A includes the following items:

• y and z coordinates.

• Cartridge manufacturer code and CSN of the cartridge.

• Family and subfamily.

• Group and ordinal in group.

• Flags.

Figure 11-9 shows the format of optional report A.

yy/mm/dd hh mm.ss. PAGE

SSUSE REPORT FILE

SSUSE OPTIONAL REPORT A - STORAGE MODULE MAP FOR SM = A

FLAGS: P = CARTRIDGE EXISTS IN POOL S = RESERVED FOR SYSTEM USE C = RESERVED FOR CUSTOMER ENGINEERING

F = NOT ASSIGNED TO THIS FAMILY E = ERROR FLAG N = (Y,Z) DOES NOT EXIST IN SMMAP

R = RESERVED FOR ALTERNATE SMMAP GPORD = ORDINAL IN GROUP

Y Z CM CSN FAMILY SUBFAMILY GROUP GPORD FLAGS

2 0 F

3 0 F

4 0 F

5 0 F

6 0 F

7 0 F

8 0 F

90 F

11 NF

12 NF

13 NF

14 NF

15 NF

17 MTST

18 MTST

19 A-51680502 MTST

20 A-13727600 MTST

21 S F

>^ 5S \

Figure 11-9. SSUSE Optional Report A

11-56 NOS Version 2 Analysis Handbook Revision M

Optional Report B

0^S

Optional Report B

Optional report B includes a title line identifying the subfamily, family, and SM being

reported on. There is one line of output per cartridge within a subfamily and storage

module. The report includes the following cartridge information about entries in the

7990 catalog.

• Group and ordinal in group.

• y and z coordinates.

• Cartridge manufacturer code and CSN of the cartridge.

• Free AUs for small and large files.

• Cartridge flags.

• Error conditions.

Figure 11-10 shows the format of optional report B.

SSUSE REPORT FILE

yy/mm/dd. hh.mm.ss. PAGE 25

SSUSE OPTIONAL REPORT B - CARTRIDGE SUMMARY REPORT SM » A SUBFAMILY = 6 FAMILY = MTST

NOTES:

FA = FLAWED AND ALLOCATED

FU = FLAWED AND UNALLXATED

SF = START OF FRAGMENT

FC = FROZEN CHAIN

AC = AU CONFLICT

GPORD = ORDINAL IN GROUP

CARTRIDGE FLAGS:

M = MISSING

I = INHIBIT

F = FREE CARTRIDGE

L = LINK(FREE AU EXIST, NO OFF CARTRIDGE LINK)

P = EXCESSIVE WRITE ERRORS

E = MAP ERROR(DETECTED BY SSVAL)

GP GPORD CM CSN

FREE AU

SMALL LARGE

1 0 2 1 7 A - 1 3 2 6 8 7 9 4 1 8 0 0 0

1 1 2 0 7 A - 5 1 3 2 9 4 1 2 1 8 0 0 3

1 2 1 9 7 A - 5 1 3 2 8 8 3 9 1 8 0 0 4

CART

FLA6S

I L

-ERROR CONDITIONS

—NUMBER OF AU

F U S F F C A C

/Gw&H&'X,

Figure 11-10. SSUSE Optional Report B

Revision M Mass Storage Extended Subsystem (MSE) 11-57

Optional Report C

Optional report C is the detailed cartridge report. It lists the group and FCT ordinal

for each 7990 catalog entry and the y and z coordinates and CSN for each cartridge.

The report also includes the AU numbers of the ﬁrst AUs for small and large ﬁles and

the division point for each cartridge.

Figure 11-11 shows the format of optional report C.

yy/mm/dd. hh.mm.dd. PAGE 33

SSUSE REPORT FILE

SSUSE OPTIONAL REPORT C - DETAILED CARTRIDGE REPORT SM = A SUBFAMILY = 6 FAMILY = MTST

FLAGS: I = INHIBIT ALLOCATION

E = MAP ERROR

M =

MISSING P = EXCESSIVE PARITY ERRORS

= AVAILABLE LINK COUNT FCTORD = SFM CATALOG ORDINAL

CM CSN

A-13268794

A-51329412

A-51328839

GROUP

FCTORD

FLAGS

FIRST

SMALL

AU CARTRIDGE

FIRST DIVISION

LARGE POINT

0 1 8 0 1

1916 1801

1832 1801

CCL

Figure 11-11. SSUSE Optional Report C

11-58 NOS Version 2 Analysis Handbook Revision M

0^S

j^s

Optional Report D

Optional report D is the detailed AU status report. It lists the group and FCT ordinal

of the cartridge, the CSN, the y and z coordinates of the cartridge, and the number of

available off-cartridge links. The report also includes the octal numbers of the AUs

with ﬂawed AU, start of volume, and error ﬂags noted. This report also speciﬁcally

notes when a cartridge is empty (for use in determining the status of a cartridge that

is to be made available).

Figure 11-12 shows the format of optional report D.

0$$k®\

<0$s

SSUSE REPORT FILE yy/mm/dd. hh.mm.ss. PAGE

SSUSE OPTIONAL REPORT D - DETAILED AU STATUS REPORT SM = A SUBFAMILY = 1 FAMILY = MTST

F = FLAWED AU (DEMARK FAILURE)

V = START OF VOLUME

E = ONE OF THE ERROR FLAGS SET (AU CONFLICT, FROZEN CHAIN. START OF FRAGMENT)

F C T O R D Y Z C M C S N G R O U P

16 2 1 1 A-51328833 1

AU XXXO XXX1 XXX2 XXX3 XXX4 XXX5 XXX6 XXX7

100X 0041300731 0041270731 0041260731 0041250731 0041240731 0041230731 0041220731 0041210731'

101X 0041200731 0041170731 0041160731 0041150731 0041140731 0041130731 0041120731 0041110731

102X 0041100731 0041070731 0041060731 0041050731 0041040731 0041030731 0041020731 0041010731

103X 0041000731 0040770731 0040760731 0040750731 0040740731 0040730731 0040720731 0040710731

104X 0040700731 0040670731 0040660731 0040650731 0040640731 0040630731 0040620731 0040610731

105X 0040600731 0040570731 0040560731 0040550731 0040540731 0040530731 0040520731 0040510731

106X 0040500731 0040470731 0040460731 0040450731 0040440731 0040430731 0040420731 0040410731

107X 0040400731 0040370731 0040360731 0040350731 0040340731 0040330731 0040320731 0040310731

110X 0040300731 0040270731 0040260731 0040250731 0040240731 0040230731 0040220731 0040210731

111X 0040200731 0040170731 0040160731 0040150731 0040140731 0040130731 0040120731 0040110731

112X 0040100731 0040070731 0040060731 0040050731 0040040731 0040030731 0040020731 0040010731

113X 0040000731 V0000621214 0040611131 0040601131 0040571131 0040561131 0040551131 0040541131

114X 0040531131 0040521131 0040511131 0040501131 0040471131 0040461131 0040451131 0040441131

115X 0040431131 0040421131 0040411131 0040401131 0040371131 0040361131 0040351131 0040341131

116X 0040331131 0040321131 0040311131 0040301131 0040271131 0040261131 0040251131 0040241131

117X 0040231131 0040221131 0040211131 0040201131 0040171131 0040161131 0040151131 0040141131

120X 0040131131 0040121131 0040111131 0040101131 0040071131 0040061131 0040051131 0040041131

121X 0040031131 0040021131 0040011131 0040001131 V0001771414 0041761214 0041751214 0041741214

122X 0041731214 0041721214 0041711214 0041701214 0041671214 0041661214 0041651214 0041641214

123X 0041631214 0041621214 0041611214 0041601214 0041571214 0041561214 0041551214 0041541214

I24X 0041531214 0041521214 0041511214 0041501214 0041471214 0041461214 0041451214 0041441214

125X 0041431214 0041421214 0041411214 0041401214 0041371214 0041361214 0041351214 0041341214

126X 0041331214 0041321214 0041311214 0041301214 0041271214 0041261214 0041251214 0041241214

127X 0041231214 0041221214 0041211214 0041201214 0041171214 0041161214 0041151214 0041141214

130X 0041131214 0041121214 0041111214 0041101214 0041071214 0041061214 0041051214 0041041214

131X 0041031214 0041021214 0041011214 0041001214 0040771214 0040761214 0040751214 0040741214

132X 0040731214 0040721214 0040711214 0040701214 0040671214 0040661214 0040651214 0040641214

133X 0040631214 0040621214 0040611214 0040601214 0040571214 0040561214 0040551214 0040541214

134X 0040531214 0040521214 0040511214 0040501214 0040471214 0040461214 0040451214 0040441214

135X 0040431214 0040421214 0040411214 0040401214 0040371214 0040361214 0040351214 0040341214

136X 0040331214 0040321214 0040311214 0040301214 0040271214 0040261214 0040251214 0040241214

137X 0040231214 0040221214 0040211214 0040201214 0040171214 0040161214 0040151214 0040141214

Figure 11-12. SSUSE Optional Report D

Revision M Mass Storage Extended Subsystem (MSE) 11-59

SSDEBUG

SSDEBUG allows the analyst to update appropriate entries in the SM maps and/or

7990 catalogs and thereby resolve inconsistencies reported by the SSVAL utility.

SSDEBUG can also copy data from selected 7990 ﬁles or cartridges to disk. This

function of SSDEBUG is intended for use when errors such as unrecoverable read

errors do not allow successful staging of a 7990 ﬁle.

NOTE

SSDEBUG updates the SM map for the speciﬁed SM. (Refer to the deﬁnition of SM

map earlier in this section.) It is recommended that the SM map is copied on tape or

on another device or family immediately after every update of the SM map.

Input to SSDEBUG is via a directive ﬁle. Refer to SSLABEL, earlier in this section,

for a description of the directive ﬁle format.

The format of the SSDEBUG command is:

SSDEBUG, pi, P2.P3-

pi Description

I=ﬁlename

I omitted

L=ﬁlename

L=0

L omitted

Z omitted

File on which directives are written.

Same as I=INPUT.

Same as I = INPUT.

File on which listable output is to be written.

Same as L=OUTPUT.

No output ﬁle is to be generated.

Same as L=OUTPUT.

Directives are contained on the SSDEBUG command. The

I parameter is ignored.

Directives are contained on the ﬁle speciﬁed by the I parameter.

11-60 NOS Version 2 Analysis Handbook Revision M

/ ^ p ^ \

SSDEBUG Directives

The following directives are available with SSDEBUG. The descriptions of the

parameters for these directives follow this listing of directives.

OP=RS — Read AUs

OP=RS reads selected AUs of a cartridge identiﬁed by its csn (CN=csn) or y and z

coordinates (YI and ZI parameters). The SM parameter speciﬁes the SM where the

cartridge resides. The range of AUs to be read is speciﬁed by the SL and SU

parameters. The AUs are written to the ﬁle speciﬁed by the PF parameter. All CYBER

data plus 7990 and CRC data, including header and trailer information, is returned.

The analyst can write an application program to reconstruct the permanent ﬁle by

stripping off all CYBER or 7990 related control information. Refer to Format of Disk

Files Read From 7990 by SSDEBUG later in this section for a description of the

control information written to disk along with the disk ﬁle contents.

OP=RF — Read File

OP = RF reads the ﬁle whose alternate storage address is speciﬁed by the FO, SB, SM,

and ST parameters. The ﬁle is written to the ﬁle speciﬁed by the PF parameter. All

data, including header and trailer information, is returned (refer to Format of Disk

Files Read From 7990 by SSDEBUG later in this section).

OP=RP — Release Frozen Chain Space

OP=RP clears ﬂags in the 7990 catalog and releases 7990 space for the chain whose

alternate storage address is speciﬁed by the FO, SB, SM, and ST parameters.

OP=RL — Remove Cartridge Entry From 7990 Catalog

OP=RL removes a 7990 catalog entry that is not linked properly to the SM map. The

FO, SB, and SM parameters identify the 7990 catalog entry to be removed.

OP=RC — Remove Cartridge Entry From SM Map

OP=RC removes an SM map entry that does not have a corresponding FCT entry in

the 7990 catalog. The SM, YI, and ZI parameters identify the SM map entry to be

removed. The cartridge at the particular location (y,z) is put into the exit tray and the

SM map entry is changed to unassigned. The SM map should be backed up

immediately after the SSDEBUG run to ensure consistency with cartridge labels and

7990 catalogs (refer to the deﬁnition of SM map earlier in this section).

A scratch label is written on the cartridge prior to it being replaced in the exit tray,

but the cartridge history information remains intact.

OP=CF — Change Flag in SM Map or 7990 Catalog

OP = CF sets or clears ﬂags in the SM map or 7990 catalog. The FL and ON or OF

parameters indicate the ﬂag type and the action to be taken. The appropriate FCT

entry or SM map entry is determined from the CN, FO, or YI and ZI parameters. For

FCT ﬂags (FL=AC, FC, or SF), the analyst can specify either a single AU using the

ST parameter or a range of AUs using the SL and SU parameters.

Revision M Mass Storage Extended Subsystem (MSE) 11-61

Parameters for the SSDEBUG Directives

The descriptions of the parameters to the SSDEBUG directives follow. Not all

parameters are valid with all directives, as indicated.

.•^tf^V

CM=A-

CM omitted

CN=csn

CN omitted

FL=ﬂag

Description

Cartridge manufacturer code is A-; indicating IBM.

Same as CM=A-.

Cartridge serial number of the cartridge to be used; not valid for

OP=RS if YI=y or ZI = z is speciﬁed; and not valid for OP=CF if

FO=ord, YI=y, or ZI=z is speciﬁed.

YI=y and ZI = z must be speciﬁed for OP=RS. FO=ord or YI=y

and ZI = z must be speciﬁed for OP=CF.

Name of ﬂag in SM map or 7990 catalog to be set or cleared.

Valid only for OP=CF.

ﬂag Description

FL omitted

FM=familyname

FM omitted

AC AU conﬂict ﬂag.

EW Excessive write parity error ﬂag.

FC Frozen chain ﬂag.

FE Linkage error ﬂag (in 7990 catalog FCT).

IB Inhibit allocation ﬂag.

LC Lost cartridge ﬂag.

ME Linkage error ﬂag (in SM map).

SF Start of fragment ﬂag.

FL=ﬂag must be speciﬁed for OP = CF.

Family to be processed.

Same as FM = system default family.

^^v

11-62 NOS Version 2 Analysis Handbook Revision M

Parameters for the SSDEBUG Directives

Pi Description

FO=ord

FO omitted

PF=ﬁlename

PF omitted

SB = subfamily

SB omitted

SL=i

SL omitted

SM=id

SM omitted

ST=s

7990 catalog ordinal indicating the ﬁle to be read or the chain

whose space is to be released. Not valid for OP=CF if CN = csn,

YI=y, or ZI=z is speciﬁed.

FO=ord must be speciﬁed for OP = RF, OP = RP, and OP = RL.

CN = csn or YI=y and ZI=z must be speciﬁed for OP=CF.

Flag speciﬁed by FL=ﬂag is to be cleared. Valid only with

OP = CF.

Flag speciﬁed by FL=ﬂag is to be set. Valid only with OP=CF.

File to which the 7990 image is to be copied. This ﬁle is deﬁned

under the user's current family and user index.

Same as PF=MMMMBUG.

Subfamily to be used; 0 ^ subfamily ^ 7.

Same as SB = 0.

AU where OP=RS begins copying or OP = CF begins changing

ﬂags; 1 ^ i ^ 1931; i ^ j (refer to SU=j).

Same as SL=1.

SM identiﬁer of the SM to be used by SSDEBUG; id is a letter

from A to H.

Same as SM=A.

AU where OP = RF begins reading ﬁles or fragments or OP = RP

begins releasing frozen chain space. If speciﬁed with OP=CF,

ST=s speciﬁes the AU number of an FCT ﬂag to be changed and

takes precedence over the range of AUs speciﬁed by the SL and SU

parameters. AU numbers are meaningful with OP = CF only for

ﬂags FL=AC, FC, or SF.

Revision M Mass Storage Extended Subsystem (MSE) 11-63

Parameters for the SSDEBUG Directives

ST omitted

SU=j

SU omitted

YI=y

YI omitted

ZI=z

ZI omitted

Description

ST=s must be speciﬁed for OP=RF and OP=RP.

AU where OP=RS ends copying or OP=CF ends changing ﬂags;

1 ^ j ^ 1931; i ^ j (refer to SL=i).

Same as SU=1.

Same as SU = 1.

y coordinate of the cubicle where the cartridge to be read resides;

0 *£. y *s 21. ZI=z must also be specified. CN=csn must not be

speciﬁed with OP=RS. FO=ord or CN=csn must not be speciﬁed

with OP=CF.

CN=csn must be speciﬁed for OP=RS. FO=ord or CN=csn must

be speciﬁed for OP=CF. YI=y and ZI=z must be speciﬁed for

OP=RC.

z coordinate of the cubicle where the cartridge to be read resides;

0 ^ z ^ 15 and z*6. YI=y must also be speciﬁed. CN=csn must

not be speciﬁed with OP=RS. FO=ord or CN = csn must not be

speciﬁed with OP=CF.

CN=csn must be speciﬁed for OP=RS. FO=ord or CN = csn must

be speciﬁed for OP=CF. YI=y and ZI=z must be speciﬁed for

OP=RC.

11-64 NOS Version 2 Analysis Handbook Revision M

Using SSDEBUG

Suppose the following validation report (ﬁgure 11-13) was generated by running SSVAL

for problem reporting purposes.

SUBFAMILY =4 SM = PROBLEMS LPD1 = yy/mm/dd. hh.mm.ss

ERR IDENTIFICATION -CHAIN- ERROR DESCRIPTION

TYP FCT - AU -- A/U - H/E

5 ORPHAN 017-1737

017-1738

017-1739

017-1740

017-1741

017-1742

017-1743

017-1930

017-1931

AH-

A—

ILL-FORMED CHAIN

SCATTERED FILE (INFO)

019-0000 U~ START OF FRAGMENT

7 UNALLOCATED 019-0000 u— START OF FRAGMENT

Figure 11-13. SSDEBUG: Executing SSVAL in Problem-Fixing Mode

This partial validation report shows an error type 5 identiﬁed by ORPHAN with an

error description of ILL-FORMED CHAIN (frozen chain). You can use SSDEBUG to set

(OP=CF) and release (OP=RP) the frozen chain space or you can execute SSVAL in

problem-ﬁxing mode to set the frozen chain ﬂag.

Enter the following SSDEBUG command to set the frozen chain ﬂag for the range of

AUs speciﬁed by the SL and SU parameters.

SSDEBUG,Z./OP=CF,FL=FC,ON,FM=SMD,SB=4,SL=1737,SU=1931,F0=17.

When the SSDEBUG or SSVAL command is processed, enter the following command to

release the frozen chain space.

SSDEBUG,Z./0P=RP,SB=4,SM=A,ST=1737,FO=17.

When this command is processed, you can verify that the frozen chain has been set by

running SSVAL again to generate another validation report.

0$m\.

Revision M Mass Storage Extended Subsystem (MSE) 11-65

Restrictions to SSDEBUG

The following restrictions apply to the SSDEBUG utility.

• Only one copy of SSDEBUG can be run at one time.

y,z Relationship with SM Map Ordinal

A one-to-one functional mapping exists between coordinate pairs and the ordinal entries

within the SM map. To determine the coordinate pair (y,z) from the SM map ordinal

(ord), use the following equations:

y = (352-ord)/l6

z = 352-ord-16y

To determine the SM map ordinal (ord) from the coordinate pair (y,z), use the following

equation:

ord=352-z-16y

A^Sl.

11-66 NOS Version 2 Analysis Handbook Revision M

0ms

Format of Disk Files Read From 7990 by SSDEBUG

A local disk ﬁle created by SSDEBUG with the OP=RS or OP=RF option includes all

CYBER and 7990 header and trailer information. Figure 11-14 shows the format of a

disk ﬁle created when SSDEBUG reads a full stripe; ﬁgure 11-15 shows the format of

a disk ﬁle created when SSDEBUG reads a partial stripe.

word o

12c

CYBER DATA © \

1601,

7990 READ RAW DATA MESSAGE (64 BYTES)

(PACKED 6 BYTES PER 60-BIT WORD, LEFT-JUSTIFIED

ZERO-FILLED)

7990 BLOCK HEADER LABEL (16 BYTES)

(PACKED 7.5 BYTES PER WORD WITH 1 BYTE LEFT-

JUSTIFIED IN THIRD WORD)

(483 BYTES PER PRU)

7990 BLOCK TRAILER LABEL (16 BYTES)

(SAME FORMAT AS 7990 BLOCK HEADER LABEL)

(TO FILL STRIPE)

(f) FOR CYBER DATA. EVERY 255TH BYTE IS A CRC BYTE. THE LAST BYTE IS ALWAYS A CRC BYTE TO

TAKE CARE OF SHORT DATA BLOCKS. FULL CYBER DATA BLOCKS HAVE 26 CRC BYTES.

Figure 11-14. Full Stripe Disk File Format

yi$RS

Revision M Mass Storage Extended Subsystem (MSE) 11-67

Format of Disk Files Read From 7990 by SSDEBUG

A^ ^S .

WORD 0

12c

CYBER DATA ©

1601.

7990 READ RAW DATA MESSAGE (64 BYTES)

(PACKED 6 BYTES PER 60-BIT WORD, LEFT-JUSTIFIED,

ZERO-FILLED)

(PACKED 7.5 BYTES PER WORD WITH 1 BYTE LEFT-

JUSTIFIED IN THIRD WORD)

(483 BYTES PER PRU)

7990 BLOCK TRAILER LABEL (16 BYTES)

(SAME FORMAT AS 7990 BLOCK HEADER LABEL)

7930 BLOCK TRAILER LABEL (16 BYTES)

UNUSED BYTES ©

TAPE MARK ©

1016 FOR TAPE MARK BLOCKS.

(3) FOR CYBER DATA, EVERY 255TH BYTE IS A CRC BYTE. THE LAST BYTE IS ALWAYS A CRC BYTE TO

TAKE CARE OF SHORT DATA BLOCKS.

0 A TAPE MARK HAS A 7990 BLOCK HEADER AND TRAILER, BUT NO DATA BLOCK.

Figure 11-15. Partial Stripe Disk File Format

11-68 NOS Version 2 Analysis Handbook Revision M

Format of Disk Files Read From 7990 by SSDEBUG

0ms

7990 Stripe Header and Trailer

The 7990 stripe header is a 32-byte block that occupies the ﬁrst 32 bytes of the stripe;

the 7990 stripe trailer occupies the last 32 bytes. The 7990 stripe header and trailer

have the same format.

Bytes

0-5

6-7

8-9

A-B

C-D

E-1D

1E-1F

Description

Volume serial number. A 6-character string specifying the volume

serial number of the stripe.

Version number. Two characters specifying the version number

associated with the volume serial number.

File number. Two characters specifying the number of the ﬁle

within the volume.

Logical stripe number. Relative stripe number within the volume.

The logical stripe number may not correlate with the physical

stripe number due to demarked stripes.

Address of last block trailer. Byte number, within the stripe, of the

ﬁrst byte of the last block trailer.

Reserved.

Checksum. Chosen such that the sum, modulo 65536, of all bytes in

the header equals zero.

Revision M Mass Storage Extended Subsystem (MSE) 11-69

Format of Disk Files Read From 7990 by SSDEBUG

7990 Block Header and Trailer

Each 7990 block has a 16-byte header that precedes each block and a 16-byte trailer

that follows each block, or partial block, of data on the stripe. The 7990 block header

and trailer have the same format.

Byte(s)

0-1

2-3

4-7

8-9

A-E

Description

Block type. The following block types are valid.

Description

Block

Type

1016

A complete block is contained in this stripe.

First block segment of a multistripe block.

Intermediate block segment of a multistripe block.

Last block segment of a multistripe block.

Tape mark.

Block length. The number of bytes in a data block excluding the

block header and trailer. The block length of a tape mark is zero.

Block sequence number. The number of a given data block within a

ﬁle.

Block segment number. For complete blocks, the segment number is

0 (zero). For multistripe blocks, the segment number is the

sequence number of the block segment within the data block.

Reserved.

Checksum. Chosen such that the sum, modulo 256, of all bytes in

the header equals zero.

^*<^x

i<^^\

<>^53Sv

11-70 NOS Version 2 Analysis Handbook Revision M

0^iu_

/#^V

SSBLD

SSBLD reads a speciﬁed direct access permanent ﬁle that contains SSBLD statements.

These statements consist of 7990 component-oriented mnemonics that deﬁne the logical

mapping from component to component within the 7990 conﬁguration. SSBLD processes

these statements and generates a direct access permanent ﬁle which SSEXEC uses as

its unit device table (UDT).

The format of the SSBLD command is:

SSBLD,CF=conﬁle,BF=bldﬁle.

Parameter Description

CF=conﬁle Conﬁguration ﬁle on which SSBLD statements are written under

user index 377760s.

CF Same as CF=SUDT.

CF omitted Same as CF=SUDT.

BF=bldﬁle Build ﬁle on which SSBLD generates the UDT for SSEXEC. The

build ﬁle is under user index 377760s.

BF Same as BF=BUDT.

BF omitted Same as BF = BUDT.

SSBLD Statements

The 7990 hardware consists of a number of 7990 control units (CU) and a number of

7991 storage modules (SM). SSBLD uses two types of statements: those that link

individually controllable nodes and paths within each CU, and those that link SMs to

CUs. SSBLD statements must be processed in the following sequence or SSBLD aborts.

1. Statements that deﬁne the linkage between nodes and paths within each CU.

2. Statements that deﬁne the linkage between SMs and CUs.

NodelPath Linkage Statements

These SSBLD statements are order-dependent. There are seven types of statements for

a given CU and they must appear together in the following sequence and format:

1. CUest=da,CH01=st,CH02=st,CH03=st,CH04=st.

est EST ordinal of the CU (10s through 777s).

da Device address of CUest.

st Channel status (ON or OFF).

The CYBER channel number is associated with the 7990 channel interface (CIF)

number. The position (ﬁrst, second, and so on) of the channel in the CU

statement minus 1 deﬁnes the CIF number wired to that channel. For example,

CH02 is wired to CIF1. Also, if a single digit channel is used to connect to a

7990, it must have a leading 0 in the CU statement. For example, channel 3 is

entered as CH03.

Revision M Mass Storage Extended Subsystem (MSE) 11-71

SSBLD Statements

2. CIFn=DTIO=St,DTI1=St,DTO0=St,DT01=St.

n Device address of CIF (0 through 3).

DTI Data transfer input.

DTO Data transfer output.

st Data transfer status (ON or OFF).

3. DTIn=DIF0=St,DIF1=St.

n Device address of DTI (0 or 1).

DIF Device interface.

st Device interface status (ON or OFF).

4. DTOn=DIF0=st,DIF1=st.

n Device address of DTO (0 or 1).

st Device interface status (ON or OFF).

5. DIFn=DRC0=st,DRC1=st.

n Device address of DIF (0 or 1).

DRC Data recording controller.

st Data recording controller status (ON or OFF).

If two DRCs exist in a CU, their device addresses must be paired as 0 and 1 or 2

and 3.

6. DRCn=DRDO=st DRD7=st. or DRCm=DRD8=St....,DRD15=St.

n Device address of DRC (0 or 1).

m Device address of DRC (2 or 3).

DRD Data recording driver.

st Data recording driver status (ON or OFF).

DRCs 0 and 1 can link only to DRDs 0 to 7, and DRCs 2 and 3 can link only to

DRDs 8 to 15.

7. AIFn=ACO=st,AC1=St,AC2=st,AC3=St. or AIFn=AC4=st,AC5=St,AC6=St,AC7=st.

A^m\

AIF Accessor interface. )

n Device address of AIF (0 or 1).

AC Accessor.

st Accessor status (ON or OFF).

An AIF can link to either ACs in the range 0 to 3 or 4 to 7, but not to ACs

from both ranges.

11-72 NOS Version 2 Analysis Handbook Revision M

yfﬂ^Bfc^

0$ms

SSBLD Statements

SMICU Linkage Statements

The SM statement deﬁnes the linkage between SMs and CUs; its format is:

SMi d=st,ACn,CUest,CUest,DRDa=st,DRDa=st,DS=d,ST=s.

Parameter Description

id SM identiﬁer (A through H).

st Status (ON or OFF).

n Device address of AC (0 through 3).

est EST ordinal of the CU(s) connected to SMid.

a Device addr e ss o f t h e D RD(s) conne c t ed t o ACn.

d Number of DRDs for destaging per storage module (0 to 2).

Default is 1.

s Number of DRDs for staging per storage module (0 to 2). Default

is 2.

SSBLD and SSEXEC use the common deck COMBUDT to deﬁne the amount of

memory to reserve within SSEXEC for the unit device tables. Any attempt to change

the number of SMs or CUs speciﬁed in the common deck will cause SSBLD to abort.

SSBLD Example

The ﬁrst two SSBLD statements specify the total number of CUs and SMs in the

following conﬁguration. The ﬁrst statement speciﬁes CUs and the second statement

speciﬁes SMs. The rest of the statements deﬁne the 7990 conﬁguration.

CU41=0,,CH23=ON,f.

CIF1=DTI0=ON,,DTO0=ON,.

DTI0=DIF0=ON,DIF1=ON.

DTO0=DIF0-ON,DIF1=ON.

DIFD=DRC0=ON,DRC1=ON.

DIF1=DRC0=ON,DRC1=ON.

DRC0=DRD0=ON,DRD1=ON

DRC1=DRD0=ON,DRD1=ON,,,,,,.

AIF0=AC0=ON,,,.

SMA=ON,ACO,,CU41,DRD0=ON,DRD1=ON,DS=1,ST=2.

•COMMENT.

An asterisk (*) in the ﬁrst column of an SSBLD statement speciﬁes a comment

statement. Comment statements are not sent to the dayﬁle.

Revision M Mass Storage Extended Subsystem (MSE) 11-73

7990 Equipment Conﬁguration Constraints

Control paths and data paths within the 7990 conﬁguration have certain constraints

placed upon the allowable device addresses of the accessors and DRDs. The accessor

numbers and corresponding DRD numbers are:

Accessor

Number DRD Numbers

0 0 or 1

1 2 or 3

2 4 or 5

36 or 7

Accessors and DRDs that are to be coupled to two controllers must have the same path

numbers assigned from each controller. For example, if an accessor is plugged into ^^

accessor path 2 of CU1, then a DRD on the same storage module must be coupled to a ^^

DRC path of 4 or 5 from CU1. If a second controller is to be attached to this same

storage module, then the other accessor port must also be plugged into accessor path 2

of CU2 and a DRD must be coupled to a DRC path of 4 or 5 from CU2.

11-74 NOS Version 2 Analysis Handbook Revision M

yam**.

SSALTER

SSALTER displays the current 7990 hardware conﬁguration and allows an analyst to

change the status of certain elements in the 7990 conﬁguration. SSALTER works with

SSEXEC to make the speciﬁed changes to the existing 7990 conﬁguration.

The initial 7990 conﬁguration ﬁle is built by the SSBLD utility and stored as a

permanent ﬁle (default ﬁle name is BUDT) under user index 377760s on the default

family. This conﬁguration ﬁle is the unit device table read by SSEXEC during

initialization.

SSALTER can be run at SSEXEC initialization time or at any time after SSEXEC

has been initialized. SSALTER aborts if SSEXEC is not executing.

Console Input

SSALTER can be executed only from the system console. Input from the console

requires the following sequence of entries and responses:

1. Call SSALTER by entering:

X.SSALTER.

The B,0 display indicates the job sequence name (jsn) of SSALTER. To the right

of the entry appears the message:

REQUEST *K* DISPLAY

2. Activate the K display for that control point by entering:

K, jsn.

In this entry, jsn is the job sequence name of SSALTER as speciﬁed on the B,0

display.

3. Select the desired K display by entering one of the following:

K.SM=id.

This entry selects the SM K display (ﬁgure 11-16), which shows the 7990

conﬁguration of the storage module and DRDs complete with status information

(id is the SM identiﬁer, A through H).

K.CUest.

This entry selects the CU K display (ﬁgure 11-17), which shows the status of

various paths/nodes in the speciﬁed control unit (est is the EST ordinal of the

CU).

Revision M Mass Storage Extended Subsystem (MSE) 11-75

Console Input

DRD

7990 SUBSYSTEM - SSALTER

PCU SCU SM/DRD

ST ST ST

A42U

0 U

DRD U U

VALID COMMANDS ARE -

K.SMI ,CUNN=S.

K.SMI ,DRDN,CUNN=S.

K.+

I = SM IDENTIFIER

N = DRD DEVICE ADDRESS

NN = CU EST ORDINAL

S = STATUS - U FOR ONLINE

F FOR OFFLINE

M FOR MAINTENANCE

+ = DISPLAY NEXT SM

,A*^S\

Figure 11-16. SM K Display

When a DRD's status is changed to online, any cartridges in the DRD are placed in

service cells if space is available; otherwise, they are ejected from the storage module.

A clear cartridge is then loaded into the DRD, its cartridge label and CSN are

checked, and it is unloaded. The same cartridge is reloaded, its label and CSN are

checked again, and the cartridge is returned to its cubicle. The service cells are then

checked for cartridges. If a cartridge is found, it is either returned to the cubicle

matrix or ejected from the storage module, depending on the cartridge label's contents.

When a SM's status is changed to online, the accessor is ﬁrst cleared of any

cartridge it may contain. The cartridge is stored in a service cell if one is empty;

otherwise, the cartridge is ejected from the storage module. The accessor then picks a

clear cartridge from its cubicle, and immediately puts it back. If the SM's status is

changed to online after SSEXEC has been initialized, the service cells are then

checked for cartridges. Any cartridges found are ejected from the storage module.

11-76 NOS Version 2 Analysis Handbook Revision M

Console Input

00&S

7990 SUBSYSTEM - SSALTER

ORD ST

CU42 ON

CIF1 DTIO DTOO DIFO DRCO DRC1 AIFO

DTIOU DIFOU DIFOU DRCOU DRDOU

DRD1U

SMAOU

DTI1U DIF1U DIF1U DRC1U

DTOOU DTI1 DT01 DIF1

DT01U DIFOU DIFOU DRCOU AIF1

DIF1U DIF1U DRC1U SMAOU

CH ST CIF

24 ON

Figure 11-17. CU K Display

Revision M Mass Storage Extended Subsystem (MSE) 11-77

MSE Operational Procedures

Next is a summary of the operational procedures used in an MSE environment. These

procedures include modiﬁcations to existing procedures and new MSE procedures,

especially in the areas of permanent ﬁle backup and reloading and disk space

management. Refer to the descriptions of the appropriate utilities for details about

these operational procedures.

Initialization

MSE processing is initiated when the SSDEF utility is run to create the system ﬁles

necessary to allow storage of ﬁles on 7990. SSDEF creates both an SM map for each

SM that is to store ﬁle data on cartridges and eight 7990 catalogs (one for each

subfamily) for each family that is to have 7990-resident ﬁles. Once these system ﬁles

have been created, the SSLABEL utility can then be run to allocate portions of the SM

to each subfamily and to initialize cartridges for use by ﬁles belonging to these

subfamilies. Note that SSEXEC need not be executing when SSDEF is run, but

SSEXEC must be executing when SSLABEL is run. ^

When permanent ﬁles are deﬁned, they reside on disk until the SSMOVE utility is

run. Depending on the parameters speciﬁed, SSMOVE can destage ﬁles to 7990 and

release the disk space assigned to ﬁles so that it is available for general use.

^tfSx&V

11-78 NOS Version 2 Analysis Handbook Revision M

Disk Space Management

The availability of disk space must be managed more frequently in an MSE

environment than in a non-MSE environment. This is necessary because 7990-resident

ﬁles must be staged to disk before a user can access them, and their disk space is not

immediately released upon completion of the jobs that attached the ﬁles. Consequently,

disks are apt to approach a disk-full condition unless the SSMOVE utility is used

periodically to release disk space.

The SSMOVE utility destages selected ﬁles to 7990 and optionally releases their disk

space. It is recommended that the site analyst use SSMOVE as part of a periodic disk

space management procedure and also just prior to a full dump to reduce the amount

of data dumped to tape. (Refer to Dump Tape Management later in this section.)

The E,M display can be used to monitor the tracks on a device or set of devices.

When too few tracks (as deﬁned by the installation) are available on the direct access

devices for a family or subfamily, the SSMOVE utility should be run. SSMOVE reads

the PFC entries for the speciﬁed family, selects certain ﬁles to reside on 7990, and

decides which ﬁles should be released from disk.

Files can be destaged to 7990 without their disk space being released. This

destage-only feature can be used prior to a routine backup dump to create 7990

images for ﬁles with the BR=MD attribute and thereby eliminate the need to include

them on a backup dump tape. (Refer to Dump Tape Management later in this

section.) Having a 7990 image also provides backup for the ﬁle in special cases where

the permanent ﬁle is being updated at the same time PFDUMP is being run.

The destage-only feature can also be used to allow disk space to become available

immediately when SSMOVE is run again. That is, SSMOVE can directly release disk

space of ﬁles that have both disk and 7990 images but are to reside on 7990 only.

However, for ﬁles that have no 7990 image, there is a delay before the disk space

can be released because these ﬁles must ﬁrst be destaged to 7990.

Permanent ﬁle PFGMSE1 contains sample procedures FREEDSK and DESTAGE to

show how a site may use SSMOVE for disk space management. FREEDSK waits in

the rollout queue for a low disk space threshold ﬂag to be set. FREEDSK is then

rolled in for execution and calls SSMOVE to release disk space. The DESTAGE

procedure calls SSMOVE at set times to destage ﬁles to 7990 whose ﬁle size in PRUs

falls within the range speciﬁed by the MN = min and MX=max parameters. Each site

can change the SSMOVE parameters and directives in the sample procedures to

satisfy their requirements.

NOTE

Before using the FREEDSK and DESTAGE procedures, you must call the SSCONTL

procedure to build the programs necessary to monitor the low disk space threshold

and the time.

j * W " S

Revision M Mass Storage Extended Subsystem (MSE) 11-79

7990 Space Management

The availability of 7990 space must be managed because 7990 space is not ■>

automatically released when a permanent ﬁle is purged. This allows you to perform a

permanent ﬁle recovery to a time prior to the purging of the ﬁle. On a typical dump

tape, the entry for a 7990-resident ﬁle includes the PFC and permit entries, but not

the ﬁle data. A subsequent reload operation recovers only what was dumped; the PFC

and permit information, but not the ﬁle data. Thus, 7990 space for a purged permanent

ﬁle cannot be released until you no longer need to recover the ﬁle data, because the

ﬁle data resides on 7990 only and typically not on any dump tape.

A 7990-resident ﬁle is stored only on the cartridges within the cubicles of SMs

assigned to a subfamily. When destaging a ﬁle to 7990, SSEXEC selects the SM with

the most space available. If no SM has enough free space to store the ﬁle being

destaged, SSEXEC abandons that destage request, issues a message to the dayﬁle,

and continues to destage other ﬁles that ﬁt on the available 7990 space. To avoid

this problem, use the SSVAL utility periodically to release 7990 space for purged

ﬁles, thereby making it available for reuse. ^s^

When you want to reuse 7990 space currently allocated to ﬁles that have been

purged, he/she runs SSVAL with the RF and RL parameters speciﬁed. The RF

parameter speciﬁes an RDF, which is an auxiliary ﬁle produced by the permanent

ﬁle utility PFDUMP and which contains versions of the 7990 catalogs and PFC

entries that were current at the time of the dump. SSVAL examines the RDF to

identify unneeded 7990 ﬁles (those without PFC entries) and their 7990 space is

released. Choose which RDF to use based on the date and time of the dump that

produced it. That is, a particular RDF is used only if there is no need for a

permanent ﬁle recovery to a time prior to the dump that produced the RDF. **^

Once an RDF is used by SSVAL to release 7990 space, you should discard it and any

other RDFs for the same set of devices. These RDFs cannot be used as input to

SSVAL in any future run. The MSE software does not permit an RDF to be speciﬁed

by the RF parameter on an SSVAL call, unless it was produced by PFDUMP after

the last SSVAL run that released 7990 space was completed. This prevents the

inadvertent release of 7990 space that was previously released and reassigned to a

ﬁle.

11-80 NOS Version 2 Analysis Handbook Revision M

0^S

7990 Space Management

Example 1:

Assume that two PFDUMPs were run a week apart producing two release data ﬁles,

RDF1 and RDF2. Assume, also, that according to both RDF1 and RDF2, ﬁle AFILE

is an orphan (it has been purged but still resides on 7990). If the following call to

SSVAL is made, the 7990 space assigned to AFILE is released and made available for

reuse, if no errors were encountered.

SSVAL,RF=RDF1,RL.

Now, suppose that a new ﬁle, BFILE, which did not exist when RDF1 and RDF2 were

produced, is assigned to the 7990 space on which AFILE has resided. If SSVAL were to

accept either RDF1 or RDF2 on a subsequent SSVAL call, the 7990 space allocated to

ﬁle AFILE (which is now the space where BFILE resides) is released. However, any

RDF created after the SSVAL run that purged AFILE would not have an entry for

AFILE and, hence, SSVAL would not inadvertently release the 7990 space presently

allocated to BFILE.

After SSVAL uses a release data ﬁle to release 7990 space, the corresponding backup

tape from PFDUMP is the oldest tape that should be used for the ﬁrst reel of a

future incremental load. The ﬁrst reel determines which ﬁles are to be reloaded

according to the CIR on the tape (refer to PFLOAD in section 17, Permanent File

Utilities). Use of an older tape as the ﬁrst reel can allow the PFC entry for a purged

ﬁle to be restored even though its 7990 space may have been reassigned. The MSE

software does not prevent loading from the wrong dump tape, but does detect as an

error an attempt to stage a ﬁle whose data was overwritten (see Example 2).

Example 2:

Assume that RDF1 and RDF2 are as in the previous example, and that ﬁle CFILE is

an orphan according to RDF2, but not according to RDF1. The following SSVAL call

releases 7990 space allocated to CFILE.

SSVAL,RF=RDF2,RL.

Now suppose that the dump tape created by the PFDUMP run that produced RDF1 is

used as the ﬁrst reel of an incremental PFLOAD. Then the PFC and permit entries for

ﬁle CFILE are reloaded, but, in fact, ﬁle CFILE does not exist on 7990. This will be

detected as an error when a user attempts to attach CFILE.

Revision M Mass Storage Extended Subsystem (MSE) 11-81

Backup and Recovery

Permanent ﬁle backup and recovery methods in an MSE environment are similar to

those in a non-MSE environment. It is assumed that most sites with disk-resident

permanent ﬁles have a ﬁle backup procedure consisting of a full dump followed by a

series of incremental dumps. The assumed reload procedure is to load from tapes

starting with the most recent incremental dump tape and continuing through older

incremental dump tapes until the most recent full dump tape is processed. These

procedures can be modiﬁed as follows to accommodate 7990-resident ﬁles.

/ * ^ K

File Dumping

The normal backup dump procedure is to dump ﬁles based on the modiﬁcation date

ﬁeld in the PFC entries for the ﬁles. New parameters to PFDUMP provide the

capability to dump ﬁles that have been modiﬁed during a certain interval, which is

useful in recycling old dump tapes (refer to Dump Tape Management later in this

section). For ﬁles that reside only on 7990, you can either stage the ﬁle to disk and

thereby include it in the dump or suppress staging and include only the ﬁle's PFC and

permit information on the dump tape. If the device or family being dumped has

7990-resident ﬁles, the following procedures should be included with the normal

dumping procedures.

Full Dumps

Full dumps should be performed as usual, but OP=S should be speciﬁed on the call to

PFDUMP to suppress staging. Thus, all ﬁles on disk are dumped but only the PFC and

permit entries for 7990 ﬁles without disk images are included in the dump. It is

unnecessary and not feasible to dump the ﬁles that reside only on 7990. Specify OP=S

and SD on the call to PFDUMP; use the RD=ﬁlename parameter at your own

discretion.

Parameter

OP=S

RD=ﬁlename

Description

Selects no staging. If the SD option was inadvertently used on any

previous nonincremental dump, an incremental dump should be

taken prior to the full dump.

Updates the dump date/time ﬁeld on each master device dumped to

permit ﬁles created or modiﬁed prior to this dump to be released

when SSMOVE is run again. Refer to Destage and Release

Algorithms earlier in this section.

Speciﬁes that PFDUMP is to write a release data ﬁle to be used

later by SSVAL. RD=ﬁlename is included or excluded on a call to

PFDUMP depending on how often you want to purge unneeded

7990 ﬁles.

.<^s£v

11-82 NOS Version 2 Analysis Handbook Revision M

0$m\

0m*s

File Dumping

Incremental Dumps

Incremental dumps should be performed as usual. No staging should occur because any

ﬁles that were created or modiﬁed since the last dump will still be on disk if a tape

backup is required. The following parameters should be speciﬁed on the call to

PFDUMP.

Parameter Description

OP=M Selects incremental dumps.

SD Updates the date/time ﬁeld on each master device dumped, as

previously described for full dumps.

AD=yymmdd Selects the date just prior to the previous incremental or full dump

and of the device. Hence, all ﬁles not included in the last dump are

AT=hhmmss dumped.

The following parameters should be used with caution.

Parameter Description

OP = S Selects no staging. If the SD option was inadvertently used on any

recent partial dump, OP=S should not be speciﬁed so any ﬁle with

a BR=Y attribute that had its disk space inadvertently released

will be included in the dump.

RD=ﬁlename Speciﬁes that PFDUMP is to write a release data ﬁle to be used

later by SSVAL, as previously described for full dumps.

Dump Tape Management

In a non-MSE environment, it is possible to make all previous dump tapes available for

use after a full dump is completed. However, in an MSE environment, such dump tape

recycling cannot occur because ﬁles that have resided only on 7990 for some time are

probably not included on any recent dump tape. When a low percentage of ﬁles on a

tape are active ﬁles, it is possible to redump any remaining ﬁles in order to recycle

the tape by running PFDUMP and specifying OP=M, BD, and BT parameters, but not

specifying the OP = S parameter.4 This performs a dump of all ﬁles that were last

dumped before the date and time speciﬁed on the PFDUMP call. If necessary, ﬁles will

be staged in from 7990; after the dump is completed, the disk space for these staged

ﬁles is released. Thus, ﬁles that might not be on any dump tape made after the date

speciﬁed by the BD parameter are dumped and tapes created before this date can be

recycled.

4. It is possible to specify the AD and AT parameters also, thereby dumping ﬁles that were dumped during

an interval after the date and time speciﬁed by AD and AT and before the date and time speciﬁed by BD

and BT (refer to parameters for PFDUMP). This performs a dump of all ﬁles last dumped during the

interval speciﬁed (including 7990-resident ﬁles) and allows your site to recycle all dump tapes created

during this interval.

Revision M Mass Storage Extended Subsystem (MSE) 11-83

File Dumping

The number of dump tapes created can be reduced by using SSMOVE to reduce the

amount of data written on these dump tapes. If SSMOVE is run prior to an

incremental dump, ﬁles with the BR=MD attribute can be destaged to 7990 and

excluded from the dump. That is, the incremental dump will exclude the ﬁles with a

BR=MD attribute that are on both 7990 and disk. If SSMOVE is run prior to a full

dump, disk space for ﬁles that do not have to be on disk can be released and these

ﬁles are not written on the dump tape. The following procedure involving dumps and

SSMOVE runs can be used for efﬁciency.

1. Run SSMOVE to destage ﬁles with the BR=MD attribute.

2. Take an incremental dump to back up ﬁles with the BR=Y attribute.

3. Run SSMOVE to release the disk images of ﬁles backed up in step 2.

4. Take a full dump to reduce the number of dump tapes involved if a reload is

required.

SM Map/7990 Catalog Backup

An SM map is created for each SM in an MSE environment and is updated by the

SSLABEL, SSVAL, and SSDEBUG utilities. (Refer to the deﬁnition of SM map earlier

in this section.) It is recommended that the SM map be copied on tape or on another

device or family immediately after every update of the SM map. PFDUMP can be used

to do this by dumping user index 377760s with FM equal the default system family

name, but only if SSEXEC is not running. When SSEXEC is executing, it has the SM

map ﬁles attached in modify mode; hence, PFDUMP will not dump the SM maps.

If the device containing subfamily 0 of the default family is reloaded, the latest SM

maps should be recovered from the backup copies. After recovering the SM maps, run

the SSVAL utility with the AM and FM parameters speciﬁed to ensure that the SM

map and 7990 catalog entries match. If they do not, corrective action should be taken

as described in Error Conditions and Corrective Actions later in this section. Failure

to retain or recover the latest SM maps can cause problems with future SSLABEL

runs (refer to the deﬁnition of SM map earlier in this section).

A 7990 catalog is created for each subfamily of a family that can have 7990-resident

ﬁles and is updated by the SSLABEL, SSMOVE, SSVAL, and SSDEBUG utilities.

When PFDUMP copies permanent ﬁles to tape for a master device or all devices in a

family, it ensures that corresponding 7990 catalogs are also dumped. (If SSEXEC has

a 7990 catalog attached in modify mode, PFDUMP interfaces with SSEXEC so that

the 7990 catalog is included on the dump tape. While PFDUMP is accessing the 7990

catalog, any ﬁle destage requests for that subfamily are delayed.) Consequently, if

PFLOAD is used to reload all ﬁles, the 7990 catalogs are automatically recovered and

entries for all 7990-resident ﬁles reﬂected in recovered PFC entries will exist in the

recovered 7990 catalog.

If PFLOAD is run after the SSLABEL utility has updated the 7990 catalog and SM

map but before the 7990 catalog has been backed up, the recovered 7990 catalog will

not match exactly the corresponding SM map. In this case, recovery procedures (as

described in 7990* Catalog/SM Map Mismatch later in this section) must be taken to

bring the SM map, 7990 catalog, and cartridge label information into agreement.

/^^s\

A^ms

r-c%\

11-84 NOS Version 2 Analysis Handbook Revision M

File Reloading

PFLOAD runs should be made as usual. PFLOAD reads the CIR records from the most

recent incremental dump tape to control loading of ﬁle data and PFC and permit

information. For 7990-resident only ﬁles, PFLOAD reloads only the PFC and permit

information (if PFDUMP dumped only this information). The following situations

involve PFLOAD in an MSE environment.

• If one device needs to be reloaded, the DN or TD parameter specifies the device.

The OP=Z and UD parameters should not be speciﬁed.

• If PFLOAD is used for other than a device recovery, both OP=Z and UD should

be speciﬁed. The OP = Z parameter ensures that the PFC entry does not link to a

7990 ﬁle that may no longer exist, and UD ensures that the ﬁle is included in

the next incremental or full dump.

• If a file has a nonzero asa value in its PFC entry (indicating that the file has a

7990 image), the SSCVER ﬂag in the PFC entry will be set when the PFC and

permit information are reloaded. If the ﬁle is successfully staged to disk when it

is attached, this ﬂag is cleared. Otherwise, the ﬂag remains set. This can be used

to indicate the possibility of an operational error, such as omitting OP=Z when a

ﬁle or set of ﬁles is reloaded from an obsolete dump tape.

• If any unscheduled device reload that includes recovery of a 7990 catalog and/or

SM map is made, SSVAL should be run with the AM and FM parameters

speciﬁed. SSVAL examines the PFC entries for 7990-resident ﬁles, the 7990

catalog, and the SM map to detect any error conditions that exist.

• If dumping a file to tape from one user and then reloading the file to a different

user, you must use the OP=Z parameter on the PFLOAD utility. Otherwise the

PFC entry still points to the old SFMCAT, which causes an error indication for

the ﬁle.

0^\\

Revision M Mass Storage Extended Subsystem (MSE) 11-85

Error Conditions and Corrective Actions

It is possible for operational, hardware, or software malfunctions to cause a variety of

error conditions to exist or appear to exist in the 7990 catalogs, SM maps, PFC

entries, or cartridge labels. The SSVAL utility can analyze the 7990 catalogs, SM maps,

and PFC entries to detect these errors, and the SSLABEL utility can identify problems

with cartridge labels. The following paragraphs describe some speciﬁc error conditions

and the actions that should be taken to correct them. Most of the errors are caused

either by incorrect operational procedures that result in the recovery of the wrong

version of an MSE system ﬁle, or by abnormal termination of SSEXEC or MSE utility

processing, which can interrupt the updating of an MSE system ﬁle.

Clearing Permanent File Error Flags

The permanent ﬁle error ﬂags that have been set by SSEXEC in the PFC can be

cleared by the CE parameter of the CHANGE command. The job must be of system

origin.

Permanent File Recovery

If a permanent ﬁle cannot be accessed or is included by SSVAL on the validation

report, try the following procedures to recover a copy of the ﬁle.

1. Attach the ﬁle. If the ﬁle has a disk image or can be staged in from 7990, the

ﬁle data will be available on disk and can be copied to another ﬁle. SSVAL

should then be run as described in step 2.

2. If the ﬁle cannot be attached, reload the ﬁle from a dump tape using PFLOAD ^"^%

with OP=Z. To identify the PFDUMP run that saved a backup copy of the ﬁle,

check the output from SSVAL or PFCAT, which includes date and time ﬁelds from

the PFC entry for the ﬁle. Depending on the BR attribute (backup requirement) of

the ﬁle, a backup copy on a dump tape may not exist.

If the ﬁle can be reloaded to disk, SSVAL should be run as follows:

SSVAL,FM=fami 1yname,SB=subfamily,FX=n,AM.

familyname and subfamily identify the family and subfamily of the recovered ﬁle

and n is sufﬁciently large that the ﬁxing/ﬂagging will be done. SSVAL sets the

obsolete ﬂag in the PFC entry because the ﬁle now has a disk image; thus the

ﬁle is disassociated from the error condition. The error condition may still exist in

the 7990 catalog or SM map, but you can analyze it and correct the error by

using the SSDEBUG utility. Before using SSDEBUG to correct the error, you

should take an incremental dump so the old PFC entry with an asa value

pointing to a problem 7990-resident ﬁle will not be reloaded if a device reload

occurs before the next scheduled backup dump.

3. If steps 1 and 2 fail, try to regenerate the ﬁle. Depending on the particular

problem and status of the cartridges to which the ﬁle was destaged, the OP=RF

or OP=RS directive to SSDEBUG might recover the ﬁle data.

4. If the ﬁle cannot be recovered, purge it so there is no PFC entry pointing to a

problem 7990-resident ﬁle.

11-86 NOS Version 2 Analysis Handbook Revision M

Cartridge Restoration and Reuse

J$$™&\

Cartridge Restoration and Reuse

If a cartridge is unexpectedly found anywhere outside its assigned cubicle, follow these

steps to restore the cartridge so it can be used again:

1. Put the cartridge in the input tray of the SM to which it is assumed to belong

and run SSLABEL with the OP=RS directive. If the cartridge label information

agrees with the SM map entry for the cubicle with the coordinates (y,z) written

on the cartridge label, the cartridge will be restored to that cubicle and the 7990

catalog entry for the cartridge is updated to clear the lost ﬂag. Otherwise, the

cartridge is put in the exit tray and the cartridge label information is included in

the output from SSLABEL.

2. If the cartridge was not restored to its SM cubicle, examine the label information

reported by SSLABEL. If the cartridge is assigned to a different SM, repeat step 1

using the correct SM. If the family name is blank, the cartridge has a scratch

label and can be assigned normally with the OP = AM directive to SSLABEL.

3. If steps 1 and 2 fail, examine the SM map entry for the cubicle with the

coordinates (y,z) written on the cartridge label to see how the cubicle is assigned.

Examine the cartridge, if any, in this cubicle. Also, it might be useful to run

SSVAL to ensure the SM map and 7990 catalog entries for the cartridge match or

to identify affected permanent ﬁles, if any, if they do not match.

4. If step 3 does not produce any useful information and the cartridge cannot be

restored, use the OP=FX directive to SSLABEL to rewrite the cartridge label and

thereby make a scratch cartridge. It might be useful to ﬁrst run SSDEBUG with

the OP = RS directive to preserve the data on the cartridge.

0$&S,

Revision M Mass Storage Extended Subsystem (MSE) 11-87

Removal of Faulty or Missing Cartridges

Follow these steps to remove a cartridge that has unrecoverable errors, or to release

the SM map entry for a cartridge that is missing.

For unallocated cartridges:

1. Attempt to remove the cartridge with the SSLABEL OP=RM directive.

2. If CARTRIDGE NOT FOUND is reported, repeat step 1 specifying the LT

parameter.

3. If CARTRIDGE NOT EMPTY is reported, at least one AU on the cartridge is still

allocated.

For allocated cartridges:

1. Use the SSLABEL OP=FC directive to inhibit further allocation of the cartridge.

2. Run SSVAL specifying the AM parameter. Examine the report for possible

discrepancies between the SM map, 7990 catalog, and PFC entries related to this

cartridge.

3. If no 7990 catalog entry corresponds to the SM map entry, use the OP = RC

directive to SSDEBUG to remove the map entry.

4. If no discrepancy is reported by SSVAL, the affected ﬁles should be attached in

write mode (if possible), reloaded using PFLOAD with the OP=Z parameter, or

purged. Then SSVAL release processing should be done to release the 7990 space

on the cartridge, and the cartridge can be removed with the SSLABEL OP=RM

directive.

>c^.

11-88 NOS Version 2 Analysis Handbook Revision M

7990 Catalog/SM Map/PFC Problems

If error conditions exist in the 7990 catalogs or SM maps, the recovery procedure is to

correct the appropriate entry rather than reload the entire ﬁle. Reloading other than

the latest copy of the SM map is not feasible because it can contain entries for several

families and these entries would not be up to date. Reloading an old copy of the 7990

catalog is not desirable because any permanent ﬁle that was destaged and released

from disk since the last incremental dump will be lost. The following paragraphs

describe corrective action for speciﬁc problems with 7990 catalogs, SM maps, or PFC

entries.

7990 CatalogISM Map Mismatch

The following examples of 7990 catalog/SM map mismatches can appear as error type 1

or 2 on the SSVAL output report, which identiﬁes the mismatch and the permanent

ﬁles or orphans affected. Before taking any corrective action, the analyst should ensure

that operational errors did not result in loading the wrong version of the SM map or

7990 catalog.

• If a cartridge has a 7990 catalog entry indicating that all AUs on the cartridge

are available for allocation but is not identiﬁed in any SM map entry (type 1

mismatch), use the OP = RL directive to SSDEBUG to remove the 7990 catalog

entry for the cartridge.

• If a cartridge is identified in an SM map entry but has no 7990 catalog entry

(type 2 mismatch), use the OP=RC directive to SSDEBUG to update the SM map

entry to unassigned and empty status. This directive will write a scratch label for

the cartridge, if any, and place the cartridge in the exit tray.

• If a cartridge has a 7990 catalog entry indicating that at least one AU on the

cartridge is allocated, recover the disk space for the affected permanent ﬁle, if

any, or purge the ﬁle. Run SSVAL to purge unneeded 7990 ﬁles and thereby

make the cartridge unallocated. Use the OP = RL directive to SSDEBUG to remove

the 7990 catalog entry for the cartridge.

NOTE

It sometimes takes two SSVAL/SSDEBUG runs to completely clear up type 1 or type

2 errors. Therefore, it is recommended that SSVAL be rerun with the AM parameter

after an SSDEBUG run using OP=RC or OP=RL.

Revision M Mass Storage Extended Subsystem (MSE) 11-89

7990 Catalog/SM Map/PFC Problems

7990 Catalog Chain Problems

AUs that are allocated to a 7990-resident ﬁle are chained together in the 7990 catalog

to identify the sequence of AUs that contain the ﬁle. (Refer to the deﬁnition of

cartridge earlier in this section.) For most ﬁles, the AUs in the chain reside on the

same cartridge. Hence, one disk access is sufﬁcient to update information for the entire

chain at one time. However, a 7990-resident ﬁle can reside on several cartridges, and

several disk accesses may be necessary to update its chain. If SSEXEC is interrupted

during such an update, the resultant chain will be incomplete. SSVAL identiﬁes

incomplete chains as error type 4, 5, or 6 on the validation report. The following are

types of incomplete chains.

• If an incomplete chain is produced while a 7990-resident file is being purged, it is

a chain fragment (one without an HOC entry).

• If an incomplete chain is produced while a file is being destaged to 7990, it is an

ill-formed chain (one without an EOC entry).

• If an ill-formed chain is not corrected, the AU that should have been linked to it -^%,

may be reused and become part of a good chain. Although the ill-formed chain

may appear to be complete, the ill-formed chain and the good chain are

intersecting.

The corrective action for any of these bad chain conditions is:

1. Run SSVAL with the FM parameter speciﬁed and use a sufﬁciently large FX=n

value so that SSVAL updates the 7990 catalog entries affected by the problem

chains.

2. Recover the permanent ﬁles, if any, that have 7990 images potentially affected by

these error conditions. (The SSVAL report identiﬁes these ﬁles.) Use the procedure

described under Permanent File Recovery earlier in this section.

3. Take an incremental or full backup dump. This prevents PFC entries with

inaccurate asa values or problem 7990 chains from being recovered if a device

reload is necessary.

4. Run SSDEBUG with the OP=RP directive to make each AU on the problem chain

available for allocation. Thus, the AUs can be included in a new chain without /SiB%

any problems. In the case of intersecting chains, all chains that have an AU in '

common with another chain should be released before any further destage

attempts; otherwise, errors will occur when a user tries to attach a ﬁle that is

contained on an intersecting chain.

PFC/7990 Catalog Mismatch

It is possible that an old version of a 7990 catalog can be reloaded because of incorrect

operational procedures. If so, the asa value in the PFC entry for a ﬁle can be invalid

according to the obsolete 7990 catalog or can point to a problem chain. If the correct

version of the 7990 catalog is available on tape, it should be reloaded. If it is not

available, the following action is appropriate. If the ﬁle also has a disk image, SSVAL

should be run to clear the asa ﬁeld in the PFC entry. If there is no disk image, the

analyst should reload the ﬁle from tape, if possible, and then run SSVAL to clear the

asa value.

11-90 NOS Version 2 Analysis Handbook Revision M

Tape Alternate Storage 12

Introduction........... 12-1

Tape Alternate Storage Terminology 12-1

Organization and Management of Destage Dump Tapes 12-3

Tape Characteristics 12-3

Tape Data Format 12-3

Destaging Files to Tape 12-3

Releasing Disk Space 12-5

R e s t o r i n g D e s t a g e d F i l e s t o D i s k R e s i d e n c e 1 2 - 6

Modifying Destaged Files .... , 12-7

Recycling Destage Dump Tapes 12-7

MAGNET Command 12-8

GENPFD Utility 12-9

Calling the GENPFD Utility 12-9

GENPFD Input Directives 12-10

File Selection Processing ...: .. 12-13

Limitations 12-13

0^*.

0m^-

j0^ahjix\

Tape Alternate Storage 12

Introduction

This section describes the use of magnetic tape as an alternate storage medium for

NOS permanent ﬁles.

Tape Alternate Storage Terminology

The following terms apply to the use of tape alternate storage.

Alternate Storage

The storage of permanent ﬁle data on external media other than mass storage

such as tape alternate storage or MSE. When a ﬁle resides on alternate storage,

the ﬁle's permanent ﬁle catalog (PFC) entry and permit data still reside on disk,

but the disk space occupied by the ﬁle data can be released.

Alternate Storage Address (ASA)

The pointers in a ﬁle's PFC entry that point to the location of the ﬁle data on

tape alternate storage or MSE.

Destage Dump Tapes

Tapes that contain ﬁles that were destaged from disk using the PFDUMP utility

with the DT parameter speciﬁed.

Destaging

The process of creating an alternate storage copy of a ﬁle on tape alternate

storage or MSE. Files are destaged to tape alternate storage when the site

executes PFDUMP with the DT parameter speciﬁed. Files are destaged to MSE

when the site executes SSMOVE with the appropriate parameters.

Permanent File Catalog (PFC)

The system catalog containing the ﬁle name, owner identiﬁcation, disk pointers,

alternate storage pointers, and other attributes of a permanent ﬁle. Refer to

Permanent File Catalog Entry in section 17, Permanent File Utilities, for the

format of the PFC entry.

Primary VSN

The volume serial number (VSN) of a single reel destage dump tape or the VSN

of the ﬁrst reel of a multireel set of destage dump tapes. The VSN must be two

alphanumeric characters followed by a four-digit decimal number; the number

must be in the range 0000 through 4095.

Revision M Tape Alternate Storage 12-1

Tape Alternate Storage Terminology

Secondary VSN

The VSN of a single reel destage dump tape or the VSN of the ﬁrst reel of a

multireel set of destage dump tapes having a sequence number (the last four

characters of the VSN) in the range 5000 through 9095. The existence of a

secondary VSN assumes the existence of a primary VSN having the same

two-character preﬁx and a sequence numbeF of 5000 less. The primary and

secondary VSN tapes, or multireel sets of tapes, are assumed to contain identical

data although the individual reels of multireel sets cannot be assumed to be

identical.

Staging

The process of restoring ﬁle data to disk residence from a copy on alternate

storage. Staging is initiated when a user executes an APPEND, ATTACH, GET, or

OLD command for a ﬁle that is not currently disk resident.

,/*®^\

12-2 NOS Version 2 Analysis Handbook Revision M

0^S

0i^s

0ms.

Organization and Management of Destage Dump Tapes

The tape characteristics and data format required for destage dump tapes are described

below, followed by some of the procedures required to manage destage dump tapes.

Tape Characteristics

Destage dump tapes must be labeled, I format, and nine track. The tapes can be

written at any nine-track tape density. It is possible to have multireel sets of tapes,

but normally single reels are created whenever possible to avoid the overhead of reel

switching when restoring ﬁles to disk. The VSN of the ﬁrst reel of a set of destage

dump tapes must be in the following format:

ppnnnn

where pp is a preﬁx consisting of any two alphanumeric characters and nnnn is a

decimal sequence number in the range from 0000 through 4095 or in the range from

5000 through 9095. If the sequence number is 5000 or greater, the secondary tape set

is assumed to be a duplicate of a primary tape set that has the same preﬁx and a

sequence number of 5000 less.

Tape Data Format

All destage dump tapes are written by PFDUMP and have exactly the same format as

other tapes written by PFDUMP including ﬁle permits. However, no PFC-only ﬁles are

present on destage dump tapes. It is also possible to load, copy, or list ﬁles from a

destage dump tape using the PFLOAD, PFCOPY, or PFATC utility, respectively.

Destaging Files to Tape

Destage dump tapes are created by executing PFDUMP with the DT (destage to tape)

parameter speciﬁed. When the DT parameter is used, PFDUMP veriﬁes the following:

• The dump file specified by the T parameter must be assigned to a nine track, I

format, labeled tape. The VSN must be in the format ppnnnn as described earlier

(refer to Tape Characteristics) and the sequence number must be in the range

from 0000 to 4095.

• If a verify file is specified (VF parameter), it too must be assigned to a nine

track, I format, labeled tape and its VSN must have the following relation to the

VSN for the primary dump tape: the two-character preﬁx must be the same and

the sequence number must be greater by 5000.

After the dump tape or tapes are successfully written, PFDUMP enters the primary

dump tape VSN in the PFC entries of the dumped ﬁles. If a verify ﬁle is written, a

ﬂag is also set in the PFC entry indicating the existence of the secondary VSN. At

this point, the disk space of the dumped ﬁles can be released, if necessary, if the ﬁle's

backup requirements are satisﬁed.

Revision M Tape Alternate Storage 12-3

Destaging Files to Tape

Although general PFDUMP selection parameters can be used to directly select ﬁles for

a destage dump, sites will normally use a separate ﬁle selection utility such as

GENPFD to perform this function. (The GENPFD utility is described later in this

section.) This procedure has the following advantages:

• Files can be prioritized by the time since they were last accessed or modified.

• Special user index or file name patterns can be special cased according to site

requirements.

• The GENPFD selection utility can select files that will reasonably fill a single

tape reel without overﬂowing to additional reels. The creation of single reel tapes

is desirable because it avoids the overhead of reel switching when ﬁles are

restored to disk.

When a ﬁle selection program such as GENPFD is used for destage processing, the

following sequence of events should be followed:

1. Execute PFDUMP with the IP (inhibit processing) and DT (destage to tape) **%,

parameters speciﬁed and the summary ﬁle selected with the S parameter.

Specifying PFDUMP parameters such as LS, US, BD, BT, etc., can be used to

screen out desired classes of ﬁles. The TA=N and CA=N parameters can be used

to exclude ﬁles that already have current alternate storage copies (from previous

destage dumps).

2. Execute the ﬁle selection program (GENPFD) using the PFDUMP summary ﬁle as

input. The ﬁle selection utility writes UI and PF utility directives for the selected

ﬁles to the utility directives ﬁle.

3. Reexecute PFDUMP (with the same parameters as speciﬁed in step 1, but without

specifying the IP parameter) using the utility directives ﬁle generated in step 2 as

an input ﬁle.

When a destage dump is performed, ﬁles on user name SYSTEMX and on the MSE

catalog user names SUBFAMO through SUBFAM7 are unconditionally excluded even if

they meet the selection criteria.

Example:

Destage enough ﬁles to ﬁll one tape (alternate storage) with ﬁles from family TEST

whose data has not been modiﬁed since 87/07/01 and have not already been destaged.

PFDUMP,FM=TEST,DT,OP=T,BD=870701,TA=N,IP,S.

REWIND,SUMMARY.

RETURN,UDIR.

GENPFD,PO=D.

REWIND,UDIR.

PFDUMP,FM=TEST,DT,OP=T,BD=870701,TA=N,I=UDIR.

In this destaging example, if GENPFD does not ﬁnd enough ﬁles to ﬁll a 2400-foot

tape at GE density, the utility will not generate a UDIR ﬁle. This will cause the

second PFDUMP to abort without dumping any ﬁles.

y**«^V

SWt$K

12-4 NOS Version 2 Analysis Handbook Revision M

0^S.

Releasing Disk Space

The disk space of a ﬁle that resides on tape alternate storage or MSE can be released

if the ﬁle's backup requirements are met. The BR parameter on the DEFINE, SAVE,

and CHANGE commands allows a user to specify a backup requirement of Y (yes), MD

(media dependent), or N (none).

Parameter

Option Description

BR=Y The disk space of a ﬁle with a backup requirement of BR=Y can be

released if either of the following conditions is satisﬁed:

• At least two copies of the file data exist on alternate storage media.

This could be one copy on MSE and one copy on tape or two copies

(primary and secondary VSNs) on tape.

• One copy of the file data exists on an alternate storage medium

(either MSE or tape) and the release inhibit date/time of either the

ﬁle's master device or the ﬁle's residency device (as set by a

PFDUMP with the SD parameter speciﬁed) is greater than the ﬁle's

utility control date/time.

BR=MD The disk space of a ﬁle with a backup requirement of BR=MD can be

released if the following condition is satisﬁed:

• One copy of the file data exists on MSE or tape alternate storage.

The PFREL utility is used to perform the releasing of disk space. As with PFDUMP,

PFREL can be used with a ﬁle selection utility such as GENPFD in the following

sequence of steps:

1. Execute PFREL with the IP (inhibit processing) parameter speciﬁed and the

summary ﬁle selected with the S parameter. Files can be excluded by specifying

selection parameters. In particular, OP=A can be used with the BD and BT

parameters to exclude ﬁles that have been accessed recently.

2. Execute the ﬁle selection program using the PFREL summary ﬁle as input. The

0^^ ﬁle selection utility writes UI and PF utility directives for the selected ﬁles to the

utility directives ﬁle.

3. Reexecute PFREL (with the same parameters as speciﬁed in step 1, but without

specifying the IP parameter) using the utility directives ﬁle generated in step 2 as

an input ﬁle.

The PFREL utility can be used to release the disk space of ﬁles that reside on MSE as

well as those ﬁles that reside on tape alternate storage. However, the MSE utility

SSMOVE performs release processing only for MSE resident ﬁles. PFREL does not

release the disk space of ﬁles cataloged on user name SYSTEMX or on the MSE

catalog user names SUBFAMO through SUBFAM7 even if the selection criteria are met.

Revision M Tape Alternate Storage 12-5

Restoring Destaged Files to Disk Residence

Example 1:

The following example illustrates releasing disk space for destaged ﬁles.

Release the disk space for all ﬁles from family TEST that have not been accessed

since 87/07/01 and for which backup requirements have been met.

PFREL,FM=TEST,OP=A,BD=870701.

Example 2:

Using the same ﬁle selection criteria as in example 1, release enough ﬁles so that

10008 tracks are available on device number 40 and 12008 tracks are available on

device number 41.

PFREL,FM=TEST,OP=A,BD=870701,S,IP.

REWIND,SUMMARY.

RETURN,UDIR.

GENPFD,PO=R,Z./RL,40,1000/RL,41,1200

REWIND,UDIR.

PFREL,FM=TEST,I=UDIR.

Restoring Destaged Files to Disk Residence

A ﬁle that does not reside on disk is restored to disk residence (staged) when a user

accesses the ﬁle with an APPEND, ATTACH, GET, or OLD command. If the RT (real

time) parameter is speciﬁed on an ATTACH, GET, or OLD command and staging is

required, the job continues executing without waiting for the staging operation to

complete. In this case, the caller must subsequently reissue the ATTACH, GET, or

OLD command to access the ﬁle after the staging operation completes.

NOTE

In order to allow staging of ﬁles from tape alternate storage to disk, the site must

enter the ENABLE,TAPE PF STAGING command (via DSD or IPRDECK) and must

specify the SJ parameter on the MAGNET command in the MAG subsystem startup

procedure.

,/«8^.

12-6 NOS Version 2 Analysis Handbook Revision M

yﬂ$^\

Modifying Destaged Files

When you modify a ﬁle having a copy or copies on alternate storage media, the data

on alternate storage is no longer current. Therefore, when a direct access ﬁle is

attached in a writable mode or an indirect access ﬁle is saved or replaced, the system

removes the pointers to all alternate storage copies in the ﬁle's PFC entry. At this

point, the ﬁle has to be destaged again (to tape or MSE) if it is to reside on alternate

storage.

Recycling Destage Dump Tapes

As ﬁles are modiﬁed and purged over a period of time, new destage dump tapes are

created and the percentage of active ﬁles on older destage dump tapes decreases.

Destage dump tape recycling consists of procedures to move the remaining active ﬁles'

data on old destage dump tapes to new destage dump tapes so that the old tapes can

be reused. Performing a destage PFDUMP (DT parameter) with staging enabled (OP=S

not speciﬁed) of all ﬁles on a tape or set of tapes copies all ﬁle data to a new tape

and changes all alternate storage references from the old tapes to the new tape.

Since only a subset of the ﬁles on tapes selected for recycling are typically active, a

site normally wants to include other ﬁles to ﬁll a new tape. The following example

illustrates this procedure.

PFDUMP,IP,S,DT,FM=TEST,TA=N,TA=AB0005,TA=AB0006.

GENPFD,OP=D,Z./SV,AB0005/SV,AB0006/DL,2200/DT,0

REWIND,UDIR.

PFDUMP,DT,FM=TEST,I=UDIR.

The TA parameters on the ﬁrst PFDUMP command select ﬁles residing on VSNs

AB0005 and AB0006 (the tapes being recycled) and ﬁles not residing on tape alternate

storage as candidates for destage. The SV directives cause GENPED to select all ﬁles

residing on VSNs AB0005 and AB0006 regardless of the priority of the modiﬁcation

date derived priority and priority group. GENPFD selects other ﬁles using the normal

priority criteria to satisfy the destage limit speciﬁed by the DL directive. The DT

(destage threshold) directive performs a recycle dump even if there are not enough ﬁles

selected to ﬁll the tape.

To help determine which destage dump tape VSNs should be recycled, you may want

to write a program that reads the PFCAT summary ﬁle and generates a list of how

many ﬁles are still active on each destage dump tape VSN.

You can determine if any current ﬁles reside on a particular destage dump tape VSN

by executing PFCAT with the TA=vsn parameter speciﬁed.

After you perform a recycle dump for a tape or a set of tapes, the tapes cannot be

reused until a full or incremental PFDUMP has been performed on the affected

devices. Without this PFDUMP, a permanent ﬁle reload could cause some of the

reloaded PFC entries to refer to tapes that no longer exist. To prevent this problem,

a tape recycling operation should always be immediately followed by a full or

incremental PFDUMP.

Revision M Tape Alternate Storage 12-7

MAGNET Command

The MAGNET command has two parameters that must be used in a tape alternate

storage environment. The SJ parameter controls the maximum number of staging jobs

that can be active concurrently to stage ﬁles to disk from tape alternate storage; and

the SV parameter controls the maximum number of staging request VSNs that can be

displayed on the DSD E,P display. Refer to the NOS Version 2 Operations Handbook

for a description of the DSD E,P display.

The MAGNET command has the following format:

MAGNET,SJ=msj,SV=mvd.

Parameter Description

SJ=msj Speciﬁes the maximum number of staging jobs that can be active at

the same time (0 ^ msj ^ 30s). The default is SJ = 0; staging is not

allowed.

NOTE

The SJ parameter must be speciﬁed or staging from tape alternate

storage to disk will not be allowed.

SV = mvd Speciﬁes the maximum number of requests for staging tape VSNs to

display on the DSD E,P display (1 ^ mvd ^ 77s). The default is

SV = 178. Requests for staging tape VSNs will appear ﬁrst on the E,P

display (before the normal tape requests) and can be distinguished by -'frmh

having a JSN of SYS and a user name of (STAGE). '

12-8 NOS Version 2 Analysis Handbook Revision M

GENPFD Utility

The GENPFD utility controls ﬁle selection processing. It reads a summary ﬁle created

by one of the permanent ﬁle utilities and uses the information to generate ﬁle

selection directives for subsequent permanent ﬁle utility operations.

Calling the GENPFD Utility

You call the GENPFD utility by executing the GENPFD command. The format of the

command is:

GENPFD,I=i nput,S=summary,UD=udi r,PO=opt,Z.di rect i ves

Parameter Description

I = input

S=summary

UD = udir

PO = opt

directives

The 1- to 7-character name of the ﬁle from which directives are read.

The default ﬁle name is INPUT.

The 1- to 7-character name of the summary ﬁle previously created by

a permanent ﬁle utility. The default ﬁle name is SUMMARY. This ﬁle

is used as input to GENPFD.

The 1- to 7-character name of the utility directives ﬁle to be created.

This ﬁle is used for subsequent permanent ﬁle utility operations such

as destaging ﬁles and releasing disk space. The default ﬁle name is

UDIR.

The processing option that speciﬁes whether to select ﬁles for destage

processing or releasing disk space. Acceptable values for opt are:

opt Description

D Select ﬁles for destage processing.

R Select ﬁles for releasing disk space.

Input directives are contained on the GENPFD command following the

terminator. If the Z parameter is speciﬁed, the I parameter is ignored.

This eliminates the need to use a separate input ﬁle for the directives

when only a few directives are needed.

When input directives appear on the GENPFD command, the ﬁrst

character following the command terminator is the separator character

for all directives on the command. Any display code character that is

not used in any of the directives, including a space, can be used as the

separator character. Each directive must be preceded by a separator.

The directives can extend to column 72 on the command. Continuation

lines are not permitted.

For example, (slant used for separator):

GENPFD, POD, Z. /D=PE/DL, 2200/SF, 3456

Revision M Tape Alternate Storage 12-9

GENPFD Input Directives

The GENPFD input directives provide the capability to selectively choose ﬁles for

processing. The GENPFD utility accepts the following input directives:

Directive Description

D=density

DL.limit

DT,threshold

PG,pi,p2,p3

Density of the destage dump tape that will be written by the

destaging operation. Acceptable values for density are:

density Description

1600

6250

800 cpi

1600 cpi

6250 cpi

1600 cpi

6250 cpi

The default density is 6250.

Destage size limit in feet of tape used with the PO=D option. (1-

to 10-digit number, assumed decimal unless a B radix is speciﬁed).

This directive speciﬁes the maximum combined size of all selected

ﬁles. It is intended to prevent writing beyond the end of reel when

selecting ﬁles for destaging with the PFDUMP utility. If selection

of a ﬁle causes the limit to be exceeded, the ﬁle is skipped and

the next lowest priority ﬁle is examined. The default limit is 2200;

this allows for a signiﬁcant number of tape errors on a 2400 foot

reel of tape.

Destage size threshold in feet of tape used with the PO = D option.

(1- to 10-digit number, assumed decimal unless a B radix is

speciﬁed). Speciﬁes the minimum combined size of all selected

ﬁles. If the combined size of all selected ﬁles is less than the

threshold, no utility directives ﬁle will be generated. This directive

is used to ensure that enough ﬁles can be selected on a destage

dump to reasonably ﬁll a tape reel. The default value is 90

percent of the limit speciﬁed for DL.

Deﬁne preferred residence priority groups. The default is T,MN,D.

The parameters pn are strings of preferred residence mnemonics

which deﬁne the priority groups used in ﬁle selection. The groups

are deﬁned in order of decreasing priority. The default value

deﬁnes three groups: 1) a high priority group consisting of ﬁles

with PR=T (tape alternate storage residence preferred); 2) a

medium priority group consisting of ﬁles having PR=M (alternate

storage residence preferred) and PR=N (no preferred residence);

and 3) a low priority group consisting of ﬁles having PR=D (disk

residence preferred). Files having PR=L (ﬁle locked to disk) are

excluded from selection.

12-10 NOS Version 2 Analysis Handbook Revision M

GENPFD Input Directives

rDirective Description

PG,pl,p2,p3 (Continued)

It is possible to deﬁne other than three groups. For example, the

following entry:

PG.TMNDL

places all ﬁles in one group, effectively causing preferred residence

to be ignored. The following entry deﬁnes four groups.

It is possible to deﬁne other than three groups. For example, the

following entry:

PG.TMNDL

places all ﬁles in one group, effectively causing preferred residence

to be ignored.

PG.T.M.N.D

If a preferred residence type is not speciﬁed, ﬁles having that value

are excluded from selection unless they are selected explicitly with

an SF or SV directive. The following entry causes ﬁles with PR=D

or PR=L to be excluded:

PG.TMN

RL,dn,tracks Deﬁne the track release limit for device number dn or device

RL,dt,tracks type dt. A device type is speciﬁed as a mass storage equipment

mnemonic followed by a unit count. Files residing on a device are

not released unless you enter an RL directive specifying either the

device number or the device's type. A device number entry takes

precedence over a device type entry if both apply. Files residing on

the applicable device or devices are released until the number of

available tracks reaches the speciﬁed limit. No release processing is

performed for the device if the number of available tracks is equal

to or greater than the speciﬁed limit. For example, the following

device number entry:

RL.40,300

speciﬁes that ﬁles should be released on device 40 until there are

300 available tracks. If 300 tracks are already available, no ﬁles

will be released.

The following device type entry:

RL.DQ2.250

speciﬁes a release limit of 250 tracks for each two-spindle 885

device for which a limit has not been speciﬁed by device number.

Revision M Tape Alternate Storage 12-11

GENPFD Input Directives

Directive Description

SF,ui,ﬁ,f2,...,fn Unconditionally select the speciﬁed ﬁles cataloged under

SF,ui user index ui, if found in the summary ﬁle. If no ﬁle names are

speciﬁed, select all ﬁles cataloged under user index ui.

SV,Vi,V2 vn Unconditionally select any ﬁles found in the summary ﬁle with

alternate storage VSNs that match the speciﬁed VSNs. This

directive is intended to be used when recycling destage dump tapes

as part of destage processing. Specifying the VSNs of tapes to be

recycled prevents ﬁles residing on those tapes from being excluded

by the priority selection process.1

XF,ui,ﬁ,f2,...,fn Exclude the selection of the speciﬁed ﬁles cataloged under

XF,ui user index ui. If no ﬁle names are speciﬁed, exclude all ﬁles

cataloged under user index ui.

1. The SF and SV directives cause GENPFD to treat the speciﬁed ﬁles as higher priority than others in the

selection process. In some cases, GENPFD satisﬁes the speciﬁed destage or release limits without selecting

all of the files. •>*^\

12-12 NOS Version 2 Analysis Handbook Revision M

0Xts

File Selection Processing

The GENPFD utility sorts the candidate ﬁles into preferred residence priority groups

as determined by the PG directive. Within each group, each ﬁle is assigned a priority

based on the time elapsed since the last access (if PO = R) or last data modiﬁcation (if

PO=D) with older ﬁles receiving higher priorities. Files selected with SF directives or

ﬁles residing on VSNs selected with SV directives are given a higher priority than

other ﬁles to force selection.

For destage processing, ﬁles are selected beginning with the highest priority until the

destage limit is reached. The permanent ﬁle utility parameters UI and PF are

written to the utility directives ﬁle for each selected ﬁle unless the combined length

of the selected ﬁles is less than the destage threshold (either the default or an

explicit nonzero value speciﬁed in the DT directive).

For release processing, ﬁles residing on devices speciﬁed by RL directives are selected

for release until the deﬁned limit for each device is reached. The permanent ﬁle

utility parameters UI and PF are written to the utility directives ﬁle for all ﬁles

selected for release.

Limitations

GENPFD can process a maximum of 80 RL directives and a total of 1000 SF, SV, and

XF directives. Each directive counts as one entry, whether or not it contains a ﬁle

name or VSN. Each ﬁle name or VSN on a directive after the intial one counts as an

additional entry.

Revision M Tape Alternate Storage 12-13

Multimainframe Operations 13

Introduction 13-1

Linked SDM Operation 13-2

Deadstart 13-3

Shared Mass Storage 13-6

Conﬁguring Shared Mass Storage Devices 13-6

Device Access Tables 13-6

Mass Storage Recovery Tables 13-7

Device Initialization 13-7

Device Unload 13-8

Device Recovery .. 13-8

Device Checkpoint 13-9

MSE Processing 13-9

Independent SDM Operation 13-10

Deadstart , 13-10

Shared Mass Storage 13-12

Conﬁguring Shared Mass Storage Devices 13-13

Device Index Tables ,... 13-14

Mass Storage Recovery Tables 13-14

Device Initialization 13-14

Device Unload 13-15

Device Recovery , 13-15

Device Checkpoint 13-16

MSE Processing .-....' 13-16

Multimainframe Operations 13

Introduction

Multimainframe operation provides mechanisms by which more than one computer can

share mass storage devices. Each mainframe in a multimainframe conﬁguration has

access to the shared mass storage devices and their preserved ﬁles. Preserved ﬁles are

those ﬁles that are retained across a level 0 deadstart (nonsystem recovery). Types of

preserved ﬁles are permanent ﬁles, queued ﬁles, and system dayﬁles.

There are two types of multimainframe conﬁgurations:

• Linked shared device multimainframe (SDM).

• Independent shared device multimainframe.

The main difference between the two SDM conﬁgurations is their method of

maintaining the critical device tables necessary to control shared mass storage and

intermainframe communications. Linked SDM conﬁgurations maintain the critical

device tables in extended memory, while independent SDM conﬁgurations maintain the

tables on each shared mass storage device. Because the independent SDM

conﬁgurations maintain the tables on the shared devices, it causes some performance

degradation.

Each mainframe in a multimainframe conﬁguration can operate in multimainframe

mode or in standalone mode; however, two machines may not access the same device

unless both are in multimainframe mode. A device is considered shared if it can be

accessed by more than one of the mainframes; it need not be accessible to all

mainframes in the complex.

The fact that a computer is operating as part of a multimainframe complex is not

apparent to the user, except for some degrading of performance for the independent

SDM. However, there are operational changes and additions that the operator should

note. These include deadstart commands, displays, online commands necessary in the

event of an interruption on one of the sharing mainframes, and the UNLOAD/MOUNT

process for removable devices.

Recovery allows a machine to either join other machines operational in a

multimainframe complex or to operate in a standalone mode. The standalone system is

not allowed to use the same mass storage devices as other machines. In the event of a

system interruption to one machine, it is possible to operate the remaining machines in

a multimainframe environment.

Automatic detection of extended memory is not provided, because it is not possible to

determine its absence and continue to run on all machine types. For example, a 6600

will hang if an attempt is made to execute an extended memory instruction without

extended memory. Extended memory status is checked during deadstart when

processing an extended memory entry in the EQPDECK.

Except where explicitly speciﬁed that controllers and/or equipment can be shared

between mainframes, assume they cannot be shared. For example, NOS does not

support sharing a two-channel tape controller between mainframes; nor does it support

sharing mass storage controllers except as speciﬁed in this section.

Revision M Multimainframe Operations 13-1

Linked SDM Operation

In a linked SDM complex, up to seven computers can access mass storage devices. Any

combination of 1 to 7 model 71, 72, 73, 74, 171, 172, 173, 174, 175, 176, 720, 730, 740,

750, 760, 865, 875, CYBER 180-class machines or 6000 Computer Systems mainframes

can be used. Most mainframes require one CPU port connected to extended memory.

The exception is CYBER 180-class machines which must be connected to extended

memory through a low-speed port (LSP) or a distributive data path (DDP) instead of a

CPU port. Shared mass storage devices supported are: 844 (single and double density),

885-11/12, and extended memory.

Extended memory is the means for controlling shared mass storage and intermainframe

communications. Each mainframe has a port into extended memory through which

system activity is controlled. In order to control shared mass storage devices, several

extended memory resident tables are required. The device access table (DAT) contains

the logical description (family name/pack name and device number) of each mass

storage device (shared or nonshared) that is accessible by any machine in the linked

SDM complex. For each device in the DAT that is to be accessed by more than one

machine, a corresponding MST (image of central memory resident MST) and TRT also

reside in extended memory. In addition, a machine recovery table (MRT) exists in

extended memory for each machine and device (that is, there are as many MRTs for

each shared device as there are mainframes in the complex).

If MSE is used, one mainframe in the linked SDM complex is physically connected to

7990 and is called the master mainframe; all other mainframes in the complex can

access 7990 ﬁles and are called slave mainframes. SSEXEC runs on the master

mainframe and is the main processing program that controls MSE activity. SSSLV runs

on each slave mainframe and communicates with SSEXEC to retrieve ﬁles from 7990

in response to ATTACH requests by jobs running on the slave mainframes.

j^^S,

13-2 NOS Version 2 Analysis Handbook Revision M

Deadstart

0m*.

Deadstart

A linked SDM environment is deﬁned at deadstart by CMRDECK and EQPDECK

entries in each machine. Use the following entries to do this.

CMRDECK Entry Description

MID=id.

EQPDECK Entry

2-character machine identiﬁcation. This

identiﬁcation is used to associate a speciﬁc machine

with its access to a shared device. It is required,

however, even if no devices are shared during

system operation. This id must be unique for each

machine in the complex. If id is omitted, the

default value is MID=AA.

Description

SHARE=esti ,est2,...,estn.

PRESET,n.

EST ordinals of the mass storage devices that will

be shared with other machines in the linked SDM

complex. The MST/TRT for each device resides in

extended memory.

Total number of devices to be shared by all

mainframes in the complex. You can specify this

entry only at the system console. You must specify

the PRESET entry on a level 0 deadstart on the

ﬁrst machine to be deadstarted in the linked SDM

complex. This entry initializes the ﬂag register and

extended memory resident tables. If n is omitted,

the default value is the number of shared devices

deﬁned for the particular machine.

These entries are also described in section 3, Deadstart Decks.

Because in a linked SDM complex two or more machines can utilize the same mass

storage devices, the device assignments and EQPDECKs of all machines are

interrelated. Care must be taken to ensure proper EQPDECK settings for each machine

to obtain the desired device conﬁguration. The following items are important in

obtaining this proper device conﬁguration.

• Each machine must specify a unique machine identification (using the MID entry in

the CMRDECK). This identiﬁcation associates a particular machine with its ﬁles on

a shared device. There are no external characteristics associated with this

identiﬁcation. However, if the machine identiﬁcation speciﬁed on a machine

deadstarting into a multimainframe complex is identical to a machine identiﬁcation

on a machine already operating, the deadstart process halts and an appropriate

message is displayed.

• The assignment of shared mass storage devices should be made properly to ensure

the recoverability of the device and the proper operation of the system (refer to

Shared Mass Storage later in this section).

• If SHARE EQPDECK entries are present, an extended memory entry (DE or DP

equipment type) must also be present in the EQPDECK. If none is found, an error

message is given to the operator indicating that no link device has been deﬁned.

Revision M Multimainframe Operations 13-3

Deadstart

• The first machine deadstarted (level 0 only) must specify the PRESET EQPDECK

entry. This command causes tables to be preset in extended memory, and in so

doing, assures that no other machine has arrived at the same point in the

deadstart sequence and is attempting the same thing. These other machines

should not have the PRESET entry speciﬁed during deadstart. All other machines

that arrive at that point in the deadstart process display a message indicating

that they are waiting for deadstart on the preset machine. A machine that does

not preset extended memory has no means of detecting whether extended memory

has been preset previously by another machine. Therefore, the operator must

ensure that extended memory has been preset by a prior deadstart before

deadstarting another machine that is to be in the multimainframe complex.

• If extended memory is initially placed in maintenance mode (half of extended

memory used for online diagnostics), the ﬁrst machine deadstarted in a complex

must have the INITIALIZE and PRESET EQPDECK entries speciﬁed when

deadstarting (level 0 only). The remaining machines must have the extended

memory size correspondingly reduced in their EQPDECKs.

When a mainframe joins a linked SDM complex, it is associated with an identiﬁcation

that it utilizes during system operation but which is independent of the machine

identiﬁcation. This association is done during deadstart, when the machine investigates

the multimainframe tables residing on the link device and places its machine

identiﬁcation in an empty slot of the seven that are available. Associated with each

slot is a unique machine index and a unique machine mask, which the machine uses

either to index itself into various MMF tables or to identify itself in these tables.

When performing a level 0 recovery deadstart on one of the machines, the MREC

utility should be run on some or all of the remaining machines before the deadstart

proceeds. The purpose of this utility is to clear interlocks held by the machine to be

deadstarted that have not been cleared by CPUMTR. It can also recover mass storage

space on a shared device that is currently not accessible because of a machine

interruption (necessitating a level 0 deadstart). MREC may have to be run from more

than one machine since it affects only shared devices (that is, devices speciﬁed on the

SHARE EQPDECK entry) of the machine on which MREC is run. If the interrupted

machine shares different devices with different machines, MREC must be run from

enough machines to account for all devices shared with the interrupted machine. The

interface to the MREC utility is described in section 8, K-Display Utilities.

If a level 3 recovery deadstart will be performed, extended memory must be intact as

well as CMR. For a level 1 or 2 recovery in a linked SDM environment, extended

memory must be intact.

Figure 13-1 illustrates the steps generally needed to deadstart a machine in a

multimainframe complex.

13-4 NOS Version 2 Analysis Handbook Revision M

Deadstart

A LEVEL 0

DEADSTART

IS REQUIRED

DETERMINE THE

LEVEL OF RECOVERY

DEADSTART

LEVEL 0 LEVEL 1,

2. OR 3

RUN THE MREC UTILITY

ON THE NECESSARY

MACHINES

ENSURE THE PROPER

EQPDECK SETTING

FOR SHARE DEVICES

DETERMINE THE

TYPE OF MMF

ENVIRONMENT

INDEPENDENT

SDM

IS THIS

MACHINE SHARING'

A DEVICE THAT HAS

JOT BEEN PRESET?.

YES YES

ENTER. DURING DEADSTART.

THE PRESET EQPDECK ENTRY NO

PROCEED WITH THE

DEADSTART PROCESS

PROCEED WITH SYSTEM

OPERATIONS

Figure 13-1. Deadstarting a Machine in a Multimainframe Complex

Revision M Multimainframe Operations 13-5

Shared Mass Storage

A major reason for operating a multimainframe complex is to be able to share mass

storage devices between machines. Thus, the users of two systems may be able to

utilize the same ﬁles if these ﬁles reside on a shared mass storage device. Having the

ability to share ﬁles between systems also means that several additional procedures are

required when operating a multimainframe complex. Additional consideration must also

be given when conﬁguring mass storage devices. This material, which is unique to

multimainframe operation, is discussed in the following paragraphs.

Conﬁguring Shared Mass Storage Devices

Some consideration should be given as to which devices should be made shared devices

and which ones should not. Since a shared device contains preserved ﬁles that can be

accessed by more than one machine, it must be physically connected to and logically

deﬁned (as shared) by each machine sharing the device. If one device of a family is

deﬁned as shared, then normally all devices in the family should be deﬁned as shared

for accessibility. If a removable device will be treated as shared, it must be deﬁned as

removable in each machine sharing it. If a shared device is not removable, additional

use of the device (beyond preserved ﬁles) is deﬁned by each machine, independently of

any other machine deﬁnition (that is, each machine speciﬁes whether the device is to

contain such things as a copy of the system or temporary ﬁles).

There is no real advantage to having nonpreserved ﬁles on a shared device, unless

spreading them across shared devices outweighs having them on a dedicated device. A

key factor is drive and controller contention. Another factor is whether or not enough

drives and accesses to those drives are available to warrant spreading temporary ﬁles ^^^

across several devices from all systems. Also, additional overhead is incurred by the *^)

system for each shared device. This includes additional CMR space for each device

(1008 CM words) and additional time that is required to maintain the device-related

tables (such as the MST or the TRT) on the link device. These factors must be

weighed to determine how best to conﬁgure shared devices.

Device Access Tables

In a linked SDM complex, mainframes use DATs to determine device usage. The DATs

identify the status of all shared and nonshared devices for the mainframes in the "^

complex. One DAT for each machine resides in extended memory.

In order to minimize conﬁguration problems, shared removable equipment should be

conﬁgured the same way on all machines in the complex. For example, if one system

deﬁnes three shared units as three single-unit devices and another system deﬁnes the

same units as one three-unit device, the ﬁrst system can accommodate a two-unit

device on these units, whereas the second system would consider it an error. Unless

the conﬁgurations are the same on all machines, any devices mounted on those drives

may not necessarily be recoverable on all machines.

NOTE

RESEX considers only the conﬁguration of the machine on which it is executing in

its overcommitment algorithm.

13-6 NOS Version 2 Analysis Handbook Revision M

0im\

Mass Storage Recovery Tables

One problem that is created by having more than one machine sharing a mass storage

device is that of recovering the mass storage space and interlocks of a machine, if it

requires recovery processing. This problem is solved by deﬁning a table that provides

the information needed to recover the mass storage space of a machine and by having

a machine recovery utility that performs the recovery. The table is called the mass

storage recovery table (MRT). There is one MRT for each mainframe per device. It tells

which tracks are interlocked and which tracks are ﬁrst tracks of ﬁles local to a

particular machine. The MRT is utilized by CPUMTR on another machine to clear

track interlocks and by the MREC utility on another machine to recover the mass

storage space of the interrupted machine. For speciﬁc information on the MRT, refer to

the NOS Version 2 Systems Programmer's Instant.

Device Initialization

To initialize a mass storage device, it is necessary ﬁrst to prevent any new activity

from starting up on the device, then to wait until all current activity has completed,

and ﬁnally to interlock the device and proceed with initialization. To accomplish this

on shared mass storage devices in a linked SDM environment, follow these steps:

1. Enter the DSD UNLOAD command. All machines sharing the device must unload

it to prevent any new activity. (This command can be used for both removable

and nonremovable shared devices.)

2. Enter the DSD INITIALIZE command on the machine from which the

initialization is to take place. If you enter the INITIALIZE command from another

machine, an error message is displayed. Refer to the NOS Version 2 Operations

Handbook for error messages.

3. The machine from which the INITIALIZE command was entered monitors the

status of the other machines that are sharing the device. Once they have unloaded

the device and user activity has ceased, the device tables are updated and the

device is checkpointed.

4. To activate the device on the other machines, enter the DSD MOUNT command

from each machine. This command clears the UNLOAD status. If initialization is

still in progress on another machine when you enter a MOUNT command, the

MOUNT process is terminated with an error. Refer to the NOS Version 2

Operations Handbook for error messages.

Revision M Multimainframe Operations 13-7

Device Unload

In a linked SDM environment, unloading a device involves more than it does under a

one-machine system. A device can be unloaded from a machine (referred to as a local

unload), or it can be unloaded from the entire multimainframe complex (referred to as

a global unload). A device can be physically removed from the complex only after a

global unload has been accomplished. This is the general procedure to complete a

global unload:

1. Enter the UNLOAD command from each machine. This is an indication to the

machine that no new accesses should be initiated. You must enter this command

from each machine sharing the device.

2. When all local unloads are set and user access has ceased, global unload status is

set if the device is a removable device. This global unload status is displayed on

all machines, indicating that there is no activity on the device from any machine

and that the device may be physically unloaded.

3. You can then switch packs and enter the MOUNT command at the console to ^^

initiate recovery of the device. The MOUNT command clears the global unload

status and the local unload status on the machine from which it was entered and

indicates that this machine is now accessing the device. All other machines

continue to ignore the device until you enter the MOUNT command on each

machine. The MOUNT command does nothing if local unload status is not set on

the machine.

Device Recovery

Deadstart and online recovery methods are similar in logic for recovering mass storage

devices.

When a device is recovered in a linked SDM complex, the DAT in extended memory

is interlocked while a check is made to see if an entry exists for this device. The

presence of an entry indicates that another machine is also accessing the device. If

an entry is found and the machine recovering the device has not been instructed to

share it, an error is indicated and recovery halts with an appropriate message

displayed. If the machine already accessing the device is not allowing it to be shared, ^^

the same error condition occurs. Therefore, if a device is being accessed, another '^^S

machine can recover that device only if the recovering machine and the accessing

machine use the device in a multimainframe mode.

NOTE

If two devices recovered on separate machines have the same family name/device

number or pack name, there is no method of determining whether or not they are the

same device if both are shared. If they are different devices, they are both destroyed

when used.

13-8 NOS Version 2 Analysis Handbook Revision M

/ ^ • N

0Bto\

Device Checkpoint

Local MST information for each machine that shares a mass storage device is

maintained on the device. MST information for other machines may also be present on

the device. The information for each machine is kept in one sector on the label track

following the TRT sectors. Entries for up to 31 unique machine identiﬁers can exist.

During checkpoint, only the local MST information of the machine performing the

checkpoint is updated. Since local MST information for many machines is kept on the

device, updating of all these areas by one machine could cause a loss of information

needed if the device were to be used with another system.

For shared devices in a linked SDM complex, duplication of checkpointing by more

than one machine is prevented. If a machine attempts to checkpoint a shared device

and determines that a second machine is performing the checkpoint (a checkpoint

request bit is set in the local MST area of the second machine), no action is taken.

Only one checkpoint bit is set at any given time for a device.

MSE Processing

Unlike most other mass storage devices, the 7990 is not shared by mainframes. Rather,

7990 is physically connected to and driven by one mainframe called the master

mainframe. All other mainframes that share permanent ﬁles with the master

mainframe are called slave mainframes. The disks to which 7990-resident ﬁles are

staged must be shared devices and accessible by all mainframes.

SSEXEC runs on the master mainframe and is the main program that controls MSE

processing activities. You can activate SSEXEC by entering the following DSD

commands at the system console.

L.SUBSYST.

ENABLE,MASTER MSE.

L.END.

MSEffff.

SSSLV runs on each slave mainframe and is the program that supports ﬁle retrieval

from a 7990 for jobs running on a slave mainframe. You can activate SSSLV by

entering the following DSD commands at the system console.

L.SUBSYST.

DISABLE,MASTER MSE.

L.END.

MSEffff.

When a job running on a slave mainframe attaches a 7990-resident ﬁle that has no

disk image, SSSLV is requested to stage the ﬁle to disk. However, SSSLV cannot access

7990 directly. SSSLV must communicate with SSEXEC, which stages the ﬁle to a

shared disk and then notiﬁes SSSLV when the ﬁle is available. SSSLV then causes the

requesting job to be resumed so that the ﬁle ATTACH can complete.

The permanent ﬁles used for communication between SSEXEC and SSSLV are

initialized and recovered automatically and require no backup or recovery processing.

The name of the permanent ﬁle used by SSSLV running on mainframe i (i = l, 2, 3,

4) to send requests to SSEXEC is STBMNOi. The name of the permanent ﬁle used by

SSEXEC to respond to all requests by all SSSLVs is MTBSPFN. These ﬁles belong to

the family that contains extended memory and are saved under the user index

3777608.

Revision M Multimainframe Operations 13-9

Independent SDM Operation

J,*^%v

Independent SDM Operation

In an independent SDM complex, up to 16 computers can access shared mass storage

devices. Any combination of from 1 to 16 model 71, 72, 73, 74, 171, 172, 173, 174, 175,

176, 720, 730, 740, 750, 760, 865, 875, CYBER 180-class machines, or 6000 Computer

Systems mainframes can be used. Shared mass storage devices supported are 844

(single and double density) and 885-11/12 (single and double density).

Each shared device contains in its label track all of the tables necessary to control

multimainframe access. These tables include the MST, TRT, device index table (DIT),

and MRT for each mainframe sharing the device. Whenever a mainframe needs to

access a shared device, it is necessary for the mainframe to ensure its copies of these

tables are up-to-date.

One or more of the mainframes can have extended memory, but each mainframe's

extended memory must be used as a nonshared device.

If MSE is used, one mainframe in the independent SDM complex is physically ^^^

connected to 7990 and only that mainframe can access it. 1

Deadstart

An independent SDM environment is deﬁned at deadstart by CMRDECK and

EQPDECK entries in each machine. Use the following entries to do this.

CMRDECK Entry Description

MID=id. 2-character machine identiﬁcation. This

identiﬁcation is used to associate a speciﬁc

machine with its access to a shared device. It is

required, however, even if no devices are shared

during system operation. This id must be unique

for each machine in the complex. If id is omitted,

the default value is MID=AA.

EQPDECK Entry Description

ISHARE = esti,est2,...,estn. EST ordinals of the mass storage devices that will ^^)

be independently shared with other machines in

the multimainframe complex. The ISHARE entry

must precede the PRESET entry. The MST/TRT

resides on each shared device.

PRESET=esti,est2,...,estn. EST ordinal of the shared mass storage device to

be preset. You can specify this entry only at the

system console. Several machines at level 0

deadstart may need the PRESET entry. The

PRESET entry is speciﬁed only on a level 0

deadstart by the ﬁrst machine accessing the device

and must follow the ISHARE entry. PRESET

initializes the tables residing on the label track of

the shared devices.

These entries are also described in section 3, Deadstart Decks.

13-10 NOS Version 2 Analysis Handbook Revision M

Deadstart

Since in an independent SDM complex two or more machines can utilize the same

rmass storage devices, the device assignments and EQPDECKs of all machines are

interrelated. Care must be taken to ensure proper EQPDECK settings for each machine

to obtain the desired device conﬁguration. The following items are important in

obtaining this proper device conﬁguration.

• Each machine must specify a unique machine identification (using the MID entry

in the CMRDECK). This identiﬁcation associates a particular machine with its

ﬁles on a shared device. There are no external characteristics associated with this

identiﬁcation. However, if the machine identiﬁcation speciﬁed on a machine

deadstarting into a multimainframe complex is identical to a machine

identiﬁcation on a machine already operating, the deadstart process halts and an

appropriate message is displayed.

• The assignment of shared mass storage devices should be made properly to ensure

the recoverability of the device and the proper operation of the system (refer to

Shared Mass Storage, next).

Each machine must specify the ISHARE=est EQPDECK entry to inform the

system that devices will be shared in the independent shared device mode.

If you need to deadstart an interrupted mainframe in an independent SDM

environment, you should put the nonfailing mainframes in STEP mode before

deadstarting the interrupted mainframe.

On a level 0 deadstart, any shared device that has not been shared prior to this

deadstart must include the PRESET entry in its deﬁnition.

On a level 0 deadstart, if a device was previously shared by the deadstarting

machine, the MRT and DIT device resident tables must be zeroed out.

- If the machines that share the device are not currently running, the device

must be preset or initialized to zero out the tables.

- If any machine that shares the device is running, the machine recovery utility

(MREC) must be run from a running machine to remove the machine

identiﬁcation of any machines that are not currently running and to clear all

interlocks held by these machines.

On a level 0 deadstart, if a device was previously shared by a machine other than

the deadstarting machine, the MRT and DIT device resident tables may or may

not have to be zeroed out.

- If another machine that is sharing the device is running, the device must not

be preset.

- If no machines that will be sharing the device are running and one or more of

them will be deadstarted at a level 1, 2, or 3, the device must not be preset.

- If all machines that will be sharing the device are not currently running and

all will be deadstarted at level 0, the device can be preset or initialized.

•

Revision M Multimainframe Operations 13-11

Shared Mass Storage

When performing a level 0 recovery deadstart on one of the machines, the MREC

utility should be run on some or all of the remaining machines before the deadstart

proceeds. The purpose of this utility is to clear interlocks held by the machine to be

deadstarted that have not been cleared by CPUMTR. It can also recover mass storage

space on a shared device that is currently not accessible because of a machine

interruption (necessitating a level 0 deadstart). MREC may have to be run from more

than one machine since it affects only shared devices (that is, devices speciﬁed on the

ISHARE EQPDECK entry) of the machine on which MREC is run. If the interrupted

machine shares different devices with different machines, MREC must be run from

enough machines to account for all devices shared with the interrupted machine. The

interface to the MREC utility is described in section 8, K-Display Utilities.

In an independent SDM complex, as in a standalone system, critical tables that reside

on the device must be intact for a level 1, 2, or 3 deadstart recovery.

Figure 13-1 illustrates the steps generally needed to deadstart a machine in a

multimainframe complex.

Shared Mass Storage

A major reason for operating a multimainframe complex is to be able to share mass

storage devices between machines. Thus, the users of two systems may be able to

utilize the same ﬁles if these ﬁles reside on a shared mass storage device. Having the

ability to share ﬁles between systems also means that several additional procedures are

required when operating a multimainframe complex. Additional consideration must also

be given when conﬁguring mass storage devices. This material, which is unique to

multimainframe operation, is discussed next.

/^at^y

13-12 NOS Version 2 Analysis Handbook Revision M