30242 90003_LAN3000_V_Diagnostic_and_Troubleshooting_Guide_Aug1987 90003 LAN3000 V Diagnostic And Troubleshooting Guide Aug1987

User Manual: 30242-90003_LAN3000_V_Diagnostic_and_Troubleshooting_Guide_Aug1987

Open the PDF directly: View PDF PDF.
Page Count: 210

Download30242-90003_LAN3000_V_Diagnostic_and_Troubleshooting_Guide_Aug1987 30242-90003 LAN3000 V Diagnostic And Troubleshooting Guide Aug1987
Open PDF In BrowserView PDF
HP AdvanceNet

lAN3000/V Diagnostic and
Troubleshooting Guide

This Manual Contains
IEEE 802.3 LAN
Diagnostic Software Information
For HP 3000 MPE-V Systems

Flio-

HEWLETT

~'~ PACKARD
HEWLETT-PACKARD COMPANY
Roseville Networks Division
8000 Foothills Boulevard
Roseville, California 95678

Manual Part Number 30242-90003
E0887
Printed in U. S. A.
August 1987

Notice
The information contained in this document is subject to change without notice.
HEWLETT-PACKARD COMPANY MAKES NO WARRANTY OF ANY
KIND WITH REGARD TO THIS MATERIAL, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
AND FITNESS FOR A PARTICULAR PURPOSE. HEWLETT-PACKARD
COMPANY shall not be lia ble for errors contained herein or for incidental or
consequential damages in connection with the furnishing, performance or use of
this material.
HEWLETT-PACKARD COMPANY assumes no responsibility for the use or
reliability of its software on equipment that is not furnished by
HEWLETT-PACKARD COMPANY.
This document contains proprietary information, which is protected by
copyright. All rights are reserved.
No part of this document may be
photocopied, reproduced or translated to another language without the prior
written consent of HEWLETT-PACKARD COMPANY Company.

Copyright

2

©

1987 by HEWLETT-PACKARD COMPANY

Printing History

New editions are complete revisions of the manual. Update packages, which are
issued between editions, contain additional and replacement pages to be merged
into the manual by the customer. The dates on the title page change only when
a new edition or a new update is published. No information is incorporated into
a reprinting unless it appears as a prior update; the edition does not change when
an update is incorporated.
The software code printed alongside the date indicates the version level of the
software product at the time the manual or update was issued. Many product
updates and fixes do not require manual changes and, conversely, manual corrections may be done without accompanying product changes. Therefore, do not
expect a one-to-one correspondence between product updates and manual
updates.

Edition 1
Edition 2

June 1985 (32189B.51.00)
August 1987 (320330.02.01)

3

Preface

This guide describes tools to troubleshoot IEEE 802.3 LAN link hardware connecting HP 3000 MPE-V systems.
The primary tool described is LANDIAG, the HP 3000 Local Area Network (LAN)
Diagnostic software utility. If you have a hardware problem on your HP 3000
LAN link, you can use LANDIAG along with the other tools to locate the Field
Replaceable Unit (FRU) at fault. You must then repair or replace the faulty
FRU.
This guide supplements other LAN troubleshooting manuals, as follows:
1.

The LA N Link Hardware Trouhleshooting Manual, IEEE 802.3 Coaxial
Cable LA N, 5955-7681, covers the troubleshooting of HP ThinLANs and
ThickLANs (thin and thick coaxial cable LANs) comprised of dissimilar systems (such as HP 1000 RTE-A, HP 9000 Series 300, 500 and 800 HP-UX, and
HP 3000 MPE-V systems). It uses a portion of the LANDIAG software information contained here. (The HP CE Handbook version of this manual is
5959-2217.)

2.

The H P ThinLA N Diagnostics and Troubleshooting Manual for PCs,
50909-90060, is directed toward HP personal computer ThinLAN networks.
When HP 3000 systems are connected, the LANDIAG information presented
here is essential.

3.

The troubleshooting process for HP StarLAN, featuring twisted-pair cables,
is covered by the HP Star/AN Diagnostics and Troubleshooting Manual for
pes, 50906-90060.
When HP 3000 MPE-V systems are connected to
StarLAN, the LANDIAG information presented here is also needed.

4.

If the problem lies in LAN software, rather then the hardware, the
NS3000/V Network Manager Relerence Manual (32344-90002) should be
consulted.

The intended audiences for this guide include the Network Manager, the
Network Consultant, the Data Comm Specialist, and the Hewlett-Packard
Customer Engineer (CE) and Systems Engineer (SE).

5

Preface (continued)

How To Use This Guide
All users should read Chapter 1 to become acquainted with general troubleshooting considerations and tools covered by this guide. As a reference manual, users
can subsequently proceed to the chapter describing the particular tool of interest.
For those users who lack troubleshooting procedures, Appendix A is provided.
Appendix A provides procedures, as a series of decision-tree flowcharts, to help
isolate HP 3000 LAN link hardware faults. These procedures utilize LANDIAG
and other tools presented in the various chapters.

Organization Of This Guide
This guide consists of the following chapters:

6

Chapter 1:

General Information

Chapter 2:

SHOWCOM

Chapter 3:

Activity LEOs

Chapter 4:

LAN Node Diagnostic

Chapter 5:

LANIC Self Test

Chapter 6:

Tracing

Chapter 7:

Software Tools

Appendix A:

Troubleshooting Flowcharts

Preface (continued)

Additional References
When using this manual, you should be familiar with the basic operating principles of HP 3000 computers and the MPE-V operating system. In addition, you
should be familiar with the subjects covered in the following manuals, as
appropriate:

• HP 3000 Computer Systems, MPE V Commands Reference Manual
(32033-90006)
• HP 3000 Computer
(32033-90007).

Systems, MPE

V

Intrinsics

Reference

Manual

• HP 3000 Computer Systems, MPE V System Operation and Resource
Management Reference Manual (32033-90005)
• Fundamental Data Communications Handbook (5979-4634)
• LAN Cable and Accessories Installation Manual (5955-7680), for coaxial
cable LANs
• HP 3000 LAN Link installation and service manuals, including
- HP 30240A LAN3000/V Link LANIC Installation and Service Manual
(part number depends on the product option selected)
- HP 30265A StarLAN/3000 Link Installation and Service Manual
(30265-90001), for Series 37 and MICRO 3000 systems
• NS3000/V Network Manager Reference Manual (Volume 1, 32344-90002;
Volume 2, 32344-90012)
• NS3000/V User/Programmer Reference Manual (32344-90001)
• NS3000/V Error Message and Recovery Manual (32344-90005)
• HP StarLAN Planning Guide for PCs (50906-90020)
• HP SiteWire Planning Guide (5959-2201)

7

j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
J

Contents

Chapter 1
General Information
Introduction
Applicable Networks
Coaxial Cable LANs
Twisted Pair LAN
Fault Isolation and Repair Strategy
Network Map
Abbreviations and Nomenclature

Chapter 2
SHOWCOM
Syntax
Interpreting Results
Transn1it Fields
Receive Fields
Errors

Page
1-1
1-1
1-3
1-3
1-3
1-4
1-5
1-6

Page
2-1
2-1

2-2
2-2
2-3
2-4

Chapter 3
Activity LEDs

Page
3-1

Cable Interface Activity LEOs
The E and L Indicators
Relationship to Cable Signals
MPU Activity LEOs
Network Fault LED Examples
Coaxial Cable LAN
StarLAN

3-5
3-6
3-7
3-9
3-12
3-12
3-15

Chapter 4
LAN Node Diagnostic
General Information
What is the LAN Node Diagnostic?
Required Hardware
Required Software
Execution Time
Tests Included
Diagnostic Limitations

Page
4-1
4-1
4-1
4-1
4-2
4-2
4-3
4-4

9

Contents (continued)

Guidelines for Setting Up and Using LANDIAG
System Infornlation
Capability Levels
System Type Check
Operation with Network Services
Reliability and Recovery from Power Failure
Integrity of the System
Running LANDIAG
User Interface
Activity Indicator
How to Get Output From LANDIAG
Station (Link-Level) Addresses
Command Set
Diagnostic Dialogue Example
Command Entry Abbreviations
Fault Messages
HELP Command
EXIT Command
GO Command
LOOP Command
NOPRINT Command
,
PRINT' Command
TEST NIALL Command
Test Descriptions
Test #1. Roll Call Test
Test #2. Channel ID Test
Test #3. Initialization Test
Test #4. Self Test.
Test #5. Interrupt Test
Test #6. Soft Reset Test
Test #7. CR-SR Loop Test
Test #8. Bus Conflict Test
Test #9. Address Offset Test
Test #10. Extended Address Test
Test #1 L FIFO Write Test.
Test #12. FIP Conflict Test.
Test #13. Coprocessor Test
Test #14. MAU Loopback Test.
Test #15. Date Code Test.
Test #16. Hood Loopback Test
Test #17. Remote Node Test.
Error Messages

10

,

4-6
4-6
4-6
4-7
4-7
4-8
4-8
.4-10
4-12
4-12
4-13
4-14
4-15
4-17
4-19
4-19
4-20
4-21
4-21
4-21
4-22
4-22
.4-22
4-23
4-24
4-25
.4-26
4-27
.4-29
.4-30
.4-31
.4-33
.4-34
.4-36
.4-38
4-40
.4-42
.4-44
.4-51
.4-52
.4-63
4-74

Contents (continued)

Chapter 5
LANIC Self Test
The Scope of LANIC Card Self Test.
Self Test Limitations
How To Read Selftest LEDs
How To Use the Self Test

Chapter 6
Tracing
General Information
Linkcontrol Trace=On
Syntax
Parameters
Discussion
Example
Linkcontrol Trace=Off
Syntax
Parameters
:
Discussion
Example
Using CSDUMP Formatting Program
Defining a Trace File for CSDUMP
Defining a CSDUMP Listing File
Initiation the CSDUMP Program
Formatted CSDUMP Trace Listing
CSDUMP Listing Header Message
Begin Tracing and Line Information Messages
Trace Record and Header Message
Trace Entry Format
Missing Entries Message
Trace Entry Mnemonics
PSCT (PRCT) Trace Entries
PCMP Trace Entries
End of Trace and Line Information Messages
CS Error Codes

Page
5-1
5-1
5-2
5-2
5-4

Page
6-1
6-1
6-2
6-2
6-2
6-3
6-3
6-4
6-4
6-4
6-4
6-4
6-5
6-6
6-6
6-7
6-7
6-8
6-9
6-10
6-11
6-12
6-12
6-14
6-17
6-19
6-20

11

Contents (continued)

Chapter 7
Software Tools

Page
7-1

M~mory

7-1
7-1
7-1
7-2
7-2
7-5
7-6
7-10
7-10
7-11
7-12
7-13
7-14

Dump
NSDPAN5/NSDUMPJ
Obtaining an NSDPAN5 Listing
LISTLOG5
Operation of LISTLOG5
NMMAINT
Example
CSLIST
Example 1
Example 2
Example 3
DSLIST
Example

'.'

Troubleshooting FIO\ychal·ts

Page
A-I

Introduction
LAN Node Diagnostic Flowcharts
Using Test 17 - The Remote Node Test (RNT)
Using Software Line Tests

A-I
A-3
A-22
A-23

Appendix A

12

'

Figures and Tables
Figure or Flowchart
Figure 1-1.
Figure 3-1.
Figure 3-2.
Figure 3-3.
Figure 4-1.
Figure 4-2.
Figure 4-3.
Figure 4-4.
Figure 4-5.
Figure 4-6.
Figure 4-7.
Figure 4-8.
Figure 6-1.
Figure 6-2.
Figure 6-3.
Figure 6-4.
Figure 6-5.
Figure 6-6.
Figure 6-7.
Figure 6-8.
Figure A-I.
Figure A-2.

Flowchart 1.
Flowchart 2.
Flowchart 3.
Flowchart 4.
Flowchart 5.
Flowchart 6.
Flowchart 7.
Flowchart 8.
Flowchart 9.

LAN Hardware Troubleshooting Complexity
LANIC Card LEOs On Series 4X/5X/6X/70 Systems
LANIC Card LEOs On Series 37 and MICR03000 Systems
LED Labels and Functions
Running LANDIAG
Program State Diagram
Selftest Error Message Example
StarLAN Loopback Connector.
StarLAN Connection Fault Isolation Using Test 14
Loopback Hood on MAU
Loopback Hood on ThinMAU
Testing AUI Cable Sections
Trace Listing Header
Begin Tracing and Line Information Messages
Trace Record Header
Sample Trace Entry
PRCT Trace Entry
PRCT and PSCT Formats
PCMP Trace Entry
End of Trace and Closing Line Information
Hardware Troubleshooting Overview
Software Line Tests

Verification of System and Hardware Configuration
LANIC Initialization and Main Module Verification
Interrupt, Soft Reset, and SIMB/IMB Interface Verification
LAN Coprocessor, Loopback and Date Code Verification
Identifying Failed FRUin Hood Loopback Test
Identifying MAU/ThinMAU Power-On Faults
Testing Coax LAN Connection AU! Interface
Testing AUI Cables
Testing MAU/ThinMAU Connection

Page
1-1
3-1
3-2
3-3
.4-10
.4-11
.4-28
.4-49
.4-50
.4-53
.4-53
4-61
6-8
6-9
6-10
6-11
6-14
6-16
6-17
6-20
A-2
A-23

A-5
A-7
A-9
A-II
A-I3
A-I5
A-I7
A-I9
A-21

13

Figures and Tables (continued)

ThWe
Table 3- I.
Table 3-2.
Table 3-3.
Table 3-4.
Table 4-1.
Table 4-2.
Table 4-3.
Table 5-1.
Table 5-2.
Table 6-1.
Table 6-2.
Table 6-3.
Table 6-4.
Table 6-5.

14

hp
LED Description During Normal Operation
3-4
Approximate Duration References
3-7
AUI Cable Activity LEOs and Possible Causes on Transmit.
3-15
StarLAN Cable Activity LEDs and Possible Causes on Transmit .. 3-17
LAN Node Diagnostic Tests
4-3
Test Dependencies
4-9
Test Dependencies
.4-16
Selftest LEDs and Subtest Descriptions
5;'7
Reporting of Unexpected Results from Self Test.
5-10
CSDUMP Error Message Parameter
6-:5
Trace Entry Type Mnemonics
6-13
Function Codes
6-18
Code Meaning of Irrecoverable Errors
6-20
Code Meaning of Recovered Errors
6-21

Conventions Used
NOTATION

DESCRIPTION

UPPERCASE

Words in uppercase or boldface text must be entered exactly as shown.
Punctuation characters other than brackets, braces and eHipses must also be entered
exactly as shown. For example:

Boldface

EXIT;
italics

Words in syntax statements that are in italics denote a parameter that must be
replaced by a user-supplied variable. For example:

CLOSE filename
lowercase
nonbold

Words in lowercase or nonbold text denote substitutable variables or user-defined
strings.

[ ]

An element inside brackets in a syntax statement is optional. Several elements
stacked inside brackets means the user may select anyone or none of these elements. For example:

[~]

User may select A, B or neither.

When brackets are nested, parameters within inner brackets can be specified only if
parameters in the outer brackets or commas (place holders) are specified. For
example:
[ pa rm 1 [ , pa rm2 [ , pa rm3] ]] can be entered as:
parm 1 ,pa rm2, pa rm3 or pa rm1 , ,pa rm3 etc.

Optional parameters that are not position-related are as follows:
[ pa rm 1]

{

}

[, pa rm2]

When several elements are stacked within braces in a syntax statement, the user
must select one of those elements. For example:

{

~} User must select A or B or C.

Vertical parallel lines indicate that any or none of the options can be used in any
sequence but none of the elements may appear more than once. For example:

A
B

Choose A,B,C, or C,A, or B, etc.

C

15

Conventions Used (continued)

NOTATION

DESCRIPTION
A horizontal ellipsis in a syntax statement indicates that a previous element can be
repeated. For example:

[, itemname] ... ;
A vertical or horizontal ellipsis may also denote omission or repetition.
A shaded delimiter preceding a parameter in a syntax statement indicates that the
delimiter must be supplied whenever (a) that parameter is included or (b) that parameter is omitted and any other parameter that follows is included. For example:

i tema [~i temb] [ ,itemc]
means that the following are allowed:

itema
itema,itemb
itema,itemb,itemo
it ema , , it emo
When necessary for clarity, the symbol fJ. can be used in a syntax statement to indicate a required blank or an exact number of blanks. For example:

SET[(modijier)]fJ.(variable);
underlining

When necessary for clarity in an example, user input can be underlined.
example:

For

NEW NAME? ALPHA
In addition, brackets, braces or ellipses appearing in syntax or format statements
that must be entered as shown will be underlined. For example:

LET var[lsubscriptl]=value
Shading represents inverse video on the terminal screen. Shading is also used to emphasize key portions of an example.
The symbol [
) indicates a key on the terminal keyboard.
[RETURN) indicates the RETURN key.
[CONTROLloha r

16

For example,

Control cha.racters are indicated by [CONTROL) followed by the character. For example, [CONTROLlY means hold the CONTROL key and press Y simultaneously.

Reader' Comment Sheet
Information Networks Group
LAN3000/V Diagnostic and Troubleshooting Guide
30242-90003

August 1987

We welcome your evaluation of this manual. Your comments and suggestions help us to improve our
publications. Please explain your answers under Comments, below, and use additional pages if necessary.

Is this manual technically accurate?

Yes

No

Are the concepts and wording easy to understand?

Yes

No

Is the format of this manual convenient in size, arrangement, and readability?

Yes

No

Comments:

This form requires no postage stamp if mailed in the U.S. For locations outside the U.S., your local company representative will ensure that your comments are forwarded.

FROM:
Name
Company
Address

Date

-----------

FOLD

FOLD

111111

NO POSTAGE
NECESSARY
IF MAILED
IN THE
UNITED STATES

BUSINESS REPLY MAIL
FIRST CLASS

PERMIT NO. 256 ROSEVILLE, CALIFORNIA

POSTAGE WILL BE PAID BY ADDRESSEE
Publications Manager
HEWLETT-PACKARD COMPANY
Roseville Networks Division
8000 Foothills Boulevard
Roseville, California 95678-6598

FOLD

FOLD

1

General Information

Introduction
An IEEE 802.3 Local Area Network (LAN) presents a relatively complex
hardware troubleshooting environment. The connection of one computer system
(i.e., node) to another may involve many hardware components.
Figure 1-1, for example, shows four HP 3000 MPE-V systems connected together
on different LANs or LAN segments. Each system connection requires hardware
that depends on the type of LAN cable. Furthermore, various combinations of
Hubs, Bridges, and Repeaters are used to extend or join the individual LANs.
As the network grows, network troubleshooting becomes more complex.
Additional systems (such as HP 9000s, HP 1000s, personal computers, or other HP
3000s), cabling, and LAN extension hardware provide additional sources of
network faults.

MPE V
System

Repeater I Bridge

~branchl ThickLAN
-) ~ csa
ThickLAN

MPE V
System

I

: AUI

I

I

AUI

U

-+---c:>-lf-----+9<::J--+----b-l=:::C:J=:f---(lo

A4-4_'

transceiver I
MAU

ThinLAN Hub

ThinLAN

AUI

Bridge

LJ

nJ - - - L-F-

StarLAN (I

MPE V
System

ThinMAU

MPE V
System

I

htJ

-------

StarLAN
Hub

StarLAN
Hub

series 37 or
MICRO 3000
Systems Only

Figure 1-1. LAN Hardware Troubleshooting Complexity

General Information
1-1

Various tools are available on MPE-V systems to assist in the LAN
troubleshooting process. As they pertain to LAN troubleshooting, the following
tools are described in this manual:
• SHOWCOM: for monitoring the communication line.
• Activity LEDs: for monitoring activity on AUI or StarLAN node cabling,
and on the LANIC card.
• LANDIAG: the LAN Diagnostic program for testing composite LAN link
hardware.
• Self Test and selftest LEDs: primarily used for testing the LANIC card
circuitry.
• Tracing: for analyzing protocol activity of the network at the link level.
• Memory Dump: for analyzing a system crash in relation to a LAN
problem.
• NSDPAN5/NSDUMPJ: for formatting memory dumps.
• LISTLOG5: for analyzing the system log.
• NMMAINT: for analyzing the NMS log.
• CSLIST: for checking the version of communications software.
• DSLIST: for obtaining a list of DS software module versions.

General Information

1-2

Applicable Networks
The tools provided in this guide are described as they pertain to troubleshooting
LANs that conform to Hewlett-Packard implementations of the IEEE 802.3
standards. These LANs feature baseband signaling, and a Carrier Sense Multiple
Access with Collision Detection (CSMA/CD) network access protocol.

Coaxial Cable LANs
Hewlett-Packard coaxial cable LANs feature 10-megabit per second burst
transfer rates over a coaxial cable bus to which each node attaches.
IEEE 802.3 Type 10BASES Standard

This LAN uses a "thick" (0.4 inch/IO mm diameter) coaxial cable. Thick cable
LANs feature connection of up to 100 nodes on a single 500 metre bus segment.
HP 3000 MPE-V systems connect to this LAN using the HP LAN3000/V link
product. Hardware included with this product consist of a Local Area Network
Interface Controller (LANIC) interface card, an Attachment Unit Interface
(AUI) cable, and an HP 30241A Medium Attachment Unit (MAU) and tap
assembly for cable access.
IEEE 802.3 Type 10BASE2 Standard

This LAN uses a "thin" This LAN uses a "thin" (0.19 inch/4.9 mm diameter) RG58
C/U coaxial cable. ThinLAN cables feature connection of up to 30 nodes on a
single 185 metre bus segment.
HP 3000 MPE-V systems connect to this LAN using the HP ThinLAN3000/V
link product. Applicable hardware consists of the LANIC interface card\ and HP
28641A ThinMAU and BNC tee connector for cable access.

Twisted Pair LAN
IEEE 802.3 Type 1BASES Standard (Proposed)

The Hewlett-Packard twisted pair cable LAN, HP StarLAN, features a I-megabit
per second burst transfer rate through a hierarchical structure of HP 27212A
StarLAN Hubs. Nodes connect to the Hubs via the twisted pair cables; each
cable can be up to 250 metres in length.
HP 3000 MPE-V systems (Series 37 and MICRO 3000 systems only) connect to
this LAN using the HP StarLAN3000/V link product. Applicable hardware
consists of the interface card, and twisted pair cable ordered separately.

General Information
1-3

Fault Isolation and Repair Strategy
LAN hardware faults must be located and corrected. Because LANs are
comprised of many pieces of hardware, faults must identified to the field
replaceable unit (FRU) level of assembly. If the faulty FRU cannot be
immediately corrected, it is replaced with a new or functional unit. (For repair,
replacement, or return procedures, consult the installation or service manual for
the failed unit.)
LAN faults can be classified into two categories: Node faults, and Network
faults. A Node fault is characterized by a single node failure that does not
affect the other nodes on the network. A Network fault is characterized by
multiple node failures, where it is likely that a piece of hardware used commonly
by the nodes has failed. Fault isolation procedures for both types of faults are
needed.
The following manuals provide both node and network fault isolation
procedures. They are differentiated by the type of network to which they apply.
For coaxial cable LANs, refer to the LA AT Link Hardware Troubleshooting
Manual, 5955-7681. (HP CE Handbook version, 5959-2217.)
For HP StarLAN, refer to the HP StarLAN Diagnostics and Troubleshooting
Guide for PCs, 50906-90060.
Appendix A of this guide provides fault isolation procedures, in flowchart
format, for HP 3000 MPE-V links. Although network considerations are not
excluded, these procedures focus on MPE-V link Node faults. They complement
the above troubleshooting manuals by providing an alternative set of procedures
that use more features of the tools described in this guide.

General Information
1-4

Network Map
will be
When trouble shootin g a networ k, the availability of a networ k map
tt-Pack ard,
Hewle
by
d
require
As
ks.
networ
x
critical, especially for large comple
of the
uration
config
the
during
ped
develo
been
the networ k map should have
cation.
modifi
or
growth
any
reflect
networ k, and mainta ined to
k, includi ng
A networ k map provides the physical layout of nodes on the networ
ent of all
distances from one to another. Physical layout means the placem
ent. In
equipm
ter
compu
k
networ
related
all
and
Hubs
cables, MAUs and Taps,
ation,
addition, the networ k map should be labeled with relevan t node inform
any locally
including node names, globally administered station addresses, and
k softwa re
administered station addresses. The config uration of relevan t networ
on each node would be helpful.
the
If a networ k map is not available, you should make one. Refer to
2;
4-9000
1,3234
e
(Volum
!
Manila
ce
Releren
er
Manag
k
NS300 0/V Networ
items
ured
Volum e 2, 32344-90012) for inform ation on addresses and other config
Guide
ng
include d in the networ k map. Also, refer to the H P SiteWire Planni
(5959-2201) for additio nal inform ation.

Genera l Inform ation
1-5

Abbreviations and Nomenclature
You may find the following terminology useful when reading this guide.

General Information
1-6

AUI

Attachment Unit Interface.

Bridge

Address filtering device connecting different LANs,
such as coaxial cable LAN to coaxial cable LAN, or
coax cable LAN to twisted-pair cable LAN.

Coax

Coaxial cable medium for 802.3 networks.

CR-SR

Control Register - Status Register.

eRC

Cyclical Redundancy Check.

DMA

Direct Memory Access.

DRT

Device Reference Table.

FRU

Field Replaceable Unit (e.g. interface card, or cable
section.

Heartbeat

After successful frame transmission, a short collision
indicator test signal.

1MB

Inter-Module Bus, the bus supporting I/O in HP 3000
systems (Series 4X/5X/6X/70).

I/O

Input/Output.

Hub

A central device to which multiple cables (hence nodes)
connect, e.g., StarLAN Hub, ThinLAN Hub.

Jabber

Excessive LANIC transmission. A jabbering node
prevents other nodes from gaining access to the
network medium.

LAN

Local Area Network.

LANIC

Local Area Network Interface Controller for IEEE
802.3 LAN I/O (the interface card in HP 3000 MPE-V
systems).

LED

Light Emitting Diode.

Loopback

Transmission and receipt of data to verify operation of
the communication path.

MAU

Medium Attachment Unit, a device that provides the
LANIC with access to a coaxial cable medium.

MAUPON

MAU Power On signal.

MAUPS

MAU Power Sense signal.

MC

A multicast message; a type of broadcast message sent to
a group of stations, but not necessarily to all the stations.

Monitor

RAM resident Z80 code used together with the self test
in order to upload detailed test results to the HP 3000.

NMI

Non-Maskable Interrupt.

MPU

A Z80 microprocessor on the LANIC.

Node

Uniquely addressable station on a LAN.

NS

HP Network Services for the HP3000.

OBII

Obtain Interrupt - An IMB/SIMB command.

RNT

Remote Node Test.

SIMB

Synchronous Inter-Module Bus, the bus connecting the
CPU, Memory, and 1/0 in HP 3000 Series 37 and
MICRO 3000 systems.

SPU

System Processor Unit on HP 3000.

STREG

Self Test Register.

Twisted-pair

Twisted-pair cable for IEEE 802.3 networks.

General Information
1-7

2

SHOWCOM

The SHOWCOM command is a useful tool for monitoring the status of a line while
it is in use. To use SHOWCOM, you normally must have operator (OP) capability,
and you must be on the system console.

Syntax
:SHOWCOM xxx [;ERRORS]

[;RESET]

Where xxx indicates the logical device (ldev) number of a Communication
System (CS) device, i.e., the LANIC card.
The ERRORS option causes SHOWCOM to display all available information. RESET
causes all fields to be reset to zero after they are displayed.
For example:

:SHOWCOM 100;ERRORS
TRANSMIT
LON - 100
RECEIVE
MESSAGES SENT
7364
MESSAGES RECVD 9941
COLLISIONS
0
BCC/CRC ERRORS 0
EXC COLLISION ERRS
30
BUFF OVERFLOWS 3676
UNDERRUNS
0
OVERRUNS
0
CLR TO SEND LOSSES
0
LENGTH ERRORS 0
# OF RECOVERABLE ERRORS
32
LAST RECOVERABLE ERROR
6
# OF IRRECOVERABLE ERRORS 4
LAST IRRECOVERABLE ERROR 201
LINE IS CONNECTED

SHOWCOM
2-1

Interpreting Results
Transmit Fields

MESSAGE SENT

This is the number of frames that the HP 3000 gives
to the LANIC, and is not necessarily the number of
frames that were actually transmitted onto the LAN
When this value increases with time, it means that the
driver continued to give frames to the LANIC card
whether or not the frames were successfully
transmitted.
This number will include protocol reply frames, such
as those in response to XID or TEST frames received
from a remote node. Such response frames are
internal to the driver/firmware and are counted even
though the responses were transparent to higher level
processes.

SHOWCOM
2-2

COLLISIONS

This is the number of times that the LANIC card
experienced a collision on transmit. A large number
here may mean excessive network traffic, a topology
violation (e.g., excessive distances), or excessive jitter.
On a coaxial cable LAN, the local MAU/ThinMAU
may be faulty, or a remote MAU/ThinMAU may be
"leaky." On a StarLAN, a Hub or LANIC card may
be faulty.

EXC COLLISION
ERRS

This statistic is incremented if, after 16 collisions, a
frame was not successfully transmitted onto the
LAN.

UNDERRUNS

The LANIC card transmits data onto the line at a
high rate. This field shows the number of times, if
any, that data could not be transmitted onto the line
at the required rate.

CLR TO SEND
LOSSES

Not meaningful for the LANIC card (applies to INP
interface card).

Receive Fields

MESSAGES RECVD

This includes frames actually passed from the LANIC
card to the 3000. Errors detected by the LANIC that
do not cross the LANIC-to-HP 3000 boundary are
not included. Therefore, frames with length or CRC
errors, and multicast frames that do not pass the
LANIC card's filtering algorithm are not counted.
However, protocol frames that are received do pass
across the LANIC-to-HP 3000 boundary and are
therefore counted in this statistic.

BCC/CRe ERRORS

A frame was received with bad checksum (CRC).
This is an indication that a bit error probably
occurred. Since the bit error rate is supposed to be
very close to zero, a large number here is probably
cause for concern. On a coaxial cable LAN, there
may be excessive jitter, too many nodes, nodes not
properly spaced on the cable, improper AUI cables,
cables of low quality, bad MAU, or bad LANIC. On
StarLAN, there may be excessive jitter, excessively
long or poor quality cables, a bad Hub, or bad
LANIC.

BUFF OVERFLOWS

This means we received a frame but didn't have a
buffer to put it into. A large number here probably
means that too few receive buffers were configured,
too few "maximum reads outstanding" were
configured, or the system is too busy or has too little
real memory. The number of buffers probably needs
to be increased.

OVERRUNS

Incremented for every inbound frame for which the
LAN controller chip attempted to write a data word
to memory. It was delayed so long by the bus latency
that inbound data was lost (the FIFO on the LANIC
card overflowed). A large number here means we
have too much bandwidth of the IMB/SIMB utilized.
This may mean that too many high-speed channels
are connected on the same IMB/SIMB.

LENGTH ERRORS

Incremented for every frame where the length-field
in the 802 frame does not match the number of bytes
actually received. This also includes frames over
1522 bytes in length (including FCS, frame check
sequence field). A very large number here probably
indicates that a non-802.3 compatible node (e.g.,
Ethernet) is transmitting, or that there is some bad
hardware transmitting to us. There could also be
someone transmitting frames longer than 1522 bytes.

SHOWCOM
2-3

Errors

# OF RECOVERABLE
ERRORS

The number of errors reported by the driver that did
not cause the link to be disconnected.

LAST RECOVERABLE
ERROR

This gives the last non-zero CS error number returned
by the driver. The completion status can be anything
other than IRRECOVERABLE ERROR (completion
status 3) or CATASTROPHE (completion status 5).
This includes CS "recoverable error codes" as well as
error numbers that are normally "irrecoverable error
codes" but were completed with good status so as not
to cause the translator to shut down the link.

# OF IRRECOVERABLE

The number of errors that would force the link to be
disconnected, e.g., on a coaxial cable LAN, could not
turn on MAU/ThinMAU power.

ERRORS
LAST IRRECOVERABLE
ERROR

This is the last non-zero CS error code that the driver
gave when it completed a request with
IRRECOVERABLE ERROR (completion status 3) or
CATASTROPHE (completion status 5).
This will not be updated for CS error codes 63 and

64, since these are not really "errors" but are normal
completion codes for ABORTIO (function code 66),
generated on all process terminations (called from
EXPIRE), and the HARD ABORT, generated when
the translator detects other errors.
The line will always be in anyone of four states: '

SHOWCOM
2-4

(1) CONNECTED

The driver is active, firmware has been downloaded
and command and response queues have been
initialized. Basically, the driver and hardware are
ready to receive and transmit frames.

(2) DISCONNECTED

The driver and hardware have not finished
initializing or an irrecoverable error has occurred.
Basically, the driver and hardware are not ready to
receive and transmit frames.

(3) CLOSED

The LDEV is not allocated (not in use).

(4) UNDEFINED

Software error.

3

Activity LEOs

There are 15 LEDs located on the edge of each LANIC card. Figures 3-1 and 3-2
show the LED locations for Series 4Xj5Xj6Xj70 and Series 37jMICR03000
systems, respectively.

A 8

co, ,
..

0

G H I

J K L .. N .

c

~I . .u.. l..o.. l.. u..l~ a! I d i

0 I:

lii

000000000000000

LANIC TEST
(RESET)
SW1

I I
LEOs

Figure 3-1. LANIC Card LEDs On Series 4Xj5Xj6Xj70 Systems

Activity LEDs
3-1

P1

,-

LEOs

RFI Clip

Figure 3-2. LANIC Card LEOs On Series 37 and MICR03000 Systems

Each of the 15 LEOs is labeled with different labels. The single alphabetic labels
provide a quick reference to the LEOs. In addition, one- and two-letter
mnemonics are provided to remind users of the LEO function. The labels and
functions of the LEOs are shown in Figure 3-3.

Activity LEOs
3-2

The LEDs can be classified into the following groups:

• Cable Interface Activity LEDs. The seven LEDs on the left, labeled "A"
through "G", monitor activity on the cable, hence the network.
• MPU Activity LEDs. During normal operation, the eight LEDs on the
right ("H" through "N" plus "*") monitor LANIC card MPU activity.
However, these LEDs are also used by card self test (refer to Chapter 5).
When used by self test, LED mnemonics do not apply.

AS

vp

co

E

F'

G

HI

I I I I
E CLL

E DOL

E CRL

J

K

L

TX RX MN DL RO

M

Nfl

Q IT

ST

~_ _ _ _

MAU POWER
(NOT USED ON
STARLAN CARD )

<

EDGE
LEVEL

<
<

EDGE
LEVEL

COLLISION

DATA OUT
CARRIER SENSE

SELF' TEST

..IDLE SELF' TEST

-----10

QUIESCENT

EDGE
LEVEL

ROM CODE

-----12

DOWN LOADING

---~3

' - - - - MONITOR MODE

-----i 4

PROCESSING A RECEIVE

N
0
A
C
T
I
V

I
T
Y

L
I
G
H

A
V

T

y

A

A

U

C
T
I

C

C

T
I

K

V

V

0

I
T

I
T

N

y

y

H

PROCESSING A TRANSMIT

E

5
-

6

S
T

EDGE

OFF BLINK ON

OFF

LEVEL

OFF

OFF

ON

ON

-

SELF' TEST BIT ,

Figure 3-3. LED Labels and Functions

Table 3-1 provides a descriptive summary of the LEDs during normal operation.

Activity LEDs
3-3

Table 3-1. LED Description During Normal Operation

LED

Mnemonic

A

VP

B

CL-E

ON when the collision line goes active.

C

CL-L

ON for steady collision detect signal from the Manchester
encoder/decoder.

0

OO-E

ON when the Data Out signal from the Manchester encoder/
decoder goes from false to true.

E

OO-L

ON if there is a steady Data Out signal from the Manchester
encoder/decoder.

F

CR-E

ON when the carrier sense signal from the Manchester encoder/
decoder goes from false to true.

G

CR-L

ON if there is a steady carrier sense signal from the Manchester
encoder/decoder.

H

TX

ON when the card is processing and transmitting a frame.

r

RX

ON when the card is processing a frame addressed to this node.

J

MN

ON when the card is monitoring all link activity, or is monitoring
activity sent to a particular address not its own.

K

DL

ON when the download command is received from the SPU to
download operating firmware.
OF F when the SPU commands the MPU to begin executing the
download feature.

L

RO

ON when ROM-resident firmware is being executed by the MPU;
OF F when RAM-resident (downloaded) firmware is being executed.

M

Q

N

IT

ON when the MPU is quiescent (waiting for a host command or I/O
completion), during which it checks for activity needing attention.
ON when the MPU is executing an idle test of internal card
circuitry, during which the MPU tests card hardware that will not
affect the readiness to process frames. The idle test also runs
before the node becomes operational on the link.

.'

*

Activity LEDs

3-4

8T

Description
For Coaxial Cable LAN: ON when power is available to the
MAU/ThinMAU.
For StarLAN: Not used, always OF F.

ON when the MPU is executing ROM-resident self test. When 5T is
lit, LEOs H through N are selftest progress and failure indicators,
rather than as defined above.

Cable Interface Activity LEOs
The cable interface LEDs monitor t.he four functions shown below:

DO

Data out. These LEDs are ON when data is transferred from
this LANIC onto the Data-Out/Transmit signal pair.

CL

Collision detect. On a coaxial cable LAN, these LEOs are ON
when a collision is detected by the MAU/ThinMAU on this
node. For receiver-based collision detection devices, collisions
are monitored continuously (whether transmitting or not) -the CL LEOs indicate virtually every collision that occurs on
the coaxial cable. For transmit-based collision detection
devices, t.he CL LEOs indicate collisions that occur only when
transmitting.
Note that the CL LEOs are blocked from lighting during the
IEEE 802.3 SQE heartbeat signals, which occur after each
transmission.
On a StarLAN, these LEDs are ON when a collision pattern is
output by the Hub to which this node is connected. The CL
LEOs will be active whether or not the LANIC is
transmitting. (Heartbeat signals are not employed on
StarLAN.)

CR

Carrier sense. These LEOs are ON when data is detected
coming into the node on the Data-In/Receive signal pair, or
when the collision function is detecting collisions. For coaxial
cable connections, the CR LEOs do not light for SQE
heartbeat signals (StarLAN connections do not employ
heartbeat).

VP

Voltage plus. For coaxial cable connections, this LED is
connected through a current limiting resistor directly to the
VP AUI lead. When ON, it indicates power (+12V) is
availa ble to the MAU /ThinMAU.
For StarLAN connections, this LED is not used and is always
OFF.

Activity LEDs

3-5

The E and L Indicators
Each of the DO, CL and CR functions consist of a pair of LEOs, labeled E and
L (for Edge and Level, respectively). The pair is driven in such a manner that all
conditions of activity -- from occasional isolated events to continuous events -are visually distinguishable. The LEOs are encoded as follows:
"E" LED. Turns ON each time a monitored event begins. It remains ON for
6 milliseconds regardless of the length of the event.
-

"L" LED. Turns ON at the beginning of the event and turns off at the end
of the event.

Following this algorithm, a single isolated event of short duration produces a 6
millisecond blink of the E LED, and the L LED is on for the length of the event,
which is short. Therefore, the L LED appears to remain off.
As the frequency of events of short duration increases, the E LED appears to be
constantly illuminated, and the L LED begins to glow.
When short duration events occur constantly, both the E and L LEDs will appear
to be constantly illuminated.
A single event of very long duration produces a single 6 millisecond blink of the
E LED at the beginning of the event, and the L LED turns on and stays on for a
long time, until the event is completed.
Events that continue for a "very long" time will cause the E LED to blink at the
beginning of each event for 6 milliseconds, and the L LED will appear to be
constantly illuminated.
Events on a normally-operating network are all of short duration. For reference,
see Table 3-2.

Activity LEOs

3-6

Table 3-2. AplJroximate Duration References
Approximate Duration
Event
Coaxial Cable

StarLAN

Maximum frame length transmission

1.2 ms

12 ms

Minimum frame length transmission

0.051 ms

0.51 ms

Collisions

0.049 ms

0.49 ms

For events of short duration, such as those in Table 3-2, note the following:
As the frequency of activity increases, the frequency of flashing of the E
LED increases while the L LED is off or very dim.
When the frequency of activity is very high, and the E LED appears to be
ON continuously, the L LED indicates further increase in activity by
becoming brighter and brighter until it reaches full intensity. This state of
the E and L LEDs indicates continuous short events.

Relationship to Cable Signals
To understand the indicat.ions given by the DO, CL and CR LEDs, it. is necessary
to understand how the signals that drive these LEDs are related to the lines of
the attached cables.
For coaxial cable LAN connections, the HP AVI cable contains three signal pairs
used by the MAU /ThinMA V: the Data-Out pair, the Data-In pair, and the
Control-In pair.
For StarLAN connections, the HP StarLAN cable contains two signal pairs: the
Transmit pair, and the Receive pair. The Receive pair is used for frame
reception and collision signals.

Activity LEDs

3-7

DO LED Events

The event indicated by the DO LEOs is the enabling of the data encoder by the
protocol controller on the LANIC. The event begins when the encoder is turned
on. While the encoder is on, a continuous stream of encoded data bits is
transmitted by theLANIC onto the Data-Out/Transmit signal pair. The event
ends when the data encoder is disabled. When the encoder is disabled, data bits
are no longer sent onto the Data-Out/Transmit pair.
The transmission of a single frame onto the cable Data-Out/Transmit pair is one
event, and will cause the E LED to blink ON for 6 ms. The L LED will be
illuminated for the length of time required to transmit the data bits onto the
Data-Out/Transmit pair.

CL LED Events

The event indicated by the CL LEOs is the occurrence of the Signal Quality
Error (Collision) signal on the Control-In/Receive pair of the attached cable.
The event begins when the first transition of the collision SQE signal is received
at the LANIC, and ends 200 nanoseconds after the last transition is received.
Coaxial Ca ble LAN Connection. On collision detection, the MAV /ThinMAU
sends the SQE signal (a 10 MHz signal) to the LANIC on the Control-In pair of
the AVI ca ble.
The SQE heartbeat, a short burst of 10 MHz signal returned on the Control-In
pair after each successful transmission, is used to test the collision detection
circuitry. SQE heartbeat does not calise the CL-E LED to blink. For a period of
5.3 microseconds after a successful transmission, the CL-E LED is blocked from
collision signals. Therefore, heartbeat and normal collisions that occur during
this period will not activate the CL-E LED.
Heartbeat will cause the illumination of the L LED for approximately 1
microsecond; however, this is too short to be seen.
StarLAN Connection. On collision detection, the StarLAN Hub to which the
LANIC is connected sends a Collision Presence Signal (CPS) -- a 1 MHz signal -to the LANIC on the Receive pair of the StarLAN cable.
Heartbeat signals are not used on StarLAN. Note that software used to monitor
card statistics may increment heartbeat errors. Such errors should be disregarded
when they occur with a StarLAN card.

Activity LEDs
3-8

CR LED Events

There are two events indicated by the CR LEOs:
- Reception of data on the Data-In/Receive lines of the attached cable, or
- Occurrence of a collision signal as described above (see "CL LED Events").
The event begins when the first data transition arrives on the Data-In/Receive
pair, or when the collision event begins. whichever occurs first. The event ends
200 nanoseconds after the last data transition on the Data-In/Receive pair, or
after the collision event ends, whichever occurs last.

MPU Activity LEOs
When the LANIC card has been reset either by power-up of the system or by the
operating software, all eight of the MPU activity indicators (LEOs "H" through
"N" and "*") will be on continuously. This indicates that the MPU is not
executing.
The cable interface LEOs \\rill all be off. For coaxial LANs, this includes the VP
LED, indicating the MAU/ThinMAU is not powered.
After the LANIC has successfully passed self test (see Chapter 5), the "*" LED
will be OFF, and the other seven MPU activity LEOs will now indicate MPU
activity. When the "*" LED is OFF, the "H" through "N" LEOs should be
interpreted according to their two-letter mnemonics.
When self test passes, the system processor unit (SPU) is interrupted and notified
of the event. Between the time that this interrupt is given and the time when the
SPU begins to access the LANIC, the "RO" and "Q" LEOs will be illuminated.
This indicates that the LANIC is executing ROM code and is quiescent, while
waiting for the SPU to take control. (For coaxial LAN connections, the VP LED
will be illuminated, indicating that the MAU is powered.)
Any activity on the network will be reflected by the state of the CL and CR
LEOs. The LANIC will never transmit in this state, and therefore, the DO LEDs
will remain inactive.

Activity LEDs
3-9

On command from operating software, the SPU prepares the LANIC for
operation. It must first download the operating firmware from system memory
to the LANIC. When this process begins, the "DL" LED turns ON, and the "Q"
and "IT" LEDs will extinguish. After each download command, the "Q" LED
lights for a few miIIiseconds. At least 7 download commands occur, but they
may not be separately distinguishable. Note that the pattern that occurs on one
working system will occur on all other working systems. So if you are suspicious
of this process on your system, compare the download pattern on the suspect
system with a system that works.
After the download is complete, the SPU will instruct the MPU to begin to
execute the downloaded firmware. When this occurs, the "RO" and "DL" LEDs
will extinguish. The "Q" and "IT" LEDs will turn ON.
A short time later the SPU will instruct the LANIC to set its individual node
(station) address. When this occurs, the LANIC performs a duplicate address
check, which is accomplished by transmitting 10 self-addressed frames onto the
network with a 500 ms separation between frames. The "TX" and "DO-E" LEDs
will both turn ON for each of the 10 frames. In addition, the "CR-E" LED will
indicate that the frames were sent onto the coax and caused carrier to come on.
If collisions occur, frame transmission will be retried up to 15 times each, with
the resultant activity indicated on the "CL" LEDs.
Because the frames are self-addressed, the "RX" LED should not light during the
duplicate address check. If the "RX" LED does light, it is due to a frame
received from a remote node. This may occur, for example, if a duplicate station
exists, or an ordinary frame is addressed to the local LANIC.
If a reply to the duplicate address check is received, the duplicate address check
fails. No further check frames will be sent. The system software wiII close the
link and clear the LANJC, forcing all the LANIC card activity LEDs to turn ON
and stay ON. The LEDs will indicate extended idle self test in progress.
If the duplicate address check passes, the link is opened, and frame transmission
and reception wiII commence. The LEDs will indicate activity as it occurs.
Presuming the network and LANIC card are idle prior to a transmit request
from the SPU to the LANIC card, frame transmission should behave as follows
during normal network operation:

Activity LEDs
3-10

While idle, "Q" and "IT" LEOs are ON. For coaxial LAN connections, the
"VP" LED is ON.
When the MPU begins processing the transmit command, the "Q" and "IT"
LEOs will extinguish, and the "TX" LED will light. The LANIC begins the
transmit process by reading the frame from system to on-card memory.
Once the frame is in LANIC card local memory, and the network is free, the
serial transmission process begins. This causes the "DO-E" LED to light. The
"DO-L" LED will also light for the duration of the frame transmission, but
this mayor may not be visible depending upon the length of the individual
frame being sent.
For coaxial LAN connections, the serial data reaches the MAU/ThinMAU
and is transmitted onto the coaxial cable. The MAUjThinMAU receives its
own signal off of the coax, and sends it back down the AUI cable on the
Data-In signal pair.
For StarLAN, the serial data is transmitted on the StarLAN cable. The Hub
receives the data and sends it down the Receive signal pair.
The LANIC card detects data arriving on the Data-In/Receive pair of the
attached cable, resulting in the "CR-E" LED turning ON. The "CR-L" LED
will also be illuminated for the duration of the frame, but this mayor may
not be visible. If the "DO-L" LED is visible, the "CR-L" LED will also be
visible for approximately the same length of time.
If no collision is encountered, the "CR-E" and "DO-E" LEOs will go OFF
after 6 milliseconds, followed quickly by the "TX" LED going OFF, and the
"Q" and "IT" LEOs turning ON.
If a collision is encountered, the "CL-E" LED will turn ON, and up to 15
additional attempts to transmit the frame will occur. From the
retransmission attempts, the "DO" and "CR-E" LEOs may appear to be ON,
and the "DO" and "CR-L" LEDs will probably appear to be partially
illuminated. The intensity of the "-L" LEOs will be determined by frame
length, the number of retransmissions, and the time separation of the
retransmissions. The "CL" LEDs will also display behavior similar to the
"CR" and "DO" LEOs if multiple retransmissions are required before the
frame is successfully transmitted.
In the collision case, it must be remembered that other network activity may
also cause the "CL" and "CR" LEDs to light, and the activity caused by the
LANIC will be superimposed upon the network activity being displayed in
the "CR" and "CL" LEOs.

Activity LEOs
3-11

Network Fault LED Examples
This section provides examples of LAN faults and the resulting display of a
LANIC card's cable interface activity LEDs.
We presume that the LANIC card is enabled for operation on the network and
the LANIC card driver is turned on. This can be verified by the S HOWCOM
command (see Chapter 2): the line must be "CONNECTED".

NOTE
The LANIC card LEOs will not reflect network activity if the line is not
CONNECTED. For some faults, system software may shut down and reset the card,
resulting in a DISCONNECTED line.

Due to hardware differences, coaxial cable connection faults differ from
StarLAN connection faults.

Coaxial Cable LAN
The following faults and cable interface activity LED displays pertain to coaxial
ca ble LAN connections.
Open Tap
If the tap on the coax is not making contact with either the center conductor or
shield, the LEOs will indicate no network activity.
When the LANIC tries to transmit, a coJIision will occur. Thus, the OO-E, CR-E
and CL-E LEOs will all light.

Activity LEOs
3-12

Open Coax

Open coax faults include missing or loose terminators, loose barrel connectors, or
even breaks in the cable. For open coax faults, attempted transmission by any
node attached to the cable will result in a collision. For nontransmitting nodes,
the CR-E and CL-E LEDs may light. For transmitting nodes, the DO-E, CR-E
and CL-E LEOs will light.
Note that, unlike an open tap fault where only the single node is affected, all
HP 3000 nodes connected to the open coax cable will display these symptoms.

Shorted Coax

For a shorted coax, the voltage associated with any transmission attempt will be
clamped to zero (loss of carrier). If the coax is shorted close to the
MAUjThinMAU, the CR-E and DO-E LEDs will flash when the node attempts
to transmit.

Short On MAU/ThinMAU Power Circuit (VP)

If there is a short on the power lines going to the MAUjThinMAU, the LANIC
overcurrent protect switch will turn the power off, and the VP LED will turn
OFF. If this occurs during the self test, a failure code, 24H, will result.
If the short occurs during normal operation of the board, card firmware will
attempt to turn the power back on. This may occur up to 20 times. If this fails,
the firmware will report a fatal error to the driver and enter a soft reset state in
which the RO and Q LEDs will be lit. When the upper level software recognizes
the fatal error, it will do a hard reset on the card, leaving LEOs "H" through "*"
lit.

Continuous 'Transmission From a Remote Node

If some other node is transmitting continuously, and its MAUjThinMAU fails to
terminate its transmission, the local LANIC card will detect carrier. Thus, its
CR-L LED will be ON.

Constant Collision on the Network

Excessive voltages on .the coax are interpreted as collisions, for example, when
multiple transmissions simultaneously exist on the cable, or a faulty
MAU jThinMAU is leaking a DC voltage onto the coax. The local
MAUjThinMAU detects such voltages and actuates the Control-IN (CI) signal
line. This will cause the CL-L and CR-L LEDs to turn ON.

Activity LEDs
3-13

Open or Shorted Data Out Signal Lines

If the Oata Out signal pair in the AVI is open or shorted, the LANIC will not be
able to transmit. Ouring transmit attempts, the OO-E LED will flash. However,
since no transmission occurs, carrier will not be detected and the CR LEOs will
not turn on.
Note, however, that the LANIC will be able to receive frames. Received frames
will light the CR LEOs.

Open or Shorted 01 Pair in AUI

If the Data-In signal pair is disabled due to a short or a open, normal network
activity will not be detected by the CR LEOs.
However, the CR LEOs are lit for collision signals on the Control-In signal pair,
so a collision will cause the CR-E LEO along with the CL-E LED to blink ON.

External Loopback

Table 3-3 provides possible causes of network errors based on cable activity
LEOs that turn ON during a transmit. A transmit can be invoked by conducting
a loopback test, Test 14, of the LAN Node diagnostic (LANDIAG3000/V). See
Chapter 4.

NOTE
The causes listed should not be construed as being complete. They only serve as
suggestions on where to look for possible faults. Note that only single faults are
presumed. Multiple faults may result in other symptoms.

Activity LEDs

3-14

Table 3-3. AUI Cable ActiYity LEOs and Possible Causes on Transmit
LEDs that
Turn ON
During Loopback

Possible Causes from
a Single System

NONE

Bad LANIC

DO only

Bad AVI or connection;
Bad MAV;
Bad LANIC;
Shorted Coax

Shorted Coax

DO and
CR

Bad MAV;
Bad LANIC;
Bad Coax

Bad Coax

Bad Tap;
Bad terminator;
Bad barrel

Bad terminator;
Bad barrel

DO and
CR and
·CL
VP not lit

Possible Causes from
Multiple Systems

Shorted AVI;
Bad MAV;
Bad LANIC

StarLAN
The following faults and symptoms pertain to HP StarLAN connections.

Bad Hub

If the StarLAN Hub is not able to process network traffic, neither collisions nor
frames will be sensed by any LANIC card. Each card's CR or CL LEOs will be
OFF.
If any LANIC transmits, the the Transmit line will be active and the OO-E LEO
will light.
Note that these symptoms apply to all HP 3000 nodes connected to the Hub.

Activity LEOs
3-15

Bad Connection

If the cable is not mated properly at each end, or the cable is severed, the
symptoms are the same as for a bad Hub, but apply only to the affected node.
Open or Shorted Transmit Pair

If the cable's transmit pair is open or shorted, the LANIC will not be able to
transmit. Ouring transmit attempts, the OO·E LED will flash. However, since no
transmission occurs, carrier will not be detected and the CR LEOs will not turn
ON for this attempted transmission.
Note, however, that the LANIC will be able to receive frames and detect
collisions. These events will light the CR and CL LEOs.
Open or Shorted Receive Pair

If the receive signal pair is disabled due to a short or open, normal network
activity will not be detected by the CL or CR LEOs.
Continuous Transmission From a Remote Node

If some other node is transmitting continuously and its Hub fails to terminate its
transmission, the local LANIC card will detect carrier via the receive signal pair.
Therefore, the CL-L and CR-L LEDs will turn ON.
Constant Collision on the Network

For StarLAN, the Hub generates collision signals and disseminates them on the
receive signal pair. If collision signals are constant, the the CL-L and CR-L LEOs
will be ON.

Activity LEDs
3-16

External Loopback

Table 3-4 provides possible causes of network errors based on cable activity
LEDs that turn ON during a transmit. A transmit can be invoked by conducting
a loopback test, Test 14, of the LAN Node diagnostic (LANDIAG3000/V). See
Chapter 4.

NOTE
The causes listed should not be construed as being complete. They only serve as
suggestions on where to look for possible faults. Note that only single faults are
presume.d. Multiple faults may reStilt in other symptoms.

Table 3-4. StarLAN Cable Activity LEDs and Possible Causes on Transmit
LEDs that
Turn ON
During Loopback

Possible Causes from
a Single System

NONE

Bad LANIC

DO only

Bad Hub;
Bad Cable

Shorted Coax

DO and
CR

Bad Hub;
Bad LANIC;
Bad Cable

Bad Hub

DO and
CR and
CL

Bad Hub;
Bad LANIC

Bad Hub

VP not lit

Shorted AVI;
Bad MAV;
Bad LANIC

Possible Causes from
Multiple Systems

Activity LEDs
3-17

4

LAN Node Diagnostic

General Information
This chapter describes the LAN Node Diagnostic as a tool in troubleshooting an
MPE-V system link.

What is the LAN Node Diagnostic?
The LAN Node Diagnostic, or LANDIAG, is an interactive online program
designed to help identify malfunctioning LAN link hardware. The diagnostic
performs a series of tests upon the LANIC card and interface hardware. Each
test diagnoses a subset of the node's link hardware. Although it can help to
identify a particular field replaceable unit (FRU), it may may not be able to
distinguish which particular circuit within the FRU is malfunctioning.
As a program running on the HP3000, LANDIAG tests the connection between
the host computer and the main module of the LANIC card. The main module
is that part of the LANIC card hardware which performs the network interface
activities.
LANDIAG can be used to initiate card self test. The LANIC self test checks the
components on the card for operation in the IEEE 802.3 environment. For more
information on LANIC card self test, see Chapter 5.

Required Hardware
LANDIAG is provided with the link products and HP network services
(NS3000/V) software. When running this software, the system must have a
minimum of two megabytes of memory and the Expanded System Table
Microcode. (Systems that are now memory-limited should add one megabyte to
maintain current performance.)
When using LANDIAG, one of the following LAN link products should be
installed:
- ThinLAN3000/V or LAN3000/V links, for connecting HP 3000 Series
3X/4X/5X/6X/7X and MICROJOOO systems to a ThinLAN or TbickLAN
coaxial cable, respectively.
- StarLAN3000/V link, for connecting HP 3000 Series 37 or MICR03000
systems to an HP StarLAN twisted pair cable.

LAN Node Diagnostic
4-1

A maximum of one LAN hardware link per system is supported.
For general LANIC card installation guidelines, and interdependencies with other
I/O cards, refer to the LANIC installation manual provided with your particular
link product.

Required Software
LANDIAG3000/V, version A.55.27.000, runs on HP 3000 or MICR03000 systems
executing the MPE V/E operating system, version V-B-delta-! (or later). This
version of LANDIAG incorporates changes required for the support of StarLAN
connections. In addition, known bugs were corrected.
The diagnostic is provided on a master installation tape (MIT). It is provided
along with the LANIC card driver and NS3000/V.
The diagnostic is intended to be an online program, running on the MPE-V
operating system. It may be run timeshared with other programs, so that the
LAN hardware may be inspected without disrupting the activities of users who
do not need to access the network.

NOTE
All network activity, including NS3000/V, must be halted before the diagnostic
can be executed.

'System Tables

LANDIAG is normally installed with NS3000/V and the driver. Before
configuring and activating NS3000/V and the LAN link, you should check the
system tables. Consult the "System Configuration" sections in the N S3000IV
Network Manager Relerence Manual, 32344-90002, for guidelines on system table
modifications required.

Execution Time
The diagnostic can execute one complete cycle of tests in under one minute.
This is all the time needed to determine that the hardware is free from most
defects. However, if there are intermittent or unusual problems, many passes
through the diagnostic rnay be required.

LAN Node Diagnostic

4-2

Tests Included
Table 4-1 lists the primary tests run from LANDIAG. These are described in
detail later in this chapter.

Table 4-1. LAN Node Diagnostic Tests
1

Roll Call

2

ChannelID

3

Initialization

4

Self Test

5

Interrupts

6

Soft Reset

7

CR SR Loop

8

Bus Conflict

9

Address Offset

10

Extended Address

11

FIFO Write

12

F/P Conflict

13

Coprocessor

14

MAD Loopback (also used for StarLAN loopback)

15

Date Code

16

Loopback Hood (Not Applicable for StarLAN Connections)

17

Remote Node

LAN Node Diagnostic
4-3

Diagnostic Limitations
The following conditions are not tested from LANDIAG:

1.

Selftest toggle switch failure (does not apply to Series 37 and MICR03000
LANIC cards).

2.

Light emitting diode failure.
These are not tested by the diagnostic because they cannot be handled
programmatically. It would be a simple matter to verify their operation by
observation after initiating self test.

3.

Priority logic failure.
The diagnostic cannot explicitly control or implicitly predict the state of the
priority lines, so testing is not possible. However, failure recognition is
possible. If the priority logic does not work, one of the following will likely
result:
- erratic LANIC card operation,
- performance degradation, or
- SIMB/IMB deadlock.

4.

Powerfail warning holdoff of master handshake.
Testing this condition would require creating a powerfail condition in the
host, which is beyond the scope of this diagnostic.

5.

SIMB/IMB parity errors.
To test the parity error detector circuit, it would be necessary to risk system
integrity by deliberately introducing parity errors into memory.
The need to test the parity error detection circuit from LANDlAG was not
felt to be necessary. Such faults are not catastrophic to system or network
operation. In addition, other cards on the system would eventually detect a
parity error.

6.

Faulty bank lines that address out of bounds memory.
Though this condition will not be tested explicitly, it would be apparent if it
occurred. For example, on the series 39, 4x or 6x systems, the 1MB would
hang if out of bounds addressing occurred.

LAN Node Diagnostic
4-4

7.

Holdoff of reset until handshake ends.
For Series 39 through 70 systems, the LANIC is protected from data loss if
the selftest switch is inadvertently pushed during a handshake. This feature
is not tested because it requires direct manual intervention.. If the
implementation circuitry were to fail and go undetected, the functionality
of the LANIC card would not be significantly impaired.

LAN Node Diagnostic
4-5

Guidelines for Setting Up and Using LANDIAG
This section contains information that you will need to know before using
LANDIAG.

System Information
When the HP 3000 is configured, it needs to be told everything about the LANIC
card, including its LDEV number, DRT number, driver name, device type and
device subtype. The following information should be provided:
- Driver name: IOLANO. PUB. SYS
- Device type: 17
- Device subtype: 9
The LDEV number and the DRT number depend on the particular system
configuration. Refer to the MPE V System Operation and Resource
Management Reference M anllal (32033-90005) for system configuration details.
To use LANDIAG, you will need the LDEV number of your LANIC card. It is
the first item requested by the diagnostic. The diagnostic locates the LANIC
card by entering the I/O configuration table at the specified LDEV. If the table
indicates that the LDEV is not device type 17 and subtype 9, the user is told:

That LDEV is not configured as a LANIC.

Capability Levels
Because the diagnostic has direct access to every location in memory, use of the
program is controlled.
The user must have one of these three capabilities in order to run the diagnostic:
OP - system supervisor or operator
DI - diagnostician
SM - system manager
If the user's security level is not sufficient, the following error message will be
displayed on the screen:

OP, DI or 8M capability needed to run diagnostic.
For LANDIAG Test 17 (Remote Node Test), a CS capability -- allowing access to
the Communications Subsystem -- is also required.

LAN Node Diagnostic
4-6

System Type Check
For Series 30/33 systems, the diagnostic will continue to run, but the following
warning is issued:

Diagnostic not designed for HP3000 /30 or/33.

Operation with Network Services
Refer to the N S3000IV Network Manager Relerence Manual (32344-90002) for
software configuration of network services on the system.
When a communication subsystem has control of the LANIC card, any attempt
to allocate it by the diagnostic will fail. The diagnostic will report a warning.
The message will be:

The LANIC could not be allocated.
To use the diagnostic,the LANIC card must be in an "AVAILABLE" state. You
can tell if the card is "AVAILABLE" as follows:
- From MPE, issue the SHO\VDEV command. The LDEVs will be listed
along with availability information. Locate theLDEV of the LANIC card;
the status must indicate "AVAILABLE".
- From MPE, issue the SHOWCOM command, specifying the LDEV of the
LANIC card (see Chapter 2). The card is available if the LINE is
DISCONNECTED or CLOSED.
With the NS3000/V services installed, the command sequence:

:NSCONTROL STOP
:NETCONTROL STOP
will disconnect the user's node from the network. For more information on the
use of these commands, refer to the NSI3000 Network Nfanager Relerence
M anlfal (32344-90002).
The user may run the diagnostic when the communication subsystems have
released the LANIC card. The system can recover control of the LANIC if there
should be a crash while the diagnostic is running.

LAN Node Diagnostic
4-7

Reliability and Recovery from Power Failure
In the event of a power failure MPE-VIE will protect the integrity of the system
tables. For the Series 6x/70 systems, the diagnostic process will resume operation;
no reinitialization will be necessary. For the Series 4xl5x systems, you have to
restart the diagnostic.
The test which was underway when the power failure occurred will probably
report a failure, whether or not the circuit being checked was actually faulty.
This mistake will arise because the diagnostic will almost certainly timeout
waiting for a response from the LANIC. This is because the LANIC (which
derives its power from the backplane) will lose the execution context as its
microprocessor and RAM will lose power due to power failure. Also, the
LANIC performs a self test at power on, including a memory test which
obliterates the contents of local RAM.
When a power failure has taken place, you should redo any test that failed.

Integrity of the System
The diagnostic will expose any defect in the LANIC card which compromises
system integrity. If the LANIC card contains such defects, there is an
unavoidable risk of altering memory or system crashes. Because the diagnostic
performs rigorous testing of all LANIC circuits, including those that implement
direct memory access, intermittent problems will be exposed.
The diagnostic will not cause any crashes or damage to system memory when
good LANIC cards are tested. The diagnostic is no more likely to cause system
problems than normal use of the LANIC by the LAN link and NS3000/V
products.
When a LANIC is installed or replaced, the diagnostic should be executed and
system integrity verified before user processes are permitted to execute. If
system failures are observed during normal operation and the LANIC is
suspected to be the cause of the failures, users should be removed from the
system prior to diagnostic execution.
If a LANIC card malfunction is suspected, but LANIC usage does not appear to
have compromised subsystem integrity, it is unlikely that diagnostic execution
will cause system integrity to be compromised. In this case the diagnostic can be
executed without removing the users from the system.
Each time LANDIAG is run, it enforces conditional execution of its tests in
order to significantly reduce the probability of system halts. Table 4-2 shows
various test completion states and their impact on other tests. (Note that
intermittent failures on the LANIC card may defeat the protection provided.)

LAN Node Diagnostic
4-8

Table 4-2. Test Dependencies
Control Test (State)

Dependent Test (State)

respond.

No other test is allowed to run because a
non-responding channel will cause a a system
halt for addressed 1MB instructions.

Channel ID (test 2) ran and found that
responding channel at specified LDEV does not
appear to be a LANIC.

Test 3 thru 17 are not allowed to run because
certain LANIC instructions may cause a system
halt if issued to other than a LANIC.

Network Service is still active as determined
during the diagnostic initialization process.

10 (test 2) are prevented from executing since

Roll Call (test 1) ran and found that the channel
# corresponding to the specified LDEV did not

All tests except Roll Call (test 1) and Channel
they will alter the state of the LANIC, thus
disrupting Network Service.

Address offset (test 9) ran and encountered a
failure. This most likely indicates the LANIC is
unable to address memory correctly while
reading.
Extended address (test 10) encountered a failure.
This most likely indicates the LANIC is unable
to address memory correctly while reading.

The FIFO write (test 11), F jP conflict (test 12)
and Date Code (test 15) tests are not allowed to
run because each of these tests write into system
memory. If an addressing failure exists then the
write may alter memory not allocated to the
diagnostic.

LAN Node Diagnostic
4-9

Running LANDIAG
To run LANDIAG, type the following at the MPE prompt:

run landiag.pub.sys
LANDIAG will ask you for the LDEY number of your LANIC card. If it is a
valid LDEY, LANDIAG will be ready to use. LANDIAG uses the ">" character
to prompt you for input.
Figure 4-1 illustrates a typical user running LANDIAG.

:run landiag
LAN Node Diagnostic, version A.55.27.000 HP1984 (c).
Please enter ldev number of LANIC to be tested.
36

Enter 'H' for help.
>
Figure 4-1. Running LANDIAG

Figure 4-2 is a high level state diagram for the use of LANDIAG. It
demonstrates the path that a user will follow to execute the diagnostic. Further
details will be provided later in this chapter.

LAN Node Diagnostic
4-10

MPE Entry State
- NS3000N active
- no line printer equate
:NSCONTROL STOP
:NETCONTROL STOP

C)1

MPE Entry State
- NS3000N nonactive
- no line printer file equate

MPE Entry State
- NS3000N nonactive
- line printer file equate established
;FILE LP;DEV=LP
:RUN LANDIAG.PUB.SVS

:RUN LANDIAG.PUB.SVS

/1\

\11
_------:L.--------,~
Program Initiation State
"".........
;..------'

:FCOPY FROM = DIAGLIST ; TO = *LP
or :FCOPV FROM = DIAG# ; TO = *LP

- Ust file is initialized
- Program emits title message
- Prompts for LANIC LDEV and
tests for valid device
program
ready

> EXIT

Program Input Sta te ( > )
...........
/

Select Options
default: Loop = 1

l)

Printon
Test = ALL

L-.....---------/'r\l
>GO
( resumes
execution)

> LOOP (n

I

> NOPRINT. PRINT
> TEST ( n I ALL )
> HELP

Control-V

\1/

Program Execution State
- Executes program per default or TEST options
- Emits messages per PRINT option

1

Tests done

L-

NO - - LOOP ON ? -

~

to first
selected
test

YES

Figure 4-2. Program State Diagram

LAN Node Diagnostic
4-11

User Interface
The diagnostic is an interactive program. The user determines the type of
operation, performs a diagnosis, and is permitted to change the operating
condition and test again, as many times as desired. The diagnostic is divided into
three sections:
- Initialization, in which setup takes place, transparently to the user, except
for being asked LDEV number and getting initialization error messages if
they occur.
- Command entry, in which the user specifies the type of diagnostic
operation.
- Test execution, in which the hardware is inspected, faults are detected, and
the results are output.
After test execution is complete, or the user aborts test execution with a
the program will return to the command entry mode. Subsequently,
the user may change the type of operation and test again, or exit from
LANDIAG to MPE.
[CONTROLIY,

The LAN Node Diagnostic can be used in either of two ways, Active mode or
Standard mode:
- Standard mode: a predetermined sequence of tests is run; the user is not

able to specify which tests to run. (Refer to the GO command, later in this
chapter.)
- Active mode: tests are user specified. (Refer to the TEST command, later in
this chapter.)

Activity Indicator
LANDIAG time requirements for test completion depend on the test(s) run. If
tests are repeated (LOOP command), completion time is extended. To provide
status of test progress, the diagnostic provides activity indicators. Before any test
is begun, the number of the test is displayed on the screen. As each step within
the test is begun, an asterisk (*) is displayed on the screen.
If the system hangs, the user can determine which test and step was running just
prior to the hang. This information may prove usefuJ for identifying and
correcting a problem. After the last step in a test is completed, the message

end of
is displayed.

LAN Node Diagnostic
4-12

pa~s

How to Get Output From LANDIAG
While all prompts, echoes, activity indicators and other useful information will
be sent to the user's screen, a summary of diagnostic results will be directed to
both the screen and the file known as DIAGLIST, if the diagnostic was invoked
with: RUN LANDIAG.
The file can also be named DIAG# where #=LDEV# if the diagnostic is invoked
with:
:RUN LANDIAG;parm=LDEV~
For example, if : RUN LANDIAG; parm=36 is used, diagnostic results are directed
to a file DIAG36.
The user will not be able to use a file equation to associate the diagnostic output
file with another file or device. In order to obtain hard copy of the diagnostic
activity, the user should direct the diagnostic output to a printer after the process
is complete.
The command sequence is as follows:
:FILE LP;DEV=LP
:FCOPY FROM =DIAGLIST;TO=*LP
:FCOPY FROM =DIAG#;TO=*LP

,or

NOTE
DIAGLIST (or DIAG#) is reinitialized every time LANDIAG is invoked, hence
results from a previous diagnostic pass are lost. The user may save the results by
either printing a hardcopy as shown above or renaming the previous diagnostic
list file as follows:
:RENAME DIAGLIST, filename
:RENAME DIAG#, filename

,or

Where #=LDEV number of LANIC being tested.

LAN Node Diagnostic
4-13

Station (Link-Level) Addresses
Each node on the LAN is identified at the link level by a unique station address.
The station address is a 12 digit hexadecimal number stored on the LANIC card
in ROM. It is normally labeled on the card, and documented on the network
map. (Do not confuse the station address with the IP address of the node. The
IP address is an upper layer address for connecting processes.)
For some tests, specifically the -Remote Node Test, the station address of various
nodes may be needed. For HP 3000 systems, the station address can be obtained
through LANDIAG. However, you must be on the system to get the station
address of the LANIC card installed in that system.
If tests 1, 2, and 13 of LANDIAG have been successfully run, the station address
will be displayed at the bottom of the screen resulting from the HELP command.
The HELP command is described later in this chapter.

LAN Node Diagnostic

4-14

Command Set
Seven commands are available to the user: EXIT, GO, HELP, LOOP, PRINT,
NOPRI NT and TEST. Single letter abbreviations wiJ) be accepted and both upper
case and lower case letters will be understood. There is another input the user
can give, the [CONTROL)Y, which returns control from test execution back to the
command entry mode.

NOTE
During the execution of most LANDIAG tests, response to a [CONTROLlY may be
delayed until after execution of a test step.
Due to programmatic protection, aborting a Remote Node Test (LANDIAG Test
17) may require multiple [CONTROL)Y entries (i.e., you may be required to enter
[CONTROLiY several times). The test will normally abort with [CONTROLIY after it has
received a loopback response frame, or after ithas timed out waiting for a
response frame.

During the command entry phase of the program, the user may issue instructions
to control the operation of the diagnostic. Whenever a ">" prompt is provided, a
command is expected. If the command set is used incorrectly, a error message to
help the user will be given. Refer to the "Error Messages" section at the end of
this chapter.
The diagnostic is designed with default settings for all parameters except the
LDEV to be tested. In most cases, use of commands other than TEST, GO, and
EXI T will be rarely needed.
The tests are designed to be executed in sequence. Each time LANDIAG is run,
test dependencies are enforced to ensure system integrity. See Table 4-3.

LAN Node Diagnostic
4-15

Table 4-3. Test Dependencies

Test must be run and pass

before this testes) is allowed to execute.

Roll Call (Test 1)

ALL other tests

Channel ID (Test 2)

Tests 3 through 17

Address offset (Test 9)

FIFO write (Test 11)
F/P conflict (Test 12)
Date code (Test 15)
Remote Node (Test 17)

Extended address (Test 10)

FIFO write (Test 11)
F/P conflict (Test 12)
Date code (Test 15)
Remote Node (Test 17)

In Table 4-3, the last two dependencies ensure that system memory "reads" work
properly before system memory "writes" are allowed.
The Hood Loopback test and Remote Node test must be individually specified
with the TEST command. They are special tests used for identifying FRU's and
network faults. They should only be run after all previous tests have executed
successfully. The individual commands are explained in greater detail on the
following pages.

LAN Node Diagnostic
4-16

Diagnostic Dialogue Example
An example of running and using LANDlAG is provided below.

:NSCONTROL STOP
:NETCONTROL STOP
:RUN LANDIAG.PUB.SYS

« shuts down network services »
« shuts down lower level NS »
« invoke LAN Node Diagnostic »

LAN Node Diagnostic (version A.55.27.000) HP1984 (c)
Please enter LDEV number of LANIC to be tested.
36
« LDEV may be found in I/O
configuration table »
Type 'H' for HELP

> TEST 9
> GO
9

«
«

indicates only test 9 is to be run »
initiates last test entered »

«

test dependencies invoked »

Tests 1 & 2 must pass for this test to execute.

«
«

> TEST
> GO
1

*

End of Pass

2

*

End of Pass

*****

End of Pass

run test 2 only »
initiates last test entered »

« test 2 completed successfully »
«
«

> TEST 9
> GO
9

« test 1 completed successfully »
«
«

> TEST 2
> GO

run test 1 only »
initiates last test entered »

run test 9 only »
initiates last test entered »

« test 9 completed successfully »

(Continued on the next page)

LAN Node Diagnostic
4-17

« run test 5 only»
« sets the test loop parameter to 3 »
« ·initiates last test entered »

> TEST 5
> LOOP 3
> GO
5 **** End of Pass

« test 5 pass

5 ****

« test 5 pass

Error in test 5, step
End of Pass

4 (Interrupt Test)

completed successfully »
2 detected an error »

MON, APR 6, 1987, 10:17 AM

« test 5 pass 3 completed successfully »

5 **** End of Pass

« user input an invalid command »

> run

Bad input - Try again or enter "Help "

« initiates last test entered »

> GO
5 ****

End of Pass

« test 5 pass

completed successfully »

5 ****

End of Pass

« test 5 pass

2 completed successfully »

5 ****

End of Pass

« test 5 pass 3 completed successfully »
« stop execution of LAN Node Diagnostic »

> EXIT

End of LAN Node Diagnostic.
~

...

LAN Node Diagnostic
4-18

Command Entry Abbreviations
Command entries may be abbreviated. The first character of the command is
normally sufficient, followed by the appropriate parameters. For example, the
following command entries are equivalent:

HELP = H
LOOP
L
LOOP 5 = LOOP5
TEST 1 = TEST1
GO = G

=
=

L5

=

L5

T 1 = T1

Fault Messages
If a fault is detected during a test, an error message is returned to both the user's
terminal and the output file. This message takes the following form:

Error in test , step  «testname»
Where,



- Identifies the number of the test that was
running when the fault was detected.



- The number of the step in the test where the
fault was detected.

 - Identifies the name of the test that was running
when the fault was detected.
The message will also contain the date and time.

LAN Node Diagnostic
4-19

HELP Command
HELP provides a list of commands and tests. The HELP screen is shown here:

Node Diagnostic Commands and Tests
terminates this program and returns to MPE.
initiates execution of the test set.
displays this screen.
HELP
LOOP [N] chooses the number of times the test set is executed.
(Default = 1)
NOPRINT suppresses terminal and file output.
resumes generation of output. (Default)
PRINT
TEST < n / ALL> selects a single test .or tests 1-15.
(Default = ALL)
EXIT

GO

type < control - y > to discontinue testing.
#7 CR SR Loop Test
#8 Bus Conflict Test
Initialization Test #9 Address Offset Test
#10 Extended Address Test
Self Test
#11 Fifo Write Test
Interrupt Test
Soft Reset Test
#12 F/P Conflict Test

#1 Roll Call Test

#2 Channel 10 Test

#3
#4
#5
#6

#13
#14
#15
#16
#17

LAN Coprocessor Test
MAU Loopback Test
Date Code Test
Hood Loopback Test
Remote Node Test

Local station address is given in help menu after test 13 is run.

In a given LANDIAG session, the HELP command can be used to identify the
local station (link-level) address stored on the LANIC card, but only after Test
#13 is run. The local station address will be displayed at the bottom of the HELP
screen as follows:

Local station address is nnnn nnnn nnnn nnnn
where

LAN Node Diagnostic
4-20

n is a hexadecimal digit.

EXIT Command
EX I T terminates the diagnostic and returns to MPE.

Execution of the EX I T command is the only way to return control to MPE.
When the user stops the test execution with the [CONTROLlY character, the
diagnostic will not terminate. It merely returns to the command mode.

GO Command
GO begins or continues execution of tests. The tests that are performed when the
GO command is entered are determined by the test selected with the latest
TEST  command. The number of times the test is looped is set by the
LOOP command. If neither command was given, GO will cause the initiation of
one pass of the complete test set (Test 1-15).

Once testing has begun, testing may be stopped before completion of the selected
test (with the specified number of loops) by entering a [CONTROLlY character. This
will return the user to the command entry mode, identified by the> prompt.
After a [CONTROL)Y, a subsequent GO command will simply restart the same test.

LOOP Command
The LOOP command establishes the number of times that a test, specified by the
TEST command, is continuously run. Using this command, a test may be run a
number of times without repeated command entry.
If LOOP is not specified during a LANDIAG session, its value defaults to "1". The
number of times a test is repeated is determined by the parameter N following
the LOOP command. This is illustrated as follows:

> LOOP 5
> LOOP 1

> LOOP

N
N
N

5, the test will be run 5 times.
1, the test will be run once.
= 0, the test will be run continuously.
=
=

The LOOP command is used primarily to find intermittent problems, or to study
the link hardware with an oscilloscope. It is recommended that oscilloscope
loops be performed with "print off" (see the NOPRINT command).
Once testing has begun, the only way to terminate execution before N iterations
of the test is with [CONTROL)Y. This will return the user to the command entry
mode, identified by the ">" prompt.

LAN Node Diagnostic

4-21

NOPRINT Command
NOPRI NT suppresses terminal and file output. It is useful during loop tests with
an oscilloscope since it eliminates overhead associated with such output and
increases the speed by which an iteration is made.

PRINT Command
PRIN T resumes generation of terminal and file output, which is the default state.
The filename being logged to depends on the way in which you invoked the
diagnostic.
If PARM is used in theLANDIAG run string to specifythe card's LDEV, then the
log file name is DI AG#, where #=PARM=L DEV. This is illustrated below:
: RUN LANDIAG. PUB. SYS; PARM=LDEV

If PARM is not used, then the log file is DIAGLIST.
Note that DIAG# or DIAGLIST will be in the group.account from which you
invoked LANDIAG.

TEST NIALL Command
The TEST command is used to select the LANDIAG test to be run. The desired
test is specified by a parameter, N, in the command run string. Note the
following:
TEST
TEST 0
TEST 1
TEST 2

N = default (see "N = ALL", below)
N = default (see "N = ALL", below)
N = I, Test I is run.
N = 2, Test 2 is run.

TEST
TEST
TEST
TEST

N = 15, Test 15
N = ALL, Tests
N = 16, Test 16
N= 17, Test 17

15
ALL

16
17

is run.
1 through 15 are run.
is run (coax LANs only).
is run.

There are a couple of noteworthy points:
• If N is "0" or not specified, the default is ALL.
• Tests 16 or 17 can be executed only with the commands TEST 16 or
TEST 17, respectively, followed by the GO command.

LAN Node Diagnostic
4-22

Test Descriptions
LANDIAG provides a number of Tests, each consisting of multiple Steps.
A Test is a sequence of interactions between the host system and and the LANIC
card designed to verify the correctness of data and control paths.
A Step is a single interaction between the host and the LANIC intended to gather
some information about a particular data or control path, contributing to the
evaluation of that path.
Each test will begin with a reset of the LANIC card, to bring it into a known
state. In some of the tests, the next step will be to download a small piece of
Z80 software. This code will prepare the LANIC cardto perform subsequent
steps.
This section describes each LANDIAG test.

LAN Node Diagnostic
4-23

Test #1. Roll Call Test
The roll call test verifies that the channel specified by the user, via the LDEV,
responds to the IMB\SIMB roll call instruction. The diagnostic requires the user
to run this test prior to other tests since a nonresponding channel will cause a
system halt. If this test fails, all subsequent tests are not allowed to fun.
Possible causes of failure include:
- The system I/O table does not match the hardware configuration.
- The I/O configuration table has been corrupted.
- The LANIC card is faulty.
If Test #1 fails, you should acquire a listing of the system I/O configuration (may
use SYSDUMP $NULL ,$STDLIST) and verify that the mapping of LDEVs,
DRT#s, and thumbwheel settings of all I/O channels are consistent. For help,
refer to HP 3000 Computer Systems, MPE V System Operation and Resource
Management Reference M anllal (32033-90005).
Correct any discrepancies found between the hardware channel settings and the
system I/O configuration table. If there are no discrepancies, try replacing the
LANIC card.

NOTE
The success of this test does not prove that there is a LANIC card at the correct
channel number, only that there is a channel present. The next test verifies that
the channel is a LANIC card.

LAN Node Diagnostic
4-24

Test #2. Channel 10 Test
Test #2 verifies that the channel specified by the user, via the LDEV, is indeed a
LANIC card. First, a check is made on the results of the Roll Call test (Test #1).
If a responding channel is not present at the specified LDEV, Test #2 is not
allowed to run. Next, register 14 (configuration register) of the specified channel
is read and compared to a special ID code assigned to the LANIC card. If the
channel responds· with the correct ID code, the test passes.
If Test #2 test fails, a flag is set that prevents all subsequent tests from running.
This prevents the possibility of system halts.
Possible causes of test failure include:
- The channel specified by the LDEV is not a LANIC card. This assumes
that the system has been configured correctly.
- There are non-LANIC card channels with the same channel number as the
LANIC card being tested (i.e., thumbwheel switches set to the same
number).
- The system I/O table does not match the hardware configuration.
- The I/O configuration table has been corrupted.
- The LANIC card is faulty.
If Test #2 fails, you should acquire a liStitlg of the system I/O configuration (may
use: SYSDUMP $nu 11, $std 1ist). Verify that the mapping of LDEVs and
DRT#s and thumbwheel settings of all I/O channels are correct.
If there are discrepancies between the hardware channel settings and the system
I/O configuration table, correct them. If there are no discrepancies, try replacing
the LANIC card.

LAN Node Diagnostic
4-25

Test #3. Initialization Test
Test #3 determines if the LANIC responds to the IMB/SIMB INIT instruction by
halting the 280 and clearing the CRFULL bit (an indication that an internal reset
pulseis generated). There are 3 steps to this test:
Step 1:

The Z80 is launched into a self test to ensure that it is not initially in a
halted state. A write to the control register is then performed to set
the CRFULL bit. Subsequently, the CRFULL bit is checked; if it is not
set, Step 1 fails.
A Step 1 failure most likely indicates a LANIC failure. Replace the
LANIC.

Step 2:

An INIT is issued to the LANIC card. This should halt the Z80 and
clear the CRFULL bit. Subsequently, the CRFULL bit is checked; if it
is not cleared, Step 2 fails.
A failure in Step 2 most likely indicates a LANIC card failure. Replace
the LANIC.

Step 3:

This step verifies that the 280 was actually halted by the INIT in Step
2. Another write to the Control register is performed to set the
CRFULL bit. The diagnostic then pauses for 5 seconds; if the Z80 was
not halted by Step 2, the CRFULL bit will be cleared. Subsequently,
the CRFULL bit is checked; if it has been cleared, Step 3 fails.
A failure in step 3 most likely indicates a LANIC card fault. Replace
the LANIC card.

LAN Node Diagnostic
4-26

Test #4. Self Test
Test #4 programmatically initiates the LANIC card selftest program contained in
ROM on the card. The self test is a comprehensive verification of the LANIC
main module, that is, all circuitry not associated with the IMB/SIMB interface.
Refer to Chapter 5 for selftest LEOs and failure codes.
There are three steps to this test:
Step 1:

This step issues an INIT to the LANIC card, then writes to the control
register the code indicating that a full self test is to be fun. Then it
reads the status of the CRFULL bit, which should be set to
acknowledge the control register write. If the CRFULL bit is not set,
this step fails.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

This step consists of the following sequence.
a. A write to the selftest control register is made. This starts the Z80
and clears the interrupt mask.
b. The interrupt mask is set.
c. The diagnostic waits up to 10 seconds for a system interrupt 0,
which indicates selftest completion.
d. The selftest result register is read. The least significant bit (LSB) of
this register is set only while the self test is running.
If the system int~rrupt is not detected, Step 2 fails.
A Step 2 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 3:

If the LSB of the selftest register is cleared, then self test has
completed. The contents of the selftest restIlt register are checked to
see if self test completed without detecting a failure. If a failure was
detected, Step 3 fails and the result is reported on the screen.
A Step 3 failure does not necessarily mean a LANIC card fault.

LAN Node Diagnostic
4-27

If a selftest error is detected, it is displayed as an encoded decimal number and
referred to as a "step number". For example, in Figure 4-3, "step number 46"
implies the decimal error code is 46.

> TEST 4
> GO
4

***

Error in test 4, step 3 (Self Test)
Self test step number is 46
End of Pass

MaN, APR 6, 1987, 11:00 AM

Figure 4-3. Selftest Error Message Example

For a coaxial cable connection, selftest steps number 36 (MAU power failure) or
number 46 (externalloopback failure) may result from a fault external to the
LANIC card. To isolate such faults, refer to Test 16 (Hood Loopback Test).
For a StarLAN connection, MAUs are not used. Therefore, a step number 36
code should not be returned unless there is a card fault. If error code 36 occurs,
replace the LANIC card.

NOTE
For selftest step number 46 to pass, the LANIC card must be properly connected
to the network, or to an appropriate loopback hood. Before detailed
troubleshooting, you should check the cables and their connections.

If any other selftest error codes (i.e., step numbers) are returned, replace the
LANIC card.

LAN Node Diagnostic
4-28

Test #5. Interrupt Test
Test #5 verifies that the LANIC system interrupt mask is functioning properly
and that the LANIC can issue a system interrupt 1. Note that system interrupt 0
was checked in test 4. These are the only two system interrupts that the LANIC
ca rd can issue.
This test is comprised of 4 steps:
Step 1:

The diagnostic issues an IMB/SIMB INIT instruction to place the
LANIC card into a known state. It then writes the skip selftest code to
the control register and starts the Z80 by writing to the selftest control
register. This should put the LANIC in the kernel state. Finally, the
diagnostic reads the osn register. If the device number field of the
OBn register is not equal to 1 then Step 1 fails.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

The diagnostic then checks the IRQ bit of the OBn register. Since the
LANIC interrupt mask is cleared, the IRQ bit should be unasserted. If
the IRQ bit is asserted then Step 2 fails, otherwise it passes.
A Step 2 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 3:

The diagnostic issues a NOP kernel command to the LANIC card
which should cause it to respond with a system interrupt 1. However,
since the LANIC interrupt mask is still not set, the interrupt should be
internally pending. The diagnostic reads the OBn register and verifies
that the IRQ bit is still unasserted. If the IRQ bit is asserted, Step 3
fails.
A Step 3 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 4:

The diagnostic does the following:
a. Checks whether any interrupts have been detected since the test
began,
b. sets the LANIC interrupt mask, and then
c. checks to see if the pending interrupt from step 3 is detected.
If either an interrupt was detected before the mask was set, or no
interrupt is detected after the mask is set, Step 4 fails.
A Step 4 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

LAN Node Diagnostic
4-29

Test #6. Soft Reset Test
Test #6 verifies that a host write to the soft reset register does indeed cause a
soft re;et in the LANIC. card. There are 3 steps to this test:
Step I:

The diagnostic performs the following:
a. Issues an INIT to the LANIC card,
b. Writes to the control register the code indicating that a full self
test is to be run,
c. Checks the status of the CRFULL bit, which should be set,
acknowledging the control register write.
If the CRFULL bit is not set, Step I fails.
A Step I failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

The diagnostic performs the following:
a. Writes to the selftest control register. This causes a hard reset
which starts the Z80.
b. Sets the interrupt mask which was cleared by the hard reset.
c. Reads the selftest register and verifies that bit 0 is set, indicating
the self test is running.
If bit 0 of the selftest register is not set, Step 2 fails.
A Step 2 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 3:

The diagnostic performs the following:
a. Writes OIH to the soft reset register. This should cause a soft reset,
which aborts the self test and starts the KERNEL firmware.
b. Writes a NOP to the control register, which causes the KERNEL to
respond with a system interrupt 1.
If a system interrupt I is not detected, Step 3 fails.
A Step 3 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

LAN Node Diagnostic
4-30

Test #7. CR-SR Loop Test
This test verifies that writes to the LANIC card control register (CR), and reads
from the status register (SR), are processed correctly by the LANIC card.
The LANIC CR-SR kernel command is used for this test. This command causes
the last word read from the control register to be written back out to the status
register.
There are 6 steps to this test:
Step 1:

Reset the LANIC card and issue the CR-SR loop kernel command with
the test pattern == AAAAH. If an error is detected while issuing the
kernel command, Step 1 fails.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

After the kernel command completes, a system interrupt 1 should be
issued indicating to the host that the command has completed. If a
system interrupt 1 is not detected then Step 2 fails.
A Step 2 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 3:

The diagnostic then reads the status register. This register should
contain the alternating 1's and O's test pattern (AAAAH) written to the
control register in Step 1. If the statns register does not contain
AAAAH step 3 fails.
A Step 3 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 4:

The diagnostic again issues the CR-SR loopback kernel command, this
time with 5555H as the test pattern. If an error is detected while
issuing the command, Step 4 fails.
A Step 4 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

LAN Node Diagnostic
4-31

Step 5:

After the kernel command completes, a system interrupt I should be
issued indicating to the host that the command has completed. If a
system interrupt I is not detected, Step 5 fails.
A Step 5 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 6:

The diagnostic reads the status register. It should contain the test
pattern 5555H, the last word written to the control register in the
CR-SR kernel command. If the status register does not contain 5555H,
Step 6 fails.
A Step 6 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

LAN Node Diagnostic
4-32

Test #8. Bus Conflict Test
Test #8 verifies that simultaneous requests made on the LANIC card backplane
interface from both the LANIC card main module and the IMB/SIMB are
properly arbitrated. The interlock kernel command is used to invoke write
alternating AAAAH and 5555H test patterns to the status register. Concurrently,
the host is reading the status register. If there is an arbitration error, one or more
status register reads by the host will be other than the test patterns described
above.
There are 2 steps to this test:
Step 1:

The LANIC is first reset, which should place it in the kernel state,
ready to accept commands from the host. The interlock command is
then issued and the return code checked. If an error was detected
while issuing the kernel command, Step 1 fails.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

The diagnostic reads the status register 500 times with interrupts
disa bled. The interrupts are disabled to ensure that the 500 status
register reads take place while the LANIC is executing the interlock
kernel command. The diagnostic then verifies that each of the status
register reads was either AAAAH or 5555H. If an invalid test pattern
is detected, Step 2 fails.
A Step 2 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

LAN Node Diagnostic
4-33

Test #9. Address Offset Test
Test #9 verifies the operation of the LANIC card IMB/SIMB address drivers.
The download kernel command is an integra) part of the test. It is used to
download a block 'of words from host memory. The success of the transaction is
checked by comparing its computed checksum to that computed by the
diagnostic. Note that 'Data Bus' failures can also cause a bad checksum;
however, these failures should be exposed by Test #4 (Self Test) or Test #7
(CR-SR Loopback), depending on the location of the fault.
There are 5 steps to this test:
Step 1:

The diagnostic first places the LANIC in the kernel state, ready to
accept interactive commands from the host. It then enables the
LANIC master handshake circuitry, thus allowing the LANIC to read
form host memory. This step fails if either the LANIC card was not
successfully placed in the kernel state, or the master handshake could
not be enabled.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

A block of words from bank 1 in host memory is read aQd a checksum
computed by the diagnostic. The LANIC card is then instructed to
read the same block of words via the download kernel command.
Finally, the diagnostic once again reads the same block of words from
host memory. If the first and second read by the diagnostic are not the
same, the sequence of three host memory reads -- by the diagnostic and
LANIC card -- is repeated. This is done up to 3 additional times, after
which the diagnostic aborts Step 2 and reports a failure.
A failure of Step 2 does NOT indicate a LANIC card fault. It
indicates that the diagnostic was unable to use the predefined test
block for this test -- either it changed too often or a memory failure
exists. Try to rerun this test when the system is under less load.

Step 3:

After it issued the download kernel command in Step 2, the diagnostic:
- waited for a system interrupt I from the LANIC card indicating
that the download had completed, and
- read the status register which contained the completion code from
the kernel indicating whether the checksums matched.
If either a system interrupt I was not detected, or the completion code
did not indicate a successful download, Step 3 fails.
The successful completion of Step 2 and a failure in Step 3 most likely
indicates a LANIC card fault. Replace the LANIC card.

LAN Node Diagnostic
4-34

Step 4:

The diagnostic reads a block of words in bank 1 of memory, however,
the bank offset is chosen so that the address lines are the complement
of those used in Step 2. The diagnostic then instructs the LANIC card
to read the same block of words and compare the the checksum to
that computed by the diagnostic. The diagnostic reads the test block
again and compares it to the first read. If the first and second read do
not match, the sequence of three host memory reads between the
diagnostic and LANIC is repeated. This is done up to 3 additional
times, after which the diagnostic aborts Step 4 and reports a failure.
A failure in Step 4 does NOT indicate a failure of the LANIC. It
indicates that the diagnostic was unable to use the predefined test
block for this test because either it changed too often, or a memory
failure exists. Try to rerun this test when the system is under less load.

Step 5:

After it issued the download kernel command in Step 4, the diagnostic:
- waited for a system interrupt 1 from the LANIC card indicating
that it had completed the download, and
- read the status register which contained the completion code from
the kernel indicating whether the checksums matched.
If either a system interrupt I was not detected, or the completion code
did not indicate a successful download, Step 5 fails.
If Step 4 completes successfully while Step 5 fails, a LANIC card fault
is likely. Replace the LANIC card.

LAN Node Diagnostic
4-35

Test #10. Extended Address Test
Test #10 verifies the operation of the LANIC card extended address bus drivers.
Like the Address Offset test (Test #9), the diagnostic uses the download kernel
command to download a block of words from host memory, and checks the
success of the transaction. The diagnostic starts with bank 1 and proceeds bank
by bank until the next bank to be tested exceeds the maximum bank of
configured memory.
Note that this test depends on the amount of memory configured in the host.
Only those extended address bits required to access this memory are tested.
Also, note that failures of the 'Data Bus' or 'Lower (Offset) Address Bus' on the
LANIC card could cause this test to fail. However, such failures should be
exposed in previous tests (e.g., Self Test, CR-SR Loopback and/or Address Offset
tests).
Step I:

The diagnostic performs the following:
- prepares the LANIC card for this test by placing it in the kernel
state, and
- enables its master handshake.
If the diagnostic detects an error either while placing the LANIC card
in the kernel state, or attempting to enable master handshake, Step 1
fails.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

The following steps are systematically repeated (i.e., first Step X, then Step Y) for
each memory bank tested:
Step X:

The diagnostic reads a block of words from bank N (where N = 1,2,4,
8, ..., until the next bank to be tested exceeds the maximum bank of
configured memory) in host memory and computes the checksum. The
LANIC card is then instructed to read the same block of words via the
download kernel command. Finally, the diagnostic again reads the
same block of words. If the first and second reads by the diagnostic
are not the same, the entire sequence is repeated. This is done up to 3
additional times, after which the diagnostic aborts the test on this bank
and reports a failure in this step.
A failure in Step "X" does NOT indicate a failure of the LANIC. It
indicates that the diagnostic was unable to use a specific bank of host
memory for this test beca use either it changed too often, or a memory
failure exists. Try to rerun this test when the system is under less load.

LAN Node Diagnostic
4-36

Step Y:

After it issued the download kernel command in the previous step
(Step X), the diagnostic:
- waited for a system interrupt I from the LANIC card indicating
that the transaction completed, and
- read the status register that contained the completion code from
the kernel indicating whether the checksums matched.
If either a system interrupt 1 was not detected, or the completion code
did not indicate a successful download, this step fails.
A failure in Step Y (presuming Step X passed) most likely indicates a
LANIC card fault. Replace the LANIC card.

LAN Node Diagnostic
4-37

Test #11. FIFO Write Test
Test #11 exercises the LANIC card master handshake logic not tested by. the
Address Offset test (Test #9). Firmware is downloaded to the LANIC card that
will write a predefined test frame into the extra data segment allocated to the
diagnostic. The diagnostic then compares its copy of the test frame to that
written by the LANICcard. An error is reported if the frames are not the same.
The test includes 5 steps:
Step 1:

The LANIC card is reset and the master handshake is enabled, thereby
preparing it for the download process. If either the reset or handshake
enable were not successful then Step 1 fails.
A Step I failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

The diagnostic moves the firmware to the LANIC card via the
download kernel command. Then, it waits for a system interrupt 1
that indicates the download has completed.
After the interrupt is detected, or a time out occurs, the diagnostic
reads the download completion code from the status register. If either
the interrupt is not detected, or an unsuccessful completion code is
encountered, Step 2 fails.
A failure of Step 2 most likely indicates a LANIC card fault. Replace
the LANIC card.
If a failure was detected while attempting to download the firmware,
the diagnostic will not continue the test, that is, Steps 3 thru 5 are not
performed.

Step 3:

If the download was successful, the LANIC is instructed to start
execution of the code. The diagnostic then waits for a system
interrupt 1 that indicates firmware code completion. If the interrupt is
not detected within the time allotted, Step 3 fails.
A failure of Step 3 most likely indicates a LANIC card fault. Replace
the LANIC card.

LA N Node Diagnostic

4-38

Step 4:

Before issuing the system interrupt 1, the firmware writes a completion
code into the status register. The diagnostic then reads the completion
code from the status register.
If the completion code does not indicate the test completed without
error, Step 4 fails.
A failure of Step 4 most likely indicates a LANIC card fault. Replace
the LANIC card.

Step 5:

In this step, the diagnostic compares its copy of the downloaded test
frame with that returned by the LANIC card. If the frames do not
match, Step 5 fails.
A failure of Step 5 most likely indicates a LANIC card fault. Replace
the LANIC card.

LAN Node Diagnostic
4-39

Test #12. F/P Conflict Test
Test #12 verifies that the LANIC card is able to properly arbitrate master
handshake requests between its "first-in-first-out" (FIFO) buffers and processor.
The following sequence is performed:
- The FIFOs are loaded with 16 words of a test frame and are held off from
writing to system memory until instructed by the host.
- After the FIFOs are full, the Z80 processor notifies the host that the
LANIC is ready to start the test.
- Upon receiving the acknowledgment, the Z80 will simultaneously enable
the FIFO writes to system memory and write its own test frames to
another section of system memory.
- The diagnostic will pause to allow the processor and FIFOs time to
complete their writes to host memory.
- The Z80 processor checks the FIFO out/in status before its first and second
write to host memory and after its last write.
- The Z80 processor then reports the results in the status register.
There are 5 steps to this test:
Step 1:

The LANIC card is reset and its master handshakes enabled in
preparation for the download process. If an error was detected during
this initialization Step 1 fails.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

In this step, the following occurs:
- The diagnostic erases any previous data from the extra data
segment to prevent interpretation of erroneous data.
- The diagnostic dow.nloads firmware to the LANIC card, then waits
for a system interrupt 1 (download completion) or a time out.
- After the interrupt or time out, the diagnostic reads the status
register. The register should contain the completion code of the
download process.

LAN Node Diagnostic
4-40

If either the interrupt was not detected, or the completion code
indicates an unsuccessful transaction, Step 2 fails.
A failure in Step 2 most likely indicates a LANIC card fault. Replace
the LANIC card.
The following steps are executed only if Step 2 passes:
Step 3:

In this step, the following occurs:
- The diagnostic issues the start code kernel command to start
execution of the downloaded firmware.
- The diagnostic waits for a system interrupt 1, indicating the
firmware has finished setting up the test.
- The status register is read for the appropriate initialization
completion code provided by the firmware.
If the interrupt is not detected after a predefined interval, or the
completion code indicates the LANIC did not set up correctly, Step 3
fails.
A failure of Step 3 most likely indicates a LANIC card fault.
Replace the LANIC card.

Step 4:

If a system interrupt 1 is correctly received in Step 3, the diagnostic
writes the GO command to the control register. This causes the FIFO
and Z80 processor to write to host memory in a round robin fashion.
Subsequently, the diagnostic waits for a system interrupt 1, indicating
the LANIC card has completed the test.
After an interrupt or time out, the diagnostic reads the completion
code in the status register. If either the interrupt was not detected, or
the completion code does not indicate successful completion of the
firmware, Step 4 fails.
A Step 4 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 5:

Finally, the diagnostic moves the test frames (written by the FIFOs and
Z80 processor) from the extra data segment to the stack, and checks
their contents. If they contain errors, Step 5 fails.
~

A Step 5 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

LAN Node Diagnostic
4-41

Test #13. Coprocessor Test
Test #13 verifies that the LAN coprocessor (INTEL 82586) can detect/report a
CRC error and correctly process multicast frames received by it.
There are 5 steps to this test:
Step 1:

In this step, the following occurs:
- The diagnostic resets the LANIC card and downloads firmware.
- The diagnostic waits for a system interrupt 1, indicating the
download has completed, or a time out.
- After the interrupt or timing out, the diagnostic reads the status
register for the proper completion code of the download process.
If the interrupt was not detected, or the completion code indicates an
unsuccessful transaction, Step 1 fails.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

If the download was successful,
- the diagnostic initiates the downloaded code and waits for a system
interrupt 1 from the LANIC card.
- It reads the status register and checks for the proper status code

that indicates the firmware is ready to continue testing.
If a system interrupt I is not detected, or the code in the status register
indicates an error, Step 2 fails.
A failure in Step 2 most likely indicates a LANIC card fault. Replace
the LANIC card.

LAN Node Diagnostic
4-42

The following steps are executed only if Steps I and 2 pass.
Step 3:

Coprocessor Initialization. In this step, the following occurs:
- The diagnostic instructs the firmware to continue the test.
- The diagnostic waits for another system interrupt 1, which
indicates that the firmware has completed the test and has placed
the first result in the status register.
- On detecting the interrupt, or if a time out occurs, the diagnostic
reads the first test result from the status register.
If the interrupt is not detected, or the first result reported (Coprocessor
initialization step) has failed, Step 3 fails.
A failure in Step 3 most likely indicates a LANIC card fault. Replace
the LANIC card.

Step 4:

Multicast Test. As the firmware test code continues, the diagnostic
reads the status register again, which should now contain the results of
the multicast test. If test results indicate a failure, then Step 4 fails.
A failure in Step 4 most likely indicates a LANIC card fault. Replace
the LANIC card.

Step 5:

CRC Test. As the firmware test code continues, the diagnostic reads
the status register again, which should now contain the result of the
CRe test. If test results indicate a failure, Step 5 fails.
A failure in Step 5 most likely indicates a LANIC card fault. Replace
the LANIC card.

LAN Node Diagnostic
4-43

Test #14. MAU Loopback Test
Test #14 verifies operation of the connection to the LAN. For coaxial cable
connections, this includes the LANIC card, AUI cabling, and MAU/Tap or
ThinMAU/BNC-Tee, as appropriate. For HP StarLAN connections, this includes
the LANIC card and twisted-pair cable.

NOTE
Despite its name, Test #14 is used for testing connections to an HP StarLAN as
well as a coaxial cable LAN.
For Test #14 to pass, the LANIC cards must be properly connected to a
functional coaxial cable or StarLAN Hub, or to appropriate loopback hoods.

Externalloopback test firmware is downloaded to the LANIC card. This code
will sequentially transmit and receive (i.e., loopback) 8 test frames to and from
the LAN medium. Results are reported to the diagnostic in the status register.
This includes:

LA N Node Diagnostic
4-44

1.

Whether MAU/ThinMAU power could be turned on (for the StarLAN
card, the power line is tested even though there is no MAU/ThinMAU
attached).

2.

The number of attempts required to turn the MAU/ThinMAU power on
(tested for StarLAN cards even though there is no MAU/ThinMAU
attached).

3.

The number of frames successfully looped back.

4.

The number of heartbeats detected from the MAU/ThinMAU (not tested
on StarLAN cards).

5.

The number of collisions detected.

6.

The number of times a transmit was aborted due to excessive collisions.

7.

Whether the MAU/ThinMAU jabbed too soon (not tested for StarLAN
cards).

8.

Whether the MAU did not jab at all (not tested for StarLAN cards).

9.

Whether the LANIC could reset the MAU from a jabber condition (not
tested for StarLAN cards).

10. Whether the SQE disable function on the LANIC is operational (not tested
for StarLAN cards).
There are 9 steps to this test:
Step 1:

The diagnostic resets the LANIC card and downloads the external
loopback test firmware. Subsequently, the diagnostic waits for a
system interrupt I indicating the download has completed. A read of
the status register is made -- it should contain the download completion
code.
If the interrupt is not detected within a predefined time interval, or the
completion code indicates an unsuccessful transaction, Step I fails.

A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.
Step 2:

If the download completed successfully, the diagnostic initiates
execution of the downloaded firmware. The diagnostic waits for a
system interrupt I, indicating the LANIC card has completed
initialization and is ready to execute the tests. After receiving the
interrupt, or if a time out occurs, the diagnostic reads the status register
for a ready code returned by the firmware.
If the system interrupt 1 was not detected, or the status register does
not contain the appropriate ready code, Step 2 fails.
A Step 2 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

LAN Node Diagnostic
4-45

The following steps are executed only if Steps 1 and 2 passed.
Step 3:

MAU/ThinMAU power line tests. If the download and start code
(Steps 1 and 2) were correctly returned, the LANIC card is instructed
to turn the MAU power supply line on. The diagnostic then waits for
a system interrupt 1, indicating completion of this task. Subsequently,
the diagnostic reads the status register for results returned by the
firmware.
The results indicate the status of two internal signals, MAUPWR and
VI2SENSE, specifically, whether or not MAU/ThinMAU power is on
and how many attempts were required to turn the power on.
If the task completion interrupt was not detected, or either of the
internal signals indicate lack of MA U/ThinMAU power, Step 3 fails.
For coaxial cable LAN connections, a Step 3 failure can result from a
LANIC card, AUI cable, or MAU/ThinMAU fault. Refer to Test #16
(Hood Loopback Test) to find the faulty unit(s).
For StarLAN connections, the LANIC card most likely contains a fault
and should be replaced. Test #16 does not operate for StarLAN
connections.

Step 4:

External loopback tests. The diagnostic instructs the LANIC card to
loop 8 test frames to the LAN and back. The firmware transmits and
receives the test frames.
During these loopback tests, the firmware monitors the number of
collisions, aborts due excessive collisions, heartbeats received from the
MAU/ThinMAU (if applicable), and frames received without error.
The diagnostic waits for the LANIC card to issue a system interrupt 1
indicating that it has completed the external loopback tests.
Subsequently, the status register is read for results returned by the
firmware. If the interrupt is not received, or the number of frames
received without error is less than four, Step 4 fails.
For a coaxial cable LAN, a Step 4 failure may indicate a faulty LANIC
card, AUI cable, MAU/Tap or ThinMAU/BNC-tee (as appropriate), or
the coax cable. Using a loopback hood and this test, or Test #16, the
fault may be isolated.
For StarLAN, a Step 4 failure may indicate a faulty card, StarLAN
cable, or StarLAN Hub. Using a StarLAN loopback connector and this
test, the fault can be isolated. See Figure 4-4.
For either type of LAN, a Step 4 failure may indicate excessive traffic
on the LAN.

LAN Node Diagnostic
4-46

NOTE
For HP StarLAN connections, the following steps in Test #14 are not performed.

Step 5:

For coaxial cable LANs, the diagnostic checks the number of times a
heartbeat was detected in Step 4. A heartbeat should be detected after
each successful transmission. If a predefined number of heartbeat
errors occur, Step 5 fails.
A Step 5 failure most likely indicates a MAU /ThinMAU fault.
However, it can also occur for AUI cable or LANIC card faults. Use a
loopback hood with this test, or Test #16 (Hood Loopback Test), to
help identify the failure.

Step 6:

For coaxial cable LAN connections, the diagnostic instructs the LANIC
to run the jabber test. The jabber test involves transmitting frames that
exceed allowable frame lengths, and monitoring the Control In (CI)
signal pair for MAU/ThinMAU jabber operation.
The SQE disable function of the LANIC card is also tested at this time.
Finally, the firmware resets the MAU /ThinMAU by cycling the power.
The CI signal pair is checked to verify that jabber operation in the
MAU/ThinMAU has been reset.
While the firmware executes this test, the diagnostic is waiting for a
system interrupt 1, indicating that the test has completed. After it
detects the interrupt, or a time out occurs, the status register is read for
test results returned by the firmware.
If the interrupt was not detected within a predefined time interval, or
a ja bber condition was detected too early in the test process, Step 6
fails.
A Step 6 failure suggests a LANIC card or MAU/ThinMAU fault. A
loopback hood with this test, or Test #16 (Hood Loopback Test) should
be used to isolate the faulty unit.

LAN Node Diagnostic
4-47

Step 7:

From Step 6, the diagnostic checks whether the jabber condition was
detected in the jabber test. If the test completion system interrupt was
not detected, or the ja bber condition did not occur, Step 7 fails.
A Step 7 failure most likely indicates a MAU/ThinMAU fault.
However, LANIC card or AUI cable faults are possible. Use a
loopback hood with this test, or Test #16 (Hood Loopback Test), to
isolate the faulty unit.

Step 8:

The diagnostic checks whether the jabber condition could be reset by
cycling the power to the MAU/ThinMAU. If the jabber condition was
never detected then the results from this test are invalid.
If the test completion system interrupt I was not detected, or the
jabber condition appears to be active after MAU/ThinMAU power
cycling, Step 8 fails.
A Step 8 failure suggests that LANIC card or MAU/ThinMAU are
faulty. Use a loopback hood with this test, or Test #16 (Hood
Loopback Test) to help isolate the fault.

Step 9:

The diagnostic checks whether the SQE disable function is operating
properly. If either the system interrupt I was not detected in Step 6, or
the firmware reports a failure in the SQE disable test, Step 9 fails.
A Step 9 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

LAN Node Diagnostic
4-48

StarLAN Connection Fault Isolation Procedures
Using a StarLAN loopback connector (see Figure 4-4), part number 5061-4977,
Test 14 is used to isolate faulty FRUs on StarLAN connections. The procedure
is illustrated in Figure 4-5, and described below.

l~]
,,::::==::::1'>-- MALE CONNECTORS ~ FEMALE ADAPTER
I-....-JF===e::::::=:::::=:::J==g]
[J

t€J=1
1
2

I ------1
...--+----+-1... - - "--+I---....f--

3

I
2

3

4

.-

5
6

.-------------

-------

----------

--.

4

7

• -

- - - - - - - - - - - - - - - - - - • - -.

7

8

.---------------------~-.

8

-------~-. 5

6

LOOPBACK
WIRING
(Dashed Lines
Indicate Permitted
Connections)

Figure 4-4. StarLAN Loopback Connector

Test A. With the connection set up as in Figure 4-5, "Test A Configuration", Test
14 is run. If Test 14 fails, the LANIC card, StarLAN cable, or StarLAN Hub may
be faulty. Proceed to Test B.
Test B. Disconnect the StarLAN cable from the LANIC, and replace it with the
StarLAN loopback connector, as shown in the "Test B Configuration". Test 14 is
again run. If Test 14 fails, the LANIC card is faulty and should be replaced. If
it passes, the LANIC card is good, so proceed to Test C.
Test C. If Test 14 passes the "Test B Configuration", try the configuration in the
"Test C Configuration". The StarLAN cable is reconnected to the LANIC card.
At the Hub end, the StarLAN cable is disconnected from the Hub, and attached
to the StarLAN loopback connector. Then, Test 14 is repeated.

LAN Node Diagnostic
4-49

NOTE
Because the maximum distance between a StarLAN node and Hub is 250 metres,
Test C results are reliable for cables up to 125 metres only. Where StarLAN
cables exceed this distance, use of the loopback connector in the "Test C
Configuration" is not supported. Instead, use of a known good Hub substituted
for the existing Hub may be required.

If Test 14 passes, the LANIC card and cable are good, and a fault exists on the
Hub or some other part of the StarLAN. Refer to the H P StarLAN Diagnostics
and Troubleshooting Manual lor pes (50906-90060) for isolating StarLAN
hardware faults.

Test 14 Loopback Test
Series 37 I
MICR03000
System

StarLAN
Loopback
Connector

HUB

.0

--e

Test A
Configura tion

0
StarLAN Cable

Test B
Configuration

.0

~'"

Loopback connector to test

Result

Possible Fault

Test A

Bad LANIC
Bad Cable

Failed

Test B
Failed

Configura tion

0

IICI

Loopback connector to test

0-

0

Test B

Bad Cable

Passed

Bad Hub or Port

Test C

Bad Cable

Failed

-------Test C
Passed

LANIC card and cable

Figure 4-5. StarLAN Connection Fault Isolation Using Test ] 4

LAN Node Diagnostic
4-50

Bad LANIC

--------

LANIC card only

Test C

Bad Hub or Port

Bad Hub or Port

Test #15. Date Code Test
Test #15 verifies that the LANIC card and ROM firmware are current. Because
this test depends on a properly operating LANIC card, it is performed after Tests
1 through 14.
This test uses the dump hardware kernel command to read the ROM and LANIC
card date codes. It consists of 3 steps:
Step 1:

The diagnostic first resets the LANIC card and enables the LANIC
card master handshake. It then issues the dump hardware kernel
command and waits for a system interrupt 1, which indicates that the
dump is complete. The diagnostic finally moves the dump from the
extra data segment to the stack.
If any of the following occurs, Step 1 fails:
- the interrupt was not detected.
- the kernel command was not issued successfully, or
- the data move from the data segment to the stack failed.
A Step 1 failure most likely indicates a LANIC card fault. Replace the
LANIC card.

Step 2:

The diagnostic checks the ROM date code against the expected date
code hardcoded in the diagnostic. If the ROM date code is earlier in
the time then the date code expected, Step 2 fails.
A Step 2 failure indicates that the LANIC is either an outdated card, or
the dump failed. In either case replace the LANIC card.

Step 3:

The diagnostic checks the board date code against expected date code
hardcoded in the diagnostic. If the board date code is earlier in time
then the expected date code, Step 3 fails.
A Step 3 failure indicates that the LANIC is either an outdated card, or
the dump failed. In either case replace the LANIC card.

LAN Node Diagnostic
4-51

Test #16. Hood Loopback Test

NOTE
Test #16 is not executed with HP StarLAN connections. Attempts to do so will
result in the following message:

Test 16 is not valid for HP30265A StarLANIC

Test #16 is used on coaxial cable LAN connections, and helps identify the faulty
units associated with an external loopback (Test #14) failure. This section
contains the following information:
• Required Hardware
• Functional Description
• Using Test 16

Required Hardware
For isolating faults with Test 16, the following hardware is required:
Part
Number

LA N Node Diagnostic
4-52

Description

30241 A
28641 A

Known good MAU or ThinMAU, for ThickLAN or
ThinLAN connections, respectively.

92257B
92227Q

Terminated loopback hoods for ThickLAN or ThinLAN
connections, respectively. See Figures 4-6 and 4-7.

30241-60009

Known good AUI cable for connecting the
MA U/loopback hood.

30241-60002

For Series 39/40/42/52, known good Internal Cable,
LANIC card to Junction panel. (Used to distinguish
LANIC card faults from internal cable faults.)

30241-60003

For Series 44/48/58/6X/70, known good Internal Cable,
LANIC card to Junction panel. (Used to distinguish
LANIC card faults from internal cable faults.)

Figure 4-6. Loopback Hood on MAU

~

ThinLAN Loopback Hood

Install
Loopback Hood
ThinMAU

Figure 4-7. Loopback Hood on ThinMAU

LAN Node Diagnostic
4-53

Functional Description
Test 16 is an interactive test with its own user interface. After Test 16 is
specified, and the GO command is issued, the following menu is displayed,
followed by the Test 16 prompt:

1.
2.
3.
4.

Test
Test
Test
Hood

HOOD lOOPBACK TEST
LANIC interface and AUI (25 ohm termination)
for collision detection (50 ohm termination)
coax interface (MAU on coaxial cable)
Test done, return to main menu
Please select test format by number

TEST 16 >

A test or action is selected by entering its identification number at the Test 16
prompt. For example,

TEST 16 > 1
will select the "LANIC interface and AUI (25 ohm termination)" test.
The tests listed above are described briefly in the following paragraphs.

LAN Node Diagnostic
4-54

1. Test LANIC interface and AUI (25 ohm termination)

This test is useful for evaluating the LANIC card to AU] cable
individual sections of AUI cable.

interface~

and

To perform this test~ a a known good MAU/ThinMAU with terminated
loopback hood is substituted for the installed MAU/ThinMAU. Fifty (50) ohm
terminators must be installed on each end of the loopback hood, resulting in an
effective resistance of 25 ohms to the MAU/ThinMAU.
This test is comprised of the following subtests:
a.
b.
c.

LANIC power switch (supplies power to the MAU/ThinMAU)
Test frame loopback from MAU/ThinMAU
MAU/ThinMAU jabber detection

A typical output from this test is as follows:

MAU POWER SWITCH STATUS
The MAUPON signal is TRUE, expecting TRUE
The MAUPS (v12_Sense) signal is TRUE, expecting TRUE
The number of power-on tries is 1, expecting fewer than five.

The
The
The
The

number
number
number
number

FRAME LOOPBACK STATUS
of correctly received frames is 8, expecting 8
heartbeats detected is 7, expecting 7
collisions detected is 0, expecting
of times the backoff retry limit was exhausted is 0, expecting

°

°

JABBER STATUS
MAU Jabber was detected correctly.

LAN Node Diagnostic
4-55

2. Test for collision detection (50 ohm termination)
This test is useful for verifying the collision detection capability of the node.
To perform this test, a loopback hood is attached to either the installed
MAU/ThinMAU, or a known good MAU/ThinMAU substituted for the installed
unit. One of the 50 ohm terminators on the loopback hood is is removed, leaving
the remaining 50 ohm terminator in series between the coaxial cable conductors.
Collision detection tests include:

1.

Detection of collision signal levels on the coax by the MAU /ThinMAU,

2.

Reporting of collisions by the MAU/ThinMAU via SQE signals on the CI
pair,

3.

Transmission of SQE signals from the MAU /ThinMAU to the LANIC card
via the CI pair,

4.

Detection and recognition of SQE signals by the LANIC. This test is
composed of the following subtests:
a.
b.

LANIC card power switch to MAU/ThinMAU,
Te.st frame loopback from MAU/ThinMAU (note that a collision is
expected for every attempted transmission).

A typical output from this test is as follows:

COLLISION STATUS

The number of correctly received frames is 0, expecting 0
The number of collisions detected is 128, expecting 128
The number of times the backoff retry limit was exhausted is 8, expecting 8

LAN Node Diagnostic
4-56

3. Test coax interface (MAU on coaxial cable)
This test serves as a final evaluation of the node connected to the network coax.
Faults associated with a coaxial cable Tap (for ThickLAN connections), or faults
external to the node (i.e., network faults), can be detected.
This test is comprised of the f oHowing subtests:
a.
b.

LANIC power switch (supplies power to the MAU/ThinMAU).
Test frame loopback from a MAU/ThinMAU connected to the network.

A typical output from this test is as follows:

MAU POWER SWITCH STATUS
The MAUPON signal is TRUE, expecting TRUE
The MAUPS (V12_Sense) signal is TRUE, expecting TRUE
The number of power-on tries is 1, expecting fewer than five.
FRAME LOOPBACK STATUS
The number of correctly received frames is 8, expecting 8
The number of heartbeats detected is 7, expecting 7
The number of collisions detected is 0
(expected value depends on the amount of network traffic)
The number of times the backoff retry limit was exhausted is 0
(expected value depends on the amount of network traffic)

LAN Node Diagnostic
4-57

Using Test 16
In this section, general procedures are provided for using Test 16. Once
understood, these procedures can be adapted to the particular installation.

NOTE
The following discussion presumes that Test 14, loopback test, failed. Be sure to
check the node connections before proceeding with the procedures below.

Procedure 1 - Part A
Attach a terminated Joopback hood to a known goodMAU/ThinMAU. Connect
this assembly to the LANIC card. For MAU connections to Series
39/4X/5X/6X/70systems, a known good AUI cable is required to connect to the
system junction box (internally connected to the internal cable and LANIC card).
Run test 1 of Test 16 ("25 ohm termination" test), to exercise the following
hardware:
a.
b.
c.
d.

LANIC power switch
LANIC AUI driver
LANIC AUI receivers
Internal cable (if applicable)

Test passes. If test frames are successfully looped back, and MAU heartbeats are
detected, the LANIC-to-A VI interface circuitry (including the internal cable, if
applicable) is functioning properly.
Proceed to Part B of Procedure l.

LAN Node· Diagnostic
4-58

Test fails. Check for the following failure conditions:
- The number of correctly received frames is less than 8 « 8), but heartbeats
detected equals 8 (= 8). This suggests a faulty LANIC card Data In (DI)
receiver or a faulty internal cable DI signal pair.
- The number of heartbeats detected < 7, but the correctly received frames =
8. This suggests a faulty LANIC card Control In (CI) receiver, or a faulty
internal cable CI pair.
- The number of heartbeats detected < 7, and the correctly received frames <
8. This suggests the following faults may exist:

*
*
*
*

LANIC card power switch,
LANIC card Data· Out (DO) driver,
Internal cable DO pair,
Internal cable power pair (assuming only a single failure).

- The number of received frames < 8, and number of collisions> O. This
suggests the MAU/ThinMAU and loopback hood are not operating
properly.
Since the MAU/ThinMA U and loopback hood are presumed to be good,
check the loopback hood connection and verify that both 50 ohm
terminators are securely fastened.
- Either the MAUPON signal was false, the MAUPS signal was false, or the
number of power-on retries> 5. This suggests the following faults may
exist:

*
*
*

There is a short between the VP & VC signal lines,
The Z80 is unable to turn the power on,
The Z80 is unable to detect that the power is on.

For the failures listed, replace the LANIC card, internal cable, or both. By
systematically replacing one or the other, the faulty unit can be uniquely
identified.

NOTE
Only authorized service personnel, such as Hewlett-Packard Customer Engineers
(HP CE) are permitted to access and replace hardware internal to HP 3000
computer systems.

LAN Node Diagnostic
4-59

Procedure 1 - Part B
Remove one of the terminators, and run test 2 of Test 16 ("50 ohm
termination" test). This is a crude verification of the collision detection and
indicator circuitry of the node.
Test passes. If the number of times the back off retry limit is exhausted is 8 then
collision detection circuits are functioning properly. Go on to Procedure 2.
Test fails. If the retry limit is exhausted is not 8, then the collision detection
circuits are faulty. Since the MAU/ThinMA U and loopback hood are presumed
good, check their connection. Since they must operate properly for remaining
tests, replace them if there are doubts. If the test continues to fail, systematically
replace the LANIC card (and internal cable, if applicable) until the test passes.

Procedure 2
Procedure 2 primarily addresses ThickLAN connections where multiple A VI
cables may be installed.
Attach the known good MAU and loopback hood to t.he end each section of
AUI cable starting wit.h the one connected directly to the host. For each section
of AUI cable, run test 1 of Test 16 (the "25 ohm terminator" test). Figure 4-8
illustrates this process.

LAN Node Diagnostic
4-60

Known Good MAU
with Terminated
Loopback Hood

r-----..., Internal
Cable

Installed
MAU

~

LANIC

LAN Cable

/

HP3000
Testing AUI Cable Sections
,..--_ _..,& Internal Cable

~

o

LANIC

HP3000

D

LANIC

'--i...-.------..-------+------...

HP3000

Figure 4-8. Testing AUI Cable Sections
Test passes. If the number of correctly received frames = 8, and the number of
heartbeats detected = 7, the test passes. Consequently, the AUI cable under test is
presumed good.

LAN Node Diagnostic
4-61

Test Fails. Check for the following failure conditions:
- The number of correctly received frames is less than 8 « 8), and the
number of heartbeats detected equals 7 (= 7). This suggests a faulty DJ
signal pair of the AUf cable.
- The number of correctly received frames = 8, and the number of
heartbeats detected < 7. This suggests a faulty CI signal pair of the AVI
cable.
- The number of correctly received frames < 8, and numberof heartbeats
detected < 7. This suggests a faulty DO signal pair, or faulty power pair
the AUI cable.
- MAUPON or MAUPS indicates "fa 1se", or the number of power-on
retries> 5. This suggests a short circuit between the VP & VC lines in the
AUI cable under test.
After all AUI cable sections have been tested and presumed good, proceed to
Procedure 3.

Procedure 3
Procedure 3 addresses MAU/ThinMAU or coaxial cable faults.
Attach a loopback connector to the original MAU/ThinMAU. Using this
assembly in place of the known good MAU/ThinMAU, repeat Procedure 1, Parts
A and B, to verify operation of the MAU/ThinMAU.
Test passes. If errors are not indicated, the MAU/ThinMAU is presumed good,
and the fault is presumed to be elsewhere on the LAN cable (e.g., a Tap/BNC tee,
cable or remote node fault). You will need to perform other tests. (see "Other
tests", below).
Test fails. If errors occur, replace theMAU/ThinMAU, reconnect the node to
the network; and run test 3 of Test 16.
Other tests. To test the original Tap, try moving to some other node location (i.e.,
some other Tap) on the same coax, and run test 3 of Test 16. If no errors are
indicated, the original Tap is probably faulty. If errors persist, the original Tap is
probably good, and the fault exists elsewhere (coax cable or other node).
To troubleshoot the coax network, refer to the LAN Link Hardware
Troubleshooting Manual (5955-7681).
After the network fault is corrected, reconnect the local node and verify
operation by running test 3 of Test 16.

LAN Node Diagnostic
4-62

Test #17. Remote Node Test
Test #17 allows the user to send frames to, and receive frames from, remote
nodes on the network. This test assists in diagnosing network faults that lie
outside of the local node hardware. Such faults include:
- topological problems
- faulty hardware in the remote nodes
- network performance degradation
This section contains the following information:

• Requirements
• Functional Description
• Interpreting Test 17

Requirements
Before running Test 17, there are various requirements that must be met.

Local Node Requirements.
- LDEV number. You must know the LDEV number of your LANIC card.
- "AVAILABLE" state. If your LANIC card is not "AVAILABLE", that is,
the line is "DISCONNECTED" or "CLOSED", you must shut down LAN
activity on your computer.
- Good node hardware. Using other diagnostic tests, the operation of the
local node hardware has been verified (i.e., good LANIC card, AUI or
StarLAN cables, MAU/ThinMAU as appropriate).
- LANDIAG tests I - 15. In a given LANDIAG session, the default series of
tests, Tests 1 through 15, must have been run prior to Test 17.

LAN Node Diagnostic
4-63

Remote Node Requirements.
- Station (link-level) address. You must know the 12-digit hexadecimal
station address of the remote LANIC card. The best source for this
information is the Network Map.
The station address of the remote node can be determined through the use
of software on the remote node. Since you cannot use LANDJAG3000/V
to remotely determine this address, you must be at the remote computer to
identify the station address.
If your remote computer is a Hewlett- Packard personal computer (or
equivalent) supported on the network, use the DIAGNET or DIAGLINK
utility to find the station address (refer to the H P StarLAN Diagnostics
and Troubleshooting Manual lor PCs, 50906-90060, or the H P ThinLAN
Diagnostics and Troubleshooting Manual lor PCs, 50909-90060).
If the remote computer is an HP 3000 MPE-V system, use the
LANDIAG3000/V "HELP" command.
- If the remote computer is an HP 3000 MPE-V system, its response to the
RNT requires its driver be active. The driver is active if

LAN Node Diagnostic
4-64

a.

a network service must be running (e.g., NS3000/V) and have control
of the LANIC card, or

b.

a Test 17 must ha ve been initiated but held in a wait state (waiting
for a carriage return that actually starts the transmit/receive process).

Functional Description
Remote Node Test examples are provided below. The examples presume that
Tests 1 through 15 have been completed. Underlined text indicates user input.
Refer to these examples during the discussion that follows.
EXAMPLE. Remote Node Test - Normal Completion

> TEST 17
> GO
Please enter the station address of the destination node.
MARKETING
A twelve digit hexadecimal number is required.
Please enter the station address of the destination node.
0800 0901 1806
Attempting to open LANIC driver ... please wait
Driver now ready to echo link level packets.
is displayed for each test packet looped back successfully.
# is displayed for each failure to loop back a test packet.
Hit return to initiate looping of test frames (1000 max),
use  to stop.
[RETURNJ

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !
!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!
! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!
tty!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!tty!!!!!!!!!!!

.......................... ,
,
.
"""""""""""""""""",."""""""""""""""
! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !

100% of the test frames were acknowledged. (1000/1000)
The average response time was 208 milliseconds.
End of Pass

LAN Node Diagnostic
4-65

EXAMPLE. Remote Node Test - Nonexistent Remote Node

> GO
Please enter the station address of the destination node.
0800 0900 A208
Attempting to open LANIC driver ... please wait
Driver now ready to echo link level packets.
is displayed for each test packet looped back successfully.

# is displayed for each failure to loop back a test packet.
Hit return to initiate looping of test frames (1000 max),
use  to stop.
[RETURN)

##### ICONTRoLiY
0% of the test frames were acknowledged.
End of Pass

LAN Node Diagnostic
4-66

(0/5)

EXAMPLE. Remote Node Test - Degraded Performance or Network Problem

> GO
Please enter the station address of the destination node.
0800 0900 EF08
Attempting to open LANIC driver ... please wait
Driver now ready to echo link level packets.
is displayed for each test packet looped back successfully.

# is displayed for each failure to loop back a test packet.
Hit return to initiate looping of test frames (1000 max),
use  to stop.
[RETURN)

!! !##!!! !#!!!! !#! !###!!!! !#!!!!!! [CONTROLlY

76% of the test frames were acknowledged. (25/33)
The average response time was 201 milliseconds.
End of Pass

General Information
You must. be able to access and exit the test, select the remote node with which
to run the test, and obtain test results. Note the following:
- Driver statistics are not provided; they can be obtained via the SHOWCOM
command from MPE (see Chapter 2).
- All terminal I/O is echoed to the list file, consistent with the rest of the
diagnostic. This presumes the NOPR I NT facility was not activated.
- The LOOP and NOPR I NT facilities are disabled for Test 17.

LAN Node Diagnostic
4-67

Accessing the Remote Node Test
Before accessing the Remote Node Test, ensure all requirements listed in the
"Requirements" section have been met.
To run the RNT, specify Test 17 in a TEST command ( > TEST 17), and
follow it with a GO command. Consistent with the diagnostic, subsequent GO
commands will rerun Test 17 if it was the last test specified.
After the GO command is issued, the user is prompted for the station address of
the remote node. The diagnostic expects a 12-digit hexadecimal number for the
station address (upper or lower case characters are accepted). Leading, trailing
and embedded blanks are permitted, as illustrated by the following examples:
hh hh hh hh hh hh
hhhh hhhh hhhh
hhhhhhhhhhhh

where h is a hexadecimal digit.
If an improper user entry is made, the following message is returned:

A twelve digit hexadecimal number is required.
followed by another prompt for a station address.
If a proper station address is not supplied after four attempts, the diagnostic
issues the following message:

Bad input - I assume you want the main menu.
and returns to the LANDIAG command prompt, ">".
Nonexistent Nodes. If a 12 digit hexadecimal number is entered, no further error
checking is performed. Thus, a proper entry may not be a valid one, that is,
there may be no node on the network with the station address specified. The
diagnostic will still try to send test packets to the address entered. Response
frames will not occur, and the "no-reply" timer will expire, resulting in a failure.

NOTE
Test 17 does not distinguish between remote nodes that are broken, too busy to
respond, isolated by a network fault, or nonexistent.

LAN Node Diagnostic

4-68

Initiating Transmissions. After a proper station address is entered, the LANIC
card driver is activated, and the user is directed to "H it retu rn" to start test
frame transmission and reception.

NOTE
If [RETURN) is not entered, Test 17 is idle. The ability to have Test 17 idle, that is,
to open the driver without actually transmitting test frames, is a necessary
feature for MPE V systems. This is especially true when the system does not
have network services installed and is the remote node in a Remote Node Test.
On an MPE V system, the LANIC driver must be active with buffers ready
before frames can be received or transmitted. If a remote HP 3000 does not
have network services installed, it can run Test 17 to activate the driver yet
remain in an idle state. In this way, it can respond to test frames transmitted by
the local node on which Test 17 is active.

Once [RETURN) is entered, the diagnostic begins sending test frames to the node
specified and checks for proper response frames.
A test frame transmission and associated response frame reception constitutes a
single test cycle. Each test cycle should be identical to all others. Up to 1000
test cycles are attempted unless the diagnostic is interrupted with a [CONTROLjY. If
a [CONTROLjY is issued during an individual test cycle, the diagnostic will allow that
test cycle to complete before exiting the test.

Activity Indicator
During each test .cycle, either a response frame is received, or a timer within the
diagnostic expires. A response frame properly received results in a "! ", while an
improper frame or frame not received results in a "#" character.
The return of these characters allows the user to know that the diagnostic is
running properly.

LAN Node Diagnostic

4-69

Test Results
Test 17 Results. When Test 17 completes. either normally or by a "[CONTROLJY",
the results of the test are summarized. The following is returned:
- Percent of test frames transmitted that were properly acknowledged. The
actual numbers are also indicated in a ratio format.
- Average response time for response frames acknowledged. This is returned
only if there were response frames received.
Driver Statistics. The LANIC card driver collects useful statistics about the
node's use of the network. This includes counts of overruns, underruns, CRC
errors, retransmissions required, etc. LANDIAG does not provide this data to the
user; the data is accessed via the SHOWCOM command with the ERROR parameter
from MPE, as follows:
: SHOWCOM LDE'V; ERROR

Test 17 resets all the statistics. Therefore, for long term intermittent error
analysis, use SHOWCOM prior to Test 17.
During a network fault isolation process, use SHOWCOM after Test 17. The
statistics will describe only those network events that have taken place since the
last time the dri vel' was opened.

NOTE
For more information on SHOWCOM, see Chapter 2.

LAN Node Diagnostic
4-70

Interpreting Test 17
Performance degradation between two nodes can occur from hardware or
software faults, extremely busy nodes, or excessive traffic on the network. Test
17 can indicate degraded performance if either of the following result:
- Less than 95% of properly transmitted test frames result in proper
responses.
- The average response time is more than 550 milliseconds.
On a small network, Test 17 can be run using all remote nodes. On a large
network, however, a sampling of nodes might be more appropriate. When
sampling, select nodes that are both near and far, where distance is measured by
lengths of network cable a frame must traverse. For problems with a specific
node, run the Remote Node Test with that node.
For network fault isolation, you should refer to the appropriate LAN
troubleshooting manual:
- LA N Link Hardware Troubleshooting Manual, 5955-7681, for IEEE 802.3
coaxial cable LANs,
- H P Star LA N Diagnostics and Troubleshooting N! anual lor pes,
50906-90060, for IEEE 802.3 twisted pair HP StarLAN.

However, the results of Test J 7 tests may suggest possible faults that can be
further investigated directly. For example, consider the following:
1.

A series of Remote Node Tests indicates that only one remote node is
faulty.
At the faulty node, determine whether NS3000 is up.
Run LANDIAG on that node. Verify that all node FRUs on that system are
operating properly.
If the remote node is connected to a coaxial cable LAN, check for a
"marginal" MAUjThinMAU. A node containing a marginal
MAUjThinMAU can communicate with nearby nodes, but cannot
communicate with distant nodes. If LANDIAG Test 16 (MAU loopback)
passes while Test 17 (Remote Node) fails for distant nodes only, the
MAU jThinMAU should be replaced.

LAN Node Diagnostic
4-71

2.

Communication is difficult or impossible with every node.
The local node may have a defective FRU. Ensure other tests of
LANDIAG have already been run and passed.
If the local node is connected to a StarLAN, check for a damaged cable or
faulty Hub.
If the local node is connected to a coaxial cable LAN, check for a marginal
MAU/ThinMAU (Test 16 passes, Test 17 fails for distant nodes only).
Check for loose connectors or terminators. Check the coaxial cable for
damage -- it may be severed, crimped or frayed.
If you have one, a Time Domain Reflectometer (TDR) can be used to test
the coax to locate cable faults. Also, useful information can be gained by
measuring the resistance across the contacts at the nearest Tap:
a.

A resistance of about 25 to 30 ohms between the center conductor and
braided shield is acceptable.

b.

A resistance of 50 ohms suggests that one term4nator is absent or that
there is an open circuit somewhere in the coax.

c.

A very large resistance suggests that both terminators are absent.

d.

A very small resistance suggests a short between the shield and the
center conductor.

For resistance faults that are difficult to find, TDR testing is recommended
(see the LA N Link Hardware Trollbleshooang M anllal, 5955-7681, for more
information).
3.

Test 17 results indicate poor response from several adjacent nodes.
For StarLAN connections, the Hub to which the remote nodes are
connected is probably faulty.
For coax connections, the coax may be partially damaged between a node
that successfully responds and one that does not..

LAN Node Diagnostic
4-72

4.

For coax connections, Test 17 results indicate performance degrades steadily
with distance from the local node.
The local node probably has a marginal transmitter. Replace the
MAUjThinMAU at the local node.

5.

All nodes were checked with Test 17, and there was no trouble
communicating with any of them.
If a communication problem exists, it is likely due to
user application.

6.

a software fault

in a

Test 17 could not send out any test frames.
a.

For some reason, LANDIAG may not have been able to open the
driver. (A CS error code will be returned if this is the case.) Check the
following:
- Ensure that NS3000/V is down, and that no other processes are using
the LANIC card.
- Ensure the LANIC card can pass self test, which is a prerequisite for
opening the driver. Selftest failures are commonly due connection
faults between the LANIC card and the LAN.

b.

7.

The diagnostic may have been unable to obtain an extra data segment
to use as a frame buffer. This condition points to a software error.

If Test 17 returns a message indicating that Tests 1, 2, 9 and 10 before Test
17 can be run, perform these LANDIAG tests.

LAN Node Diagnostic
4-73

Error Messages
The following is a list of error messages associated with LANDIAG.

• There is no LANIC present at that location.
The LDEV specified by the user does not respond to a roll call during
initialization.

• The diagnostic needs driver version 8011000 or newer.
Software compatibility must be maintained between the diagnostic and the
driver. However, it may be possible to continue diagnostic testing in spite of this
warning.

• The LANIC could not be allocated.
Network Services (NS) is probably running. NS has already allocated the LANIC
and has not released it.

• An extra data segment could not be created.
This reflects an operating system problem or a system resource shortage.

• That LDEV is not configured as a LANIC.
The LDEV entry in the I/O configuration table is not a LANIC of type 17,
subtype 9.

• OP, DI, or 8M capability needed to run diagnostic.
Have the system manager provide you DJ (diagnostician) capability.

• The output file cou ld not be opened.
Logging of diagnostic results to an output file is disabled, but the diagnostic can
still be run.

LAN Node Diagnostic

4-74

• Diagnostic not designed for HP3000j30 or 33.
Diagnostic test results on Series 30/33 systems may not be valid; LAN services are
not supported on Series 30s or 33s.

• Bad input - Try again or enter "Help".
Command entry error. The HELP message includes a list of all the commands.

• Test's parameter must be 1 to 17 or "ALL

II
•

Parameter error on a TEST command. Note that "ALL II includes Tests 1-15 only.
IIALL" is default if no parameter is provided.

• Exit failed to return all resources.
This reflects an operating system problem.

• Bad input - I assume you want the main menu.
Improper parameter entered in Tests 16 or 17 more than three consecutive times.
Returns to LANDIAG prompt II> II.

• A small integer input was expected.

Please retry.

Test 16 prompt expects input range 1 through 4. Non-numeric character entry
was made.

• Test 1 must run for this test to run.
Test dependency message.

• Test 1 and 2 must pass for this test to run.
Test dependency message.

• Test 1,2,9 and 10 must pass for this test to run.
Test dependency message.

LAN Node Diagnostic
4-75

• A twelve digit hexadecimal number is required.
For Test 17 (Remote Node Test), a proper station address must be entered.
Leading, trailing and embedded blanks are permitted. Upper and lower case
characters are accepted.

• MPE commands are not permitted.
Access to MPE, via a colon (:) followed by an MPE command, during a
LANDlAG session is not permitted.

• This test can't run while NS/3000 is up.
With NS running and in control of the LANIC card, Tests 3 through 17 cannot
be initiated.

• You must have CS capability to run this test.
For Test 17, capability to use the Communication Subsystem is necessary.

• You must specify an individual address.
(The first byte of the address must be even).
For Test 17, a proper station address must be entered. The address cannot be a
group address, that is, its first byte cannot be an odd value.

Test 16 is not valid for the HP30265A StarLANIC.
Test 16 does not apply to systems with HP StarLAN connections. Use Test 14,
instead.

LAN Node Diagnostic
4-76

5

LANIC Self Test

The LANIC card contains a self test routine in ROM. An understanding of this
self test can help you identify whether or not a LANIC card is faulty and
requires replacement. For more information on LANIC card self test, refer to
your LANIC card installation and service manual.

The Scope of LANIC Card Self Test
The LANIC card self test checks the operation of the card and LAN connection
hardware.
The hardware tested depends partially on the type of LAN card connection:
coax cable LAN, or HP StarLAN. The following circuitry is common to both
types of LAN cards:
•
•
•
•
•
•

Microprocessor and associated circuitry
ROM
RAM
Timer [Chip]
FIFOs [Receive Data Buffers]
Protocol Controller [82586 Chip]

Card specific circuitry tested includes:
Coax LAN card
• MAU Power Circuitry
• Analog Driver [8023]
StarLAN card
• StarLAN media interface [7960]
By testing the above circuitry, the operation of the LANIC card's main module
(all circuitry not associated with the SIMB/JMB interface) is verified.
In addition, the self test performs an externalloopback test to verify the
card-to-LAN interface. If the card is not connected to its respective LAN, the
self test will fail this portion of the test.

LANIC Self Test
5-1

Self Test Limitations
The following LANIC card circuitry or functions are not tested by self test.
However, these items can be exercised by using LANDIAG3000/V (see Chapter
4).

•
•
•
•
•
•
•
•

1MB/51MB Interface, including data and address lines
Master and slave handshakes
Interrupt mask flipflop
Z80/82586 to master handshake arbitration
Slave handshake vs Z80 arbitration (system accesses)
Slave handshake vs Z80 arbitration (I/O)
82586 collision recovery, and identification of bad packets
Collision detect or jabber functions

How To Read Selftest LEOs
There are 15 LEDs located on the edge of the LANIC card, eight of which are
accessed by the self test. For location and identification of these LEDs see
Chapter 3.
The eight LEDs used by self test are labeled H to N, and "*". These eight LEDs
are also labeled with mnemonics to indicate their function during normal
operation, as follows:
TX
RX
MN
DL

RO
Q

IT
8T

LANIC Self Test
5-2

ON
ON
ON
ON
ON
ON
ON
ON

when
when
when
when
when
when
when
when

a XMIT command is started
a RCV command is started
in monitor mode
the DOWNLOAD command is started
ROM-resident firmware is in control
microprocessor is waiting for a command
microprocessor is running an idle selftest routine
self test is running

Depending on your particular system, the LED labeled "ST" (the rightmost LED)
will be in one of three states: off, on, or blinking.
ST is OFF (applies to all MPE-V systems).
When the ST is off, self test is not running, and the other seven LEDs perform
their mnemonic functions as noted above.
ST is ON (applies to all MPE-V systems).
However, when 8T is on, the self test is in progress. The self test consists of
many subtests; while the self test is executing, the remaining seven LEDs are
used to display a code indicating which subtest is executing. See Table 5-1.
If the self test completes with no errors, the selftest LED will be on and the
other seven LEDs will be off (a code of all zeros) for five seconds. After five
seconds, the self test LED (ST) will go off, and the remaining LEDs perform
their normal LANIC card functions.
ST is Blinking.
If self test fails on Series 39j4X/5X/6Xj70 systems, the ST LED will blink slowly,
and the code of the subtest that failed will be displayed for about 20 seconds.
The 8T LED does not blink on cards installed on Series 37/MICR03000 systems.
If self test fails at power-on, a hexadecimal failure code is returned to the system
console under the heading "Local Area Network Interface Controller". When
self test is initiated from LANDIAG, failures are reported to LANDIAG (see
Chapter 4).

LANIC Self Test
5-3

How To Use the Self Test
Invoking Self Test
Methods of invoking LANIC card self test depends on the computer type.
LANDIAG. For all MPE-V computer systems, the self test can be invoked by
LANDIAG Test #4. Refer to Chapter 4.
Power-On. For all MPE- V computer systems, self test may be manually initiated
by powering on the computer.
For Series 39j4Xj5Xj6Xj70 systems, theself test will first blink the 8 self test
LEDs on and off slowly for approximately 13 seconds, after which self test will
run.
For Series 37jMICR03000 systems, self test will simply run at power up.
Reset Switch. For Series 39j4Xj5Xj6Xj70 systems, the LANIC card contains a
LANIC Test Reset Switch located on the card edge.

I CAUTION I
Pressing the reset switch performs a hard reset on the LANIC card before the
self test is initiated. Be sure that no network activity is in progress that may be
destroyed by pressing the reset switch.

To use the reset switch to run the self test, perform the following:

LANIC Self Test

5-4

1.

Ensure the LANIC card is not in use. Do a SHOWCOM and look for
LINE UNCONNECTED, or CLOSED.

2.

Open the computer card cage to observe the selftest LEDs. Note the present
LED indications. If the ST LED is on, and the remaining seven LEDs are
displaying a steady pattern, make a written record of which LEDs are lit.
(This information may be needed later if the problem persists beyond the
initial steps of troubleshooting.)

3.

Press the LANIC TEST RESET switch to initiate the self test. See Chapter 3
for location of this switch.

4.

Observe the selftest LEDs. Refer to Table 5-1 to interpret the various LED
patterns displayed.
If the self test completes with no errors, the selftest LED (ST /*) will be on
and the LEOs H through N will be off (a code of zeros) for about five
seconds. Then, the S T/ * LED will go off, and LEOs H through N will
perform their normal operating functions.
If the self test fails, the code of the test that failed will be displayed by
LEOs H through N and the ST/* LED will blink slowly for at least 20
seconds.
Table 5-2 provides codes for unexpected results from the self test. If such
an event occurs, the S T/ * LED will be on (not blinking) and LEOS H
through N will display one of the codes described in Table 5-1.

5.

If the LANIC fails the self test, the failure is most likely in the LANIC card.
However, failures of tests 36 (24 HEX, MAU Power) or 46 (2E HEX,
external loopback) may indicate problems with the AUI, MAU, or coax. If
in doubt, run the self test with a loopback hood, or use LANDlAG

Looping
Looping on the self test is useful for catching intermittent errors.
For all MPE-V systems, selftest looping can be invoked from LANDIAG, using
the following command sequence:

> LOOP 1000
> TEST 4
> GO
For Series 39/4X/5X/6X/70 systems, self test may be manually looped by
continuously depressing the LANIC Test Switch, such as with a clip. The looping
continues until a error is detected. Self test will halt with the failed subtest
number indicated by the selftest LEOs. The ST LED will blink until the switch is
released.

I CAUTION I
If self test is run during normal LAN activity, the LANIC card and network
operations will cease. Recovery may be possible in higher software levels;
however, avoidance of this situation is recommended. Results of self test may be
erroneous.

LANIC Self Test
5-5

Observing Self Test
While the self test is cycling through its "subtests", the LEDs will display which
subtest is running. Refer to the paragraphs under "How to Read Selftest LEDs",
presented earlier in this chapter.

Selftest Failure
With two exceptions, replace the LANIC card jf any subtests within the self test
fail. The exceptions are:
- Test 36 (24 HEX, MAD Power). Applies to coaxial cable LAN connections
only.
- Test 46 (2£ HEX, External loopback). Requires connection to the LAN or
applica ble loopback connector to pass.
Failures of these two tests may require additional troubleshooting using a
loopback hood and/or LANDIAG.

LANIC Self Test
5-6

Table 5-1. Selftest LEDs and Subtest Desc.riptions

CODE
LED
INDICATION
DEC

HEX
NO.

1

1

0 0 0

2

2

0 0 0 0 0 1 0 1 EPROM

Checksum

3

3

o0 o0

Checksum

4

4

0 0 0 0 1 0 0 1 High Byte Latch

5

5

o

0

6

6

o

0 0 0 1 1 0 1 Byte RAM Data

7

7

o0

8

8

o

9

9

0 0 0 1

H I J K L MN*

o0

o0

SUBTEST

0 1 1 Z80

0 1 1 1 Station Address PROM

1 0 1 1 Byte RAM Data

DESCRIPTION
Instruction set

Even addresses
Odd addresses

0 0 1 1 1 1 Byte RAM Address

Incrementing addresses

0 0 1 0 0 0 1 Byte RAM Address

Decrementing addresses

o0

1 1 Word RAM

Address tests

10

A 0 0 0 1 0 1 0 1 Word/Byte Address

Address mapping

11

B 0 0 0 1 0 1 1 1 Z80

Memory reference instructions

12

C 0 0 0 1 1 0 0 1 MDIAG register
SYSCON register

Proper state after reset

13

D 0 0 0 1 1 0 1 1 CTC

Data test

14

E 0

15
16

o0

1 1 1 0 1 CTC

Mode 0 counting

F

0 0 0 1 1 1 1 1 CTC

Mode 2 counting

10

0 0 1 0 0 0 0 1 CTC

Mode 4 counting

._-

17

11

o 0 1 0 0 0 1

18

12

o0

1

o

19

13

o0

1

o0

20

14

o

1

Interrupt PAL

Bit 4 set and cleared

0 1 0 1 Z80 interrupt
1 1 1 Z80 NMI

0 1 0 1 0 0 1 MHSDIS

Non-Maskable Interrupt
DMA Handshake Disabled

LANIC Self Test
5-7

Table 5-1. Selftest LEDs and Subtest Descriptions (continued)

CODE
LED
INDICATION
DEC

HEX
NO.

21

15

0 0 1 0 1 0 1 1 PADDR to BADDR bus

Low 15 bits

22

16

0 0 1 0 1 1 0 1 ZBANKL register

Low Z-80 bank bit

23

17

0 0 1 0 1 1 1 1 ZBANKH register

Eight high Z-80 bank bits

24

18

0 0 1 1 0

o

25

19

o

0 1 1 FIFO Data

BDATA(7)

26

1A

0 0 1 1 0 1 0 1 FIFO Data

BEA(7,8)

27

18

0 0 1 1 0 1 1 1 FIFO Data

BDATA(2:6)

28

1C

0 0 1 1 1 0

29

1D

0 0 1 1 1 0 1 1 FIFO Data

30

1E

0 0 1 1 1 1

31

1F

0 0 1 1 1 1 1 1 FIFO Data

32

20

0 1 0 0 0 0

33

21

0 1 0 0

34

22

0 1 0

35

23

0 1

36

24

0 1 0 0 1 0 0 1 MAU Power

37

25

0 1

o

0 1 0 1 1 R13

CR, CR Full Bit

38

26

0 1

o

0 1 1 0 1 R15

Selftest Result register

H I J K L MN *

LANIC Self Test
5-8

0 1 1

o

o

o

o

SUBTEST

0 1 Prel iminary FIFO

0

0

0

1 FIFO Data

1 FIFO Data

1 FIFO Data

0 1 1 R14

0 1 0 1 OBII register

0 0 1 1 1 COMCON register

DESCRIPTION

INREADY, ADVREADY, OUTREADY

BDATA(0,1,13:15)
BDATA(8:12)
BA ( 11 : 15)
BA(6:10)
BA (1 : 5)
Configuration regis"ter
Value; Channel number not 0
Values from reset
On/Off (AUI/MAU not required)

Table 5-1. Selftest LEDs and Subtest Descriptions (continued)

CODE
LED
INDICATION

HEX
H I J K L MN*

SUBTEST

DESCRIPTION

DEC

NO.

39

27

0 1 0 0 1 1 1 1 82586

40

28

0 1

41

29

0 1 0 1 0

42

2A

0 1 0 1 0 1 0 1 82586

RAM addressing

43

2B

0 1 0 1 0 1 1 1 82586

Diagnose

44

2C

0 1 0 1 1 0 0 1 8023

Loopback

45

2D

0 1 0 1 1 0 1 1 82586

Write to FIFOs

46

2E

0 1 0 1 1 1 0 1 External loopback

Loopback to/from LAN

o

1 0 0 0 1 82586

o

Interrupt
Reset

1 1 PBUS register
addressing

**

where "*" is theST (Self Test) LED

** Note:

Failure of this test is normal if the LANIC card is not connected to the LAN or applicable
loopback connector.

The final test in Ta ble 5-1. the External loopback test, transmits and receives a
1148 byte frame on the LAN or loopback connector. The frame contains the
f oHowing information:
Destination and Source Address fields. These fields each contain the LANIC
card's 6 byte stat.ionaddress st.ored in PROM. The first. 3 byt.es are 00 08 09
hexadec ima 1, and the last 3 byt.es are identifiable by the 6 hexadecimal digits
labeled on the PROM.
Length field. This field consists of 2 bytes indicating the length of the Data field
(in this case, 1134 bytes)

LANIC Self Test

5-9

Data field. The data field contains data to be interpreted by the receiving node.
The first 3 bytes are encoded to indicate the following:
- The frame is an IEEE 802.3 "TEST Command" frame.
- A 'TEST Response frame must be returned by the remote node.
The command and response is handled at the IEEE 802.3 Media Access
Control (I\1AC) layer. For HP 3000 systems, the MAC layer consists of the
LANIC card and driver.
ll

Next, the following 31 bytes of ASCII data are provided:
HP3000 NODE xxxxxxxxxxxx TEST
where xxxxxxxxxxxx is the station address in ASCII.
The remaining data consist of 1100 bytes with a binary incrementing pattern.

Table 5-2. Reporting of UneXIJected Results from Self Test

CODE
LED
INDICATION
DEC

HEX
NO.

122

7A

1 1 1 1

o

1

123

78

1 1 1 1

o

1 1 1

124

7C

1 1 1 1 1 0 0 1

125

7D

1 1 1 1 1

126

7E

1 1 1 1 1 1 0 1

Unexpected Z80 interrupt.

127

7F

1 1 1 1 1 1 1 1

The LANIC was reset, but self test never started,
or LED circuitry failed.

where

"*"

H I J K L MN *

o

o

1

1 1

DESCRIPTION OF FAILURE
The 82586 failed to clear its command word.
Selftest Result register (R15) bit 0 bad.
Z80 stack underflow during self test.
Unexpected Z80 Non-Maskable Interrupt (NMI).

is the ST (Self Test) LED

LANIC Self Test
5-10

Tracing

6

General Information
The CS/3000 Trace Facility is used to provide a record of the line actions, CS
states and events that occur during Network Services operation. When problems
occur during operation, the trace facility provides the means to pinpoint the
problem area. For example, suppose you are experiencing a data integrity
problem between two computers in your network; the information that you send
to one computer isn't the same information that is being received at the other
end. A trace could be enabled to monitor the data in your subsystem to try to
"trap" the data transfer error when it occurs.
The trace facility is invoked by the operator with a : LIN KCONTROL command.
Tracing can be enabled/disabled when OPENing the line, or before or after the
line is opened. Tracing can be invoked for any communication line that
Network Services uses. Once invoked for a particular communications line, the
trace facility continues to record line activity until the user issues a new
: LINKCONTROL command with the TRACE=OFF parameter. The trace facility
keeps track of actions, states and events in the form of trace entries.
The trace entries are grouped into trace records: one trace record for each CS
intrinsic called by Network Services. The trace records are permanently stored in
a system-generated file named LI NKTxxx, or in a user-specified trace file. The
contents of a CSj3000 trace file can be formatted and printed through the use of
a trace dump utility program. There are two link level trace formatting
programs for Network Services: CSDUMP and DSDUMP. CSDUMP does some
formatting and displays all trace file data in a raw form. DSDUMP allows you
to choose a subset of the trace file to be formatted, and will also analyze the
chosen data. In addition, CSDUMP will display all of the bisynchronous line
protocol, while DSDUMP only displays the DS protocol. The trace facility must
be terminated before CSDUMP and DSDUMP can be run. Refer to the
Fundamental Data Communications Handbook for additional information on the
CSTRACE.

Tracing
6-1

Linkcontrol Trace=On
LI NKCONTROl TRACE=ON activates the Trace Facility, that is, link-level tracing is
activated on a communication line's link.

Syntax

LINKCONTROL LinkName;TRACE=ON
mAll] [@Mask] L;NumEntries] [".;WRAP] [".~FileName]

Parameters
LinkName

The configured name in NMMGR of an active data
communications line.

ALL

Generate trace records for all line activity. If you omit
this parameter, only I/O errors are traced.

Mask

An Octal number preceded by a percent sign, %nnn.
The Mask is used to select the type of trace entries
generated. Refer to Table 6-2 for a detailed description
of the Mask parameter protocols.

Combine these values for Mask:
•
•
•
•
•
•
•

Tracing
6-2

%001 = Protocol Send Text (PSTX) entries.
%002 = Protocol Send Control (PSCT) entries.
%004 = Protocol Receive Text (PRTX) entries.
%010 = Protocol Receive Control (PRCT) entries.
%040 = Protocol State Transition (PSTN) entries.
%100 = INP interconnect entries.
%200 = Generate IMF (bisync only) control unit state transition entries.

numen tries

The value of entries is used to derive the size of trace
file record. Trace entries are deposited in a record in a
circular manner. A driver depend ent defaul t of 24 will
be used if the parame ter is omitted. Defaul t = 24.

WRAP

Specifies that if the trace record is full for a givenC S
intrinsic, previous entries are overlayed. Its absence
indicates that succeeding entries will be flushed. This
parame ter does not affect the EOF marker of the file.

filena me

Trace output will be sent to a specified file name which
has been previously built. If a file name is not specified,
the defaul t destination will be LIN KT nnn, where nnn is
the MPE logical number of the CS device. If a trace
file exists, it will be purged and a new trace file will be
created.

Discussion
).
Refer to NS3000/V Netwo rk Manager Reference Manual (32344-90002

Example

:LINKCONTROL NEWYORK;TRACE=ON,ALL,,24,WRAP,NYCOOO
of all line
Link level tracing is activated for the NEWYORK CS device. Records
to file
sent
and
ted
activity and all trace entries, except for PSTN are genera
than 24
more
no
has
record
NYCOOO in the logon group and account. Each trace
entries. Overflow entries overlay prior trace entries.

Tracin g
6-3

Linkcontrol Trace=Off
LI NKCONTROL TRACE=OFF deactivates the Trace Facility.

Syntax

LINKCONTROL

LinkName;TRACE=OFF

Parameters

LinkName

The configured name of an active data communications
line.

Discussion
This command deactivates link level tracing on the specified communications
line. The link must have been started before you can issue this command.
LinkName is NMMGR configuration, not sysdump.

Example
:LINKCONTROL NEWYORK; TRACE=OFF
Link level tracing is deactivated for the NEWYORK LANIC.

Tracing
6-4

Using CSDUMP Formatting Program
:RUN CSDUMP.PUB.SYS[,OCTAL] [,HEXJ[;PARM={!} ]
The trace dump program uses the CSTRACE file as input and produces a
formatted trace listing on the LIST file.
The secondary entry point OCTAL allows you to specify that aU raw data will be
output in octal, otherwise it will be output in hexadecimal. (The entry point HEX,
allowing you to specify hexadecimal for the output, has been retained for
backward compatibility to the time when the default was octal.) If you specify
PARM=O or 1 all entries will be output by time; however, if you specify PARM=2
only CSj3000 intrinsics will be output by time.
Various conditions can cause this program to abort. These are indicated in an
information error message, and in parameter values of the QUI T intrinsic shown
in Table 6-1.

Table 6-1. CSDUMP Error Message Parameter
Parameter
1
2
3
4

5
6

7

Meaning
Illegal dump format request
Open failure on trace file
Open failure on list file
Trace file access error
Open failure on temporary file
Temporary file access error
List file access error

Tracing
6-5

Defining a Trace File for CSDUMP
The program expects a trace file named CSTRACE. If your trace file has a
different name, such as the default file name LINKTnnn, you will need to equate
the trace file name to CS TRAC E. Use the MPE : F I LE command this way:
:FILE CSTRACE=LINKTnnn.PUB.SYS

Defining a CSDUMP Listing File
The formal file designator of the trace listing file for CSDUMP is LIST. The file
may be defined as a CRT terminal, a line printer, or a disc file. To define the list
file, enter an MPE : FI LE command prior to initiating the CSDUMP program.
Some typical examples are:
:FILE LIST;DEV=LP

LP is assumed to be the device class name for one
or more line printers.

:FILE LIST=FILENAME

FI LENAME is assumed to be the name of an old
temporary or permanent disc file.

If a list file does not exist or is not designated by a : FI LE command, and PARM
of the RUN command is not 1, the CSDUMP program employs the user's
session/job output device as the Jist file. If PARM is set to 1, then the dump
program attempts to open the file LIST as an old job or system file. If this fails
because LIS T does not exist, then LIS T is opened as a new file in the system
domain. After the CSDUMP program has run, the contents of this file may be
accessed via a text editor, such as EDIT/3000.

Tracing
6-6

Initiation the CSDUMP Program
After the CSTRACE and LIST files have been defined, enter the following
command:

:RUN CSDUMPoPUSoSYS[,OCTAL] [;PARM={!}J
The trace dump program uses the CSTRACE file as input and produces a
formatted trace listing on the LIST file. The format of the trace listing is
described in the following text. If the secondary entry point OCTAL is specified
when CSDUMP is run, the numeric codes for both control characters and data
will be printed in octal instead of hexadecimal. If you specify PARM=O or 1, all
entries will be output by time; however, if you specify PARM=2 only CS/3000
intrinsics will be output by time.

Formatted CSDUMP Trace Listing
A CSDUMP Trace listing has a specific format. The components of a Trace
listing are a header message; the beginning-of-trace message; the opening Line
Information Display box; a series of trace records, each consisting of a record
header and one or more consecutively numbered entries; an end-of-trace message;
and the closing Line Information Display box. These components are discussed
in detail on the following pages. The examples of the various components are
taken from a trace of a line connected to a LANIC.

Tracing
6-7

CSDUMP Listing Header Message

NOTE
Items under discussion are~b~(;J~g for easy identification.

At the start of the trace listing is a header message (Figure 6-1) that tells the date
and time of day when the listing was printed and the fully-qualified name of the
trace file. The meanings of the two remaining items in the header message are:

Meaning

Item

LAST OPENED ON

This tells you the date and time of day
when the trace was executed.

SYSTEM ID=v.uu.ff

This tells you the version (v), update level

(u u) and the fix level (ff) of the MPE
operating system that was being used when
the trace was performed.

CS TRACE ANALYZER (B.00.23)

MqN••~ • • ··i1:gN•• • ···R.·~.·· • ·1·~§4.~·· • • ·1·1•• ;·.'7~· . ·.t\r1

j"gA~~·.···IT.IL.~··• ··~.$··.··L·INk$6

•.·.g0B•.•.SY$

~A~fl'· • • ·qg~N.~o.·.·qN • • • MqN.;··

APR•• • 1.e., ·.··1.S)e4.~.···.·.1 • 1·.i..4p•• • ~M

ALL ENTRIES DUMPED BY TIME

$~~±~M.··.·~DFQQ

•.•• gO
Figure 6-1. Trace Listing Header

Tracing

6-8

Begin Tracing and Line Information Messages
The BEGI N TRACI NG ... message appears in the listing when the line to be
traced is opened. The message tells you the decimal logical device number of the
line (36 in the example in Figure 6-2). It indicates the line's activities are now
being monitored by the trace facility. It is followed by the Line Information
Display describing the state of the line when tracing started.
~******************************

* BEGIN

TRACING FOR DEVICE 36 *
*******************************

**************************************************
*-L-I-N-E---I-N-F-O-R-M-A-T-I-O-N---D-I-S-P-L-A-Y*
**************************************************
LOGICAL DEV. NUMBER: 36
*
* LINE NUMBER: 3
*$Q~"t~A~;n$
VER: x.55.23 *

*

*
COPTIONS:
AOPTIONS:
*
DOPTIONS:
*
* NETWORK'ID:
* NUMBUF FERS:

*

*

0123456789012345
0000100010000010
0000000100001101
0000010000000000
0000000000000000

*
*
*
*
1@I.JFR§I.4FUt l$OOn(~~m~$) *

:IN$p~qE:DiT120()00Q

oUTsP~gD~

* MISCARRAY:
RECEIVE TIMEOUT:
LOCAL TIMEOUT:
*
CONNECT TIMEOUT:
*
RESPONSE TIMEOUT:
*
LINE BID TIMEOUT:
*
NO. ERROR RETRIES:
*
CLEAR-TO-SEND DELAY:
*
DATA-SET-READY DELAY:
*
TRANSMISSION MODE:
*
MMSTAT TRACE FACILITY:
*

1200000

*

*
*
*
0
*
60
*
0
*
00.0 SECS. *
DISABLED.
*
DUPLEX
*
DISABLED
*
*P.RI.VERNAM~:1~I..ANO
*
* DOWNLOAD FILE: csdlan1.pub.sys
*
* CTRACEINFO:
ENTRIES~24
MA$~~P1111~~~0
*
TYPE OF TRACE~lliU,~PW~~A
*
*
* PHONELIST:
ENTRIES=O
INDEX=O
*
ENTRIES=O
INDEX=O
*
* IDLIST:
IRRECOVERABLE=O
*
* ERRORCODE: RECOVERABLE=O
MSGRECV: 1
*
* MSGSENT: 1
IRRECOVERRORS: 0
*
* RECOVERRORS: 0
**************************************************
20
60
900

SECS.
SECS.
SECS.
HSECS.
SECS.

Figure 6-2. Begin Tracing and Line Information Messages

Tracing
6-9

The opening Line Information Display box contains detailed information on how
the line was opened, how the communications controller was configured
(transmission speeds, timeout values, logical device number, etc.) and trace
parameters selected. In the example in Figure 6-2, we know that:
• the communications controller is an LANIC, because
type) is 17 and DRIVERNAME is IOLANO,

DEV. TYPE (device

• it is a synchronous, switched line (i.e., dial-up), because it is

SUBTYPE 9,

• BUFFSIZE is 1500 WORDS, so the configured line buffer size is 4095,
• I NSPEED and OUTSPEED (transmission speeds) are 1200000 characters per
second.

• MASK is 011111000 (%37; for LANIC ignores the three zeroes on the right),
• ENTRIES=24 is the maximum number of entries in each trace record. (24 is
the default.)
• AL L events will be traced

• Overflow record entries will be discarded

(NOWRAP).

Trace Record and Header Message
The trace listing is organized into a series of trace records, each consisting of a
series of trace entries. Every trace record pertains to a particular request.
A trace record is signified by a header message. The header message identifies
the CS intrinsic call that generated the trace record. The header (see the example
in Figure 6-3) shows the name of the CS intrinsic, where the intrinsic was called
from the program, the calling parameters and a REQUEST I Dthat is the same as
the REQUEST I Dfor the corresponding record entries.

***********************************************************************
~:~·e'A·~
ggQQe$i • ··.mD:::%Q't.4;34Z.{·Y.4$E.f.).
*
~bb~t§§~%riri6~j~~~~.~~
.
CALLER: SEGMENT=PRG %000
*
COPTIONS=%004201 DOPTIONS=%002000 *
STATE: LINE STATE=DISCONNECT
SPEC. STATION #=0
COMP #=0 *
INPUT: IN BUF=%OOOOOO LENGTH=O
RESP. STATION #=0
COMP #=0 *
OUTPUT: TRANSMISSION LOG=O
***********************************************************************

*
*
*
*
*

Figure 6-3. Trace Record Header

Tracing

6-10

Trace Entry Format
All entries in a trace listing contain a prefix consisting of four fields:

1.

An entry number (0 in the example in Figure 6-4).

2.

A "time stamp" in seconds and thousandths of seconds (17.073 in the
example).

3.

An entry-type mnemonic (PCMP in the example). Mnemonics are described
later in this chapter.

4.

A "request ID" that correlates the entry with a particular intrinsic call

(%043136 in Figure 6-4).
The first entry is numbered zero, and successive entries throughout the rest of
this trace record are numbered consecutively in ascending order (1, 2, 3 and so
on). The "time stamp" makes it possible for you to determine the elapsed time
between one trace entry and another. The mnemonic tells you what type of
trace entry you are examining. There are five types of trace entries used in
Network Services. They are summarized in Table 6-2 and described in greater
detail on the following pages. The body of each trace entry tells you the
pertinent information for the particular activity that has happened or is about to
happen.

~~''t-· •..P't-~~~M~

RsqqS§JJPrr%P4?t$p ( ! 465 E)

ERROR CODE=O
LAST RECOVERABLE ERROR CODE= 0
FUNC CODE=7
LAST FATAL ERROR CODE= 0
#MSG SENT=1
#MSG RECV=1
STATE=CONNECTED
# RECOVERABLE ERR=O
# IRRECOVERABLE ERR=O
Figure 6-4. Sample Trace Entry

Tracing
6-11

Missing Entries Message
If MISSING ENTRIES appears in the listing, it means that the record was not
large enough to accommodate all of the trace entries and some entries were lost.
If WRAP was not specified (NOWRAP), then the missing entries were at the end just
before the PCMP entry; otherwise they are missing from the beginning where
they were overlaid by the trace entries that extended past the end of the record.
If the missing entries are crucial:
1.

Purge the trace file.

2.

Invoke trace again, issuing: L I NKCONTROL with
a. a larger numentries value
b. a mask setting that will produce only those trace entries you are really
interested in.

Trace Entry Mnemonics
There are five types of trace entries created by the LANIC driver. There are no
conflicts with NMS tracing. A received frame will generate the following
sequence of entries which are summarized in Table 6-2 and described in greater
detail on the pages following this table.

Tracing
6-12

Table 6-2. Trace Entry Type Mnemonics
Mnemonic

Entry Type

Definition

PRCT

Receive
Control
Sequence

Generated each time a control character sequence is
received from the remote station. The PRCT trace
entry shows (in octal or hexadecimal) the exact
sequence of bytes that was received.

PSCT

Send
Control
Sequence

Generated each time the driver sends a control
character sequence to the remote station. The PSCT
trace entry shows (in octal or hexadecimal) the
exact sequence of bytes that was sent.

PRTX

Receive
Text

Generated each time a message is received from a
remote station. The PRTX trace entry shows (in
octal or hexadecimal) the exact sequence of bytes
received.

PSTX

Send
Text

Generated each time the driver sends a message to
the remote station. The PSTX entry shows (in octal
or hexadecimal) the exact sequence of bytes sent.
There is one PRCT entry for every frame received.

PCMP

User
Request
Completed

Generated each time an I/O operation to the
LANIC occurs. The PCMP trace entry summarizes
the line activity, such as the number of frames sent
and received and the number of errors that have
occurred.

Tracing
6-13

PSCT (PRCT) Trace Entries
The PSCT (PRCT), meaning "Sent (Received) Control", is really the first part of a
text block. The way it works is that there is one PSCT (PRCT) entry and zero or
more PSTX (PRTX), meaning "Sent (Received) Text" entries for every
transmission (received) frame traced. The PSCT (PRCT) entry will contain the
first 32 bytes of sent/received frame, and the remaining bytes will be included in
subsequent PSTX (PRTX) entries.
The first 32 bytes of a sent (received) LANIC frame will a contain a header. An
example is shown in Figure 6-5.

5

59.274

18¢~

......

REQUEST 10=%043622(!4792)

MlnplF'~bP~§m~~~~qg~g.~q~gj

~~~~q~ppg~qq~~~~
F"~(:)W~P:ffi4~~QnP~r4dO

.~~~i~i7j~:~

~~~~rr·1.~

The header data will also be included, unformatted, in the raw data:

o 8.0 0 0 9.0 0 0 8.2

I 0 8.0 0 0 9.0 0 1 8.0 9
BS NUL HT NUL BS ! BS NUL HT NUL CAN HT
1 8.1 8 0 4.2 0 0 7.D 2 F 4.0 D 0 3.8 5 0 8.0 0
CAN CAN EOT
BEL R I CR ETX ENQ BS NUL
o 9.0 0 1 8.0 9 0 0.2 B 1 8.1 8
HT NUL CAN HT NUL + CAN CAN
Figure 6-5. PRCT Trace Entry

Tracing
6-14

.

B/Rs·q

The meaning of the various items are as follows:

MU1

Mode-independent Unnumbered Information (frame type),
other options are XID (exchange Identity), TEST (test), and !xx
(undecodable).

P/F

Poll/Final bit (0 or 1).

OEST

This the destination address.

OSAP

Destination Service Access Point points to the proper protocol
the next higher level.

SRC

This is the source address.

SSAP

Source Service Access Point points from the proper protocol
of the next higher level.

C/R

Command

FLOW 10

This is a unique 48-bit quantity that is not used presently.

= 0 and

Response

= 1.

Note that, in order to facilitate tracing, the format of the 32 byte header does
not coincide with the data actually sent or received on the wire. The LANIC
driver header contains data that allows the dump reader to match the link-level
trace entries with higher level trace entries. The formats are provided on the
next page.

Tracing
6-15

o

destination
address

o

destination
address

o

source
address

6

11

11

12

dest SAP

12

13

source SAP

13

14

flow 10 to
match w/hilevel trace

14

flow 10 to
match w/hilevel trace

6

undefined

12

undefined

13
destination
address

14

15
destination
address

19

19

20

control

21

undefined

20

source
address

21
22
26

source
address

length

27

27

length

28

dest SAP

28

29

source SAP

29

30

control

30

dest SAP

31

undefined

31

source SAP

Figure 6-6. PRCT and PSCT Formats

.Tracing
6-16

flow 10 to
match w/hilevel trace

11

16

31

destination
address

5

5

5
6

byte

byte

byte

PSCT format (1)

PSCT format (0)

PRCT format

The difference in format between the two PSCT entries is whether the user data
started on an even byte [PSCT(O)] or an odd byte [PSCT( I )]. Now, in the case of
PSCT format 0 (even-byte user data), the data field will begin the first byte (i.e.,
byte 0) of the PSTX entry immediately following the PSCT entry. The data
actually transmitted onto the AVI (and therefore onto the coaxial cable) begins
at byte 14, and byte 31 will be discarded before transmission.
For the PSCT format I (odd-byte user data) and the PRCT entries, the first
PSTX/PRTX entries will contain a garbage byte (this will actually contain the
control field). In this case, the data actually transmitted on to the AVI (and
therefore onto the coaxial cable) begins at byte 16, and all bytes will be
transmitted (no bytes are discarded before transmission).

NOTE
It is important to recognize that this byte exists in all received packets and in
some transmit packets if the trace output is to be understood correctly.

PCMP Trace Entries
A PCMP trace entry is generated each time an I/O request is completed. An
example is shown in Figure 6-6.

4

72.717

I¢MB

.

REQUEST ID=%043533(!475B)
P.~~~~.· ·(]QDEE.~

FRNP90PE=QO

~M8~§ENTJ'g

lliA$T.·•• ·~e~qVs.~AS:t.~.
·.. 4~$,.LFAIAL

#MSG·RECYs6

#RECbvERABlliE ERR±:1

..~.·~g(j)g • • ··~qpg::: • ·•.O
P~R~R~QPs== 0
$"~"f:FCONNECIED

#.LIggEC;:~Y:~~Aal.~

E~~::=O

Figure 6-7. PCMP Trace Entry

Tracing
6-17

The meanings of the various items are as follows:

ERROR CODE:

The code of the request's most recent Recoverable
Error or Irrecoverable Error (see the CS trace section
of the Fundamental Data Communications Handbook
for CS error codes).

FUNC CODE:

This describes the nature of the operation whose
completion is being traced by the PCMP trace entry
(refer to table 6-3).

# MSG SENT:

The total number of text messages sent so far for this
connection.

# MSG RECV:

The total number of text messages received so far for
this connection.

STATE:

The line state after the completion of the user
request. In the example it is in the connected state.

# RECOVERABLE ERR:

The total number of Recoverable Errors that have
occurred so far for this connection.

# I RRECOVERABLE ERR: The total number of Irrecoverable Errors that have
occurred so far for this connection. Note the Last
Fatal Error Code is the Last Irrecoverable Error Code.

Table 6-3. Function Codes
FUNC CODE

FUNCTION

0

READ
I

WRITE

5

CONTROL

7

INFO TRANSFER

9

HARD ABORT

66

Tracing
6-18

ABORT 10

End of Trace and Line Information Messages
The END OF TRACE .... message appears in the listing when the trace is turned
off. The message tells you the decimal logical number of the line (36 in the
example in Figure 6-7) and indicates that the line's activities are no longer being
monitored by the trace facility. It is followed by the Line Information Display,
showing the state of the line just before tracing was stopped. Note the counts of
messages sent (1 in our example), messages received (1 in our example), number
of recoverable errors (4 in our example), and number of irrecoverable errors (0
in our example).

********************************
* END OF TRACE FORD~~~(:)E L36 *
********************************
**************************************************
*-l-I-N-E---I-N-F-O-R-M-A-T-I-O-N---D-I-S-P-l-A-Y*
**************************************************
* lINE NUMBER: 4
lOGICAL DEV. NUMBER: 36
*
* DEV. TYPE: 17
SUBTYPE: 3 VER: X.55.23 *
*
0123456789012345
*
COPTIONS: 0000100010000010
*
*
*
AOPTIONS: 0000000100001101
*
*
DOPTIONS: 0000010000000000
*
* NETWORK'ID: 0000000000000000
*
* NUMBUfFERS: 1
BUFFSIZE: 1500 (BYTES) *
* INSPEED: 1200000
OUTSPEED: 1200000
*
* MISCARRAY:
RECEIVE TIMEOUT: 20
SECS. *
*
lOCAL TIMEOUT: 60
SECS. *
*
CONNECT TIMEOUT: 900
SECS. *
*
RESPONSE TIMEOUT: 0
HSECS. *
*
lINE BID TIMEOUT: 60
SECS. *
NO. ERROR RETRIES: 0
*
*
*
CLEAR-TO-SEND DELAY: 00.0 SECS. *
DATA-SET-READY DELAY: DISABLED.
*
*
*
TRANSMISSION MODE: DUPLEX.
*
*
MMSTAT TRACE FACILITY: DISABLED.
*
* DRIVERNAME: IOlANO
*
* DOWNLOAD FILE: csdlan1.pub.sys
*
* CTRACEINFO:
ENTRIES=24
MASK=011111000
*
*
TYPE OF TRACE = ALL, NOWRAP
*
* PHONELIST: ENTRIES=O
INDEX=O
*
* IDlIST:
ENTRIES=O
INDEX=O
*
* ERRORCODE: RECOVERABLE=O IRRECOVERABLE=O *

*

M~~$R~m:~

M$~~RY~:.:1,

*~~G~~e~~~~$:.Q1~~E¢~\l~~~~~$·~U:Q

*
*

**************************************************
END OF JOB.
Figure 6-8. End of Trace and Closing Line Information

Tracing
6-19

CS Error Codes
Table 6-4. Code Meaning of Irrecoverable Errors
Meaning

Code
(Decimal)
0

Tracing
6·20

No error.

13

Text overflow (> 1500 bytes of user data).

52

Invalid operator.

63

No I/O found to abort.

64

Abort ignored because I/O already completed or aborted.

116

Software timeout.

117

Invalid interrupt.

120

Driver internal error.

121

Self test failure.

130

System fail, LANIC.

131

Power failure.

133

TRANSMIT: Transmit incomplete in absence of abnormal
condition.

134

Fatal MAU power error.

135

MA U jabber error occurred.

153

Duplicate address.

160

An internal error detected by MPE.

201

Operation aborted.

223

RECEIVE: buffer space, ran out.

Table 6-5. Code Meaning of Recovered Errors
Code
(Decimal)

Meaning

0

No error.

2

RECEIVE: frame too short, length field error, unintelligible
sequence received.

3

RECEIVE: FCS error (CRC).

16

RECEIVE: overrun (DMA).

17

TRANSMIT: underrun.

121

Self test failure.

134

MAU power, recovered.

135

MAU jabber, recovered.

136

TRANSMIT: CRS on always, SQE not on always, DI not on
always.

137

TRANSMIT: DI never idle, SQE not on always.

138

TRANSMIT: SQE on always, DI not on always.

139

TRANSMIT: SQE on always, DI never idle.

207

Excessive collisions.

Tracing

6-21

7

Software Tools

Memory Dump
When a SYSFAIL or some other severe condition occurs, refer to the M PE V
System Operation and Resource Management Relerence Manual (32033-90005)
for instructions on doinga memory dump.

NSDPAN5/NSDUMPJ
The NSDPAN5 pr~gram produces a formatted listing of main memory, based on
a memory dump taken after a system failure, HALT, or other abnormal
condition.

Obtaining an NSDPAN5 Listing
1.

Immediately after system termination, use the Software Dump Facility
(SDF) to make a main memory dump. This facility gives you the capability
of storing main memory to either a serial disc, cartridge tape, or magnetic
tape. It operates in a stand-alone environment after a system failure has
occurred or a system halt has been performed. The SDF is described in the
M PE V System Operation and Resource Management Relerence Manual
(32033-90005).

2.

When the system has been restarted, enter the command:

:STREAM NSDUMPJ.NET.SYS
This version of NSDPAN5 has several advantages:
• It is customized to show NS data structures.

• It saves virtual memory
• If the problem is in another product, this dump still applies.

Software Tools
7-1

LISTLOG5
LISTLOG5 analyzes files in the MPE system log file. An MPE log file records
events such as session or job initiation and termination, process termination, file
closure, I/O errors, and system shutdown. Refer to the M PE V System Operation
and Resource Management Reference Manual (32033-90005) for more
information on system logging.

NOTE
Refer to NS3000/V Network Manager Reference Manual (32344-90002) for
information on NMDUMP and NMS logging.

System manager (SM) capability is needed to run LISTLOG5.
Log files are named by the following convention, where nnnn is a four-digit
number:

LOGnnnn.PUB.SYS
The formal file designator of the output file is LOGLIST, with the default device
class LP. LOGLIST is opened as new, and closed as a permanent file.

Operation of LISTLOG5
1.

To find out which log files are on the system before you run LISTLOG5,
enter the following:

:LISTF LOG@.PUB.SYS
MPE returns a list of filenames numbers for all of the log files currently on
the system. These are the valid logfile numbers you can choose from when
you run LISTLOG5. For example:

FILENAME
LOG
LOG1970
LOG1976

Software Tools
7-2

LOG1971
LOG1977

LOG1972
LOG1978

LOG1973
LOG1979

LOG1974
LOG1980

LOG1975

2. To run LISTLOG5, type:

:RUN LISTLOG5.PUB.SYS
3. LISTLOG5 identifies itself and asks for the number of the first
log file to print:

LISTLOG5 G.OO.OO (C) HEWLETT-PACKARD CO., 1982
ENTER FIRST AND LAST LOG FILES TO BE ANALYZED
FIRST? 1973
Enter the four-digit numbers from the list of log files. If
you only want to analyze one file, enter it as the first file
number and press [RETURN) in response to the "LAST?"
prompt.
4. You are then prompted for the four-digit number of the last log file to
print. Press [RETURN) to list only the first file.

LAST? 1980
5. LISTLOG5 now displays a numbered list of events for which
histories can be printed:

TYPE NO.

o
1

2
3

4
5
6

7
8
9
10

11
12
13
14
15
16

17
18
19

EVENT
LOG FAILURE
SYSTEM UP
JOB INITIATION
JOB TERMINATION
PROCESS TERMINATION
FILE CLOSE
SYSTEM SHUTDOWN
POWER FAILURE
SPOOLING LOG RECORD
LINE DISCONNECTION
LINE CLOSE
I/O ERRORS
PRIVATE VOLUMES
PRIVATE VOLUMES
TAPE LABELS
CONSOLE LOG RECORDS
PROGRAM FILE EVENT
DCE PROVIDED INFO
MAINTENANCE REQUEST
DIAGNOSTIC CONTROL UNIT

Software Tools
7-3

6.

At the end "of the list of events, you are prompted for input with the
message:

ENTER EVENT NUMBERS SEPARATED BY COMMAS?
A CARRIAGE RETURN ASSUMES ALL EVENTS WILL BE EVALUATED.
Type the event numbers and press [RETURN!. LISTLOG5 creates spool files of
the events that you requested. There are no messages echoed back to your
terminal if your request is successful. If you request is not successful, one of
two messages will be displayed: 1) an error message in the format described
under "ERROR CONDITIONS", or 2) a message indicating that there are no
events for the log file numbers that you requested:

NO DESIRED EVENTS FOUND IN LOGFILE 1980
If events have been found for a log file, it's number will not appear in the
NOD ESIR ED EVE NTS list:

NO DESIRED EVENTS FOUND IN LOGFILE
NO DESIRED EVENTS FOUND IN LOGFILE

1978
1979

Events ha ve been found for all other requested log files, so they do not
appear in this list.
7.

LISTLOG5 then asks:

DO YOU WANT TO PURGE LOG FILES?
If you answer YES, then log files are printed and then purged from the
system. If you answer NO, the files are printed and also retained by the
system. You are now asked if you want to return to the program; Type
YES to continue with LOGLIST5, NO or ~ to terminate:

DO YOU WISH TO RUN AGAIN (Y OR N)?

Software Tools
7-4

~

NMMAINT
The NM Maintenance Utility (NMMAINT) is a utility program supplied with HP
3000 LAN link products as part of the Node Management Services (NMS).
NMMAINT is used to display the individual and overall version numbers for the
software modules of the data communications products that use the Node
Management Services (NMS). These products include the NS3000/V, SNA IMF,
and SNA NRJE Network Services products as well as the SNA and LAN3000/V
Network Link products.
NMS defines one or more subsystems for each of the data, communications
products. For the HP3000 LAN link, there are three subsystems defined: the
Node Management Services, the Network Transport, and the Link Services
subsystems. For NS3000/V, there is one subsystem, the Network Services. Each
software module within a subsystem, such as a program file or SL segment, has
its own version 10 number. If the version, update, and fix levels of these
modules do not match, the subsystem will not work correctly. NMMAINT can
be used to determine if you have an invalid software installation or if the
software modules of a subsystem are mismatched. The information provided by
NMMAINT should be included in any SR submitted. Refer to the NS3000/V
Network Manager Relerence Manual.
To run NMMAINT, issue the command:

:RUN NMMAINT.PUB.SYS
NMMAINT will respond with the following:

32099-11018A.01.00 NM Maintenance Utility
Co. 1985

(C) Hewlett Packard

NMMAINT then lists the version identification numbers for each software
module as well as subsystem information for each subsystem. As shown in the
example below, the NMMAINT utility displays version information for the
subsystems of the products actually installed on your system. The Node
Management Services, Link Services, and Network Transport subsystems are
displayed if the HP 3000 LAN link product is installed. The Network Services
subsystem is displayed if the NS3000/V product is instaIIed. The SNA Transport,
NRJE, and IMF subsystems are displayed if the appropriate HP to IBM data
communications products are instaIIed on your system. The example shows a
system with NS3000/V and the HP 3000 LAN link installed. The subsystems are
always displayed in the same order, which is the Node Management Services
subsystem first, followed by SNA Transport, NRJE,and IMF, then the Network
Transport, Network Services, and Link Services subsystems. The last version ID
number listed is the port software, I PCVERSION. This is not part of the
NetworkIPC user service, nor does it form a subsystem as the previous sets, but
its individual version ID number is displayed by NMMAINT for your
information.

Software Tools
7-5

As described, version ID numbers include version, update, and fix levels as well as
an internal fix level in the format vuuffiii. In the example below, where
NMVE RSOO is listed, its version ID number is AO 100016. The A is the version
level, the next two zeros represent the update level, and the following two zeros
are the fix level. The remaining numbers, 016, show the internal fix level, which
is used only within Hewlett-Packard.

Example
The following example shows the information displayed when you use the
NMMAINT utility. A discussion follows the example.

:RUN NMMAINT.PUB.SYS
32099-11018A.01.00 NM Maintenance Utility

(C) Hewlett Packard Co. 1985

Subsystem version 1D's
Node Management Services 32099-11018 module versions:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
Program fi Ie:
Program file:
Prog ram f i Ie:
Program fi Ie:
V+ form s f i 1e :
Program file:
Program fi Ie:
Catalog file:

NMVERSOO
NMVERS01
NMLOGSLVERS
NMLOGDATAVERS
NMVERS04
NMVERS05
NMVERS06
MCVERS
NMMA1NT.PUB.SYS
NMFILE.PUB.SYS
NMLOGMON.PUB.SYS
NMMGR.PUB.SYS
NMMGRF.PUB.SYS
NMMGRVER.PUB.SYS
NMDUMP.PUB.SYS
NMCAT.PUB.SYS

Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:
Version:

Node Management Services 32099-11018 overall version

(Continued on Next Page)

Software Tools
7-6

A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000
A0100000

= A.01.00

Network Transport Module Versions
Program File
Program File
Program File
Program File
Message File
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure
SL Procedure

Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
Version: AOOOOOOO
** Not Installed

NETCP.NET.SYS
NETSERVE.NET.SYS
SOCKREG.NET.SYS
NETMSG.NET.SYS
NET'SM4'VERS
NET'UI'VERS
NET'SL'VERS
NET'NI'VERS
NET'PROBE'VERS
NET'TCPO'VERS
NET'TCP1'VERS
NET'PXPO'VERS
NET'PXP1'VERS
NET'IP'VERS
NET'IPU'VERS
NET'PD'VERS
SOCKIOVERS
SOCKACCESSVERS
SOCKMISC1VERS
SUBSYS3FMTVERS
SUBSYS5FMTVERS
TRIGVERS

Network Transport Overall Version

A.OO.OO

Network Services HP32344 individual module versions:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
Prog ram f i Ie:
Program file:
Program fi Ie:
Prag ram f i Ie:

ASCXVERS
ASBUFVERS
ASENVVERS
DSUTILVERS
ASRFAVERS
VTSVERS1
VTSVERS2
ASPTOPVERS
ASRPMVERS
SUBSYS6FMTVERS
DSDAD.NET.SYS
DSSERVER.NET.SYS
IOVTERMO.PUB.SYS
NFT.NET.SYS

Vers ion:
Version:
Vers ion:
Version:
Version:
Version:
Vers ion:
Version:
Version:
Version:
Version:
Vers ion:
Version:
Version:

Network Services HP32344 overall subsystem version:

AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
AOOOOOOO
A.OO.OO

Software Tools
7-7

LINK SERVICE Subsystems VERSION
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
SL procedure:
Program file:

module vers ions:

TRAN'VERSO
TRAN'VERS1
BFMICSVERS
BFMVERS
SUBSYS8FMTVERS
LINKVERS
LINKMGR.PUB.SYS

LINK SERVICE Subsystems VERSION
SL Procedure: IPCVERSION

Version:
Version:
Version:
Version:
Version:
Version:
Version:

B0206000
B0206000
B0206000
B0206000
B0206000
B0206000
B0206000

overall vers ion = B.02.06
Version: B0204000

In the previous example, the first group of numbers listed are the version ID
numbers of the modules of the Node Management Services subsystem, part of
the HP 3000 LAN link. Notice that the first five characters of the version for
each module listed in this group are AO 100. This means that all the software
modules in the subsystem match. This must be true for all the modules of a
given subsystem. If a subsystem module is invalid, the following error message is
printed:

Program file: NMMAINT.PUB.SYS
ONE OR MORE SUBSYSTEM MODULES ARE INVALID.

** MODULE ERROR **
(NMERR 105)

This message indicates that the modules of the subsystem are not compatible.
Because the module version ID numbers match, NMMAINT displays the overall
subsystem version number for NMS as A.Ol.OO. The rest of the subsystems are
handled in a similar fashion.
NMMAINT also checks that all the modules that belong with a particular
subsystem are present. If a module is missing, NMMAINT displays the name of
the module with the following error message in place of. the version number.

SL procedure: NMVERS01
** REQ'D MODULE MISSING
ONE OR MORE REQUIRED SUBSYSTEM MODULES ARE MISSING.
(NMERR 104)
If this is a new software installation, check your installation procedure.
If the modules were correct when installed, only unusual circumstances, such as a
reload, a disc problem, or a system failure, result in missing or invalid modules.
Restore a known valid version of such modules. Refer to the N S3000IV
Network Manager Reference Manual.

Software Tools
7-8

The missing module may be optional. For example:

SL Procedure : TRIGVERS

**Not installed

This module, TRIGVERS, is not normally present on the system. It is only
installed by an HP system engineer if needed for troubleshooting.
Question marks in the overall version number indicate that the fix levels of the
individual modules do not match. Remember that the internal fix level,
represented by the last three numbers of the version 10, does not need to match
between modules for the software to be compatible; therefore it may be
disregarded. The fix numbers are requested for Service Requests; it makes a
considerable difference to HP when troubleshooting.
As each subsystem is displayed, NMMAINT checks that all the modules are
present and compatible with each other. However, NMMAINT does not
perform any cross-subsystem version verification. When a system has HP to IBM
products as well as HP to HP products installed, you need to be aware of the
fact that the Node Management Services, the Link Services, and the port
software are used by both types of data communications products. Therefore, it
is important to check that the version numbers of these common subsystems and
port software modules are correct. It is possible for the HP to IBM products to
use previous versions of the common software that are not compatible with the
HP to HP products. For more information, refer to the NS3000/V Network
Manager Relerence Manual.
NMMAINT only displays information on the subsystems for the products that
are installed on your system. Thus, the NMMAINT display from different
systems may vary, depending on which products are installed on the system. In
the example above, the SNA Link, SNA NRJE, and SNA IMF products were not
installed, so the the SNA Transport, NRJE and IMF subsystems were not
displayed.

Software Tools
7-9

CSLIST
The Communications Systems (CS/3000) subsystem consists of the software
modules used for link management and diagnostics. It is used by all HP data
communications network link products, including the HP 3000 LAN link and the
DS Compatible Links. The CSLIST utility lists the version numbers for the
software modules of the CS subsystem. CSLIST also provides detailed
information on the INP and LANIC download files on your system. The
information provided by CS LIST must be included in any Service Request
submitted to liP. Refer to the N S30()O IV Network Manager Reference Manual.
Example 1 shows how to run CSLIST.

Example 1

:RUN CSLIST.PUB.SYS
HP30131A.55.25 CSLIST/3000 SUN, MAR 17, 1984, 9:05 AM
(C) HEWLETT-PACKARD co. 1980
THIS ROUTINE HAS TWO MAJOR FUNCTIONS - ONE ASSOCIATED WITH
THE CS MODULES AND ONE ASSOCIATED WITH THE DOWNLOAD FILES.
THE FIRST PORTION REPORTS THE CSMODULES INSTALLED
ON THE SYSTEM.
NOTINSTDINDICATES THE MODULE HAS NOT BEEN INSTALLED ON THE SYSTEM.
CSLIST ALSO ALLOWS THE USER TO OBTAIN INFORMATION
CONCERNING THE HP-STANDARD DOWNLOAD FILES.
THIS INFORMATION INCLUDES PROTOCOL TYPE,BOARD TYPE,
COMPILE DATE, AND VERSION NUMBER INFORMATION.
DO YOU WANT A COMPLETE LISTING OF INSTALLED VUFS? YES
DO YOU WANT THE DOWNLOAD FILE INFORMATION?

NO

SHOULD OUTPUT BE DIRECTED TO THE LP? YES

After the the RUN command for CSLJST is issued, CSLJST displays a header and
a description, followed by a series of prompts. In the first example, the user
requested a complete listing of the CS module version numbers,without the
download file information, and specified that the output should be directed to
the lineprinter (device LP).

Software Tools

7-10

Example 2 shows a typical listing of CS modules produced by CSLIST.

Example 2

HP30131v.uu.ff CSLIST/3000 SUN, MAR 17, 1985, 9:05 AM
(C) HEWLETT-PACKARD co. 1980
COMPLETE LISTING OF INSTALLED VUFS NOW BEING PRODUCED.
OUTPUT GOING TO DEVICE LP.
COMSYS1
COMSYS2
COMSYS3
COMSYS4
COMSYS5
CSUTILITY
CSDUMMY
CSDUMP
TRACPROG
IOINPO
DSM
INPDPAN
NETCONF
CSLIST
IOLAND
LANDPAN
LANDIAG

INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED
INSTALLED

VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF
VUF

IS
IS
IS
IS
IS
IS
IS
IS
IS
IS
IS
IS
IS
IS
IS
IS
IS

v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff
v.uu.ff

In the first example, the download file information was not selected. Since
CSLIST lists information on all of the download files for all of the HP 3000
products installed on your system, requesting the download file information may
produce a lengthy listing.
The alternative, recommended if you want to check a specific download file, is
to use CSLIST with an entrypoint. The entrypoint, either LAN or IN P, is
appended to the run command, as shown in Example 3 below. When you use
CSLIST with an entrypoint, CSLIST displays an abbreviated description and
prompts for the download file for which you need information.

Software Tools
7-11

Example 3

:RUN CSLIST.PUB.SYS,LAN
CSLIST ALLOWS THE USER TO OBTAIN INFORMATION
CONCERNING USER-SPECIFIED DOWNLOAD FILES. THIS
INFORMATION INCLUDES PROTOCOL TYPE, BOARD TYPE,
COMPILE DATE, AND VERSION NUMBER INFORMATION.
SHOULD OUTPUT BE DIRECTED·TO THE LP?
DOWNLOAD FILE NAME =CSDLAN1.PUB.SYS
LAN DOWNLOAD FILE = CSDLAN1.PUB.SYS
LAST MODIFIED WED, NOV 21, 1984, 9:41 AM
DRIVER OPTIONS = %000000
DATE CODE = B.00.014.077
DOWNLOAD FILE NAME =CSDLAPB2.PUB.SYS
DOWNLOADFILE= CSDLAPB2.PUB.SYS
PROTOCOL TYPE= X.25
BOARD TYPE= INP 20B COMPILE DATE= WED, JUL 4, 1984, 6:26 PM
IC VERSION = 01.02
PROTOCOL VERSION = 01.04
TRACE VERSION = 02.06
RAMCP VERSION = 05.04
DOWNLOAD FILE NAME =CSDBSC2.PUB.SYS
DOWNLOADFILE= CSDBSC2.PUB.SYS
PROTOCOL TYPE= BISYNC (DS,RJE,X.21)
BOARD TYPE= INP 20B COMPILE DATE= THU, OCT 25, 1984, 1:56 PM
IC VERSION = 01.02
PROTOCOL VERSION = 01.11
TRACE VERSION = 02.06
RAMCP VERSION = 05.05
DOWNLOAD FILE NAME =
END OF PROGRAM
The download files specified in Example 3 were for a LAN link (LANIC), a
Point-to-Point Link (JNP). and an X25 Link (INP). Notice that CSLIST provided
information on all three download files, even though two are fOT the INP.

Software Tools

7-12

DSLIST
For the modules of the OS Services subsystem of NS3000jV, and for any of the
OS Compatible Links installed on your system, use the OSLIST program installed
in the PUB group of the SYS account to obtain a list of the versions of the
software modules.
In the OSLIST display, shown in the example below, notice that most of the
modules are grouped under a product number heading. There is also a common
module heading for the utilities, including this one, that apply to all the DS
compatible links.
All modules shown, except those for OSNjX.25, should be displayed on your
system. The DSNjOS HP32189B modules are used by the Point-to~Point Links
and by the DS Services subsystem of NS3000jV. The DSNjX.25 HP32191B
modules are installed with the X.25 Network Link; they only appear if an X.25
Link is installed on your system. The CS HP30131A modules show a subset of
the display provided by CSLIST. The COMSYS module is an overall version
number for the CS modules. The NETCONF module is the utility used for the
network configuration of the X.25 Network Link, described in the DSN/ X.25
for the H P 3000 Reference Manual.
It is essential that:

- all the OS software modules installed on the system have the same version
identification;
- all the X.25 software modules (if installed) have the same version; and,
- all the common and CS modules have the same version
in order to ensure successful operation. The information provided by DSLIST
should be included in any SR submitted. Refer to the NS3000/V Network
Manager Reference Manual.

Software Tools
7-13

Example

:RUN DSLIST.PUB.SYS
HEWLETT PACKARD 32189v.uu.ff DSLISTj3000 SUN, MAR 17, 1985,
DSNjDS HP32189B:
VERSION
MODULE
v.uu.ff,
SL DSSEGS
SL DSRTECALL v.uu.ff,
v.uu.ff,
DSMON
v.uu.ff,
DSTEST
v.uu.ff,
DS2026
DS2026CN' v.uu.ff,
v.uu.ff,
DSCOPY
v.uu.ff,
IODSO
v.uu.ff,
IODSTRMO
v.uu.ff,
IODSTRMX

INTERNAL
INTERNAL
INTERNAL
INTERNAL
INTERNAL
INTERNAL
INTERNAL
INTERNAL
INTERNAL
INTERNAL

FIX
FIX
FIX
FIX
FIX
FIX
FIX
FIX
FIX
FIX

xxx
xxx
xxx
xxx
xxx
xxx
xxx
xxx
xxx
xxx

DSNjX.25 HP32191B:
VERSION
MODULE
v.uu.ff,
DSMONX
v.uu.ff,
IODSX
v.uu.ff,
IOPADO
v.uu.ff,
IOPAD1

INTERNAL
INTERNAL
INTERNAL
INTERNAL

FIX
FIX
FIX
FIX

xxx
xxx
xxx
xxx

7:07 PM

COMMON MODULES:
VERSION
MODULE
INTERNAL FIX xxx
v.uu.ff,
SL DSIOM
INTERNAL FIX xxx
v.uu.ff,
DSDUMP
FIX xxx
INTERNAL
v.uu.ff,
DSLIST
CS SUBSYSTEM HP30131A:
VERSION
MODULE
INTERNAL FIX xxx
v.uu.ff,
SL COMSYS
INTERNAL FIX xxx
v.uu.ff,
NETCONF
END OF PROGRAM
In the example, both Point-to-Point and X.25 Network Links were installed on
the system. If X.25 was not installed, the DSLIST program would display a
message IINOT INSTALLED II under the DSN/X.25 HP32191B heading, and the CS
module NETCONF would not be displayed.

Software Tools
7-14

Troubleshooting Flowcharts

A

Introduction
This appendix suggests some procedures, in flowchart format, for use in
troubleshooting your IEEE 802.3 LAN link. These procedures can help you find
a faulty field replaceable unit (FRU).
A LAN comprised of HP 3000 MPE-V based nodes is presumed. For a more
general IEEE 802.3 LAN troubleshooting approach, refer to the LAN Link
Hardware Trolfbleshooting M anlfal (Coaxial Cable LANs), 5955-7681 (HP CE
Handbook version 5959-2217), or the H P StarLAN Diagnostics and
Troubleshooting M anlfal lor PCs, 50906-90060.
Figure A-I is an overview of the procedures, followed by the following:

• the troubleshooting flowcharts,
• supplemental Remote Note Testing information, and
• Use of "Software Line Tests".
The troubleshooting procedures can be modified as the conditions require.
Experienced users may wish to alter the sequence (or actual implementation) of
each procedure, especially for obvious faults. However, because each path in the
procedures depends on previous outcomes, deviating from the procedures may
cause troubleshooting errors that can be misleading.

NOTE
Recall that the LAN Node Diagnostic (LANDIAG) Test 16, Hood Loopback
Test, does not apply to HP StarLAN connections. For StarLAN connections,
refer to LANDIAG Test 14 (see Chapter 4).

Troubleshooting Flowcharts
A-I

Record SHOWCOM
statistics. {Refer
to Chapter 2'>

Run LAN Node
Diagnostics.

Yes

No

Replace Faulty
Hardware

Fix error.

Use loopback
Connector;
Find Node Faults.

Yes

Run Remote Node
Tests; find
LAN cable faults.

Yes

~NO
Go to
Remote
Node.

Verify test
components
are good.
Yes

Figure A-t. Hardware Troubleshooting O"erview

Troubleshooting Flowcharts
A-2

LAN Node Diagnostic Flowcharts
The troubleshooting flowcharts are presented on the following pages. Start with
Flowchart 1.

Troubleshooting Flowcharts
A-3

Flowchart 1
The first task involves determining if the configuration table and hardware
match. For example, there is a possibility that the I/O configuration table for the
LANIC card may have been entered wrong or the wrong LDEV was specified.
Refer to Chapter 4 for details on running the LAN Node diagnostic.
On successful completion of this flowchart, you should be confident that the
host has confirmed a responding channel exists at the specified LDEV, and that
it is a LANIC card. Refer to the NS3000/V Network Manager Re/erence
Manual for information about verification of Network (NMMGR)
configuration.

Troubleshooting Flowcharts
A-4

Obtain a hardcopy
of 110 Configuration
Table.

Run LAN Node
Diagnostic.
(Refer to Chapter

•

4.)

No

No
Reconfigure
110 Table.

No

Reconfigure
hardware to
match 110 table.
No

Move LANIC to
another slot
of same priority.

Replace
the LANIC.

Flowchart 1. Verification of System and Hardware Configuration

Troubleshooting Flowcharts
A-5

Flowchart 2
Flowchart 2 det.ermines whet.her or not the LANIC card can be initialized, and
verifies that most of the LANIC card is operating properly. The LANIC card
self test is used. For more information on self test, refer to Chapter 5.
A selftest failure does NOT always imply a faulty LANIC card. For successful
selftest completion, the card must be connected to the LAN, or loopback
connector.
If Flowchart 2 yields no faults, you can be confident that the LANIC card can
be initialized, and the microprocessor and LAN connection appear to operate
properly.

Troubleshooting Flowcharts
A-6

Replace the LANIC;
then rerun the
Diagnostic.

No

Replace. the LANIC;
then rerun the
Diagnostic.

No

Yes

Connect LANIC
to LAN or
Loopback Hood.

No

Flowchart 2. LANIC Initialization and Main Module Verification

Troubleshooting Flowcharts
A-7

Flowchart 3
Flowchart 3 verifies operation of the LANIC card to 51MB/1MB interface. A
subtle dependency exists between system operation and test execution. Refer to
detailed descriptions of LANDIAG Tests 9 and 10 if these tests fail (see Chapter
4).

If Flowchart 3 successfully passes, the following are true:
- the HP 3000 can communicate with the LANIC,
- the LANIC card responds as expected, and
- the LANIC can correctly access system memory for DMA processes.

TrOll bleshooti ng Flowcha rts
A-8

Replace the LANIC
then rerun the
diagnostic.

No

Return the diagnostic
when the system
is not busy.
Yes

No

Replace the LANIC
then rerun the
diagnostic.

Replace the LANIC
then rerun the
diagnostic.

Flowchart 3. Interrupt, Soft Reset, and SIMB/IMB Interface Verification

TrOllbleshooting Flowcharts
A-9

Flowchart 4
In Flowchart 4, LANDIAG evaluates that portion of the node directly related to
communications on the LAN. Specifically, the LAN coprocessor, card·to~LAN
interface, and LAN attachment hardware are exercised.
Other LAN features tested during self test include:
1.
. 2.
3.
4.
5.

multicast addressing
CRC generation/detection
MAU/ThinMAU jabber protection (coaxial cable connections only)
SQE disable operation (coaxial cable connections only)
heartbeat detection (coaxial cable connections only)

If Flowcharts I through 4 pass with no faults indicated, the node hardware is
presumed good.
However, marginal hardware faults may stiJJ exist, especially for coaxial LAN
connections. These include MAU/ThinMAU faults that are difficult to detect
by simply by bouncing test frames off the LAN cable.
Marginal faults generally result in network performance degradation. For
example, .slight collision detect circuitry faults may cause frame check sequence
(FeS) errors, Or, a weak MAU/ThinMAU transmitter or receiver may result in
frames not being received by the intended node.
The remaining flowcharts (Flowcharts 5 through 9) are intended to help find
marginal faults on coaxial cable LAN connections.

Troubleshooting Flowcha rts
A-tO

..,IRePI.ce the LANIC.

>N._o_ _-i. ..

Yes

Use Test 14 and
Loopback connector
to find Node Fault.

No

LANIC is outdated.
Replace with
current version.

Perform
Remote Node Tests.

Flowchart 4. LAN Coprocessor, Loopback and Date,. Code Verification

Troubleshooting Flowcharts
A-ll

Flowchart 5

NOTE
Flowcharts 5 through 9 employ LANDIAG Test #16 and apply to coaxial cable
LAN connections only.

Flowchart 5 implements LANDIAG Test #16 to directly or indirectly check the
following hardware:
1.
2.
3.
4.
5.

LANIC card,
LANIC card cable (internal AUI for Series 39/4X/5X/6X/70 systems),
AUI cabling,
MAU/ThinMAU,
Coaxial cable and associated hardware (Tap/BNC Tee connectors, barrels,
connectors, terminators, etc.).

For running Test 16, use of a known good loopback hood and MAU/ThinMAU
is required. A loopback hood and MAU/ThinMAU can be verified as known
good by testing them on a good node. Refer to Chapter 4 for the use of Test 16.
Flowchart 5 checks for liMA U power-on" errors arising from Test 16, and directs
the user to the next flowchart, as appropriate.

Troubleshooti og FJowcha rts
A-12

Run Test 16

'Coax Loopback'

subtest.

No

Flowchart 5. Identifying Failed FRU in Hood Loopback Test

Troubleshooting Flowcharts
A-i3

Flowchart 6
If MAU/ThinMAU power-on errors occur, or power-on attempts are excessive
(more than 5), the following hardware may be faulty:
1.

2.
3.
4.

LANIC card
Internal cable short circuits (Series 39/4X/5X/6X/70 systems only),
AUI cable short circuits,
MAU/ThinMAU.

Flowchart 5 provides steps to identify faulty hardware.

Troubleshooting Flowcharts
A-14

Run Test 16 '25 ohm' subtest
with a MAU Loopback hood
and good internal AUI
connected to LANIC.

Yes

Replace the LANIC;
then rerun the
Diagnostic.

Run Test 16 '25 ohm' subtest
with a MAU Loopback hood
at the end of a section
of AUI.

Replace last section
of AUf tested; then
run Diagnostic.

Proceed to next
section of AUI.

Replace original MAU
and rerun Diagnostic.

Flowchart 6. Identifying MAUjThinMAU Power-On Faults

Troubleshooting Flowcharts
A-i5

Flowchart 7
Flowchart 7 checks the ability of the LANIC card to transmit and receive
packets on the LAN cable bY_tusing a known good loopback hood and
MAU/ThinMA U.
Note that both "25 ohm" and "50 ohm" subtests are performed. The "50 ohm"
subtest verifies collision detection operation. This is important because the
inability of a node to detect collisions can cause serious degradation of network
performance.

Troubleshooting Flowcharts

A..16

Run Test 16 '25 ohm' subtest
with a MAU Loopback hood
and good LANIC cable.

Run Test 16 '25 ohm' subtest
with a MAU Loopback hood
at the end of any good

LANIC cable.

Replace the LANIC;
then rerun the
Diagnostic.

Yes

Flowchart 7. Testing Coax LAN Connection AU' Interface

TrOllbleshooting Flowcharts
A-17

Flowchart 8
Flowchart 8 systematically checks each AUI cable connecting the LANIC card
to the LAN. Only the '25 ohm' test is used.

Troubleshooting Flowcharts

A-IS

Run Test 16 '25 ohm' subtest
with a MAU Loopback hood
at the end of known
AUI cable.

Replace the LANIC;
Then rerun the
Diagnostic.

Run Test 16 '25 ohm' subtest
with a MAU Loopback hood
at the end of each section
of AUI.

Yes

Proceed to next
section of AUI.

Replace last section
of AUI tested.
Rerun Diagnostic.

No

Flowchart 8. Testing AUI Cables

Troubleshooting Flowcharts
A-19

Flowchart 9
Using Test 16, Flowchart 9 checks the operation of the coaxial cable. Here. the
loopback hood is removed, and the known good MAU/ThinMAU is attached to
the coaxial cable in place of the originally installed MAU/ThinMAU.
If the test now runs successfully, the originaJly instaJled MA U/ThinMAU is
faulty.
If the test still fails, the coaxial cable contains a fault. An open Tap, that is, one
that does not make contact with the coaxial cable conductors, can be verified by
repeating the test on a nearby Tap. A shorted Tap is not easily verified, since the
entire cable will appear faulty.
If the coaxial cable appears to be faulty, perform LANDIAG Test #17, Remote
Node Tests, or refer to the LAN L;nk Hardware Trollbleshoor;ng Manual,
5955-7681 (HP CE Handbook version 5959-2217).

Troubleshooting Flowcharts
A-20

Run Test 16 'Coax loopback'
subtest with good MAU
connected to network.

;>-N_O

leave new MAU/ThinMAU or
....., network and replace original
MAUlThinMAU.
Rerun Dia nostic.

Yes

Move MAU to another Tap
and run Test 16 'Coax
loopback' subtest.

No

Original Tap is Faulty:
1. Repair or replace original Tap.
2. Add new Tap, or
3. leave Node attached to
functional Tap.

Perform Remote
Node Tests.
(LANDIAG Test 17.)

Flowchart 9. Testing MAUjThinMAU Connection

Troubleshooting Flowcharts
A-21

Using Test 17 - The Remote Node Test (RNT)
LANDIAG Test 17, the Remote Node Test, aJlows the user to send frames to and
receive frames from remote nodes on the network. It can be llsed to isolate
faults on the LAN between operational nodes. (Refer to Chapter 4, LAN Node
Diagnostic, for details on performing LANDIAG Test 17.)

Coaxial Cable LAN. For coax LANs, a binary search is used in conjunction with
Test 17. In a binary search, the coax cable is first divided into two smaller LANs,
which are each tested using Test 17. Each smaller LAN that fails is further
divided, and the process continues until the LAN cable sections can no longer be
conveniently divided. At that point, either the fault is found and corrected, or
the faulty cable section is replaced.
For other alternatives, refer to the LA N L;nk Hardware Troub/eshooll'ng M anua/,
5955-7681 (HP CE Handbook version 5959-2217).

HP StarLAN. For a StarLAN with multiple Hubs, the Hubs are systematically
isolated from one another while Test 17 is run between connected nodes. In this
way, faults associated with a Hub or remote node can be quickly found.
For more information, refer to the H P Stm'LA N Dhlgnosth's and
Trolfb/eshoot;ngM anual lor pes, 50906-90060.

Troubleshooting Flowcharts
A-22

Using Software Line Tests
In this section, a brief discussion of software line tests used to troubleshoot
Network Services application software is provided. Figure A-2 illustrates the
coverage of various testing tools available. Network Services operation and use
of troubleshooting tools are fully discussed in the N S3000 IV N e/work Manager
Reference Manual, (Volume 1,32344-90002; Volume 2, 32344-90012).

2-Node Line Tests

1-Node Tests

NODE2

NODE1
QuickVal

QuickVal
Network Services

Network Services
IPC Test
or
XPT Test

XPT Test
NetIPC

NetIPC
Network Transport

Network Transport

Link Services

Link Services
LANDIAG 17
LANDIAG
1-16

~

===~:='.l..====+-====l
..

===

Figure A-2. Software Line Tests

TrOllbleshooting Flowcharts
A-23

NOTE
Before running the software line tests, use the LAN Node Diagnostic
(LANDIAG) to ensure that there are no hardware problems.
Also, the software should be reset -- shut down and restart the Network
Transport and Network Services. Issue the command

SHOWCOM ldev; RESET
to set the values of the SHO\VCOM status display to zero.
Be sure to refer to the NS3000/V Network Manager Re/erence Manual before
attempting to perform software line tests.

The HP 3000 LAN link includes three software line tests that can be used to
verify the operation of the NS3000/V and LAN link software. The line tests are
XPT, IPC and QuickVal. They can be run on a single node (software loopback),
and between nodes.
The XPT and IPC line tests are both interactive and use the NetlPC intrinsics.
They verify operation of the Network Transport Level.
QuickVal, which runs in batch mode, checks the netvi-'ork services.

Troubleshooting Flowcharts
A-24

Procedures
the local
Single Node Testing. On a single node, XPT verifies operati on of
m has
proble
LAN
the
y,
properl
ning
functio
not
is
it
Netwo rk Transport. If
ry.
necessa
be
not
may
testing
line
further
probab ly been found and
operat ion of
If the local Transp ort is operati ng properly, run QuickVal to verify
the services on the node.
Finally, perform a two-no de XPT line test.
Transp ort
Two-N ode Testing. The two-no de XPT line test checks the remote
to the
refer
level, and the local and remote links. If this test reveals an error,
results.
the
NS300 0/V Networ k Manag er Referen ce Manua l to interpr et
Perfor m any hardwa re trouble shootin g required.
tests reveal no
If both the single-node softwa re loopback and two-no de XPT line
single node,
a
on
test
this
errors, run the IPe line test betwee n nodes. (Runni ng
transpo rt.)
the
d
checke
i.e., softwa re loopback, is not necessary since you already
ort and
Transp
rk
Netwo
the
The two-no de IPe line test, in addition to checki ng
ement
Manag
s
Proces
e
both links, uses Netwo rk Services, specifically Remot
within the
(RPM). This means that if this test fails, the proble m probab ly lies
RPM capabilities of the Netwo rk Services.
al betwee n
If the IPe two-no de test reveals no errors, you should nm QuickV
the services
tests
al
QuickV
RPM.
nodes to check all Netwo rk Services other than
.
service
each
by executing the intrinsics and comma nds of
D and its
If more than one service is not working, it could mean that DSDA
l process
contro
the
is
D
DSDA
y.
associated modules are not functio ning properl
for all Netwo rk Services.

Trou blesho oting Flowch arts
A-25

Index
A

Abbreviations, 1-6
Activity LEDs, 3-1

C

Cable interface activity LEDs, 3-3,3-5
Coaxial cable LAN
fault isolation procedures, 4-58
IEEE 802.3 Standards, ]-3
loopback hoods, 4-53
LED faults, 3-12
Remote Node Test, A-22
Commands, LANDIAG, 4-15
CS error codes, 6-20
CS trace facility, 6-1
CSDUMP,6-1
trace listing format, 6-7
using, 6-5
CSLIST, 7-10

D

DSDUMP,6-1
DSLIST,7-13

E

Error codes
CS facility, 6-20
self test, 5-7
Error messages
LANDIAG operation, 4-74
LANDIAG tests, 4-19
External loopback, 3- J 4
LANDIAG Test 14, 4-44
LANDIAG Test 16, 4-52
StarLAN, 4-49
EXIT command, LANDIAG, 4-2 J

F

Faults
coaxial cable connections, 4-58
isolation strategy, 1-4
LED examples, 3-12
software, A-23
StarLAN connections, 4-49
Field Replaceable Unit, troubleshooting strategy, 1-4

Index-l

Flowcharts
LANDIAG and, A-3
troubleshooting, A-I

G

GO command, LANDIAG, 4-21

H

HELP command, LANDIAG, 4-20

IEEE 802.3 Standards, 1-3

L

Index-2

LAN, IEEE 802.3 Standards, 1-3
LANDIAG
commands, 4-15
EXIT command, 4-21
general information, 4-1
GO command, 4-21
HELP command, 4-20
limitations, 4-4
LOOP command, 4-21
network services with, 4-7
NOPRINT command, 4-22
operation error messages, 4-74
output from, 4-13
PRINT command, 4-22
Remote Node Test, 4-63, A-22
setting up and using, 4-6
summary of tests, 4-3
system power failure and, 4-8
test dependencies, 4-9, 4-16
test descriptions, 4-23
test error messages, 4-19
troubleshooting flowcharts, A-3
TEST N/ALL command, 4-22
LANIC card, self test, 5-1
LEDs
examples of fault indications, 3-12
labels and definitions, 3-3
location of, 3-1
monitoring cable activity, 3-5
monitoringMPU activity, 3-9
self test, 5-7
selftest, 5-2
Limitations
LANDIAG,4-4
LANIC card self test, 5-2

Link-level address, 4-14
LISTLOG5, 7-2
Loopback hood
coaxial cable LANs, 4-53
LANDIAG Test 14 and, 4-44
LANDIAG Test 16 and, 4-52
LANIC self test and, 4-28
StarLAN, 4-49
LOOP command, LANDIAG, 4-21

M

Memory dump, 7-1
MPU activity LEDs, 3-3, 3-9

N

Network map, 1-5
Network services, LANDIAG with, 4-7
NMMAINT, 7-5
Nomenclature, 1-6
NOPRINT command, LANDIAG, 4-22
NSDPAN5/NSDUMPJ, 7-1

P

Power failure, LANDIAG and, 4-8
Procedures
LANDIAG and, A-3
troubleshooting, A-I
PRINT command, LANDIAG, 4-22

R

Remote Node Test, A-22
LANDIAG, 4-63
Repair, strategy, 1-4

S

Self test
codes, 5-7
from LANDIAG, 4-27
general information, 5-1
limitations, 5-2
loopback hood and, 4-28
LEDs, 5-2
SHOWCOM, 2-1
LANDIAG with, 4-70
Software fault tests, A-23
StarLAN
fault isolation procedures, 4-49
IEEE 802.3 Standard, 1-3

Index-3

100pback hood, 4-49
LED faults, 3-15
Remote Node Test, A-22
Station address, 4-14

T

Index-4

Terminology, abbreviations and nomenclature, 1-6
Tests, software faults, A-23
Tests, LANOIAG
dependencies, 4-9, 4-16
description, 4-23
summary, 4-3
TEST N/ALL command, LANOIAG, 4-22
Tools
CS trace facility, 6- I
CSLIST utility, 7-10
OSLIST utility, 7-13
LANOIAG,4-1
LANIC self test, 5-1
LEOs, 3-1
LISTLOG5 utility, 7-2
memory dump, 7-1
NMMAINT utility, 7-5
NSOPAN5/NSOUMPJ, 7-1
software line tests, A-23
SHOWCOM, 2-1
troubleshooting, J-2
Tracing
activating, 6-2
deactivating, 6-4
general information, 6-1
Troubleshooting
CS trace facility, 6-1
CSLIST utility, 7-10
OSLIST utility, 7-13
flowchart procedures, A-I
general, I-I
LANOIAG, 4-1
LANIC self test, 5-1
LEOs, 3-1
LISTLOG5 utility, 7-2
memory dump, 7-1
NMMAINT utility, 7-5
software faults, A-23
. strategy, 1-4
SHOWCOM, 2-1
tools summa'ry, 1-2

SALES & SUPPORT OFFICES
Arranged alphabetically by country

Canberra, Australia
Capital Territory
Office

Product Line Sales/Support Key
Key Product Line
A Analytical
CM Components
C Computer Systems
E Electronic Instruments' Measurement Systems
M Medical Products
P Personal Computation Products
Sales only for specific product line
•• Support only for specific product line

Hewlett-Packard Australia LId.
121 Wollongong Street
FYSHWICK, A.C.T. 2609
Tel: 80 4244
Telex: 62650
Gable: HEWPARD Canberra
C,CM,E,P

.

IMPORTANT:These symbols designate general product line
capability.They do not insure sales or support availability for all
products within aline, at all locations.Contact your local sales
office for information regarding locations where HP support is
available for specific products.

Melbourne, Victoria
Office
Hewlett-Packard Australia LId.
31-41 Joseph Street
BUCKBURN, Victoria 3130
Tel: 895·2895
Telex: 31·024
Cable: HEWPARD Melbourne
A,C,CM,E,M,P

HEADQUARTERS OFFICES
If there is no sales office listed for your area, contact one of these
headquarters offices.

NORTH/CENTRAL AFRICA
Hewlett-Packard SA
7, rue du Bois-du-Lan
CH-1217 MEYRIN 1, Switzerland
Tel: (022) 83 12 12
Telex: 27835 hmea
Cable: HEWPACKSA Geneve

ASIA
Hewlett-Packard Asia LId.
47IF, 26 Harbour Rd.,
Wanchai, HONG KONG
G.P.O. Box 863, Hong Kong
Tel: 5-8330833
Telex: 76793 HPA HX
Cable: HPASIAL TO

CANADA
Hewlett-Packard (Canada) LId.
6877 Goreway Drive
MISSISSAUGA, Ontario L4V 1M8
Tel: (416) 678·9430
Telex: 610·492·4246

EASTERN EUROPE
Hewlett-Packard Ges.m.b.h.
Lieblgasse 1
P.O.Box 72
A-1222 VIENNA, Austria
Tel: (222) 2500-0
Telex: 1 3 4425 HEPA A

NORTHERN EUROPE
Hewlett-Packard SA
Uilenstede 475
P.O.Box 999
NL-1183 AG AMSTELVEEN
The Netherlands
Tel: 20437771
Telex: 18919 hpner nl

SOUTH EAST EUROPE
Hewlett-Packard SA
World Trade Center
110 Avenue Louis Casal
1215 Cointrin, GENEVA, Switzerland
Tel: (022) 98 96 51
Telex: 27225 hpser

MEDITERRANEAN AND
MIDDLE EAST
Hewlett-Packard SA
Mediterranean and Middle East
Operations
Atrina Centre
32 Kifissias Ave.
Paradissos-Amarousion, ATHENS
Greece
Tel: 68288 11
Telex: 21-6588 HPAT GR
Cable: HEWPACKSA Athens

UNITED KINGDOM
Hewlett-Packard LId.
Nine Mile Ride
Easthampstead, WOKINGHAM
Berkshire, IRGII 3LL
Tel: 0344773100
Telex: 848805

EASTERN USA
Hewlett-Packard Co.
4 Choke Cherry Road
ROCKVILLE, MD 20850
Tel: (301) 258-2000

MIDWESTERN USA
Hewlett-Packard Co.
5201 Tollview Drive
ROLLING MEADOWS, IL 60008
Tel: (312) 255-9800

SOUTHERN USA
Hewlett-Packard Co.
2000 South Park Place
P.O. Box 105005
ATLANTA, GA 30348
Tel: (404) 955-1500

WESTERN USA
Hewlett-Packard Co.
3939 Lankershim Blvd.
P.O. Box 3919
LOS ANGELES, CA 91604
Tel: (213) 506-3700

OTHER INTERNATIONAL
AREAS
Hewlett-Packard Co.
Intercontinental Headquarters
3495 Deer Creek Road
PALO ALTO, CA 94304
Tel: (415) 857-1501
Telex: 034-8300
Cable: HEWPACK

ANGOLA

Perth, Western Australia
Office

Telectra Angola LDA
Empresa T~cnica de Equipamentos
Rua Conselheiro Julio de Vilhema, 16
Caixa Postal 6487
LUANDA
Tel: 35515,35516
Telex: 3134
E,C·

Hewlett-Packard Australia LId.
261 Stirling Highway
CLAREMONT, WA 6010
Tel: 383-2188
Telex: 93859
Cable: HEWPARD Perth
A,C,CM,E,M,P

ARGENTINA

Sydney, New South
Wales Office

Hewlett-Packard Argentina SA
Montaneses 2140 I 50
1428 BUENOS AIRES
Tel: 783-4886/4836/4730
Cable: HEWPACKARG
A,C,CM,E,P
Biotron S.A.C.l.e.1.
Av. Paso Colon 221, Piso 9
1399 BUENOS AIRES
CM
Laboratorio Rodriguez
Corswant S.R.L.
Misiones, 1156 - 1876
Bernal, Oeste
BUENOS AIRES
Tel: 252-3958, 252-4991
A
Argentina Esanco S.R.L.
Avasco 2328
1416 BUENOS AIRES
Tel: 541-58-1981, 541-59-2767
A

AUSTRALIA
Adelaide, South Austrlill
Office
Hewlett-Packard Australia Ltd.
153 Greenhill Road
PARKSIDE, SA 5063
Tel: 272-5911
Telex: 82536
Cable: HEWPARD Adelaide
A· ,C,CM,E,M,P

Brlsblne, QUMnsllnd
Office
Hewlett-Packard Australia Ltd.
10 Payne Road
THE GAP, Queensland 4061
Tel: 30·4133
Telex: 42133
Cable: HEWPARD Brisbane
A,C,CM,E,M,P

Hewlett-Packard Australia LId.
17·23 Talavera Road
P.O. Box 308
NORTH RYDE, N.S.W. 2113
Tel: 888-4444
Telex: 21561
Cable: HEWPARD Sydney
A,C,CM,E,M,P

AUSTRIA
Hewlett-Packard Ges.m.b.h.
VerkauisbOro Graz
Grottenhofstrasse 94
A·8052 GRAZ
Tel: (0316) 291 5660
Telex: 32375
C,E
Hewlett-Packard Ges.m.b.h.
Lieblgasse 1
P.O. Box 72
A·1222 VIENNA
Tel: (0222) 2500-0
Telex: 134425 HEPA A
A,C,CM,E,M,P

BAHRAIN
Green Salon
P.O. Box 557
MANAMA
Tel: 255503-255950
Telex: 8441
P
Wael Pharmacy
P.O. Box 648
MANAMA
Tel: 256123
Telex: 8550 WAEL BN
E,M
Zayani Computer Systems
218 Shaik Mubarak Building
Government Avenue
P.O. Box 5918
MANAMA
Tel: 276278
Telex: 9015

P

BELGIUM
Hewlett-Packard Belgium SAIN.v.
Blvd de la Woluwe, 100
Woluwedal
B-1200 BRUSSELS
Tel: (02) 762-32-~
Telex: 23-494 paloben bru
A,C,CM,E,M,P

BERMUDA
Applied Computer Technologies
Atlantic House Building
Par-La-Ville Road
Hamilton 5
Tel: 295-1616
P

BRAZIL
Hewlett-Packard do Brasil
l.e.C. L1da.
Alameda Rio Negro, 750
Alphaville
06400 BARUERI SP
Tel: (011) 421.1311
Telex: (011) 33872 HPBR-BR
Cable: HEWPACK Sao Paulo
A,C,CM,E,M,P
Hewlett-Packard do Brasil
l.e.C. L1da.
Praia de Botafago 228
6° Andar-conj 614
Edificio Argentina - Ala A
22250 RIO DE JANEIRO
Tel: (021) 552-6422
Telex: 21905 HPBR-BR
Cable: HEWPACK Rio de Janeiro
A,C,CM,E,P·
ConvexIVan Den
Rua Jose Bonifacio
458 Todos Os Santos
CEP 20771
RIO DE JANEIRO, RJ
Tel: 591-0197
Telex: 33487 EGLB BR
A
ANAMED I.C.E.I. L1da.
Rua Bage, 103
04012 SAO PAULO, SP
Tel: (011) 572-6537
Telex: 24720 HPBR-BR
M
Datatronix Electronica L1da.
Av. Pacaembu 746-C11
SAO PAULO, SP
Tel: (118) 260111
CM

CAMEROON
Beriac
B. P. 23
DOUALA
Tel: 420153
Telex: 5351
C,P

CANADA
Alberti
Hewlett-Packard (Canada) LId.
30303rd Avenue N.E.
CALGARY, Alberta T2A 6T7
Tel: (403) 235-3100
A,C,CM,E*,M,P*
Hewlett-Packard (Canada) LId.
11120-178th Street
EDMONTON, Alberta T5S 1P2
Tel: (403) 486-6666
A,C,CM,E,M,P

o SALES & SUPPORT OFFICES

~lJ

Arranged alphabetically by country

CANADA (Cont'd)
British Columbia
Hewlett-Packard (Canada) Ltd.
10691 Shellbridge Way
RICHMOND,
British Columbia V6X 2W7
Tel: (604) 270-2277
Telex: 610-922-5059
A,C,CM,E* ,M,P*
Hewlett-Packard (Canada) Ltd.
121 - 3350 Douglas Street
VICTORIA, British Columbia V8l 3L 1
Tel: (604) 381-6616
C

Manitoba
Hewlett-Packard (Canada) Ltd.
1825 Inkster Blvd.
WINNIPEG, Manitoba R2X 1R3
Tel: (204) 694·2777
A,C,CM,E,M,P'

New Brunswick
Hewlett-Packard (Canada) Ltd.
814 Main Street
MONCTON, New Brunswick E1C 1E6
Tel: (506) 855-2841
C

Nova Scotia
Hewlett-Packard (Canada) Ltd.
Suite 111
900 Windmill Road
DARTMOUTH, Nova Scotia B3B 1P7
Tel: (902) 469-7820
C,CM,E',M,P*

Ontario
Hewlett-Packard (Canada) Ltd.
3325 N. Service Rd., Unit 3
BURLINGTON, Ontario L7N 3G2
Tel: (416) 335-8644
C,M'
Hewlett-Packard (Canada) Ltd.
496 Days Road
KINGSTON, Ontario K7M 5R4
Tel: (613) 384-2088
C
Hewlett-Packard (Canada) Ltd.
552 Newbold Street
LONDON, Ontario N6E 2S5
Tel: (519) 686-9181
A,C,CM,E* ,M,P*
Hewlett-Packard (Canada) Ltd.
6877 Goreway Drive
MISSISSAUGA, Ontario L4V 1M8
Tel: (416) 678-9430
A,C,CM,E,M,P
Hewlett-Packard (Canada) Ltd.
2670 Queensview Dr.
OTTAWA, Ontario K2B 8K1
Tel: (613) 820-6483
A,C,CM,E* ,M,P*
Hewlett-Packard (Canada) Ltd.
The Oaks Plaza, Unit #9
2140 Regent Street
SUDBURY, Ontario, P3E 5S8
Tel: (705) 522-0202
C
Hewlett-Packard (Canada) Ltd.
3790 Victoria Park Ave.
W1UOWDALE, Ontario M2H 3H7
Tel: (416) 499-2550
C

Quebec
Hewlett-Packard (Canada) Ltd.
17500 Trans Canada Highway
South Service Road
KIRKLAND, Quebec H9J 2X8
Tel: (514) 697-4232
A,C,CM,E,M,P*
Hewlett-Packard (Canada) Ltd.
1150 rue Claire Fontaine
QUEBEC CITY, Quebec G1R5G4
Tel: (418) 648-0726
C
Hewlett-Packard (Canada) Ltd.
130 Robin Crescent
SASKATOON, Saskatchewan S7L 6M7
Tel: (306) 242-3702
C

CHILE
ASC Ltda.
Austria 2041
SANTIAGO
Tel: 223-5946,223-6148
Telex: 340192 ASC CK
C,P
Isical Ltda.
Av. Italia 634 Santiago
Casilla 16475
SANTIAGO 9
Tel: 222-0222
Telex: 440283 JCYCL Cl
CM,E,M
Metrolab SA
Monjitas 454 of. 206
SANTIAGO
Tel: 395752, t398296
Telex: 340866 METLAB CK
A
Olympia (Chile) Ltda.
Av. Rodrigo de Araya 1045
Casilla 256-V
SANTIAGO 21
Tel: 225-5044
Telex: 340892 OLYMP
Cable: Olympiachile Santiagochile
C,P

CHINA, People'.
R.publlco'
China Hewlett-Packard, Ltd.
47/F China Resources Bldg.
26 Harbour Road
HONG KONG
Tel: 5-8330833
Telex: 76793 HPA HX
Cable: HP ASIA LTO
A*,M*
China HeWlett-Packard, Ltd.
P.O. Box 9610, Beijing
4th Floor, 2nd Watch Factory Main
Bldg.
Shuang Yu Shu, Bei San Huan Rd.
Hai Dian District
BEIJING
Tel: 28-0567
Telex: 22601 CTSHP CN
Cable: 1920 Beijing
A,C,CM,E,M,P

COLOMBIA
Instrumentaci6n
H. A. Langebaek & Kier SA
Carrera 4A No. 52A-26
Apartado Aereo 6287
BOGOTA 1, D.E.
Tel: 212-1466
Telex: 44400 INST CO
Cable: AARIS Bogota
CM,E,M
Nefromedicas Ltda.
Calle 123 No. 9B-31
Apartado Aereo 100-958
BOGOTA DE, 10
Tel: 213-5267, 213-1615
Telex: 43415 HEGAS CO
A

ECUADOR
CYEDE Cia. Ltda.
Avenida Eloy Alfaro 1749
y Belgica
Casilla 6423 CCI
QUITO
Tel: 450-975, 243-052
Telex: 22548 CYEDE ED
CM,E,P
Medtronics
Valladolid 524 Madrid
P.O. 9171, QUITO
Tel: 223-8951
Telex: 2298 ECKAME ED
A

Compumundo
Avenida 15 # 107-80
BOGOTAD.E.
Tel: 214-4458
Telex: 45466 MARICO
P

Hospitalar SA
Robles 625
Casilla 3590
QUITO
Tel: 545-250, 545-122
Telex: 2485 HOSPTL ED
Cable: HOSPITALAR-Quito
M

Carvajal, SA
Calle 29 Norte No. 6A-40
Apartado Aereo 46
CALI
Tel: 368-1111
Telex: 55650
C,E,P

Ecuador Overseas Agencies CA
Calle 9 de OCtubre #818
P.O. Box 1296, Guayaquil
QUITO
Tel: 306022
Telex: 3361 PBCGYE ED
M

CONGO
Seric-Congo
B. P. 2105
BRAZZAVILLE
Tel: 815034
Telex: 5262

COSTA RICA
Cientifica Costarricense SA
Avenida 2, Calle 5
San Pedro de Montes de Oca
Apartado 10159
SAN JOse
Tel: 24-38-20, 24-08-19
Telex: 2367 GALGUR CR
CM,E,M

CYPRUS

EGYPT
Sakrco Enterprises
70, Mossadak Str.
Dokki, Giza
CAIRO
Tel: 706440
Telex: 93146
C
International Engineering Associates
24 Hussein Hegazi Street
Kasr-el-Ain
CAIRO
Tel: 23829, 21641
Telex: 93830 lEA UN
Cable: INTEGASSO
E,M*

Telerexa Ltd.
P.O. Box 4809
14C Stassinos Avenue
NICOSIA
Tel: 62698
Telex: 2894 LEVIDO CY
E,M,P

S.S.C. Medical
40 Gezerat EI Arab Street
Mohandessin
CAIRO
Tel: 803844, 805998, 810263
Telex: 20503 SSC UN
M*

DENMARK

ELSALVADOR

Hewlett-Packard A/S
Datavej 52
DK-3460 B1RKEROD
Tel: (02) 81-66-40
Telex: 37409 hpas dk
A,C,CM,E,M,P
Hewlett-Packard AlS
Rolighedsvej 32
DK-8240 RISSKOV, Aarhus
Tel: (06) 17·60-00
Telex: 37409 hpas dk
C,E

IPESA de EI Salvador SA
29 Avenida Norte 1223
SAN SAlVADOR
Tel: 26-6858, 26-6868
Telex: 20539 IPESA SAL
A,C,CM,E,P

ETHIOPIA
Seric-Ethiopia
P.O. Box 2764
ADDIS ABABA
Tel: 185114
Telex: 21150
C,P

DOMINICAN REPUBLIC
Microprog SA
Juan Tomas MejIa y Cotes No. 60
Arroyo Hondo
SANTO DOMINGO
Tel: 565-6268
Telex: 4510 ARENTA DR (RCA)
P

FINLAND
Hewlett-Packard Oy
Piispankalliontie 17
02200 ESPOO
Tel: 00358-0-88721
Telex: 121563 HEWPA SF
A,C,CM,E,M,P

FRANCE
Hewlett-Packard France
l.l. Mercure B
Rue Berthelot
13763 Les Milles Cedex
AIX·EN·PROVENCE
Tel: (42) 59-41-02
Telex: 410770F
A,C,E,M,P*
Hewlett-Packard France
64, rue Marchand Saillant
61000 ALENCON
Tel: (33) 29 04 42
Hewlett-Packard France
28 rue de la Republique
Boite Postale 503
25026 BESANCON Cedex
Tel: (81) 83-16-22
Telex: 361157
C,M
Hewlett-Packard France
Chemin des Mouilles
Boite PostaIe 162
69130 ECULlY Cedex (Lyon)
Tel: (78) 833-81-25
Telex: 310617F
A,C,E,M
Hewlett-Packard France
Pare d'activitlls du Bois Briard
2, avenue du Lac
91040 EVRY Cedex
Tel: 6077-9660
Telex: 692315F
E
Hewlett-Packard France
5, avenue Raymond Chanas
38320 EYBENS (Grenoble)
Tel: (76) 62-57-98
Telex: 980124 HP GRENOB EYBE
C
Hewlett-Packard France
Rue Fernand. Forest
l.A. Kergaradec
29239 GOUESNOU
Tel: (98) 41-87-90
Hewlett-Packard France
Centre d'aftaires Paris-Nord
Batiment Am~re
Rue de la Commune de Paris
Boite Postale 300
93153 LE BLANC·.SNIL
Tel: (1) 865-44-52
Telex: 211032F
C,E,M
Hewlett-Packard France
Pare d'activitlls Cadera
Quartier Jean-Mermoz
Avenue du Prllsident JF Kennedy
F-33700 MERIGNAC (Bordeaux)
Tel: (56) 34-00-84
Telex: 550105F
C,E,M
Hewlett-Packard France
Immueble "les 3 B"
Nouveau chemin de la Garde
lAC du Bois Briand
44085 NANTES Cedex
Tel: (40) 50-32-22
Telex: 711085F
C"
Hewlett-Packard France
125, rue du Faubourg Bannier
45000 ORLEANS
Tel: (38) 6801 63

SALES & SUPPORT OFFICES
Arranged alphabetically by country

lANCE (Cont'd)

GERMAN FEDERAL
REPUBLIC

Nlett-Packard France
lelndustrielle de Courtaboeuf
~nue des Tropiques
~47 Les Ulis Cedex ORSAY
: (6) 907-78-25
ex: 600048F
;,CM,E,M,P

Hewlett-Packard GmbH
GeschAftsstelle
Keithstrasse 2-4
D-l000 BERLIN 30
Tel: (030) 21 99 04-0
Telex: 0183405 hpbln d
A,C,E,M,P

wlett-Packard France
ris Porte-Maillot
, boulevard de L'Amiral-Bruix
782 PARIS Cedex 16
I: (1) 502-12-20
lex: 613663F
IA.P

Hewlett-Packard GmbH
Vertriebszentrun SOdwest
Schickardstrasse 2
D-7030 BOBLINGEN
Tel: (07031) 645-0
Telex: 7265 743 hep
A,C,CM,E,M,P

!wlett-Packard France
4, Boulevard Tourasse
000 PAU
I: (59) 80 38 02

Hewlett-Packard GmbH
Vertriebszentrum West
Berliner Strasse III
D-4030 RATINGEN 3
Tel: (02102) 494-0
Telex: 589 070 hprad
A,C,E,M,P

lwlett-Packard France
Allee de la Bourgonnette
,100 RENNES
il: (99) 51-42-44
ilex: 740912F
CM.E,M,P*
~wlett-Packard France
I avenue de Bretagne
;100 ROUEN
ll: (35) 63-57-66
llex: 770035F

ewlett-Packard France
rue Thomas-Mann
oite Postale 56
7033 STRASBOURG Cedex
~I: (88) 28-56-46
~Iex: 890141F
,E,M,P'

Hewlett-Packard GmbH
Geschaftsstelle
Schleefstr. 28a
D-4600 DORTMUND-41
Tel: (0231) 45001
Telex: 822858 hepdad
A,C,E
Hewlett-Packard GmbH
Vertriebszentrum Mitte
Hewlett-Packard-Strasse
D-6380 BAD HOMBURG
Tel: (06172) 400-0
Telex: 410844 hpbhg
A,C,E,M,P

Hewlett-Packard GmbH
Vertriebszentrum Nord
Kapstadtring 5
0-2000 HAMBURG 60
ewlett-Packard France
Tel: (040) 63804-1
a Peripole III
Telex: 021 63032 hphh d
O. chemin du Pigeonnier de la cepi~re A.C,E,M,P
-31083 TOULOUSE Cedex
Hewlett-Packard GmbH
el: (61) 40-11-12
GeschAftsstelle
elex: 531639F
Heidering 37-39
••C.E,P'
D-3000 HANNOVER 61
lewlett-Packard France
Tel: (0511) 5706-0
, rue Baudin
Telex: 092 3259
'6000 VALENCE
A,C,CM,E,M,P
el: (75) 42 76 16
Hewlett-Packard GmbH
tewlett-Packard France
Geschaftsstelle
:arolor
Rosslauer Wag 2-4
~AC de Bois Briand
D-6800 MANNHEIM
i7640 VlGY (Metz)
Tel: (0621) 70 OS-O
'el: (8) 771 20 22
Telex: 0462105
A.C,E
iewlett-Packard France
)arc d'activite des Pres
I, rue Papin
,9658 VILLENEUVE D'ASCQ Cedex
fel: (20) 47 78 78
felex: 160124F
~,E.M,P'

~ABON

:;ho Gabon
).0. Box 89

.IBREVILLE
rei: 721 484
relex: 5230

Hewlett-Packard GmbH
GeschAftsstelle
Messerschmittstrasse 7
D-7910 NEU ULM
Tel: (0731) 70 73-0
Telex: 0712816 HP ULM·D
A,C,E*
Hewlett-Packard GmbH
Geschaftsstelle
Emmericher Strasse 13
D-8500 NORNBERG 10
Tel: (0911) 5205-0
Telex: 0623 860 hpnbg
C,CM,E,M,P

Hewlett-Packard GmbH
Vertriebszentrum Sad
Eschenstrasse 5
D-8028 TAUFKIRCHEN
Tel: (089) 61 207-0
Telex: 0524985
A,C,CM,E,M,P
Hewlett-Packard GmbH
Geschaftsstelle
Ermlisallee
7517 WALDBRONN 2
Tel: (07243) 602-0
Telex: 782838 hepk
A,C,E

GREAT BRITAIN
See United Kingdom
GREECE
Hewlett-Packard A.E.
178, Kifissias Avenue
6th Floor
Halandri-ATHENS
Greece
Tel: 6471543. 6471673, 6472971
Telex: 221 286 HPHLGR
A,C,CM' *,E,M,P
Kostas Karaynnis S.A.
8, Omirou Street
ATHENS 133
Tel: 32 30 303, 32 37 371
Telex: 215962 RKAR GR
A,C*,CM,E
Impexin
Intelect Div.
209 Mesogion
11525 ATHENS
Tel: 6474481/2
Telex: 216286
P
Haril Company
38, Mihalakopoulou
ATHENS 612
Tel: 7236071
Telex: 218767
M*
Hellamco
P.O. Box 87528
18507 PIRAEUS
Tel: 4827049
Telex: 241441
A

GUATEMALA
IPESA
Avenida Reforma 3·48, Zona 9

GUATEMALA CITY
Tel: 316627, 314786
Telex: 30557651PESA GU
A.C,CM,E,M.P

HONOKONG
Hewlett-Packard Hong Kong, Ltd.
G.P.O. Box 795
5th Floor, Sun Hung Kai Gentre
30 Harbour Road

HONG KONG
Tel: 5-8323211
Telex: 66678 HEWPA HX
Cable: HEWPACK Hong Kong
E,C,P
CET Ltd.
10th Floor, Hua Asia Bldg.
64-66 Gloulester Road

HONG KONG
Tel: (5) 200922
Telex: 85148 CET HX
CM

Schmidt & Co. (Hong Kong) Ltd.
18th Floor, Great Eagle Centre
23 Harbour Road
HONG KONG
Tel: 5-8330222
Telex: 74766 SCHMC HX
A,M

ICELAND
Hewlett-Packard Iceland
Hoefdabakka 9
110 Reykjavik
Tel: (1) 67 1000
A,C,CM,E,M,P

Blue Star Ltd.
15/16 C Wellesley Rd.
PUNE 411011
Tel: 22775
Cable: BLUE STAR
A
Blue Star Ltd.
2-2-47/1108 Bolarum Rd.
SECUNDERABAD 500 003
Tel: 72057, 72058
Telex: 0155645
Cable: BLUEFROST
A,E

Blue Star Ltd.
T.C. 7/603 Poornima
INDIA
Maruthunkuzhi
Computer products are sold through TRIVANDRUM 695 013
Blue Star Ltd.AII computer repairs and Tel: 65799, 65820
maintenance service is done through Telex: 0884-259
Computer Maintenance Corp.
Cable: BLUESTAR
E
Blue Star Ltd.
Sabri Complex 2nd Floor
Computer Maintenance Corporation
24 Residency Rd.
Ltd.
BANGALORE 560 025
115, Sarojini Devi Road
Tel: 55660, 578881
SECUNDERABAD 500 003
Telex: 0845-430
Tel: 310-184, 345-774
Telex: 031-2960
Cable: BLUESTAR
C"
A,C*,CM,E
Blue Star Ltd.
Band Box House
Prabhadevi
BOMBAY 400025
Tel: 4933101, 4933222
Telex: 011-71051
Cable: BLUESTAR
A,M
Blue Star Ltd.
Sahas
414/2 Vir Savarkar Marg
Prabhadevi
BOMBAY 400025
Tel: 422-6155, 422-6556
Telex: 011-71193 BSSS IN
Cable: FROSTBLUE
A.C',CM,E.M
Blue Star Ltd.
Kalyan, 19 Vishwas Colony
Alkapuri, BORODA, 390 005
Tel: 65235. 65236
Cable: BLUE STAR
A
Blue Star Ltd.
7 Hare Street
CALCUTTA 700 001
Tel: 230131, 230132
Tele~: 021-7655
Cable: BLUESTAR
A,M
Blue Star Ltd.
133 Kodambakkam High Road
MADRAS 600 034
Tel: 472056. 470238
Telex: 041-379
Cable: BLUESTAR
A,M
Blue Star Ltd.
13 Community Genter
New Friends Colony
NEW DELHI 110 065
Tet: 633773, 634473
Telex: 031-61120
Cable: BLUEFROST
A,C' ,CM,E.M

INDONESIA
BERCA Indonesia P.T.
P.O. Box 496/Jkt.
JI. Abdul Muis 62

JAKARTA
Tel: 21-373009
Telex: 46748 BERSAL IA
Cable: BERSAL JAKARTA

P
BERCA Indonesia P.T.
P.O. Box 2497/Jkt
Antara Bldg., 12th Floor
JI. Medan Merdeka Selatan 17

JAKARTA·PUSAT
Tel: 21-340417, 341445
Telex: 46748 BERSAL IA
A,C.E,M
BEReA Indonesia P.T.
Jalan Kutai 24

SURABAYA
Tel: 67118
Telex: 31146 BERSAL SB
Cable: BERSAL·SURABAYA
A*,E,M,P

IRAQ
Hewlett-Packard Trading SA
Service Operation
AI Mansoor City 9B/3/7

BAGHDAD
Tel: 551-49-73
Telex: 212-455 HEPAIRAQ IK

C

IRELAND
Hewlett-Packard Ireland Ltd.
82/83 Lower Leeson Street
DUBLIN 2
Tel: 0001 608800
Telex: 30439
A.C.CM,E,M.P
Cardiac Services Ltd.
Kilmore Road
Artane
DUBLIN 5
Tel: (01) 351820
Telex: 30439
M

~ SALES & SUPPORT OFFICES

~

Arranged alphabetically by country

ISRAEL
Eldan Electronic Instrument Ltd.
P.O.Box 1270
JERUSALEM 91000
16, Ohaliav S1.
JERUSALEM 94467
Tel: 533 221, 553242
Telex: 25231 AB/PAKRD IL
A,M
Computation and Measurement
Systems (CMS) Ltd.
11 Masad Street
67060
TEL·AVIV
Tel: 388388
Telex: 33569 MotillL
C,CM,E,P

Hewlett-Packard Italiana S.p.A.
Viale C. Pavese 340
1-00144 ROMA EUR
Tel: (06) 54831
Telex: 610514
A,C,E,M,P'
Hewlett-Packard Italiana S.p.A.
Via di Casellina 571C
1-50018 SCANDlCCI·F1RENZE
Tel: (055) 753863
C,E,M
Hewlett-Packard Italiana S.pA
Corso Svizzera, 185
1-10144 TORINO
Tel: (011) 74 4044
Telex: 221079
A',C,E

ITALY

IVORY COAST

Hewlett-Packard Italiana S.p.A
Traversa 99C
Via Giulio Petroni, 19
1-70124 BARI
Tel: (080) 41-07-44
C,M

S.I.T.E.L.
Societe Ivoirienne de
Telecommunications
Bd. Giscard d'Estaing
Carrefour Marcory
Zone 4.A.
Boite postale 2580
ABIDJAN 01
Tel: 353600
Telex: 43175
E

Hewlett-Packard Italiana S.pA
Via Emilia, 51/C
1-40011 BOLOGNA Anzola DeIl'Emilia
Tel: (051) 731061
Telex: 511630
C,E,M
Hewlett-Packard Italiana S.p.A.
Via Principe Nicola 43G/C
1-95126 CATANIA
Tel: (095) 37-10-87
Telex: 970291
C
Hewlett-Packard Italiana S.pA
Via G. Di Vittorio 9
1-20063 CERNUSCO SUL
NAVIGLIO
(Milano)
Tel: (02) 4459041
Telex: 334632
A,C,CM,E,M,P
Hewlett-Packard Italiana S.p.A.
Via C. Colombo 49
1-20090 TREZZANO SUL
NAVIGUO
(Milano)
Tel: (02) 4459041
Telex: 322116
C
Hewlett-Packard Italiana S.p.A.
Via Nuova San Rocco a
Capodimonte, 62/A
1-80131 NAPOLI
Tel: (081) 7413544
Telex: 710698
A",C,E,M

S.I.T.1.
Immeuble "Le General"
Av. du General de Gaulle
01 BP 161
ABIDJAN 01
Tel: 321227
C,P

JAPAN
Yokogawa-Hewlett-Packard Ltd.
152-1,Onna
ATSUGI, Kanagawa, 243
Tel: (0462) 25-0031
C,CM,E
Yokogawa-Hewlett-Packard Ltd.
Meiji-Seimei Bldg. 6F
3-1 Hon Chiba-Cho
CHI8A,280
Tel: 472257701
C,E
Yokogawa-Hewlett-Packard Ltd.
Yasuda-Seimei Hiroshima Bldg.
6-11, Hon-dori, Naka-ku
HIROSHIMA, 730
Tel: 82-241-0611
Yokogawa-Hewlett-Packard Ltd.
Towa BUilding
2-3, Kaigan-dori, 2 Chome Chuo-ku
KOIE,650
Tel: (078) 392-4791
C,E

Hewlett-Packard Italiana S.p.A.
Viale G. Modugno 33
1-16156 GENOVA PEQU
Tel: (010) 68-37-07
Telex: 215238
C,E

Yokogawa-Hewlett-Packard Ltd.
Kumagaya Asahi 82 Bldg
3-4 Tsukuba
KUllAGAYA, Saitama 360
Tel: (0485) 24-6563
C,CM,E

Hewlett-Packard Italiana S.p.A.
Via Pelizzo 15
1-35128 PADOVA
Tel: (049) 664888
Telex: 430315
A,C,E,M

Yokogawa-Hewlett-Packard Ltd.
Asahi Shinbun Daiichi Seimei Bldg.
4-7, Hanabata-cho
KUMAMOTO, 860
Tel: (096) 354-7311
C,E

Yokogawa-Hewlett-Packard Ltd.
Shin-Kyoto Genter Bldg.
614, Higashi-Shiokoji-cho
Karasuma-Nishiiru
Shiokoji-dori, Shimogyo-ku
KYOTO, 600
Tel: 075-343-0921
C,E
Yokogawa-Hewlett-Packard Ltd.
Mito Mitsui Bldg
4-73, Sanno-maru, 1Chome
MITO, Ibaraki 310
Tel: (0292) 25-7470
C,CM,E
Yokogawa-Hewlett-Packard Ltd.
Meiji-Seimei Kokubun Bldg. 7-8
Kokubun, 1Chome, Sendai
MIYAGI,980
Tel: (0222) 25-1011
C,E
Yokogawa-Hewlett-Packard Ltd.
Nagoya Kokusai Center BUilding
47-1, Nagono, 1Chome
Nakamura-ku
NAGOYA, 450
Tel: (052) 571-5171
C,CM,E,M
Yokogawa-Hewlett-Packard Ltd.
Saikyoren BUilding
1-2 Dote-machi, OHMIYA
Saitama 330
Tel: (0486) 45-8031
Yokogawa-Hewlett-Packard Ltd.
Bldg.,
4-20 Nishinakajima, 5 Chome
Yodogawa-ku
OSAKA, 532
Tel: (06) 304-6021
Telex: YHPOSA 523-3624
A,C,CM,E,M,P'

~Chuo

Yokogawa-Hewlett-Packard Ltd.
27-15, Yabe, 1Chome
SAGAMIHARA Kanagawa, 229
Tel: 042759-1311

Yokogawa-Hewlett-Packard Ltd.
Yasuda Seimei Yokohama Nishiguchi
Bldg.
30-4 Tsuruya-cho, 3 Chome
YOKOHAMA 221
Tel: (045) 312-1252
C,E

JORDAN
Scientific and Medical Supplies Co.
P.O. Box 1387
AMMAN
Tel: 24907, 39907
Telex: 21456 SABCO JO
C,E,M,P

KENYA
ADCOM Ltd., Inc., Kenya
P.O.Box 30070
NAIROBI
Tel: 331955
Telex: 22639
E,M

KOREA
Samsung Hewlett-Packard Co. Ltd.
Dongbang Yeoeuido Building
12-16th Floors
36-1 Yeoeuido-dong
Yongdeungpo-ku
SEOUL
Tel: 784-2666, 784-4666
Telex: 25166 SAMSAN K
A,C,CM,E,M,P
Young In Scientific Co., Ltd.
Youngwha BUilding
547 Shinsa Dong, Kangnam-ku
SEOUL 135
Tel: 5467771
Telex: K23457 GINSCO
A

KUWAIT
AI-Khaldiya Trading & Contracting
P.O. Box 830
SAFAT
Tel: 424910, 411726
Telex: 22481 AREEG KT
Cable: VISCOUNT
E,M,A

LIBERIA
Unichemicals Inc.
P.O. Box 4509
MONROVIA
Tel: 224282
Telex: 4509
E

MADAGASCAR
Technique et Precision
12, rue de Nice
P.O. Box 1227
101 ANTANANARIVO
Tel: 22090
Telex: 22255
P

LUXEMBOURG
Hewlett-Packard Belgium SA/N.v.
Blvd de la Woluwe, 100
Woluwedal
B-1200 BRUSSELS
Tel: (02) 762-32-00
Telex: 23-494 paloben bru
A,C,CM,E,M,P

MALAYSIA
Hewlett-Packard Sales (Malaysia)
Sdn. Bhd.
9th Floor
Chung Khiaw Bank BUilding
46, Jalan Raja Laut
KUALA LUMPUR
Tel: 03-986555
Telex: 31011 HPSM MA
A,C,E,M,P'
Protei Engineering
P.O. Box 1917
Lot 6624, Section 64
23/4 Pending Road
Kuching, SARAWAK
Tel: 36299
Telex: 70904 PROMAL MA
Cable: PROTELENG
A,E,M

MALTA
Philip Toledo Ltd.
Birkirkara P.O. Box 11
Notabile Rd.
MRiEHEL
Tel: 447 47, 455 66
Telex: 1649
E,M,P

Yokogawa-Hewlett-Packard Ltd.
Dalichi Seimei Bldg.
7-1, Nishi Shinjuku, 2 Chome
Shinjuku-ku,TOKYO 160
Tel: 03-348-4611
C,E

Gulf Computing Systems
P.O. Box 25125
SAFAT
Tel: 435969
Telex: 23648
P

Yokogawa-Hewlett-Pa,ckard Ltd.
29-21 Takaido-Higashi, 3 Chome
Suginami-ku TOKYO 168
Tel: (03) 331-6111
Telex: 232-2024 YHPTOK
A,C,CM,E,M,P'

Photo & Cine Equipment
P.O. Box 270
SAFAT
Tel: 2445111
Telex: 22247 MATIN KT
Cable: MATIN KUWAIT
P

Blanche Birger Co. Ltd.
18, Jules Koenig Street
PORT LOUIS
Tel: 20828
Telex: 4296
P

Yokogawa Hokushin Electric Corp.
9-32 Nokacho 2 Chome
2 Chome Musashino-shi
TOKYO, 180
Tel: (0422) 54-1111
Telex: 02822-421 YEW MTK J
A

W.J. Towell Computer Services
P.O. Box 5897
SAFAT
Tel: 2462640
Telex: 30336 TOWELL KT
C

MEXICO

Yokogawa-Hewlett-Packard Ltd.
Meiji-Seimei
Utsunomiya Odori BUilding
1-5 Odori, 2 Chome
UTSUNOMIYA, Tochigi 320
Tel: (0286) 33-1153
C,E

LEBANON
Computer Information Systems SAL.
Chammas BUilding
P.O. Box 11-6274 Dora
BEIRUT
Tel: 894073
Telex: 42309
C,E,M,P

MAURITIUS

Hewlett-Packard de Mexico, SA
Francisco J. Allan #30
Colonia Nueva
Los Angeles 27140
COAHUILA, Torreon
Tel: 37220
P
Hewlett-Packard de Mexico, SA
Monti Morelos 299
Fraccionamiento Loma Bonita 45060
GUADALAJARA, Jalisco
Tel: 316630/314600
Telex: 0684 186 ECOME
P

SALES & SUPPORT OFFICES
Arranged alphabetically by country

Hewlett-Packard Nederland B.V.
Microcomputadoras Hewlett-Packard, Pastoor Petersstraat 134-136
S.A.
NL 5612 LV EINDHOVEN
Monti Pelvoux 115
P.O. Box 2342
LOS LOMAS, Mexico, OJ.
NL 5600 CH EINDHOVEN
Tel: 520-9127
Tel: (040) 326911
P
Telex: 51484 hepae nl
A,C,E,M,P
Hewlett-Packard Mexicana, S.A.
de C.V.
NEW ZEALAND
Av. Periferico Sur No. 6501
Hewlett-Packard (N.l.) Ltd.
Tepepan, Xochimilco
5 Owens Road
16020 MEXICO D.F.
P.O. Box 26-189
Tel: 6-76-46-00
Epsom, AUCKLAND
Telex: 17-74-507 HEWPACK MEX
Tel: 687-159
A,C,CM,E,M,P
Cable: HEWPAK Auckland
Hewlett-Packard De Mexico (Polanco) C,CM,E,P'
Avenida Ejercito Nacional #579
Hewlett-Packard (N.l.) Ltd.
2day3er piso
4-12 Cruickshank Street
Colonia Granada 11560
Kilbirnie, WELLINGTON 3
MEXICO D.F.
P.O. Box 9443
Tel: 254-4433
Courtenay Place, WELLINGTON 3
P
Tel: 877-199
Hewlett-Packard De Mexico, SA
Cable: HEWPACK Wellington
de C.V.
C,CM,E,P
Czda. del Valle
Northrop
Instruments & Systems Ltd.
409 Ote. 4th Piso
369 Khyber Pass Road
Colonia del Valle
P.O. Box 8602
Municipio de Garza GarciA
AUCKLAND
66220 MONTERREY, Nuevo Le6n
Tel: 794-091
Tel: 784241
Telex: 60605
Telex: 038410
A,M
P

MEXICO (Cont'd)

MOROCCO
Etablissement Hubert Dolbeau & Fils
81 rue Karatchi
B.P. 11133
CASABLANCA
Tel: 3041-82, 3068-38
Telex: 23051, 22822
E
Gerep
2, rue Agadir
Boite Postale 156
CASABLANCA 01
Tel: 272093, 272095
Telex: 23739
P
Sema-Maroc
Dept. Seric
6, rue Lapebie
CASABLANCA
Tel: 260980
Telex: 21641
C,P

NETHERLANDS
Hewlett-Packard Nederland B.V.
Startbaan 16
1187 XR AMSTELVEEN
P.O. Box 667
NL1180 AR AMSTELVEEN
Tel: (020) 547·6911
Telex: 13216 HEPA NL
A,C,CM,E,M,P
Hewlett-Packard Nederland B.V.
Bongerd 2
NL 2906VK CAPELLE AID IJSSEL
P.O. Box 41
NL 2900AA CAPELLE AID IJSSEL
Tel: (10) 51-64-44_
Telex: 21261 HEPAC NL
C,E

Northrop Instruments & Systems Ltd.
110 Mandeville SI.
P.O. Box 8388
CHRISTCHURCH
Tel: 488-873
Telex: 4203
A,M
Northrop Instruments & Systems Ltd.
Sturdee House
85-87 Ghuznee Street
P.O. Box 2406
WELLINGTON
Tel: 850-091
Telex: Nl 3380
A,M

NIGERIA
Elmeco Nigeria Ltd.
46, Calcutta Crescent Apapa
P.O. Box 244and
LAGOS
E

NORTHERN IRELAND
See United Kingdom
NORWAY
Hewlett-Packard Norge A/S
Folke Bernadottes vei 50
P.O. Box 3558
N·5033 mLlNGSDALEN (Bergen)
Tel: 0047/5/16 55 40
Telex: 76621 hpnas n
C,E,M
Hewlett-Packard Norge A/S
Osterndalen 16-18
P.O. Box 34
N-1345 OiTERAs
Tel: 0047/2/17 11 80
Telex: 76621 hpnas n
A,C,CM,E,M,P

OMAN

PORTUGAL

Khimjil Ramdas
P.O. Box 19
MUSCAT/SULTANATE OF OMAN
Tel: 745601
Telex: 5289 BROKER MB MUSCAT
P

Mundinter Intercambio
Mundial de Com~rcio S.A.R.L.
Av. Antonio Augusto Aguiar 138
Apartado 2761
LISBON
Tel: (19) 53-21-31,53-21-37
Telex: 16691 munter p
M

SUhaii & Saud Bahwan
P.O.Box 169
MUSCAT/SULTANATE OF OMAN
Tel: 734201
Telex: 5274 BAHWAN MB
E
Imtac LLC
P.O. Box 8676
MUTRAHISULTANATE OF OMAN
Tel: 601695
Telex: 5741 Tawoos On
A,C,M

PAKISTAN
Mushko & Company Ltd.
House No. 16, Street No. 16
Sector F-6/3
ISLAMABAD
Tel: 824545
Cable: FEMUS Islamabad
A,E,M,P'
Mushko & Company Ltd.
Oosman Chambers
Abdullah Haroon Road
KARACHI 0302
Tel: 524131, 524132
Telex: 2894 MUSKO PK
Cable: COOPERATOR Karachi
A,E,M,P'

PANAMA
Electronico Balboa, SA
Calle Samuel Lewis, Ed. Alfa
Apartado 4929
PANAMA 5
Tel: 64-2700
Telex: 3483 ELECTRON PG
A,CM,E,M,P

PERU
Cia Electro M~dica SA
Los Flamencos 145, Otc. 30112
San Isidro
Casilla 1030
LIMA 1
Tel: 41-4325, 41-3705
Telex: Pub. Booth 25306 PEC PISIDR
CM,E,M,P
SAMS
Arenida Republica de Panama 3534
San Isidro, LIMA
Tel: 419928/417108
Telex: 20450 PE L1BERTAD
A,C,P

Soquimica
Av. da Liberdade, 220-2
1298 L1SIOA Codex
Tel: 56-21-82
Telex: 13316 SABASA
A
Telectra-Empresa T6cnica de
Equipmentos EI~ctricos S.A.R.L.
Rua Rodrigo da Fonseca 103
P.O. Box 2531
LISBON 1
Tel: (19) 68-60-72
Telex: 12598
CM,E
C.P.C.S.I.
Rua de Costa C!ibral 575
4200 PORTO
Tel: 499174/495173
Telex: 26054
C,P

PUERTO RICO
Hewlett-Packard Puerto Rico
101 Mu~oz Rivera Av
Esu. Calle Ochoa
HATO REY, Puerto Rico 00918
Tel: (809) 754-7800
A,C,CM,M,E,P

QATAR
Computer Arabia
P.O. Box 2750
DOHA
Tel: 428555
Telex: 4806 CHPARB
P
Nasser Trading & Contracting
P.O.Box 1563
DOHA
Tel: 422170
Telex: 4439 NASSER DH
M

SAUDI ARABIA

PHILIPPINES

SCORAND
See United Kingdom
SENEGAL
Societe Hussein Ayad & Cie.
76, Avenue Georges Pompidou
B.P.305
DAKAR
Tel: 32339
Cable: AYAD-Dakar
E
Moneger Distribution SA
1, Rue Parent
B.P.148
DAKAR
Tel: 215671
Telex: 587
P
Systeme Service Conseil (SSe)
14, Avenue du Parachois
DAKAR ETOILE
Tel: 219976
Telex: 577
C,P

SINGAPORE
Hewlett-Packard Singapore (Sales)
Pte. Ltd.
08-00 Inchcape House
450-2 Alexandra Road
Alexandra P.O. Box 58
SINGAPORE,9115
Tel: 4731788
Telex: 34209 HPSGSO RS
Cable: HEWPACK, Singapore
A,C,E,M,P
Dynamar International Ltd.
Unit 05-11 Block 6
Kolam Ayer Industrial Estate
SINGAPORE 1334
Tel: 747-6188
Telex: 26283 RS
CM

SOUTH AFRICA

Modern Electronics Establishment
Hewlett-Packard Division
P.O. Box 281
Thouqbah
AL-KHOBAR 31952
Tel: 895-1760, 895-1764
Telex: 671 106 HPMEEK SJ
Cable: ELECTA AL·KHOBAR
C.E,M

Hewlett-Packard So Africa (Ply.) Ltd.
P.O. Box 120
Howard Place CAPE PROVINCE 7450
Pine Park Center, Forest Drive, Pinelands
CAPE PROVINCE 7405
Tel: (021) 53 7954
Telex: 57·20006
A,C,CM,E,M,P

Modern Electronics Establishment
Hewlett-Packard Division
P.O. Box 1228

Hewlett-Packard So Atrica (Pty.) Ltd.
2nd Floor Juniper House
92 Overport Drive
DURBAN 4067
Tel: (031) 28·4178
Telex: 6-22954
C

oIEDDAH
The Online Advanced Systems Corp.
2nd Floor, Electra House
115-117 Esteban Street
Legaspi Village, Makati
P.O. Box 1510
Metro MANILA
Tel: 815-38-10 (up to 16)
Telex: 63274 ONLINE PN
A,C,E,M,P

Abdul Ghani EI Ajou Corp.
P.O. Box 78
RIYADH
Tel: 4041 717
Telex: 200 931 EL AJOU
P

Tel: 644 96 28
Telex: 4027 12 FARNAS SJ
Cable: ELECTA JEDDAH
A,C,CM,E,M,P
Modern Electronics Establishment
Hewlett-Packard Division
P.O.Box 22015
RIYADH 11495
Tel: 476-3030
Telex: 202049 MEERYD SJ
A,C,CM,E,M,P

Hewlett-Packard So Africa (Pty.) Ltd.
6 Linton Arcade
511 Cape Road
Linton Grange
PORT ELIZABETH 6001
Tel: 041-301201
Telex: 24·2916
C

0

~;]

0

~il

SALES & SUPPORT OFFICES
Arranged alphabetically by country

SOUTH AFRICA (Cont'd)
Hewlett-Packard So Africa (Pty.) Ltd.
Fountain Center
Kalkden Str.
Monument Park Ext 2
PRETORIA 0105
Tel: (012) 45 57258
Telex: 3-21063
C.E
Hewlett-Packard So Africa (Pty.) Ltd,
Private Bag Wendywood
SANDTON 2144
Tel: 802-5111, 802-5125
Telex: 4·20877 SA
Cable: HEWPACK Johannesburg
A,C,CM,E,M,P

SPAIN
Hewlett-Packard Espanola SA
Calle Entenza, 321
08029 BARCELONA
Tel: 3/3222451, 321 7354
Telex: 52603 hpbee
A,C,E,M,P
Hewlett·Packard Espanola SA
Calle San Vicente SIN
Edificio Albia 11-7B
48001 BILBAO
Tel: 4/4238306
A,C,E,M
Hewlett-Packard Espanola SA
Crta. de la Coruna, Km. 16,400
Las Rozas
E-MADRID
Tel: (1) 637.00,11
Telex: 23515 HPE
C,M
Hewlett-Packard Espanola SA
Avda, S. Francisco Javier, SIN
Planta 10. Edificio Sevilla 2
41005 SEVILLA
Tel: 54/64 44 54
Telex: 72933
A,C,M,P
Hewlett-Packard Espanola SA
Isabel La Catolica, 8
46004 VALENCIA
Tel: 0034/6/351 5944
C,P

SWEDEN
Hewlett-Packard Sverige AB
Ostra Tullgatan 3
S-21128 MALMO
Tel: (040) 70270
Telex: (854) 17886 (via SpAnga
office)
C,P
Hewlett-Packard Sverige AB
Skalholtsgatan 9, Kista
Box 19
S-16393 SPANGA
Tel: (08) 750-2000
Telex: (854) 17886
Telefax: (08) 7527781
A,C,CM,E,M,P
Hewlett-Packard Sverige AB
Frotallsgatan 30
S-42132 VA8TRA-FR6LUNDA (Gothenburg)
Tel: (031) 49-09-50
Telex: (854) 17886 (via SpAnga
office)
A,C,CM,E,M,P

Bangkok Business Equipment Ltd.
SUDAN
Mediterranean Engineering & Trading 5/5-6 Dejo Road
Co. Ltd.
BANGKOK
P,O. Box 1025
Tel: 234-8670, 234-8671
KHARTOUM
Telex: 87699-BEQUIPT TH
Tel: 41184
Cable: BUSIQUIPT Bangkok
Telex: 24052
P
C,P

TOGO
SWITZERLAND
Hewlett-Packard (Schweiz) AG
Clarastrasse 12
CH·4058 BASEL
Tel: (61) 33-59-20
A
Hewlett-Packard (Schweiz) AG
7, rue du Bois-du-Lan
Case postale 365
CH-1217 MEYRIN 1
Tel: (0041) 22-83-11-11
Telex:27333 HPAG CH
C,CM
Hewlett-Packard (Schweiz) AG
Allmend 2
CH-8967 WIDEN
Tel: (0041) 5731 21 11
Telex: 53933 hpag ch
Cable: HPAG CH
A,C,CM,E,M,P

SYRIA
General Electronic Inc.
Nuri Basha Ahnaf Ebn Kays Street
P.O. Box 5781
DAMASCUS
Tel: 33-24-87
Telex: 411 215
Cable: ELECTROBOR DAMASCUS
E
Middle East Electronics
P,O.Box 2308
Abu Rumaneh
DAMASCUS
Tel: 334592
Telex: 411 771
M

TAIWAN
Hewlett-Packard Taiwan
Kaohsiung Office
11/F, 456, Chung Hsiao 1st Road
KAOHSIUNG
Tel: (07) 2412318
C,E
Hewlett-Packard Taiwan
8th Floor, Hewlett-Packard Building
337 Fu Hsing North Road
TAIPEI
Tel: (02) 712-0404
Telex: 24439 HEWPACK
Cable:HEWPACK Taipei
A,C,CM,E,M,P
Ing Lih Trading Co,
3rd Floor, 7 Jen-Ai Road, Sec. 2

TAIPEI 100
Tel: (02) 3948191
Cable: INGLIH Taipei
A

THAILAND
Unimesa Co. Ltd.
30 Patpong Ave., Suriwong
BANGKOK 5
Tel: 235-5727
Telex: 84439 Simonco TH
Cable: UNIMESA Bangkok
A,C.E,M

Societe Africaine De Promotion
lmmeuble Sagap
22, Rue d'Atakpame
B.P.4150
LOME
Tel: 21-62-88
Telex: 5304
P

TRINIDAD' TOBAGO
Caribbean Telecoms Ltd.
Corner McAllister Street &
Eastern Main Road, Laventille
P,O. Box 732
PORT-oF·SPAlN
Tel: 624-4213
Telex: 22561 CARTEL WG
Cable: CARTEL, PORT OF SPAIN
CM,E,M,P
Computer and Controls Ltd.
P.O. Box 51
66 Independence Square
PORT·OF·SPAIN
Tel: 62·279-85
Telex: 3000 POSTLX WG, ACCT
L0090 AGENCY 1264
A,P
Feral Assoc.
8 Fitzgerald Lane
PORT-oF·SPAlN
Tel: 62-36864, 62-39255
Telex: 22432 FERALCO
Cable: FERALCO
M

TUNISIA
Precision Electronique S.A.R.L.
31 Avenue de la Liberte
TUNIS
Tel: 893937
Telex: 13238
P
Tunisie Electronique S.A.R.L.
94, Av, Jugurtha, Mutuelleville
1002 TUNIS·BELVEDERE
Tel: 280144
Telex: 13238
C,E,P
Corema SA
23, bis Rue de Marseille
TUNIS
Tel: 253-821
Telex: 14812 CABAM TN
M

TURKEY
EM.A
Mediha Eidem Sokak No. 41/6
Yenisehir

ANKARA
Tel: 319175
Telex: 46912 KTX TR
Cable: EMATRADE ANKARA
M

Teknim Company Ltd,
Iran Caddesi No, 7
Kavaklidere
ANKARA
Tel: 275800
Telex: 42155 TKNM TR
E,CM

Hewlett-Packard Ltd.
Bridewell House
9 Bridewell Place
LONDON EC4V 6BS
Tel: 01 5836565
Telex: 298163
C,P

Saniva Bilgisayar Sistemleri A.S.
Buyukdere Caddesi 103/6
Gayrettene
ISTANBUL
Tel: 1727030
Telex: 26345 SANI TR
C,P

Hewlett-Packard Ltd.
Pontefract Road
NORMANTON, West Yorkshire WF6 lRN
Tel: 0924895566
Telex: 557355
C,P

Best Inc.
Esentepe, Gazeteciler Sitesi
Keskin Kalemy
Sokak 6/3, Gayrettepe
ISTANBUL
Tel: 1721328
Telex: 42490
A

UNITED ARAB
EMIRATES
Emitac Ltd.
P,O. Box 1641
SHWAH
Tel: 591181
Telex: 68136 EMITAC EM
Cable: EMITAC SHARJAH
E,C,M,P,A
Emitac Ltd.
P.O. Box 2711
ABU DHABI
Tel: 820419-20
Cable: EMITACH ABUDHABI
Emitac Ltd.
P.O. Box 8391
DUBAI,
Tel: 377591
Emitac Ltd.
P,O. Box 473
RAS AL KHAIMAH
Tel: 28133, 21270

UNITED KINGDOM
GREAT BRITAIN
Hewlett-Packard Ltd.
Trafalgar House
Navigation Road
ALTRINCHAM
Cheshire WA 14 lNU
Tel: 061 928 6422
Telex: 668068
A,C,E,M,P
Hewlett-Packard Ltd.
Miller House
The Ring, IRACKNEU
Berks RG12 1XN
Tel: 0344424898
Telex: 848733
E

Hewlett-Packard Ltd.
The Quadrangle
106-118 Station Road
REDHILL, Surrey RH 1 1PS
Tel: 0737 68655
Telex: 947234
C,E,P
Hewlett-Packard Ltd.
Avon House
435 Stratford Road
Shirley, SOLIHUU, West Midlands
BOO 4BL
Tel: 021 7458800
Telex: 339105
C,E,P
Hewlett-Packard Ltd.
West End House
41 High Street, West End
SOUTHAMPTON
Hampshire S03 3DO
Tel: 0703 476767
Telex: 477138
C,P
Hewlett-Packard Ltd.
Harmon House
No. 1George Street
UXBRIDGE, Middlesex UX8 1YH
Tel: 895 720 20
Telex: 893134/5
C,CM,E,M,P
Hewlett-Packard Ltd,
King Street Lane
Winnersh, WOKINQHAM
Berkshire RG 11 5AR
Tel: 0734784774
Telex: 847178
A,C,E,M,P

IRELAND

NORTHERN IRELAND
Hewlett·Packard (Ireland) Ltd.
Carrickfergus Industrial Centre
75 Belfast Road, Carrickfergus
BELFA8T BT38 8PH
Tel: 0960367333
Telex: 747626
C,E

SCOTLAND

Hewlett·Packard Ltd.
Elstree House, Elstree Way
BOREHAMWOOD, Herts WD6 lSG
Tel: 01 2075000
Telex: 8952716
C,E

Hewlett-Packard Ltd.
8 Woodside Place
GLA8GOW, G3 7QF
Tel: 041 332 6232
Telex: 779615
C,E

Hewlett-Packard Ltd.
Oakfield House, Oakfield Grove
.Clifton BRISTOL, Avon BS8 2BN
Tel: 0272 736806
Telex: 444302
C,E,P

Hewlett-Packard Ltd.
80UTH QUEEN8FERRY
West Lothian, EH30 9TG
Tel: 0313311188
Telex: 72682
C,CM,E,M,P

SALES & SUPPORT OFFICES
Arranged alphabetically by country

UNITED STATES
Alabama
Hewlell-Packard Co.
700 Century Park South, Suite 128
BIRMINGHAM, AL 35226
Tel: (205) 822-6802
A,C,M,P*
Hewlell-Packard Co.
420 Wynn Drive
HUNTSVILLE, AL 35805
Tel: (205) 830-2000
C,CM,E,M*

Alaska
Hewlell-Packard Co.
3601 CSt., Suite 1416
ANCHORAGE, AK 99503
Tel: (907) 563-8855
C,E

Arizona
Hewlell-Packard Co.
8080 Pointe Parkway West
PHOENIX, AZ 85044
Tel: (602) 273-8000
A,C,CM,E,M,P
Hewlell-Packard Co.
3400 East Britannia Dr.
Bldg. C, Suite 124
TUCSON, AZ 85706
Tel: (602) 573-7400
C,E,M**

California
Hewlell-Packard Co.
99 South Hill Dr.
BRISBANE, CA 94005
Tel: (415) 330-2500
C
Hewlell-Packard Co.
5060 E. Clinton Avenue, Suite 102
FRESNO, CA 93727
Tel: (209) 252-9652
C,M
Hewlell-Packard Co.
1421 S. Manhallan Av.
FULLERTON, CA 92631
Tel: (714) 999-6700
C,CM,E,M
Hewlell-Packard Co.
7408 Hollister Ave. fA
GOLETA, CA 93117
Tel: (805) 685-6100
C,E

Hewlett-Packard Co.
9606 Aero Drive
SAN DIEGO, CA 92123
Tel: (619) 279-3200
C,CM,E,M
Hewlell-Packard Co.
5725 W. Las Positas Blvd.
Pleasanton, CA 94566
Tel: (415) 460-0282
C
Hewlell-Packard Co.
3003 Scoll Boulevard
SANTA CLARA, CA 95054
Tel: (408) 988-7000
Telex: 910-338-0586
A,C,CM,E
Hewlell-Packard Co.
2150 W. Hillcrest Dr.
THOUSAND OAKS, CA 91320
(805) 373-7000
C,CM,E

Colorado
Hewlell-Packard Co.
2945 Center Green Court South
Suite A
BOULDER, CO 80301
Tel: (303) 938-3005
A,C,E
Hewlell-Packard Co.
24 Inverness Place, East
ENGLEWOOD, CO 80112
Tel: (303) 649-5000
A,C,CM,E,M

Connecticut

Georgia

Kan..1

Hewlell-Packard Co.
2000 South Park Place
ATLANTA, GA 30339
Tel: (404) 955-1500
Telex: 810-766-4890
A,C,CM,E,M,P*

Hewlell-Packard Co.
7804 East Funston Road, 203
WICHITA, KS 67207
Tel: (316) 684-8491
C,E

Hewlell-Packard Co.
3607 Parkway Lane
Suite 300
NORCROSS, GA 30092
Tel: (404) 448-1894
C,E,P

Hawaii

Maryland

Hewlell-Packard Co.
11309 Chinden Blvd.
BOISE, 10 83707
Tel: (208) 323-2700
C

Hewlell-Packard Co.
3701 Koppers Street
BALTIMORE, MD 21227
Tel: (301) 644-5800
Telex: 710-862-1943
A,C,CM,E,M

Illinoll

Hewlell-Packard Co.
2 Choke Cherry Road
ROCKVILLE, MD 20850
Tel: (301) 948-6370
A,C,CM,E,M

Hewlett-Packard Co.
47 Barnes Industrial Road South
WALI.INQFORD, CT 06492
Tel: (203) 265-7801
A,C,CM,E,M

Hewlell-Packard Co.
1200 East Diehl Road
NAPERVILLE, IL 60566
Tel: (312) 357-8800
C
Hewlell-Packard Co.
5201 Tollview Drive
ROLLING MEADOWS, IL 60008
Tel: (312) 255-9800
Telex: 910-687-1066
A,C,CM,E,M

Hewlell-Packard Co.
6800 South Point Parkway
Suite 301
~ACKSONYW, FL 32216
Tel: (904) 398-0663
C*,MOO

Indiana

Hewlell-Packard Co.
6177 Lake Ellenor Drive
ORLANDO, FL 32809
Tel: (305) 859-2900
A,C,CM,E,P*

Hewlell-Packard Co.
3702 Rupp Drive
FT. WAYNE, IN 46815
Tel: (219) 482-4283
C,E

Iowa

Hewlell-Packard Co.
3155 Porter Drive
PALO ALTO, CA 94304
Tel: (415) 857-8000
C,E

Hewlell-Packard Co.
4700 Bayou Blvd.
Building 5
PENSACOLA, FL 32503
Tel: (904) 476-8422
A,C,M

Hewlell-Packard Co.
4070 22nd Av. SW
CEDAR RAPIDS, IA 52404
Tel: (319) 390-4250
C,E,M

Hewlell-Packard Co.
4244 So. Market Court, Suite A
SACRAMENTO, CA 95834
Tel: (916) 929-7222
A*,C,E,M

Hewlell-Packard Co.
5550 W. Idlewild, 150
TAMPA, FL 33614
Tel: (813) 884-3282
C,E,M,P

Hewlell-Packard Co.
4201 Corporate Dr.
WEST DESIIOINES, IA 50265
Tel: (515) 224-1435
AOO,C,MOO

Hewlell-Packard Co.
5400 W. Rosecrans Blvd.
LAWNDALE, CA 90260
Tel: (213) 643-7500
Telex: 910-325-6608
C,M
Hewlell-Packard Co.
2525 Grand Avenue
Long Beach, CA 90815
Tel: (213) 498-1111
C

Hewlell-Packard Co.
11911 N. Meridian St.
CARMEL, IN 46032
Tel: (317) 844-4100
A,C,CM,E,M

Hewlell-Packard
10824 Old Mill Rd., Suite 3
OMAHA, NE 68154
Tel: (402) 334-1813
C,E,M

New Jersey
Louiliana

Idaho

Hewlell-Packard Co.
525 W. Monroe, 1308
CHICAGO, IL 60606
Tel: (312) 930-0010
C

Hewlett-Packard Co.
2901 N.W. 62nd Street
FORT LAUDERDALE, FL 33309
Tel: (305) 973-2600
C,E,M,P*

Hewlell-Packard Co.
10300 Linn Station Road, 100
LOUISVILLE, KY 40223
Tel: (502) 426-0100
A,C,M

Hewlell-Packard Co.
160 James Drive East
ST. ROSE, LA 70087
P.O. Box 1449
KENNER, LA 70063
Tel: (504) 467-4100
A,C,E,M,P

Hewlell-Packard Co.
500 Sylvan Av.
BRIDGEPORT, CT 06606
Tel: (203) 371-6454
C,E

Florida

Nebralka
Kentucky

Hewlett-Packard Co.
Kawaiahao Plaza, Suite 190
567 South King Street
HONOlULU, HI 96813
Tel: (808) 526-1555
A,C,E,M

Hewlell-Packard Co.
304 Eldorado Road
P.O. Box 1607
BLOOMINGTON, IL 61701
Tel: (309) 662-9411
C,MOO

Hewlell-Packard Co.
13001 Hollenberg Drive
BRIDGETON, MO 63044
Tel: (314) 344-5100
A,C,E,M

Malsachusettl
Hewlell-Packard Co.
1775 Minuteman Road
ANDOVER, MA 01810
Tel: (617) 682-1500
A,C,CM,E,M,P*
Hewlell-Packard Co.
32 Hartwell Avenue
LEXINGTON, MA 02173
Tel: (617) 861-8960
C,E

Michigan
Hewlell-Packard Co.
4326 Cascade Road S.E.
GRAND RAPIDS, MI 49506
Tel: (616) 957-1970
C,M
Hewlell-Packard Co.
39550 Orchard Hill Place Drive
HOVI, MI 48020
Tel: (313) 349-9200
A,C,E,M
Hewlell-Packard Co.
1771 W. Big Beaver Road
TROY, MI 48084
Tel: (313) 643-6474
C

Minnesota
Hewlell-Packard Co.
2025 W. Larpenteur Ave.
ST. PAUL, MN 55113
Tel: (612) 644-1100
A,C,CM,E,M

Miliourl
Hewlett-Packard Co.
1001 E. 101st Terrace Suite 120
KANSAS CITY, MO 64131-3368
Tel: (816)941-0411
A,C,CM,E,M

Hewlell-Packard Co.
120 W. Century Road
PARAMUS, NJ 07653
Tel: (201) 265-5000
A,C,CM,E,M
Hewlell-Packard Co.
20 New England Av. West
PISCATAWAY, NJ 08854
Tel: (201) 562-6100
A,C,CM,E

New Mexico
Hewlell-Packard Co.
7801 Jefferson N.E.
ALBUQUERQUE, NM 87109
Tel: (505) 292-1330
C,E,M

New York
Hewlell-Packard Co.
5 Computer Drive South
ALBANY, NY 12205
Tel: (518) 458-1550
A,C,E,M
Hewlell-Packard Co.
9600 Main Street
CLARENCE, NY 14031
Tel: (716) 759-8621
C,E
Hewlell-Packard Co.
200 Cross Keys Office Park
FAIRPORT, NY 14450
Tel: (716) 223-9950
A,C,CM,E,M
Hewlell-Packard Co.
7641 Henry Clay Blvd.
UVERPOOL, NY 13088
Tel: (315) 451-1820
A,C,CM,E,M
Hewlell-Packard Co.
No. 1 Pennsylvania Plaza
55th Floor
34th Street & 8th Avenue
MANHAnAN NY 10119
Tel: (212) 971-0800
C,M*
Hewlell-Packard Co.
15 Myers Corner Rd.
Hollowbrook Park, Suite 20
WAPPINGER FALLS, NY 12590
CM,E
Hewlell-Packard Co.
250 Westchester Avenue
WHITE PLAINS, NY 10604
Tel: (914) 684-6100
C,CM,E
Hewlell-Packard Co.
3 Crossways Park West
WOODBURY, NY 11797
Tel: (516) 682-7800
A,C,CM,E,M

[2]
l6!l

o SALES & SUPPORT OFFICES
~il

Arranged alphabetically by country

UNITED STATES (Cont'd)
North Carolina
Hewlett-Packard Co.
305 Gregson Dr.
CARY, NC 27511
Tel: (919) 467-6600
C,CM,E,M,P*

Oregon
Hewlett-Packard Co.
9255 S. W. Pioneer Court
WILSONVILLE, OR 97070
Tel: (503) 682-8000
A,C,E*,M

Pennsylvania

Hewlett-Packard Co.
9600-H Southern Pine Blvd.
CHARLOm, NC 28210
Tel: (704) 527-8780
C*

Hewlett-Packard Co.
50 Dorchester Rd.
HARRISBURG, PA 17112
Tel: (717) 657-5900
C

Hewlett-Packard Co.
5605 Roanne Way
GREENSBORO, NC 27420
Tel: (919) 852-1800
A,C,CM,E,M,P*

Hewlett-Packard Co.
111 Zeta Drive
PITTSBURGH, PA 15238
Tel: (412) 782-0400
A,C,E,M

Ohio
Hewlett-Packard Co.
2717 S. Arlington Road
AKRON, OH 44312
Tel: (216) 644-2270
C,E
Hewlett-Packard Co.
23200 Chagrin Blvd #100
BEACHWOOD, OH 44122
Tel: (216) 292-4677
C,P
Hewlett-Packard Co.
9920 Carver Road
CINCINNAn,OH 45242
Tel: (513) 891-9870
C,M
Hewlett-Packard Co.
16500 Sprague Road
CLEVELAND, OH 44130
Tel: (216) 243-7300
A,C,CM,E,M
Hewlett-Packard Co.
9080 Springboro Pike
MIAMISBURG, OH 45342
Tel: (513) 433-2223
A,C,CM,E*,M
Hewlett-Packard Co.
One Maritime Plaza, 5th Floor
720 Water Street
TOLEDO,OH43604
lei: (419) 242-2200
C
Hewlett-Packard Co.
675 Brooksedge Blvd.
WESTERYUE, OH 43081
Tel: (614) 891-3344
C,CM,E*

Oklahoma
Hewlett-Packard Co.
3525 N.W. 56th SI.
Suite C·loo
OKLAHOMA CITY, OK 73112
Tel: (405) 946-9499
C,E*,M
Hewlett-Packard Co.
3840 S. 103rd E. Ave., 100
TULSA, OK 74146
Tel: (918) 665-3300
A**,C,E,M*,P*

SEPT. 18B!

Hewlett-Packard Co.
2750 Monroe Boulevard
YALLEY FORGE, PA 19482
Tel: (215) 666-9000
A,C,CM,E,M

South Carolina
Hewlett-Packard Co.
Brookside Park, Suite 122
1Harbison Way
COLUMBIA, SC 29210
Tel: (803) 732-0400
C,M
Hewlett-Packard Co.
555 N. Pleasantburg Dr.
Suite 107
GREENVILLE, SC 29607
Tel: (803) 232-8002
C

Texas

Washington

Hewlett-Packard Co.
1826-P Kramer Lane
AUSnN, TX 78758
Tel: (512) 835-6771
C,E,P*

Hewlett-Packard Co.
15815 S.E. 37th Street
BELLEVUE, WA 98006
Tel: (206) 643-4000
A,C,CM,E,M

Hewlett-Packard Co.
5700 Cromo Dr
EL PASO, TX 79912
Tel: (915) 833-4400
C,E*,M**

Hewlett-Packard Co.
708 North Argonne Road
SPOKANE, WA 99212-2793
Tel: (509) 922-7000
C

Hewlett-Packard Co.
3952 Sandshell Drive
FORT WORTH, TX 76137
Tel: (817) 232-9500
C

West Virginia

Hewlett-Packard Co.
10535 Harwin Drive
HOUSTON, TX 77036
Tel: (713) 776-6400
A,C,E,M,P*
Hewlett-Packard Co.
511 E. John W. Carpenter Fwy.
Royal Tech. Center 100
IRYlNG, TX 75062
Tel: (214) 556-1950
C,E
Hewlett-Packard Co.
109 E. Toronto, Suite 100
McALLEN, TX 78503
Tel: (512) 630-3030
C
Hewlett-Packard Co.
930 E. Campbell Rd.
RICHARDSON, TX 75081
Tel: (214) 231-6101
A,C;CM,E,M,P*

Tennessee
Hewlett-Packard Co.
One Energy Centr. 200
Pellissippi Pkwy.
KNOXYW, TN 37932
Tel: (615) 966-4747
A,C,M

Hewlett-Packard Co.
1020 Central Parkway South
SAN ANTONIO, TX 78216
Tel: (512) 494-9336
A,C,E,M,P*

Hewlett-Packard Co.
3070 Directors Row
Directors Square
MEMPHIS, TN 38131
Tel: (901) 346-8370
A,C,M

Utah

Hewlett-Packard Co.
220 Great Circle Road, Suite 116
NASHVILLE, TN 37228
Tel: (615) 255-1271
C,M,P*

Hewlett-Packard Co.
3530 W. 2100 South
SALT WE CITY, UT 84119
Tel: (801) 974-1700
A,C,E,M

Virginia
Hewlett-Packard Co.
4305 Cox Road
GLEN ALLEN, VA 23060
Tel: (804) 747-7750
A,C,E,M,P*
Hewlett-Packard Co.
Tanglewood West Bldg.
Suite 240
3959 Electric Road
ROANOKE, VA 24018
Tel: (703) 774-3444
C,E,P

Hewlett-Packard Co.
50156th
CHARLESTON, WV 25304
Tel: (304) 925-0492
A,C,M

Wisconsin
Hewlett-Packard Co.
275 N. Corporate Dr.
BROOKFIELD, WI 53005
Tel: (414) 784-8800
A,C,E*,M

URUGUAY
Pablo Ferrando SAC. e I.
Avenida lIalia 2877
Casilla de Corrao 370
MONTEVIDEO
Tel: 80-2586
Telex: 802586
A,CM,E,M
Olympia de Uruguay SA
Maquines de Oficina
Avda. del Libertador 1997
Casilla de Correos 6644
MONTEVIDEO
Tel: 91-1809, 98-3807
Telex: 6342 OROU UY
P

Hewlett-Packard de Venezuela CA
Residencias Tia Betty Local 1
Avenida 3 y con calle 75
MARACAIBO, Estado Zulia
Apartado 2646
Tel: (061) 75801-75805-7580680304
Telex: 62464 HPMAR
C,E*
Hewlett-Packard de Venezuela C.A.
Urb. Lomas de Este
Torre Trebol - Piso 11
YALENCIA, Estado Carabobo
Apartado 3347
Tel: (041) 222992/223024
C,P

YUGOSLAVIA
OC Hermes
General Zdanova 4
YU-ll000 BEOGRAD
Tel: 340327,342641
Telex: 11433
A,C,E,P
Hermes
Titova 50
YU-61000 UUBLJANA
Tel: 324856, 324858
Telex: 31583
C,E,M,P
Elektrotehna
Titova 51
YU-61000 UUBUANA
CM

ZAIRE
Computer & Industrial Engineering
25, Avenue de la Justice
B.P. 12797
KINSHASA, Gombe
Tel: 32063
Telex: 21552
C,P

VENEZUELA
Hewlett-Packard de Venezuela C.A.
3A Transversal Los Ruices Norte
Edificio Segre 2 & 3
Apartado 50933
CARACAS 1071
Tel: 239-4133
Telex: 251046 HEWPACK
A,C,CM,E,M,P
Hewlett-Packard de Venezuela, C.A.
Centro Civdad Comercial Tamanaco
Nivel C-2 (Nueva Etapa)
Local 53H05
Chuao, CARACAS
Tel: 928291
P
Albis Venezolana S.R.L.
Av. Las Marias, Ota. Alix,
EI Pedregal
Apartado 81025
CARACAS 1080A
Tel: 747984, 742146
Telex: 24009 ALBIS VC
A
Tecnologica Medica del Carlbe, C.A.
Multicentro Empresarial del Este
Ave. Libertador
Edl!. Libertador
Nucleo "C" - Oficina 51·52
CARACAS
Tel: 339867/333780
M

ZAMBIA
R.J. Tilbury (Zambia) Ltd.
P.O. Box 32792
LUSAKA
Tel: 215590
Telex: 40128
E

ZIMBABWE
' Field Technical Sales (Private) Limited
45, Kelvin Road North
P.O. Box 3458
HARARE
Tel: 705 231
Telex: 4-122 RH
E,P

Manual Part Number: 30242-90003
Printed in U.S.A., August 1987
Edition 2
E0887

Ft:' HEWLETT

aft.

PACKARD



Source Exif Data:
File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.4
Linearized                      : No
XMP Toolkit                     : Adobe XMP Core 4.0-c321 44.398116, Tue Aug 04 2009 14:24:30
Modify Date                     : 2013:07:08 07:36:43-07:00
Create Date                     : 2013:07:03 13:58:17-07:00
Metadata Date                   : 2013:07:08 07:36:43-07:00
Creator Tool                    : Adobe Acrobat 8.31
Format                          : application/pdf
Document ID                     : uuid:7a17a0bc-2c32-b644-9e93-07ef1e5c9e65
Instance ID                     : uuid:2741a78e-fa0b-c444-8b09-6494b59a35ac
Producer                        : Adobe Acrobat 8.31 Paper Capture Plug-in
Page Count                      : 210
Creator                         : Adobe Acrobat 8.31
EXIF Metadata provided by EXIF.tools

Navigation menu