Compaq Version 7.3 DECamds, Data Analyzer, Data Provider User’s Guide

DECamds User’s Guide

Order Number: AA–Q3JSE–TE

April 2001

This guide explains how to use DECamds software to detect and fix system availability problems. It also explains how to install DECamds.

Revision/Update Information:

This guide supersedes the DECamds

User’s Guide, Version 7.1.

Operating System and Version:

Data Analyzer:OpenVMS Alpha and

VAX Version 7.2 or later

Data Provider: OpenVMS Alpha and

VAX Version 6.2 or later

Software Version:

Compaq DECamds Version 7.3

Compaq Computer Corporation

Houston, Texas

© 2001 Compaq Computer Corporation

Compaq, VAX, VMS, and the Compaq logo Registered in U.S. Patent and Trademark Office.

OpenVMS is a trademark of Compaq Information Technologies Group, L. P. in the United States and other countries.

Motif, OSF/1, and UNIX are trademarks of The Open Group in the United States and other countries.

All other product names mentioned herein may be trademarks of their respective companies.

Confidential computer software. Valid license from Compaq required for possession, use, or copying.

Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software

Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor’s standard commercial license.

Compaq shall not be liable for technical or editorial errors or omissions contained herein. The information in this document is provided "as is" without warranty of any kind and is subject to change without notice. The warranties for Compaq products are set forth in the express limited warranty statements accompanying such products. Nothing herein should be construed as constituting an additional warranty.

ZK5929

The Compaq OpenVMS documentation set is available on CD-ROM.

This document was prepared using DECdocument, Version 3.3-1b.

Contents

Preface

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 Overview of DECamds

1.1

1.2

1.3

1.3.1

1.3.2

1.3.2.1

1.3.2.2

1.3.2.3

1.3.3

1.3.4

How Does DECamds Work? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Where to Install the DECamds Data Analyzer . . . . . . . . . . . . . . . . . . . . .

DECamds Security Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Understanding DECamds Security Files . . . . . . . . . . . . . . . . . . . . . . .

Customizing Security Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting Up Node Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Defining Data Exchange Access Between Nodes . . . . . . . . . . . . . .

Limiting Specific Users to Read Access . . . . . . . . . . . . . . . . . . . . .

Sending Messages to OPCOM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting Broadcast Intervals for Node Availability Messages . . . . . . . .

2 Getting Started

2.1

2.2

2.2.1

2.2.2

2.2.3

2.3

2.3.1

2.3.2

2.3.3

2.3.4

2.3.5

Starting DECamds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Using the System Overview Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Expanding and Collapsing Group Information . . . . . . . . . . . . . . . . . . .

Displaying Additional Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Stopping Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Using the Event Log Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Displaying Information About an Event Log Entry . . . . . . . . . . . . . . .

Performing Corrective Action on an Event Log Entry . . . . . . . . . . . . .

Sending Event Information to OPCOM . . . . . . . . . . . . . . . . . . . . . . . .

Removing an Event from the Event Log Window . . . . . . . . . . . . . . . . .

Retaining and Releasing an Event in the Event Log Window . . . . . . .

3 Managing DECamds Data Windows

3.1

3.2

3.3

3.4

3.5

3.6

3.7

3.8

3.9

Disk Status Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Volume Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Single Disk Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Page/Swap File Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Node Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Process I/O Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CPU Modes Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CPU Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Memory Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.10

Single Process Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.11

Lock Contention Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.12

Single Lock Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.13

Cluster Transition/Overview Summary Window . . . . . . . . . . . . . . . . . . . . .

ix

1–2

1–3

1–3

1–5

1–7

1–8

1–8

1–9

1–9

1–10

2–1

2–2

2–5

2–5

2–6

2–7

2–9

2–9

2–10

2–10

2–10

3–2

3–5

3–6

3–8

3–10

3–12

3–14

3–15

3–17

3–19

3–25

3–28

3–30 iii

3.13.1

Data Displayed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.13.2

Notes About the Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.14

System Communications Architecture Summary Window . . . . . . . . . . . . .

3.14.1


3.15

NISCA Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.15.1

3.15.2

Data Displayed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


4 Performing Fixes

4.1

4.2

4.2.1

4.2.2

4.2.3

4.2.4

4.2.5

4.2.6

4.2.7

4.2.8

4.3

4.3.1

4.3.2

Understanding Fixes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Performing Fixes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Adjust Quorum Fix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Adjust Process Quota Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Adjust Working Set Fix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Change Process Priority Fix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Crash Node Fix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Exit Image and Delete Process Fixes . . . . . . . . . . . . . . . . . . . . . . . . . .

Purge Working Set Fix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Suspend Process and Resume Process Fixes . . . . . . . . . . . . . . . . . . . .

Examples for Fixing Low Memory Availability . . . . . . . . . . . . . . . . . . . . . .

Performing a Fix Using Automatic Fix Settings . . . . . . . . . . . . . . . . .

Performing a Fix Using Manual Investigation . . . . . . . . . . . . . . . . . . .

5 Customizing DECamds

5.1

5.1.1

5.1.2

5.1.3

5.2

5.2.1

5.2.2

5.3

5.4

5.5

5.5.1

5.5.1.1

5.5.1.2

5.5.1.3

5.5.2

5.5.2.1

5.5.2.2

5.5.3

Customizing DECamds Defaults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting Default Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting Automatic Event Investigation . . . . . . . . . . . . . . . . . . . . . . . .

Setting Automatic Lock Investigation . . . . . . . . . . . . . . . . . . . . . . . . .

Filtering Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Filtering Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Customizing Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Sorting Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting Collection Intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Optimizing Performance with System Settings . . . . . . . . . . . . . . . . . . . . .

Optimizing DECamds Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting Process Quotas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting LAN Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting Window Customizations . . . . . . . . . . . . . . . . . . . . . . . . . . .

Optimizing System Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting Data Link Read Operations . . . . . . . . . . . . . . . . . . . . . . . .

Setting the Communications Buffer . . . . . . . . . . . . . . . . . . . . . . . .

Optimizing Performance with Hardware . . . . . . . . . . . . . . . . . . . . . . .

A Installing the DECamds Data Analyzer

A.1

A.2

A.3

A.4

A.5

A.5.1

A.6

General Installation Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Data Analyzer Installation Requirements . . . . . . . . . . . . . . . . . . . . . . . . .

Obtaining the Data Analyzer Software . . . . . . . . . . . . . . . . . . . . . . . . . . .

Installing Data Analyzer Software from a PCSI Kit . . . . . . . . . . . . . . . . .

Postinstallation Tasks on Data Provider Nodes . . . . . . . . . . . . . . . . . . . . .

Starting, Stopping, and Reloading DECamds . . . . . . . . . . . . . . . . . . . .

Postinstallation Tasks on a Data Analyzer Node . . . . . . . . . . . . . . . . . . . .

3–32

3–33

3–33

3–35

3–36

3–38

3–40

4–9

4–9

4–10

4–10

4–11

4–1

4–2

4–4

4–5

4–6

4–6

4–7

4–8

5–19

5–19

5–20

5–20

5–20

5–20

5–21

5–1

5–3

5–4

5–4

5–4

5–6

5–8

5–14

5–16

5–18

5–18

A–1

A–2

A–4

A–4

A–7

A–8

A–8 iv

Figures

3–3

3–4

3–5

3–6

3–7

3–8

3–9

3–10

3–11

3–12

3–13

3–14

3–15

3–16

3–17

1–1

2–1

2–2

2–3

2–4

2–5

3–1

3–2

A.7

A.8

A.9

Starting to Use the Data Provider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Determining and Reporting Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Running the Installation Verification Procedure Separately . . . . . . . . . . .

B DECamds Files and Logical Names

B.1

B.2

B.3

B.4

B.5

B.6

Files and Logical Names for the Data Analyzer Node . . . . . . . . . . . . . . . .

Files and Logical Names for Data Provider Nodes . . . . . . . . . . . . . . . . . . .

Log Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Event Log File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Lock Contention Log File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

OPCOM Log . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Glossary

Index

Examples

A–1

B–1

B–2

Sample OpenVMS Alpha Installation . . . . . . . . . . . . . . . . . . . . . . . . .

Sample Event Log File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Sample Lock Contention Log File . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DECamds Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

System Overview Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

System Overview Window Menus . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Event Log Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Event Log Window Menus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Event Display Choice Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DECamds Data Window Hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . .

Disk Status Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Volume Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Single Disk Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Page/Swap File Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Node Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Process I/O Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CPU Modes Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CPU Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Memory Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Single Process Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Lock Contention Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . .

Filtering Lock Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Single Lock Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Cluster Transition/Overview Summary Window . . . . . . . . . . . . . . . . . .

SCA Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

NISCA Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

A–9

A–10

A–10

B–1

B–2

B–4

B–5

B–5

B–6

A–6

B–5

B–6

3–5

3–7

3–8

3–10

3–12

3–14

3–15

3–17

3–19

3–25

3–27

3–28

3–31

3–34

3–37

1–3

2–3

2–5

2–7

2–8

2–9

3–1

3–3 v

Tables

2–2

3–1

3–2

3–3

3–4

3–5

3–6

3–7

1–1

1–2

1–3

1–4

2–1

3–8

3–9

3–10

3–11

3–12

3–13

3–14

3–15

3–16

3–17

3–18

4–11

5–1

5–2

5–3

5–4

5–5

5–6

5–7

4–1

4–2

4–3

4–4

4–5

4–6

4–7

4–8

4–9

4–10 vi

FIX Adjust Quorum Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

FIX Adjust Process Quota Limit Dialog Box . . . . . . . . . . . . . . . . . . . .

FIX Adjust Working Set Size Dialog Box . . . . . . . . . . . . . . . . . . . . . . .

FIX Process Priority Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

FIX Crash Node Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

FIX Process State Dialog Box — Exit Image or Delete Process . . . . . .

FIX Purge Working Set Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . .

FIX Process State Dialog Box — Suspend or Resume Process . . . . . . .

Sample Fix Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DECamds Memory Summary Window . . . . . . . . . . . . . . . . . . . . . . . . .

DECamds Node Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . .

DECamds Application Customizations Dialog Box . . . . . . . . . . . . . . . .

Event Qualification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CPU Summary Filtering Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . .

Customize Events Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

LOWSQU Event Customization Window . . . . . . . . . . . . . . . . . . . . . . .

Memory Summary Sorting Dialog Box . . . . . . . . . . . . . . . . . . . . . . . . .

Memory Summary Collection Interval Dialog Box . . . . . . . . . . . . . . . .

4–13

5–2

5–5

5–7

5–9

5–10

5–15

5–16

4–4

4–5

4–6

4–7

4–7

4–8

4–9

4–10

4–11

4–12

Security Triplet Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Security Triplet Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DECamds Logical Names for OPCOM Messages . . . . . . . . . . . . . . . . .

Broadcast Availability Logical Names . . . . . . . . . . . . . . . . . . . . . . . . .

System Overview Window Display Fields . . . . . . . . . . . . . . . . . . . . . .

Event Log Window Display Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DECamds Data Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Disk Status Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . .

Volume Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . . . . .

Single Disk Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . .

Page/Swap File Summary Window Data Fields . . . . . . . . . . . . . . . . . .

Node Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . . . . . . .

Process I/O Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . .

CPU Modes Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . .

CPU Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . . . . . . .

Memory Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . . . . .

Single Process Summary Window Data Fields . . . . . . . . . . . . . . . . . . .

Lock Contention Summary Window Data Fields . . . . . . . . . . . . . . . . .

Single Lock Summary Window Data Fields . . . . . . . . . . . . . . . . . . . . .

Data Items in the Summary Panel of the Cluster Transition/Overview

Summary Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Data Items in the Cluster Members Panel of the Cluster

Transition/Overview Summary Window . . . . . . . . . . . . . . . . . . . . . . . .

Data Items in the SCA Summary Window . . . . . . . . . . . . . . . . . . . . . .

Data Items in the Transmit Panel . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Data Items in the Receive Panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3–32

3–32

3–35

3–38

3–38

1–5

1–7

1–9

1–10

2–4

2–7

3–1

3–3

3–6

3–8

3–9

3–11

3–13

3–15

3–16

3–18

3–20

3–26

3–29

5–6

5–7

A–1

B–1

B–2

B–3

B–4

3–19

3–20

3–21

3–22

4–1

5–1

5–2

5–3

5–4

5–5

Data Items in the Congestion Control Panel . . . . . . . . . . . . . . . . . . . .

Data Items in the Channel Selection Panel . . . . . . . . . . . . . . . . . . . . .

Data Items in the VC Closures Panel . . . . . . . . . . . . . . . . . . . . . . . . .

Data Items in the Packets Discarded Panel . . . . . . . . . . . . . . . . . . . . .

Summary of DECamds Fixes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DECamds Application Defaults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Event Log Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CPU, I/O, and Memory Class Definitions . . . . . . . . . . . . . . . . . . . . . . .

Memory Summary Collection Interval Fields . . . . . . . . . . . . . . . . . . . .

Default Window Collection Intervals . . . . . . . . . . . . . . . . . . . . . . . . . .

LAN Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Monitoring Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Recommended System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . .

Files on the Data Analyzer Node . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Logical Names Defined for the Data Analyzer . . . . . . . . . . . . . . . . . . .

Files on Nodes Running the Data Provider . . . . . . . . . . . . . . . . . . . . .

Logical Names Defined on Nodes Running the Data Provider . . . . . . .

5–19

5–21

A–2

B–1

B–2

B–3

B–3

3–39

3–39

3–40

3–40

4–3

5–3

5–6

5–12

5–17

5–17 vii

Preface

Intended Audience

This guide is intended for system managers who install and use Compaq

DECamds software.

Document Structure

This guide contains the following chapters and appendixes:

• Chapter 1 describes an overview of DECamds software, where to install

DECamds, security features, and customizing security files.

• Chapter 2 describes how to start DECamds and use online help. It also describes the System Overview window and the Event Log window.

• Chapter 3 describes how to use the DECamds data windows.

• Chapter 4 describes how to take corrective actions, called fixes, to improve system availability.

• Chapter 5 describes the tasks you can perform to filter, sort, and customize the display of system data using DECamds. It also describes how some of these tasks can optimize the performance of DECamds.

• Appendix A contains instructions for installing DECamds.

• Appendix B contains a description of all files and logical names created when

DECamds is installed and gives examples of the log files that DECamds writes.

• The Glossary defines DECamds terminology.

Related Documents

The following manuals provide additional information:

• OpenVMS Version 7.3 Release Notes describes features and changes that apply to DECamds software.

• OpenVMS System Manager’s Manual describes tasks you perform to manage an OpenVMS system. It also describes installing a product with the

POLYCENTER Software Installation utility.

• OpenVMS System Management Utilities Reference Manual describes utilities you use to manage an OpenVMS system.

• OpenVMS Programming Concepts Manual explains OpenVMS lock management concepts.

•

OpenVMS System Messages: Companion Guide for Help Message Users

explains how to use help messages.

ix

x

• POLYCENTER Software Installation Utility User’s Guide describes the features you can request with the PRODUCT INSTALL command when starting an installation.

For additional information about Compaq OpenVMS products and services, access the Compaq website at the following location: http://www.openvms.compaq.com/

Reader’s Comments

Compaq welcomes your comments on this manual. Please send comments to either of the following addresses:

Internet

Mail

[email protected]

Compaq Computer Corporation

OSSG Documentation Group, ZKO3-4/U08

110 Spit Brook Rd.

Nashua, NH 03062-2698

How to Order Additional Documentation

Use the following World Wide Web address to order additional documentation: http://www.openvms.compaq.com/

If you need help deciding which documentation best meets your needs, call

800-282-6672.

Conventions

The following conventions are used in this guide:

Ctrl/x

PF1 x

Return

. . .

A sequence such as Ctrl/x indicates that you must hold down the key labeled Ctrl while you press another key or a pointing device button.

A sequence such as PF1 x indicates that you must first press and release the key labeled PF1 and then press and release another key or a pointing device button.

In examples, a key name enclosed in a box indicates that you press a key on the keyboard. (In text, a key name is not enclosed in a box.)

In the HTML version of this document, this convention appears as brackets, rather than a box.

Horizontal ellipsis points in examples indicate one of the following possibilities:

• Additional optional arguments in a statement have been omitted.

• The preceding item or items can be repeated one or more times.

• Additional parameters, values, or other information can be entered.

.

.

.

Vertical ellipsis points indicate the omission of items from a code example or command format; the items are omitted because they are not important to the topic being discussed.

( )

[ ]

|

{ }

bold text

italic text

UPPERCASE TEXT

Monospace text

numbers

In command format descriptions, parentheses indicate that you must enclose choices in parentheses if you specify more than one.

In command format descriptions, brackets indicate optional choices. You can choose one or more items or no items.

Do not type the brackets on the command line. However, you must include the brackets in the syntax for OpenVMS directory specifications and for a substring specification in an assignment statement.

In command format descriptions, vertical bars separate choices within brackets or braces. Within brackets, the choices are optional; within braces, at least one choice is required. Do not type the vertical bars on the command line.

In command format descriptions, braces indicate required choices; you must choose at least one of the items listed. Do not type the braces on the command line.

This typeface represents the introduction of a new term. It also represents the name of an argument, an attribute, or a reason.

Italic text indicates important information, complete titles of manuals, or variables. Variables include information that varies in system output (Internal error number), in command lines (/PRODUCER=name), and in command parameters in text (where dd represents the predefined code for the device type).

Uppercase text indicates a command, the name of a routine, the name of a file, or the abbreviation for a system privilege.

Monospace type indicates code examples and interactive screen displays.

In the C programming language, monospace type identifies the following elements: keywords, the names of independently compiled external functions and files, syntax summaries, and references to variables or identifiers introduced in an example.

A hyphen at the end of a command format description, command line, or code line indicates that the command or statement continues on the following line.

All numbers in text are assumed to be decimal unless otherwise noted. Nondecimal radixes—binary, octal, or hexadecimal—are explicitly indicated.

xi

1

Overview of DECamds

This chapter describes the following:

• Overview of DECamds

• Where to install the DECamds Data Analyzer

• DECamds security features

Compaq DECamds is a real-time monitoring, diagnostic, and correction tool that helps you improve OpenVMS system and OpenVMS Cluster availability.

DECamds also helps system programmers/analysts to target a specific node or process for detailed analysis, and system operators and service technicians to determine hardware and software issues.

DECamds simultaneously collects and analyzes system data and process data from multiple nodes and displays the output on a DECwindows Motif display.

Based on the analyzed data, DECamds detects events and proposes actions to correct resource availability and system denial issues in real time.

DECamds helps improve OpenVMS system and OpenVMS Cluster availability in the following ways:

Availability

Centralized management

Intuitive interface

Correction capability

Customization

Scalability

Alerts users to resource availability problems, suggests paths for further investigation, and recommends actions to improve availability.

Provides centralized management of remote nodes within an extended local area network (LAN).

Provides an easy-to-learn and easy-to-use DECwindows Motif user interface.

Allows real-time intervention, including adjustment of node and process parameters, even when remote nodes are hung.

Adjusts to site-specific requirements through a wide range of customization options.

Makes it easier to monitor multiple OpenVMS systems and OpenVMS

Cluster systems over a single site or over multiple sites.

1–1


1.1 How Does DECamds Work?


DECamds is a client/server application. It is installed in two parts as follows:

1.

The Data Provider gathers system data and transmits it to the Data Analyzer.

2.

The Data Analyzer receives data from the Data Provider, analyzes the data, and displays it.

A node that has the DECamds Data Provider installed announces its availability, using a multicast LAN message, to any DECamds Data Analyzer that is installed and running. The Data Analyzer receives the Data Provider’s availability announcement and a communications link is established.

Note

The Data Provider runs at a high interrupt priority level (IPL), so it gathers data and transmits it to the Data Analyzer even if the Data

Provider is on a remote node that is hung. However, because of the high

IPL collection, the Data Provider cannot collect nonresident memory data, restricting some data collection in process space.

The Data Analyzer portion of DECamds is a DECwindows Motif application that runs on any OpenVMS Version 6.2 or later system. Although you can run the Data Analyzer as a member of a monitored cluster, it is typically run on an

OpenVMS system that is not a member of the cluster being monitored. You can have more than one Data Analyzer application executing in a LAN, but only one can be running at a time on each OpenVMS system.

System data is analyzed and translated into meaningful values and rates that are displayed in DECwindows Motif windows. The data is screened for data points that exceed thresholds that might cause system or OpenVMS Cluster availability problems. The Data Analyzer can also implement various system correction options if authorized to do so.

The Data Analyzer and Data Provider nodes communicate over an Extended LAN using an IEEE 802.3 Extended Packet format protocol. Once a secure connection is established, the Data Analyzer instructs the Data Provider to gather specific system and process data.

Figure 1–1 illustrates the interaction of the Data Analyzer and Data Provider on nodes in a cluster.

Nodes A, C, D, E, F, and H can exchange information with the Data Analyzer.

Node B has defined its security to exclude the Data Analyzer from accessing its system data. Node G has not installed DECamds and does not communicate with the Data Analyzer.

1–2

Figure 1–1 DECamds Processing

A

Data Provider

Node

H

Data Provider

Node



B

Data Provider

Node

Security triplet does not match.

G

DECamds not installed on this node.

Data Analyzer Node

C

Data Provider

Node

F

Data Provider

Node

D

Data Provider

Node

E

Data Provider

Node

ZK−7946A−GE

1.2 Where to Install the DECamds Data Analyzer

This section discusses where to install the DECamds Data Analyzer software. You can install and run the DECamds Data Analyzer from either a cluster member or a standalone system outside the cluster. However, Compaq recommends that you run the Data Analyzer from outside a cluster because then you can monitor system information even if the nodes in the cluster pause or hang.

Generally, you can install and run the DECamds Data Provider on any OpenVMS

Version 6.2 or later system. Appendix A describes the specific system hardware and software requirements for installing and running the DECamds Data

Analyzer.

1.3 DECamds Security Features

DECamds has several security features, including the following:

• Private LAN transport

1–3



The DECamds protocol is based on the 802.3 Extended Packet Format (also known as SNAP). The IEEE DECamds protocol values are as follows:

Protocol ID: 08-00-2B-80-48

Multicast Address: 09-00-2B-02-01-09

If you filter protocols for bridges or routers in your network, add these values to your network protocols.

• DECamds data transfer security

Each node running DECamds as a Data Analyzer or a Data Provider has a file containing a list of three-part codes, called security triplets. See

Section 1.3.1 for more information about security triplets.

For Data Analyzer and Data Provider nodes to exchange data, at least one security triplet must match between the files on each system. DECamds Data

Provider nodes that have read access allow system data to be viewed by the

Data Analyzer node. Data Provider nodes that have write access also allow fixes to be performed by the Data Analyzer node.

• DECamds security log

The Data Provider logs all access denials and executed write instructions to the operator communication manager (OPCOM). Each log entry contains the network address of the initiator. If access is denied, the log entry also indicates whether a read or write was attempted. If a write operation was performed, the log entry indicates the process identifier (PID) of the affected process.

• OpenVMS file protection and process privileges

When the DECamds Data Analyzer and Data Provider are installed, they set directory and file protections on system directories so that only

SYSTEM accounts can read the files. For additional security on these system directories and files, you can create access control lists (ACLs) to restrict and set alarms on write access to the security files. For more information about creating ACLs, see the OpenVMS Guide to System Security.

The AMDS$CONFIG logical translates to the location of the default security files, including the following:

• The AMDS$DRIVER_ACCESS.DAT file is installed on all Data Provider nodes. The file contains a list of Data Analyzer nodes to which system data can be sent. It also contains the type of access allowed for each of those nodes.

• The AMDS$CONSOLE_ACCESS.DAT file is installed on only those nodes that run the Data Analyzer portion of DECamds. It contains a list of passwords to identify itself to Data Provider nodes.

You can create additional security files in the directory associated with the

AMDS$CONFIG logical name. By default, this logical name is assigned to

AMDS$SYSTEM. As you customize DECamds, you can change the logical assignment of AMDS$CONFIG to read input files from other locations.

The following sections describe what a security triplet is, where to find the security files, and how to set up your security files.

1–4



1.3.1 Understanding DECamds Security Files

A security triplet determines which systems can access system data from the node. The AMDS$DRIVER_ACCESS.DAT and AMDS$CONSOLE_ACCESS.DAT

files on the Data Analyzer and Data Provider systems list security triplets.

A security triplet is a three-part record that is separated by backslashes ( \ ). A triplet consists of the following fields:

• A network address (DECnet address, hardware address, or a wildcard character)

• An 8-character (alphanumeric) password

The password is not case sensitive, so the passwords ‘‘testtest’’ and

‘‘TESTTEST’’ are considered to be the same.

• A read or write (R or W) access verification code

For the Data Analyzer, the security triplets that allow write access are listed last in the AMDS$CONSOLE_ACCESS.DAT security file.

The exclamation point ( ! ) is a comment delimiter; any characters after the comment delimiter are ignored.

Table 1–1 describes the detailed format of each portion of the security triplet and then gives some examples for different situations.

Table 1–1 Security Triplet Format

Item Description

DECnet address

(area.number)

Although DECnet is not required to run DECamds, the DECnet address is used to determine a node’s physical address. The DECnet address is created by using the area.number format, where area is a value from 1 to 63, and number is a value from 1 to 1023. This address is modified into a physical address of the form AA-00-04-00xx-yy to conform to the standard IEEE 802.3 protocol for network addressing. The AA-00-04-00 prefix is associated with the Compaqowned address. The xx-yy suffix is the hexadecimal representation of the address formula: area*1024+number

Note

If you are running on a system with more than one LAN adapter or are running DECnet-Plus networking software, then this format is not valid for you. Instead, you must use the hardware address or wildcard address format for this field.

1–5



Item

Hardware address

(08-00-2B-xx-xx-xx)

Wildcard address

( * )

Description

The hardware address field is the physical hardware address in the

LAN adapter chip. It is used if you have multiple LAN adapters or are running the DECnet-Plus networking software on the system

(as opposed to the DECnet for OpenVMS Phase IV networking software).

For adapters provided by Compaq, the hardware address is in the form 08-00-2B-xx-xx-xx, where the 08-00-2B portion is Compaq’s valid range of LAN addresses as defined by the IEEE 802 standards and the xx-xx-xx portion is chip specific.

To determine the value of the hardware address on a system, use the OpenVMS System Dump Analyzer (SDA) as follows:

$ ANALYZE/SYSTEM

SDA> SHOW LAN

The previous commands display a list of available devices. Choose the template device of the LAN adapter you will be using and then enter the following command:

SDA> SHOW LAN/DEVICE=xxA0

The wildcard character allows any incoming triplet with a matching password field to access the Data Provider node. Use the wildcard character to allow read access and to run the console application from any node in your network.

Because the Data Analyzer does not use this field, you should use the wildcard character in this field in the AMDS$CONSOLE_

ACCESS.DAT file.

Caution

Use of the wildcard character for write access security triplets enables any system to perform system-altering fixes.

The following steps show how DECamds uses the security triplets to ensure security among DECamds nodes:

1.

A message is broadcast at regular intervals to all nodes within the LAN indicating the availability of a Data Provider node to communicate with a

Data Analyzer node.

2.

The node running the Data Analyzer receives the availability message and returns a security triplet that identifies it to the Data Provider and requests system data from the Data Provider.

3.

The Data Provider examines the security triplet to determine if the Data

Analyzer is listed in the AMDS$DRIVER_ACCESS.DAT file to permit access to the system.

• If the AMDS$DRIVER_ACCESS.DAT file lists Data Analyzer access information, then the Data Provider and the Data Analyzer can exchange information.

• If the Data Analyzer is not listed in the AMDS$DRIVER_ACCESS.DAT

file, or does not have appropriate access information, then access is denied and a message is logged to OPCOM; the Data Analyzer receives a message stating that access to that node is not permitted.

1–6



Table 1–2 describes how the Data Provider node interprets a security triplet match.

Table 1–2 Security Triplet Verification

Security Triplet Interpretation

08-00-2B-12-34-

56\ HOMETOWN\ W

2.1\ HOMETOWN\ R

*\ HOMETOWN\ R

The Data Analyzer has write access to the node only when the

Data Analyzer is run from the node with this hardware address

(multiadapter or DECnet-Plus system) and with the password

HOMETOWN.

The Data Analyzer has read access to the node when run from a node with DECnet for OpenVMS Phase IV address 2.1 and the password HOMETOWN.

Any Data Analyzer with the password HOMETOWN has read access to the node.

1.3.2 Customizing Security Files

Security files define which Data Analyzers can access data on nodes that have a

Data Provider. The security files let you group nodes according to specific criteria.

Note

Compaq recommends that you group nodes according to OpenVMS Cluster membership. A node can be in only one group at a time. All nodes in a cluster must also be in the same group.

Installing DECamds initially assigns all nodes to one group. Each node that is assigned to a group is listed under the group name heading in the System

Overview window.

Consider the following items when you set up customized groups:

• OpenVMS Cluster and data integrity

– All nodes in a cluster must be in the same group for data in the disk volume and lock contention windows to be complete and accurate.

It is possible to include two clusters in one group, but if a cluster is divided between two groups or only partially included, the data might not be accurate.

– Adding standalone nodes to the group will affect only the accuracy of disk volume and lock contention data.

• Partitioning for analysis

Specific users can have read or write access to certain subsets of nodes.

For example, one Data Analyzer can be designated to monitor a certain hardware type or cluster. This is entirely independent of the group to which

the nodes of that hardware type or cluster are assigned. Apart from strict security considerations, this mechanism is often used to partition systems for convenience.

Your site might already have criteria relevant to defining groups. These could include a system management division of labor, hardware type, physical location, or work function.

1–7



Compaq recommends that you correlate your security files to your group definitions so that all nodes in the group are visible in the System Overview window. Section 1.3 explains how to set up security files.

1.3.2.1 Setting Up Node Groups

Assign nodes in a cluster to the same group.

To assign a node to a group, perform the following steps on each Data Provider node that is to be part of the group:

1.

Assign a unique name of up to 15 alphanumeric characters to the AMDS$GROUP_NAME logical name in the

AMDS$SYSTEM:AMDS$LOGICALS.COM file. For example:

$ AMDS$DEF AMDS$GROUP_NAME FINANCE ! Group FINANCE; OpenVMS Cluster alias

2.

Apply the logical name by restarting the Data Provider, as follows:

$ @SYS$STARTUP:AMDS$STARTUP.COM START

For more information about the other logical names in

AMDS$LOGICALS.COM, see Appendix B.

1.3.2.2 Defining Data Exchange Access Between Nodes

The Data Provider stores access security triplets in a file called AMDS$DRIVER_

ACCESS.DAT, which indicates the Data Analyzer nodes that are allowed to request that data be provided. If a Data Analyzer node is not listed in the file, access is denied.

Examples

All Data Provider nodes in Group FINANCE have the following AMDS$DRIVER_

ACCESS.DAT file:

*\FINGROUP\R ! Let anyone with FINGROUP password read

!

2.1\DEVGROUP\W ! Let only DECnet node 2.1 with

! DEVGROUP password perform fixes (writes)

!

2.2\FINGROUP\W ! Let DECnet node 2.2 perform fixes

All Data Provider nodes in Group DEVELOPMENT have the following

AMDS$DRIVER_ACCESS.DAT file:

*\GROUPBRD\R ! Let anyone with GROUPBRD password read

!

2.1\DEVGROUP\W ! Let only DECnet node 2.1 with

! DEVGROUP password perform fixes

AMDS$CONSOLE_ACCESS.DAT file for a Data Analyzer

For a Data Analyzer to access information on any node in Groups FINANCE or

DEVELOPMENT, the following access security triplets must be listed in the Data

Analyzer node’s AMDS$CONSOLE_ACCESS.DAT file:

1–8



*\FINGROUP\R ! To access data on nodes in Group FINANCE

!

*\GROUPBRD\R ! To access data on nodes in Group DEVELOPMENT

!

*\DEVGROUP\W ! Assumes you are the owner of DECnet

! address 2.1 so you can access data and

! perform fixes on both Group FINANCE and

! Group DEVELOPMENT nodes.

!

*\FINGROUP\W ! Assumes you are the owner of DECnet

! address 2.2 so you can access data and

! perform fixes on Group FINANCE nodes.

After you modify the AMDS$CONSOLE_ACCESS.DAT security file, restart the Data Analyzer with the AVAIL command to use the changes. For more information about starting DECamds, see Chapter 2.

1.3.2.3 Limiting Specific Users to Read Access

You can restrict write access for certain users by performing the following steps:

1.

Assign a search list of directories to the AMDS$CONFIG logical name in the

AMDS$SYSTEM:AMDS$LOGICALS.COM file. For example:

$ DEFINE AMDS$CONFIG SYS$LOGIN,AMDS$SYSTEM

Execute the procedure as follows:

$ @AMDS$SYSTEM:AMDS$LOGICALS

2.

Copy the AMDS$CONSOLE_ACCESS.DAT security file to the SYS$LOGIN directory of a user and edit the file for that user.

3.

Restart the Data Analyzer with the AVAIL command. For more information about starting the Data Analyzer, see Chapter 2.

The next time the user starts DECamds, the new security file will be found in their SYS$LOGIN directory and will be used. The security file found in

AMDS$SYSTEM will not be read.

1.3.3 Sending Messages to OPCOM

The logical names shown in Table 1–3 control the sending of messages to OPCOM and are defined in the AMDS$LOGICALS.COM file.

Table 1–3 DECamds Logical Names for OPCOM Messages

AMDS$RM_OPCOM_READ

AMDS$RM_OPCOM_WRITE

A value of TRUE logs read failures to OPCOM.

A value of TRUE logs write failures to OPCOM.

To use the changes, restart the Data Analyzer with the following command on each system or use the System Management utility (SYSMAN) to run the command on all systems within the OpenVMS Cluster:

$ @SYS$STARTUP:AMDS$STARTUP RESTART

1–9



1.3.4 Setting Broadcast Intervals for Node Availability Messages

Availability messages are broadcast by the Data Provider on nodes at regular intervals until a node establishes a link with the Data Analyzer. After a link has been established, the interval varies depending on the amount of data collection

(and other factors) occurring between nodes.

You can modify the logical names in the AMDS$LOGICALS.COM file (shown in

Table 1–4) to change the broadcast availability intervals.

Table 1–4 Broadcast Availability Logical Names

AMDS$RM_DEFAULT_INTERVAL

AMDS$RM_SECONDARY_INTERVAL

Defines from 15- to 300-second intervals between availability message broadcasts.

Defines from 15- to 1800-second intervals between availability message broadcasts after a link has been established between nodes.

To use the changes, restart the Data Analyzer with the following command on each system or by using SYSMAN to run the command on all systems within the

OpenVMS Cluster:

$ @SYS$STARTUP:AMDS$STARTUP RESTART

1–10

2

Getting Started

This chapter describes the following:

• How to start DECamds

• How to use the System Overview window to monitor resource availability problems on your system

• How to use the Event Log window to correct resource availability problems on your system

2.1 Starting DECamds

To start the DECamds Data Analyzer, enter the following command and any of the following qualifiers:

AVAIL /qualifiers

Note

If you have a recent version of DECamds or if you have Availability

Manager installed, you must use the following command to invoke

DECamds:

$ AVAIL/ MOTIF

Qualifiers

/CONFIGURE

Specifies the directories from which input files are read. This can be a search list of directories or a logical defining a search list of directories.

/LOG_DIRECTORY

Specifies the directory to which log files are written. Output files can be directed to the null device, NLA0:.

/GROUP

A comma-separated list of the groups of Data Provider nodes that you want the

Data Analyzer to access.

Note

If you have not already set up a group hierarchy of nodes during

DECamds installation, refer to Section 1.3.2.1 for information about setting up node groups.

2–1

Getting Started

2.1 Starting DECamds

The following examples of commands start DECamds with input files read first from SYS$LOGIN, and then from AMDS$SYSTEM (if the files are not found in

SYS$LOGIN). All output files are written to the SYS$LOGIN directory. Only data from the group you enter (such as KUDOS) is collected.

$ DEFINE/JOB AMDS$CONFIG SYS$LOGIN,AMDS$SYSTEM

$ AVAIL/CONFIGURE=AMDS$CONFIG/LOG_DIRECTORY=SYS$LOGIN/GROUP=(KUDOS)

When DECamds starts, it displays the System Overview and Event Log windows.

To obtain help about DECamds, choose a menu item from the Help menu.

2.2 Using the System Overview Window

The System Overview window allows you to focus on resource usage activity at a high level and to display more specific data when necessary. The System

Overview window displays CPU, memory, I/O data, number of processes in CPU queues, operating system version, and hardware model for each node and group

DECamds recognizes.

Figure 2–1 shows a sample System Overview window displaying the nodes that

DECamds can reach and is monitoring.

2–2

Getting Started


Figure 2–1 System Overview Window

System Overview

File Control Customize View Collect Help

Group (node cnt)

NodeName

EVMS (29)

2BOYS

4X4TRK

ALTOS

ARUSHA

AZSUN

BARNEY

CALPAL

CHOBE

CLAIR

CLAWS

CRNPOP

DFODIL

ETOSHA

FARKLE

FCMOVE

GLOBBO

GNRS

LOADQ

MACHU

MILADY

ORNOT

PITMOD

RUMAD

SUB4

TSAVO

VAX5

VMSRMS

ZAPNOT

ZOON

4

0

3

3

6

11

0

2

3

1

1

10

34

0

4

5

1

32

8

0

2

10

1

6

21

24

4

4

1

2

% Utilization

CPU MEM

71

70

55

64

88

69

25

69

68

63

49

35

28

72

73

70

70

30

37

50

47

29

71

37

79

70

34

40

71

72

Rate / Sec / CPU

BIO DIO

# procs in

CPU Qs

1

1

1

0

0

32

1

0

0

0

0

0

3

0

0

0

3

0

17

0

0

0

0

0

1

0

5

0

0

105

2

0

1

2

4

5

1

1

29

2

0

0

3

0

0

19

9

13

1

2

0

0

0

0

0

0

13

0

0

8

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

0

0

1 dir

O. S.

Version

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V6.2

V7.0

V6.2

V6.2

V6.2

V6.2

V6.2

V7.0

V6.2

V6.2

V7.0

V6.2

V6.2

V6.2

V6.2

Hardware

Model

DEC 3000 Model 400

VAXstation 3100−M76/SPX

DEC 7000 Model 630

DEC 7000 Model 630

DEC 4000 Model 620

VAXstation 3100/SPX

VAX 6000−430

DEC 7000 Model 630

AlphaStation 400 4/233

DEC 4000 Model 610

VAXstation 4000−VLC

VAXstation 3100/GPX

DEC 7000 Model 630

VAXstation 3100−M75/SPX

VAXstation 3100/GPX

VAXstation 3100/GPX

DEC 3000 Model 500

VAX 7000−620

VAXstation 4000−VLC

VAXstation 3100/SPX

VAX 6000−440

VAXstation 3100/GPX

DEC 3000 Model 400

VAXstation 3100/GPX

DEC 7000 Model 630

VAX 6000−540

VAXstation 3100/GPX

VAX 6000−440

VAXstation 3100/SPX

The System Overview window contains two kinds of information:

• Group information, displayed in the row next to the group name, shows averages for all nodes in the group.

• Node information, displayed in the row next to the node name, shows averages for the node.

If the View menu is set to Hide Nodes, node information is not displayed.

Table 2–1 explains the fields displayed in the System Overview window.

ZK−8543A−GE

2–3

Getting Started


Table 2–1 System Overview Window Display Fields

Field Description

Group

NodeName

CPU (CPU usage)

Displays the group names in alphabetical order and the number of nodes recognized by DECamds. A group is a defined set of nodes that appear together in the System Overview window. A group can be defined by type of hardware, physical location, function, or

OpenVMS Cluster alias.

Displays the name of the node in a node row.

In a group row, displays the average of the percentage of CPU time used by all processors weighted toward the present.

In a node row, displays the percentage of CPU time used by all processes on the node, expressed as an exponential average, weighted toward the present.

On Symmetric Multiprocessing (SMP) nodes, rates for CPU time are added and divided by the number of CPUs.

MEM (Memory rate) In a group row, displays the average of the sampled values (over time) for all processes on all nodes in a group.

In a node row, displays the percent of space in physical memory that all processes on the node are currently occupying. The value represents 100 percent minus the amount of free memory.

BIO (Buffered I/O rate)

DIO (Direct I/O usage)

# procs in CPU Qs

(Number of processes in CPU queues)

In a group row, displays the average of BIO operations of all processes on all nodes.

In a node row, displays the BIO rate for all processes on the node across the number of CPUs.

In a group row, displays the average of DIO operations of all processes on all nodes.

In a node row, displays the DIO rate for all processes on the node.

Represents the number of processes the Node Summary data collection found in the COM, COMO, MWAIT, and PWAIT CPU queues.

O.S. Version (Version of the operating system)

Hardware Model

Lists the currently loaded version of OpenVMS on the node being monitored (not the node doing the monitoring).

Lists the hardware model of the node being monitored.

A percentage of a used resource is shown both by number and a dynamic status bar. For group rows, the values are averaged for all nodes in the group when node summary data collection is active. (Node summary data collection is active by default on DECamds startup.)

Resource availability problems are indicated by highlighting. When an event occurs, DECamds highlights the status bar that represents the resource.

Highlighting is shown in red on color monitors, by default; it is bold on monochrome monitors. You can change the highlight color. (See Chapter 5 for more information.)

When data appears dimmed, the data is more than 60 seconds old due to a user action that stopped node summary data collection. When the data is updated, the display returns to normal resolution.

Figure 2–2 shows the System Overview window options. Note that on the View menu, the Hide Nodes item toggles with Show Nodes; on the Control menu, the

Disable menu choices toggle with Enable choices.

2–4

Getting Started


Figure 2–2 System Overview Window Menus

File

Quit

Exit

View

Hide Nodes

Control

Disable Automatic Data Collection

Enable Automatic Event Investigation

Stop All Data Collection

Close All Displays

Customize

DECamds Customizations

Save DECamds Customizations

Save Geometry

Use System Defaults

Collect

All Node Summary

All CPU Summary

All Memory Summary

All Process I/O Summary

All Disk Status Summary

All Disk Volume Summary

All Page/Swap File Summary

All Lock Contention Summary

All Cluster Transition Summary


Help

On Context

On Window

On Version

On Help

2.2.1 Expanding and Collapsing Group Information

Use the View menu to display group or group and node status in the System

Overview window. Typically, a group is an OpenVMS Cluster. Groups are displayed in alphabetical order. Nodes within a group are also displayed in alphabetical order.

You can also expand and collapse specific group displays by clicking MB3 while the cursor is on the selected group and choosing either the Hide Nodes or Show

Nodes menu item.

2.2.2 Displaying Additional Data

By default, the Data Analyzer collects, analyzes, and displays four categories of data from Data Provider nodes:

• Node Summary

• Page/Swap File Summary

• Lock Contention Summary

• Cluster Transition Summary

In addition to the default data, you can choose any of these categories of additional data to be collected, analyzed, and displayed:

• CPU Summary

• Memory Summary

• Process I/O Summary

• Disk Status Summary

• Disk Volume Summary

ZK−7968A−GE

2–5

Getting Started


You can change the default data windows that are displayed with the DECamds

Application Customizations dialog box. For more information about customizing

DECamds, see Chapter 5.

Note

Data gathering and display consume CPU time and network bandwidth.

Request only the data you need to conclude an investigation, and then stop collecting the data (see Section 2.2.3). Whenever possible, collect data for just one node, not the entire group.

To request a specific data category, do one of the following:

• For data on a single node or a group, in the System Overview window, click

MB3 on a selected node or group, then choose Collect from the menu, and then choose a category from the submenu.

• For data on all nodes, in the System Overview window, choose a category from the Collect menu.

• In the Event Log window, click MB3 on a selected event and choose Display from the menu. (See Section 2.3 for information on the Event Log window.)

2.2.3 Stopping Data Collection

To stop collecting data, do one of the following:

• Choose Stop All Data Collection from either of the following:

Collect menu or Control menu of the System Overview window

Control menu of the Event Log window

This stops collecting for all nodes. Events are removed from the Event Log, and data values in the System Overview window go to zero and are dimmed.

Use this item if you lose track of data you are collecting in the background.

Then restart data collection as needed; new events appear once data collection resumes.

• Click MB3 on a group or node name of the System Overview window to display the Collect submenu. Select Stop All Data Collection.

This stops all data collection for the group or node you select. Node or group data in the System Overview window is zeroed.

• From the File menu of any data window, select Stop Collecting.

If the data window is specific to a node or group, this option stops collecting for the node or group. (Data windows are discussed in Chapter 3.)

Note

Choosing Close Display from the File menu of any data window closes the window but continues data collection as a background task.

• From the File menu of the System Overview window, select Exit or Quit.

2–6

Getting Started

2.3 Using the Event Log Window


The Event Log window allows you to identify and correct a system problem. The

Event Log window displays a warning message whenever DECamds detects a resource availability problem. Figure 2–3 shows an Event Log window.

Figure 2–3 Event Log Window

Event Log

File Control Customize Help

........

Time

........

Sev Event

......

Description

....................................................................................................

13:29:12.21

13:28:33.12

13:28:20.92

13:28:22.21

13:28:20.39

13:28:03.97

13:27:50.79

12:52:32.17

12:30:52.04

60 HIBIOR,

60 HIBIOR,

80 LOMEMY,

75 HIHRDP,

80 LCKCNT,

80 LCKCNT,

80 LCKCNT,

80 LOMEMY,

80 LOMEMY,

AMDS buffered I/O rate is high

GALAXY buffered I/O rate is high

GALAXY free memory is low

ETOSHA hard page fault rate is high

ORNOT possible contention for resource F11B$s{...

AJAX possible contention for resource F11B$vWORK213

CALPAL possible contention for resource PHASE1

HELENA free memory is low

DELPHI free memory is low

ZK−7951A−GE

DECamds writes all events to a log file (AMDS$LOG:AMDS$EVENT_LOG.LOG).

You can read this file in the Event Log window while the application is running.

Note

Ignore event messages that report the system process ‘‘SWAPPER’’ as having used all its quotas. The SWAPPER process is the OpenVMS memory management process; it does not have its quotas defined in the same way other system and user processes do.

Table 2–2 explains the fields displayed in the Event Log window.

Table 2–2 Event Log Window Display Fields


Time

Sev

(Severity)

Event

Displays, in real time, the time that an event is detected.

Displays a value from 0 to 100. By default, events are listed in the Event

Log window in order of decreasing severity. 0 is an informational message;

100 is a severe event. An event severity of 80 is high and indicates a potentially serious problem. Events with a severity of less than 50 appear dimmed, to indicate that they are less important. See Chapter 5 for information about how to change the display of severe events. Events that are critical are also sent to the OpenVMS operator communication manager (OPCOM).

Displays an alphanumeric identifier of the type of event.

(continued on next page)

2–7

Getting Started


Table 2–2 (Cont.) Event Log Window Display Fields


Description Displays the node or group name and a short description of the resource availability problem.

When an event ‘‘times out’’ by an improvement in availability, it is removed from the display. Events that are not triggered by a condition are timed out after 30 seconds (for example, the ‘‘CFGDON, node configuration done’’ event). When you select an event, the event remains displayed for 15 seconds (or until you initiate another task in the window), even if the event times out.

Figure 2–4 shows the Event Log window options.

Figure 2–4 Event Log Window Menus

File

Quit

Exit

Control

Disable Automatic Data Collection

Enable Automatic Event Investigation


Close All Displays

Customize

DECamds Customizations

Save DECamds Customizations

Save Geometry

Use System Defaults

Customize Events

Save Event Cutomizations

Sort Data

Filter Data

Use Last Saved Settings

Save Sort Changes

Save Filter Changes

Help

On Context

On Window

On Version

On Help

ZK−7952A−GE

For information about customizing event log information, see Section 5.2.1.

2–8

Getting Started


2.3.1 Displaying Information About an Event Log Entry

To display more information about an event, click MB3 on the event in the Event

Log window, and then choose Display. Depending on the event, you have one or more event display choices that give you more information about the event.

Figure 2–5 shows a sample event display choice dialog box.

Figure 2–5 Event Display Choice Dialog Box

DISPLAY − HIHRDP, DIMOND hard page fault rate is high

Event Display Choices

Memory Summary

Node Summary

OK Apply

Cancel

ZK−7950A−GE

2.3.2 Performing Corrective Action on an Event Log Entry

To take corrective action on an event, click MB3 on the event in the Event Log window, and then choose Fix. Depending on the type of event, one or more of the following event fix choices are displayed (not all events have all fix options):

Adjust process working set

Crash node

Delete a process

Exit an image

Lower process priority

Purge process working set

Raise process priority

Resume a process

Suspend a process

See Chapter 4 for detailed information about performing fixes.

2–9

Getting Started


2.3.3 Sending Event Information to OPCOM

DECamds sends critical events to the operator communication manager

(OPCOM).

By default, events that meet both of the following criteria are sent to OPCOM:

• Have a severity level of 90 or above

• Occur continuously for 600 seconds (10 minutes)

You can change either criterion by choosing Filter Data... from the Customize menu of the Event Log window. For more information on changing Event Log filters, see Chapter 5.

2.3.4 Removing an Event from the Event Log Window

To remove an event from the Event Log window, click MB3 on an event, and choose Remove from the menu. An event reappears if DECamds routine sampling detects the same situation that caused the original log entry.

2.3.5 Retaining and Releasing an Event in the Event Log Window

Event Log entries are removed when the underlying cause is removed, so an event might disappear from the Event Log window. To retain the selected event in the Event Log window, click MB3 on an event and choose Freeze. When an event is frozen, the Time field is highlighted.

To release the selected event, click MB3 on the event and choose Unfreeze.

2–10

3

Managing DECamds Data Windows

This chapter describes the DECamds data windows that you can display from the

System Overview and Event Log windows.

Figure 3–1 shows the hierarchy of the DECamds data windows.

Figure 3–1 DECamds Data Window Hierarchy

Event

Log

System

Overview

Disk

Status

Summary*

Volume

Summary*

Page/Swap

File

Summary*

Node

Summary

Lock

Contention

Summary**

Cluster

Transition/

Overview

Summary

Single

Disk

Summary

Process

I/O

Summary

CPU

Modes

Summary

CPU

Summary

Memory

Summary

Single

Lock

Summary

System

Communication

Architecture

Summary

* Available for individual nodes and groups of nodes.

** Available for groups only.

Single

Process

Summary

NISCA

Summary

ZK−7970A−GE

Table 3–1 describes the data windows and their functions.

Table 3–1 DECamds Data Windows

Window Reference

Opened from

Window

Disk Status

Summary

Section 3.1

Event Log

System Overview

Volume

Summary

Section 3.2

Event Log

System Overview

Displays

Disk device data including path, volume name, status, and mount, transaction, error, and resource wait counts.

Disk volume data, including path, volume name, disk block utilization, queue length, and operation count rate.


3–1


Table 3–1 (Cont.) DECamds Data Windows

Window Reference

Opened from

Window

Single Disk

Summary

Page/Swap

File Summary

Node

Summary

Section 3.3

Section 3.4

Section 3.5

Disk Status

Summary

Volume Summary

Event Log

System Overview

Event Log

System Overview

Process I/O

Summary

Section 3.6

Event Log

Node Summary

System Overview

Single Disk

Summary

Node Summary CPU Modes

Summary

CPU

Summary

Memory

Summary

Section 3.7

Section 3.8

Section 3.9

Single Process

Summary

Section 3.10

Event Log

Node Summary

System Overview

Event Log

Node Summary

System Overview

Event Log

Any data window

Lock

Contention

Summary

Single Lock

Summary

Section 3.11

Section 3.12

Cluster Section 3.13

Transition/Overview

Summary

Section 3.14

System

Communication

Architecture

Summary

NISCA

Summary

Section 3.15

Event Log

System Overview

Event Log

Lock Contention

Summary

Event Log

System Overview

Cluster

Transition/Overview

Summary

System

Communication

Architecture

Summary

Displays

Summary data about each node in a group in which a disk is available.

Data about page and swap names and utilization, including free, used, and reserved pages.

Overview of a specific node’s resource demand on the CPU state queues and processor modes, memory utilization, page faults, and I/O.

Statistics about I/O utilization by process, including buffered I/O, direct I/O, and page write I/O; also lists various I/O quotas.

A graphic representation of each CPU’s processor modes, listing the process currently executing in the CPU.

Statistics about CPU utilization by process, including process state, priority, execution rate,

CPU time, and wait time.

Statistics about memory usage by process, including process working set count, quota and extent, and paging rates.

Specific data about a process, basically a combination of data elements from the CPU,

Memory, and Process I/O displays, as well as data for specific quota utilization, current image, specific process information, and wait queue time.

Data about each resource for which a potential lock contention situation exists.

Specific data about the blocking lock and any other locks in the granted, conversion, or waiting queues.

Summary information about each node’s membership in an OpenVMS Cluster.

System Communication Architecture (SCA) information about a selected node’s connection or connections to other nodes in a cluster.

Summary information about the Network

Interconnect System Communication Architecture

(NISCA) protocol, which is responsible for carrying messages to other nodes in the cluster.

3.1 Disk Status Summary Window

The Disk Status Summary window shown in Figure 3–2 displays data about availability, count, and errors of disk devices on the system.

3–2



Figure 3–2 Disk Status Summary Window

AMDS Disk Status Summary

File View Fix Customize

Device Name Path Volume Name

DAD18

DAD19

DAD20

DAD15

DAD16

DAD17

DAD22

DAD26

DAD27

DAD28

DAD23

DAD24

DAD25

DAD3

AMDS

AMDS

AMDS

AMDS

AMDS

AMDS

AMDS

AMDS

AMDS

AMDS

AMDS

AMDS

AMDS vs0121

DECLEARN010

OPTMOD

VAXDOCJUL942

VAXDOCJUL943

VAXDOCJUL944

V4RESD

V46_RESD

V47RES

ESS_KITSDISK

V43_RESD

V44RES

V45RES

VAXBINJUL942

Status

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Mounted wrtlck

Error

Count

Trans Mount Rwait

0

0

0 1

1

1

1

1

1

1

1

0

0

0

0

0

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

0

0

0

0

0

0

0

0

0

Help

ZK−7947A−GE

To open a Disk Status Summary window, do one of the following:

• In the System Overview window, click MB3 on a node or group line, choose

Display from the menu, and choose Disk Status Summary from the submenu.

• In the Event Log window, click MB3 on any disk status-related event, and choose Display from the menu.

Table 3–2 describes the Disk Status Summary window data fields.

Table 3–2 Disk Status Summary Window Data Fields

Field Displays

Device

Name

Path

Volume

Name

The standard OpenVMS device name that indicates where the device is located, as well as a controller or unit designation.

The primary path (node) from which the device receives commands.

The name of the media that is currently mounted.


3–3



Table 3–2 (Cont.) Disk Status Summary Window Data Fields

Field Displays

Status

Errors

Trans

1

Mount

1

Rwait

1

1

One or more of the following disk status values:

Alloc

Disk is allocated to a specific user

CluTran

Disk status is uncertain due to a cluster state transition in progress

Dismount

Foreign

Disk in process of dismounting; may be waiting for a file to close

Disk is mounted with the /FOREIGN qualifier

Invalid

MntVerify

Mounted

Offline

Online

Shadow Set

Member

Unavailable

Wrong

Volume

Disk is in an invalid state (likely Mount Verify Timeout)

Disk is waiting for a mount verification

Disk is logically mounted by a MOUNT command or service call

Disk is no longer physically mounted in device drive

Disk is physically mounted in device drive

Disk is a member of a shadow set

Disk is set /UNAVAILABLE

Disk has been mounted with the wrong volume name

Wrtlck

Disk is mounted and write locked

The number of errors generated by the disk (a quick indicator of device problems).

The number of currently-in-progress file system operations for the disk.

The number of nodes that have the specified disk mounted.

An indicator that a system I/O operation is stalled, usually during normal connection failure recovery or volume processing of host-based shadowing.

1

For the group window, the sum of the node window values is displayed.

DECamds detects the following disk status-related events and displays them in the Event Log window. Node refers to the name of the node that is signaling the event. Disk refers to the name of the disk to which the event is related.

DSKERR, node disk disk error count is high

DSKINV, node disk disk is in an invalid state

DSKMNV, node disk disk mount verify in progress

DSKOFF, node disk disk is off line

DSKRWT, node disk disk Rwait count is high

DSKUNA, node disk disk is unavailable

DSKWRV, node disk wrong volume mounted

3–4


3.2 Volume Summary Window


The Volume Summary window shown in Figure 3–3 displays summary data about disk volumes mounted in the system. Volume summary data is accurate when every node in an OpenVMS Cluster environment is in the same group. Multiple clusters can share a group, but clusters cannot be divided into different groups without losing accuracy.

Figure 3–3 Volume Summary Window

EVMS Volume Summary


Device Name Path

DSA4010

DSA4006

DSA67

$64$DUA113

$64$DUA178

DSA4005

DSA64

$64$DUA203

$64$DUA114

DSA4009

DSA4004

DSA4007

Volume Name

Disk Space (blocks)

Used

EVMS

EVMS

EVMS

CALPAL

HICLIT

EVMS

EVMS

LOADQ

CALPAL

EVMS

EVMS

EVMS

WORK9

WORK5

FOLKLORE

SCRTCH.1

EVMS_SYS_061

WORK4

VMSCMSMASTER

AXPVMS061

SCRTCH.2

WORK6

WORK3

WORK7

3815570

3920478

2252727

2890557

3850054

3020402

3097410

3917270

2881599

3207386

3449114

2924410

% Used Free

104380

12

688218

50394

70404

900088

823080

3220

59352

713104

471344

996080

Queue

0.00

0.00

0.00

0.00

0.00

0.00

1.26

1.00

0.00

0.89

0.00

0.00

OpRate

45.14

27.14

7.66

6.35

5.16

4.90

4.38

3.37

2.02

0.83

0.29

0.17

Help

ZK−7948A−GE

Note

The group value for Free blocks used is determined from the node with the mastering lock on the volume resource.

To open a Volume Summary window, do one of the following:

• In the System Overview window, click MB3 on a node or group line, choose

Display from the menu, and choose Volume Summary from the submenu.

• In the Event Log window, click MB3 on any volume-related event, and choose

Display from the menu.

Note

DECamds does not collect Volume Summary data on remote disks mounted using the VAX Distributed File Service (DFS).

Table 3–3 describes the Volume Summary window data fields.

3–5



Table 3–3 Volume Summary Window Data Fields

Field Displays

Device

Name

Path

Volume

Name

Used

% Used

The standard OpenVMS device name that indicates where the device is located, as well as a controller or unit designation.

The primary path (node) from which the device receives commands.

The name of the mounted media.

Free

Queue

OpRate

The number of volume blocks in use.

The percentage of the number of volume blocks in use in relation to the total volume blocks available.

The number of blocks of volume space available for new data.

The average number of I/O operations pending for the volume (an indicator of performance; less than 1.00 is optimal).

The rate at which the operations count to the volume has changed since the last sampling. The rate measures the amount of activity on a volume. The optimal load is device-specific.

DECamds detects the following volume-related events and displays them in the

Event Log window. Node refers to the name of the node that is signaling the event. Disk refers to the name of the disk to which the event is related. Group refers to the name of the group to which the event is related.

DSKQLN, node disk disk volume queue length is high

LOVLSP, group disk disk volume free space is low

3.3 Single Disk Summary Window

The Single Disk Summary window shown in Figure 3–4 displays summary data about each node in the group in which a disk is available. This window is a node-by-node display of the data that is summarized in the Disk Status Summary and Volume Summary windows. The values displayed are those you would see if you displayed Disk Status Summary or Volume Summary for each node within the group.

You can use this display to determine both of the following:

• Which node in the group has a disk with high I/O rates

Determining which node has a high I/O rate to the disk is useful because you can sort by direct I/O rate and learn which process or processes are causing the high I/O rates to the disk.

• If a disk is in a state that is inconsistent with other nodes

Determining which node or nodes might be in an abnormal state is useful because you can then discover if, for some reason, one node believes that the disk is in the MntVerify or CluTran state, thus holding up processing in the cluster in which the node resides.

3–6



Figure 3–4 Single Disk Summary Window

$64$DUA208(V15SNAPSHOTS) Single Disk Summary for EVMS

File View Fix Customize Help

Node Status Errors Trans Rwait Free QLen OpRate

............................................................................................................................................................

ARUSHA

2BOYS

CHOBE

VMSRMS

CLAWS

RUMAD

ZAPNOT

BARNEY

MILADY

MACHU

LOADQ

DFODIL

FARKLE

ZOOH

ALTOS

CRNPOP

GLOBBO

GNBS

Mounted

Mounted

Mounted

Mounted

Mounted

Mounted

Mounted

Mounted

Mounted

Mounted unknown

Mounted

Mounted

Mounted

Mounted

Mounted

Mounted

Mounted

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

1

1

0

1

1

1

1

1

1

1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

909

0

909

909

0

909

909

909

909

909

0

909

909

909

909

909

909

909

(M)

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

0.00 0.00

ZK−8544A−GE

To open a Single Disk Summary window, follow these steps:

1.

In the System Overview window, click MB3 on a group or node name.

The system displays a pop-up menu.

2.

Choose Display from the menu and Disk Status Summary (or Volume

Summary) from the submenu.

The system displays the Disk Status Summary window (or Volume Summary window).

3.

In the Disk Status Summary window (or Volume Summary window), click

MB3 on a device name.


4.

Choose Display Disk.

The system displays the Single Disk Summary window.

As an alternative to steps 3 and 4, you can can double-click MB1 on a line in the

Disk Status Summary or Volume Summary window to display the Single Disk

Summary window.

Note that when you click on an item, DECamds temporarily stops updating the window for 15 seconds or until you choose an item from a menu.

3–7



Table 3–4 lists the Single Disk Summary window data fields.

Table 3–4 Single Disk Summary Window Data Fields

Field Displays

Node

Status

Errors

Trans

Rwait

Free

QLen

OpRate

Name of the node

Status of the disk: mounted, online, offline, and so on

Number of errors on the disk

Number of currently-in-progress file system operations on the disk

(number of open files on the volume)

Indication of an I/O stalled on the disk

Count of free disk blocks on the volume

An (M) after the free block count indicates this node holds the lock on the volume that DECamds uses to obtain the true free block count on the volume. Other nodes might not have accessed the disk, so their free block count might not be up to date.

Average number of operations in the I/O queue for the volume

Count of rate of change to operations on the volume

From the Single Disk Summary window, you can display the Process I/O

Summary window. See Section 3.6 for more information.

3.4 Page/Swap File Summary Window

The Page/Swap File Summary window shown in Figure 3–5 displays data about a node’s page/swap file usage and identifies page or swap files that are overused or underconfigured. It also displays nodes that lack a page or swap file.

Figure 3–5 Page/Swap File Summary Window

STAR Page/Swap Files Summary


File Usage (blocks)

Node Name File Name Used

DELPHI DISK$DELPHI_PAG66:[SYSEXE]PAGEFILE_DELPHI_2.SYS;1

HELENA

DELPHI

DISK$HELENA_PAG65:[SYSEXE]PAGEFILE_HELENA_1.SYS;1

DISK$DELPHI_PAG64:[SYSEXE]PAGEFILE_DELPHI_1.SYS;1

HELENA DISK$HELENA_PAG67:[SYSEXE]PAGEFILE_HELENA_2.SYS;1

ADRIC DISK$ADRIC_20400:[SYS13.SYSEXE]PAGEFILE.SYS;2

BRYTT DISK$BRYTT_19565:[SYS1.SYSEXE]PAGEFILE.SYS;1

BARODA DISK$BARODA_65093:[SYS39.SYSEXE]PAGEFILE.SYS;1

HELENA DISK$PAGEDISK4:[SYSEXE]PAGEFILE_HELENA_4.SYS;1

SPRGLU DISK$SPRGLU_65118:[SYS1.SYSEXE]PAGEFILE.SYS;1

DELPHI

DAG

BOOM

DISK$PAGEDISK4:[SYSEXE]PAGEFILE_DELPHI_4.SYS;1

DISK$DAG_19613:[SYS1.SYSEXE]PAGEFILE.SYS;2

DISK$VAXVMSV055:[SYS0.SYSEXE]PAGEFILE.SYS;1

138842

132163

123951

111784

106415

48170

45713

39130

37876

37224

35394

30285

%Used Total Reservable

499992

499992

499992

499992

169992

199992

149992

149992

149992

149992

149992

74000

121063

110758

121895

108575

−4604

81519

44993

32993

54985

37170

72112

−15934

ZK−7964A−GE

3–8


3.4 Page/Swap File Summary Window

To open a Page/Swap File Summary window, do one of the following:

• In the Event Log window, click MB3 on any event, and choose Display from the menu displayed. Then choose Page/Swap File Summary from the submenu displayed.

• In the System Overview window, click MB3 on any node or group line, and choose Display from the menu displayed. Then choose Page/Swap File

Summary from the submenu displayed.

Table 3–5 describes the Page/Swap File Summary window data fields.

Table 3–5 Page/Swap File Summary Window Data Fields

Field Displays

Node

Name

File Name

The name of the node on which the page/swap file resides.

The name of the page/swap file. For secondary page/swap files, the file name is obtained by a special AST to the job controller on the remote node.

DECamds makes one attempt to retrieve the file name.

Used

% Used

The number of used pages or pagelet blocks within the file.

A graph representing the percentage of the blocks from the available page or pagelet blocks in each file.

The total number of pages or pagelet blocks within the file.

Total

Reservable The number of pages or pagelet blocks that can be logically claimed by a process for future physical allocation. This value can be listed as a negative value, because it is merely a value of a process’s interest in getting pages from the file. If every process currently executing needed to use the file, then this value is the debt that is owed.

DECamds detects the following page and swap file-related events and displays them in the Event Log window. Node is replaced by the name of the node to which the event is related.

LOPGSP, node file page file space is low

LOSWSP, node file swap file space is low

NOPGFL, node has no page file

NOSWFL, node has no swap file

3–9


3.5 Node Summary Window


The Node Summary window shown in Figure 3–6 displays a high-level graphic summary of node resource demands on the CPU, memory, and I/O.

Figure 3–6 Node Summary Window

DELPHI Node Summary


Model:

O.S.:

Uptime:

Memory:

CPUs:

DEC 7000 Model 630

OpenVMS V7.0

10 19:00:21.98

192.00 Mb

4

CPU Process State Queues

0 2 4 6 Curr Peak

COM

WAIT

0.08

5.40

1.89

3.79

Page Faults (per second)

0 192

Total

Hard

System

Memory (Pages in thousands)

0 78

Free

Used

Modified

Bad

157

384

235

C

M

N

I

K

E

S

U

CPU Modes (Avg all processors)

0 25 50 75 100 Curr Peak

14

0

9

33

25

14

1

5

20

0

19

4

50

14

5

38

576

314

768

393

Curr

269.43

25.35

0.00

Curr

22838

354729

15649

0

Peak

1363.80

63.97

0.38

Peak

19375

358194

15679

0

I/O (per second)

0

WIO

DIO

BIO

48 96 144 192 Curr

0.96

98.10

37.71

Peak

8.03

116.49

92.76

ZK−7963A−GE

To open a Node Summary window, do one of the following:

• In the System Overview window, double-click on any node name. You can also click MB3 on any node name, and choose Display from the menu.

• In the Event Log window, double-click on any node name. You can also click

MB3 on an event that is related to node summary data, and choose Display from the menu.

3–10



Dynamic bar graphs display the current values for each field. Peak values are also displayed from when DECamds begins collecting node summary data. A peak value is typically the highest value received; however, for the Free Memory field it is the lowest value received.

You can open the following windows from the Node Summary Window by double-clicking in the space for each category:

CPU Summary

CPU Modes Summary

Memory Summary

I/O summary

Table 3–6 describes the Node Summary window data fields.

Table 3–6 Node Summary Window Data Fields

Field Displays

Hardware Model

Operating System

Uptime

Memory

CPUs

CPU Process State

Queues

CPU Modes

Page Faults

Memory

I/O

The system hardware model name.

The name and version of the operating system.

The time since last reboot measured in days, hours, minutes, and seconds.

The total amount of physical memory found on the system.

The number of active CPUs on the node.

One of the following:

COM Sum of the queue lengths of processes in the COM and

COMO states.

WAIT Sum of the queue lengths of processes in the MWAIT,

COLPG, CEF, PFW, and FPG states.

The CPU usage by mode (kernel, executive, supervisor, user, interrupt, compatibility, multiprocessor synchronization, and null). On symmetric multiprocessing (SMP) nodes, percentages are averaged across all the CPUs and displayed as one value.

The rate of system hard and soft page faulting, as well as peak values seen during a DECamds session. System page faults are taken from kernel processes.

The histogram listing memory distribution (Free, Used, Modified,

Bad) as absolute values of number of thousands of pages or pagelets.

Peak values are also listed, with Free using lowest seen value as peak.

The histogram listing Buffer, Direct, and Page Write I/O rates per second. Also included is the peak value seen.

DECamds detects the following node events and displays them in the Event Log window. Node is replaced by the name of the node to which the event is related.

HIBIOR, node buffered I/O rate is high

HICOMQ, node many processes waiting for CPU

HIDIOR, node direct I/O rate is high

HIHRDP, node hard page fault rate is high

HIMWTQ, node process waiting in MWAIT

HINTER, node interrupt mode time is high

HIPWIO, node paging write I/O rate is high

HIPWTQ, node many processes waiting in COLPG, PFW, or FPG

HITTLP, node total page fault rate is high

3–11



HMPSYN, node MP synchronization mode time is high

HISYSP, node system page fault rate is high

LOMEMY, node free memory is low

NOPROC, node cannot find process names process

3.6 Process I/O Summary Window

The Process I/O Summary window shown in Figure 3–7 displays summary statistics about process I/O rates and quotas. Use the Process I/O Summary window to display information about I/O issues that might be caused by

I/O-intensive programs or I/O bottlenecks.

Note

DECamds does not yet support kernel threads. If you use threaded processes, DECamds displays only the top thread.

Figure 3–7 Process I/O Summary Window

EDISON Process I/O Summary


..................

PID

Process

Name

..................

I/O Rate per second

DIO BIO PIO

Open

...........

Files DIO

Remaining Quotas

...........

BIO Bytes Files

20A0005B

20A000A4

20A000A6

20A00094

DECW$SERVER_0

Dana’s Dad 2

AMDS$COMM

Dana’s Dad

20A0004D

20A0006B

20A00063

AUDIT_SERVER

SMISERVER

SQLSRV$SERVER

20A00056 EVL

20A0009A DECW$MWM

20A00050 SECURITY_SERVER

0.00

1.08

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.83

0.33

14.69

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

7

6

5

4

3

3

21

9

9

8

100

250

100

250

200

100

100

100

250

200

92

249

199

246

198

97

254

97

249

199

38224

11984

11984

75728

97632

37312

39552

37184

89168

66528

179

591

591

592

93

94

95

96

597

97

Help

ZK−7965A−GE

To open a Process I/O Summary window, do one of the following:

• In the Node Summary window, double-click in the I/O area.

• On the View menu in the Single Disk Summary window, choose Display

Process I/O Summary.

• In the System Overview window, double-click on the BIO or DIO fields for any node. You can also click MB3 on any field for any node, choose Display from the menu, and choose Process I/O Summary from the submenu.

• To open a Process I/O Summary window for every node in a group, in the

System Overview window, click MB3 on a group line, choose Display from the menu, and choose Process I/O Summary from the submenu.

• In the Event Log window, click MB3 on any process I/O-related event, and choose Display from the menu.

3–12


3.6 Process I/O Summary Window

You can open a window about a specific process in the Process I/O Summary window by double-clicking on the process name.

Table 3–7 describes the Process I/O Summary window data fields.

Table 3–7 Process I/O Summary Window Data Fields

Field Displays

PID

Process Name

Direct I/O Rate

(DIO)

Buffered I/O Rate

(BIO)

Paging I/O Rate

(PIO)

Open Files

Direct I/O Limit

Remaining (DIO)

Buffered I/O Limit

Remaining (BIO)

Byte Limit

Remaining (Bytes)

Open File Limit

Remaining (Files)

The process identifier, a 32-bit value that uniquely identifies a process.

The current process name.

The rate at which I/O transfers occur between the system devices and the pages or pagelets that contain the process buffer that the system locks in physical memory.

The rate at which I/O transfers occur between the process buffer and an intermediate buffer from the system buffer pool.

The rate of read attempts necessary to satisfy page faults (also known as Page Read I/O or the Hard Fault Rate).

The number of open files.

The number of remaining direct I/O limit operations available before the process reaches its quota. DIOLM quota is the maximum number of direct I/O operations a process may have outstanding at one time.

The number of remaining buffered I/O operations available before the process reaches its quota. BIOLM quota is the maximum number of buffered I/O operations a process may have outstanding at one time.

The number of buffered I/O bytes available before the process reaches its quota. BYTLM is the maximum number of bytes of nonpaged system dynamic memory that a process can claim at one time.

The number of additional files the process can open before reaching its quota. FILLM quota is the maximum number of files that can be opened simultaneously by the process, including active network logical links.

DECamds detects the following process I/O-related events and displays them in the Event Log window. Node is replaced by the name of the node to which the event is related. Process is replaced by the name of the process to which the event is related.

LOBIOQ, node process has used most of its BIOLM process quota

LOBYTQ, node process has used most of its BYTLM job quota

LODIOQ, node process has used most of its DIOLM process quota

LOFILQ, node process has used most of its FILLM job quota

PRBIOR, node process buffered I/O rate is high

PRDIOR, node process direct I/O rate is high

PRPIOR, node process paging I/O rate is high

3–13


3.7 CPU Modes Summary Window


The CPU Modes Summary window shown in Figure 3–8 displays more detailed summary statistics about CPU mode usage than the Node Summary window.

Use the CPU Modes Summary window to diagnose issues that may be caused by

CPU-intensive users or CPU bottlenecks.

Figure 3–8 CPU Modes Summary Window

HELTER CPU Modes Summary

File View Customize

CPU Id

Capabilities

CPU #01

PRIMARY

RUN

QUORUM

CPU #03

RUN

QUORUM

CPU #04

RUN

QUORUM

State

Mode

Run

Kernel

Executive

Supervisor

User

Interrupt

Compatiblity

MP Synch

Null

Run

Kernel

Executive

Supervisor

User

Interrupt

Compatiblity

MP Synch

Null

Run

Kernel

Executive

Supervisor

User

% Used

PID

Rate

Name

Peak

0%

0%

0%

0%

4%

0%

0%

95%

7%

4%

0%

14%

1%

0%

0%

74%

*** None ***

13%

3%

0%

36%

13%

0%

0%

54%

*** None ***

26%

7%

9%

60%

3%

0%

2%

34%

2EA031EB

8%

7%

0%

24%

APAS1_CALENDAR

42%

8%

10%

63%

Help

ZK−7940A−GE

To open a CPU Modes Summary window, do one of the following:

• In the Node Summary window, double-click MB1 in the CPU Modes area. You can also click MB3, and choose Display from the menu.

• In the Node Summary window View menu, choose Display Modes Summary.

You can open a window about a specific process in the CPU Modes Summary window by double-clicking on the process name.

Table 3–8 describes the CPU Modes Summary window data fields.

3–14



Table 3–8 CPU Modes Summary Window Data Fields

Field Displays

CPU ID A decimal value representing the identity of a process in a multiprocessing system. On a uniprocessor, this value will always be CPU #00.

Capabilities One of the the following CPU capabilities: Primary, Quorum, Run, or Vector.

State

Mode

One of the following CPU states: Boot, Booted, Init, Rejected, Reserved,

Run, Stopped, Stopping, or Timeout.

One of the following values for CPU modes supported for the architecture:

Compatibility, Executive, Interrupt, Kernel, MP Synch, Null, Supervisor, or

User. Note: Compatibility mode does not exist on OpenVMS Alpha systems.

% Used

PID

Name

Rate

Peak

A bar graph, by CPU, representing the percentage of the CPU utilization for each mode.

The process identifier value of the process that is using the CPU. If the PID is unknown to the console application, the internal PID (IPID) will be listed.

The process name of the process found in the CPU. If no process is found in the CPU, this will be listed as *** None ***.

A numerical percentage of CPU time for each mode.

The peak CPU usage determined for each mode.

3.8 CPU Summary Window

The CPU Summary window shown in Figure 3–9 displays summary statistics about process CPU usage issues that might be caused by CPU-intensive users or

CPU bottlenecks.

Figure 3–9 CPU Summary Window

VAX5 CPU Summary


..................

PID

Process ..................

Name Priority

.....................................

State Rate

CPU

Wait

.....................................

Time

41C029F6

41C02c6c

41C02c55

41C0286B

BATCH_548

SYSBAK_286B

SYSBAK_28DA

SYSBAK_2185_3

3/ 3

6/ 6

6/ 6

7/ 6

LEF 25.63%

LEF

LEF

23.77%

9.71%

LEF 5.01%

0.00%

0.00%

0.00%

0.00%

0 00:09:32.39

0 00:01:11.60

0 00:05:47.91

0 00:00:14.08

ZK−7942A−GE

3–15


3.8 CPU Summary Window

To open a CPU Summary window, do one of the following:

• In the System Overview window, double-click on the CPU field of any node.

You can also click MB3 on an event that is related to CPU usage, choose

Display from the menu, and choose CPU Summary from the list.

• In the Node Summary window, double-click on CPU Process State Queues.

• In the Event Log window, click MB3 on an event that is related to CPU usage, choose Display from the menu, and choose CPU Summary from the list.

You can open a window about a specific process in the CPU Summary window by double-clicking on the process name.

Table 3–9 describes the CPU Summary window data fields.

Table 3–9 CPU Summary Window Data Fields

Field Displays

PID

Name

Priority

State

Rate

Wait

Time


The process name.

Computable (xx) and base (yy) process priority in the format xx/yy.

One of the values listed under the Single Process Summary description in

Table 3–11.

The percent of CPU time used by this process. This is the ratio of CPU time to elapsed time. The CPU rate is also displayed in the bar graph.

The percent of time the process is in the COM or COMO state.

The amount of actual CPU time charged to the process.

DECamds detects the following CPU-related events and displays them in the

Event Log window. Node is replaced by the name of the node to which the event is related. Process is replaced by the name of the process to which the event is related.

PRCCOM, node process waiting in COM or COMO

PRCCVR, node process has high CPU rate

PRCMWT, node process waiting in MWAIT

PRCPWT, node process waiting in COLPG, PFW, or FPG

3–16


3.9 Memory Summary Window


The Memory Summary window shown in Figure 3–10 displays memory usage data for processes on a node so that you can identify processes that use large amounts of memory or have high page fault rates.

Figure 3–10 Memory Summary Window

File View Fix

AMDS Memory Summary

Customize

PID

Process

Name Count

000001D5

000001CE

000001CD

000001C5

000001C2

000001AB

00000091

00000065

0000004F

0000004A

_FTA24:

VUE$SYSTEM_4

VUE$SYSTEM_3

DECW$MWM

390

337

271

1290

DECW$SESSION

DECW$SERVER_0

DECW$TE_0091

LATACP

NETACP

QUEUE_MANAGER

490

4423

2768

395

249

500

Size

Working Set

Extent Rate

512 32768 0.00

512

512

32768

32768

0.00

0.00

5762 32768 0.00

6512 32768

5581

3264

32768

32768

680

500

0.00

0.00

0.02

2048 0.00

2048 0.00

2048 32768 0.00

Page Fault

I/O

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Help

ZK−7960A−GE

To open a Memory Summary window, do one of the following:

• In the Node Summary window, double-click on the Page Faults or Memory area. You can also click MB3 on the Page Faults or Memory area, and choose

Display from the menu.

• In the View menu of the Node Summary window, choose Display Memory

Summary.

• In the System Overview window, double-click on the Memory field for any node. You can also click MB3 on any field for any node, choose Display from the pop-up menu, and choose Memory Summary from the submenu.

• To display a memory summary of every node in a group from the System

Overview window, click MB3 on the group line, choose Display from the menu, and choose Memory Summary from the submenu.

• In the Event Log window, click MB3 on an event related to memory usage, and choose Display from the menu.

You can open a window about a specific process in the Memory Summary window by double-clicking on the process name.

3–17



Table 3–10 describes the Memory Summary window data fields.

Table 3–10 Memory Summary Window Data Fields

Field Displays

PID

Process Name

Working Set

Count

1


The process name.

The number of physical pages or pagelets of memory that the process is using. The bar graph represents the percentage of working set count used to the working set extent.

Working Set

Size

1

Working Set

Extent

1

The number of pages or pagelets of memory the process is allowed to use. This value is periodically adjusted by the operating system based on analysis of page faults relative to CPU time used. When the value increases in large units, this indicates a process is receiving a lot of page faults and its memory allocation is increasing.

The number of pages or pagelets of memory in the process’s

WSEXTENT quota as defined in the user authorization file (UAF).

The number of pages or pagelets will not exceed the value of the system parameter WSMAX.

Page Fault Rate The number of page faults per second for the process. The bar graph represents a relative number of page faults per second.

Page Fault I/O The rate of read attempts necessary to satisfy page faults (also known as Page Read I/O or the Hard Fault Rate).

1

Working Set Value = Total Physical Memory / Maximum Process Count

DECamds detects the following memory-related events and displays them in the

Event Log window. Node is replaced by the name of the node to which the event is related. Process is replaced by the name of the process to which the event is related.

LOWEXT, node process working set extent is too small

LOWSQU, node process working set quota is too small

PRPGFL, node process high page fault rate


3–18


3.10 Single Process Summary Window


The Single Process Summary window shown in Figure 3–11 displays summary data about a process, including Execution Rates, Process Quotas in Use, Wait

States, and Job Quotas in Use.

Figure 3–11 Single Process Summary Window

EDISON Process SECURITY_SERVER (DETACHED)


Process name

Username

Account

UIC

PID

Owner ID

PC

PSL

Priority

State

SECURITY_SERVER

SYSTEM

<start>

[1,4]

20A00050

00000000

7FFEDF8A

03C00000

10/ 8

HIB

WS global pages

WS private pages

WS total pages

WS size

WSdef

WSquo

WSextent

Images activated

Mutexes held

21

113

134

1178

1028

4100

21685

0

0

Help

EXECUTION RATES

CPU

Direct I/O

Buffered I/O

Paging I/O

Page Faults

0.00

0.00

0.00

0.00

0.00

WAIT STATES

Compute

Memory

Direct I/O

Buffered I/O

Control

Quotas

Explicit

0

0

0

0

0

0

68

100

100

100

100

100

100

100

Current image: Not Available − not in memory

PROCESS QUOTAS IN USE

DIOlm

BIOlm

ASTlm

CPU

0

1

8

200

200

200

JOB QUOTAS IN USE

Fillm

Pgflquo

Englm

TQElm

Prclm

Bytlm

3

3296

6

7

0

0

100

32768

2000

50

8

66528

ZK−7967A−GE

To open a Single Process Summary window, do one of the following:

• In any window that displays processes (CPU, CPU Modes, Memory, Process

I/O, and Single Lock Summary), double-click on any field. You can also click

MB3 on any field in a process line, and choose Display from the pop-up menu.

• You can also click on any field in a process line, and choose Display Process from the View menu.

• In the Event Log window, double-click on a process-related event. You can also click MB3 on a process-related event, choose Display from the menu, and choose Single Process in the dialog box.

3–19



Table 3–11 describes the Single Process Summary window data fields.

Table 3–11 Single Process Summary Window Data Fields

Field Displays

Process name

Username

Account

UIC

PID

Owner ID

PC

PSL

Priority

The name of the process.

The user name of the user owning the process.

The string assigned to the user by the system manager.

The user identification code (UIC), a pair of numbers or character strings designating the group and user.


The PID of the process that created the process displayed in the window. If 0, then the process is a parent process.

The program counter. On OpenVMS VAX systems, this is the address of the next instruction the CPU will execute. On OpenVMS Alpha systems, this value is displayed as 0, because the data is not readily available to the Data Provider node.

The processor status longword (PSL). On OpenVMS VAX systems, this indicates the current processor mode (user, kernel, and so on) and its interrupt level. On OpenVMS Alpha systems, this value is displayed as

0, because the data is not readily available to the Data Provider node.

The computable and base priority of the process. Priority is an integer between 0 and 31. Processes with higher priority get more CPU time.


3–20



Table 3–11 (Cont.) Single Process Summary Window Data Fields

Field Displays

State One of the following process states:

CEF

Common Event Flag, waiting for a Common Event Flag

COLPG

Collided Page Wait, involuntary wait state; likely indicates a memory shortage, waiting for hard page faults

COM

COMO

CUR

FPW

Computable; ready to execute

Computable Outswapped, COM, but swapped out

Current, currently executing in a CPU

Free Page Wait, involuntary wait state; likely indicates a memory shortage

Local Event Flag, waiting for a Local Event Flag

LEF

LEFO

HIB

Local Event Flag Outswapped; LEF, but outswapped

Hibernate, voluntary wait state requested by the process; it is inactive

HIBO

Hibernate Outswapped, hibernating but swapped out

MWAIT

Miscellaneous Resource Wait, involuntary wait state; possibly caused by a shortage of a systemwide resource such as no page or swap file capacity or synchronizations for single threaded code

PFW

Page Fault Wait, involuntary wait state; possibly indicates a memory shortage, waiting for hard page faults

RWAST

Resource Wait State, waiting for delivery of an asynchronous system trap (AST) that signals a resource availability; usually an I/O is outstanding or a process quota is exhausted

RWBRK

Resource Wait for BROADCAST to finish

RWCAP

Resource Wait for CPU Capability

RWCLU

Resource Wait for Cluster Transition

RWCSV

Resource Wait for Cluster Server Process

RWIMG

Resource Wait for Image Activation Lock

RWLCK

Resource Wait for Lock ID data base

RWMBX

Resource Wait on MailBox, either waiting for data in mailbox (to read) or waiting to place data (write) into a full mailbox (some other process has not read from it; mailbox is full so this process cannot write).

RWMPB

Resource Wait for Modified Page writer Busy

RWMPE

Resource Wait for Modified Page list Empty

RWNPG

Resource Wait for Non Paged Pool

RWPAG

Resource Wait for Paged Pool

RWPFF

Resource Wait for Page File Full

RWQUO

Resource Wait for Pooled Quota

RWSCS

Resource Wait for System Communication Services

RWSWP

Resource Wait for Swap File space

SUSP

Suspended, wait state process placed into suspension; it can be resumed at the request of an external process

SUSPO

Suspended Outswapped, suspended but swapped out


3–21




Field Displays

WS global pages The shared data or code between processes, listed in pages or pagelets.

WS private pages

The amount of accessible memory, listed in pages or pagelets.

WS total pages The sum of global and private pages or pagelets.

WS size

WSdef

WSquo

WSextent

The working set size, number of pages or pagelets of memory the process is allowed to use. This value is periodically adjusted by the operating system based on analysis of page faults relative to CPU time used. When it increases in large units, this indicates a process is taking a lot of page faults and its memory allocation is increasing.

The working set default, the initial limit to the number of physical pages or pagelets of memory the process can use. This parameter is listed in the user authorization file (UAF); discrepancies between the

UAF value and the displayed value are due to page/longword boundary rounding or other adjustments made by the operating system.

The working set quota, the maximum amount of physical pages or pagelets of memory the process can lock into its working set. This parameter is listed in the UAF; discrepancies between the UAF value and the displayed value are due to page/longword boundary rounding or other adjustments made by the operating system.

The working set extent, the maximum number of physical pages or pagelets of memory the system will allocate for the process. The system provides memory to a process beyond its quota only when it has an excess of free pages and can be recalled if necessary. This parameter is listed in the UAF; any discrepancies between the UAF value and the displayed value are due to page/longword boundary rounding or other adjustments made by the operating system.

The number of times an image is activated.

Images activated

Mutexes held The number of mutual exclusions (mutexes) held. Persistent values other than zero (0) require analysis. A mutex is similar to a lock but is restricted to one CPU. When a process holds a mutex, its priority is temporarily incremented to 16.

CPU

Direct I/O

Buffered I/O

Paging I/O

Page Faults

Execution Rates

The percent of CPU time used by this process. This is the ratio of CPU time to elapsed time. CPU rate is also displayed in the bar graph.

The rate at which I/O transfers take place from the pages or pagelets containing the process buffer that the system locks in physical memory to the system devices.

The rate at which I/O transfers take place for the process buffer from an intermediate buffer from the system buffer pool.

The rate of read attempts necessary to satisfy page faults. This is also known as Page Read I/O or the Hard Fault Rate.

The page faults per second for the process. The bar graph visually represents page faults per second.


3–22




Field Displays

DIOLM

BIOLM

ASTLM

CPU

Compute

Memory

Direct I/O

Buffered I/O

Control

Quotas

Explicit

Process Quotas in Use

1

Direct I/O Limit. A bar graph representing current count of DIOs used with respect to the limit that can be attained.

Buffered I/O Limit. A bar graph representing current count of BIOs used with respect to the limit that can be attained.

Asynchronous System Traps Limit. A bar graph representing current count of ASTs used with respect to the limit that can be attained.

CPU Time Limit. A bar graph representing current count of CPU time used with respect to the limit that can be attained. If the limit is 0, then this value is not used.

Wait States

2

A relative value indicating that the process is waiting for CPU time.

The included states are COM, COMO, RWCAP.

A relative value indicating that the process is waiting for a page fault that requires data to be read from disk; common during image activation. The included states are PFW, COLPG, FPG, RWPAG,

RWNPG, RWMPE, RWMPB.

A relative value indicating that the process is waiting for data to be read from or written to a disk. The included state is DIO.

A relative value indicating that the process is waiting for data to be read from or written to a slower device such as a terminal, line printer, or mailbox. The included state is BIO.

A relative value indicating that the process is waiting for another process to release control of some resource. The included states are CEF, MWAIT, LEF, LEFO, RWAST, RWMBX, RWSCS, RWCLU,

RWCSV, RWUNK, and LEF waiting for a ENQ.

A relative value indicating that the process is waiting because the process has exceeded some quota. The included states are QUOTA and

RWAST_QUOTA.

A relative value indicating that the process is waiting because the process asked to wait, such as a hibernate system service. The included states are HIB, HIBO, SUSP, SUSPO, and LEF waiting for a TQE.

FILLM

PGFLQUO

ENQLM

Job Quotas in Use

File Limit. A bar graph representing current number of open files with respect to the limit that can be attained.

Page File Quota. A bar graph representing current number of disk blocks in page file that the process can use with respect to the limit that can be attained.

Enqueue Limit. A bar graph representing current count of resources

(lock blocks) queued with respect to the limit that can be attained.

1

When you display the SWAPPER process, no values are listed in this section. The SWAPPER process does not have quotas defined in the same way other system and user processes do.

2

The wait state specifies why a process cannot execute, based on application-specific calculations.


3–23




Field Displays

TQELM

PRCLM

BYTLM

Image Name

Job Quotas in Use

Timer Queue Entry Limit. A bar graph representing current count of timer requests with respect to the limit that can be attained.

Process Limit. A bar graph representing current count of subprocesses created with respect to the limit that can be attained.

Buffered I/O Byte Limit. A bar graph representing current count of bytes used for buffered I/O transfers with respect to the limit that can be attained.

The name of the currently executing image, if available. If this field does not appear, then the data is not resident in memory.

DECamds displays them in the Event Log window. Node is replaced by the name of the node to which the event is related. Process is replaced by the name of the process to which the event is related.

LOASTQ, node process has used most of its ASTLM process quota

LOBIOQ, node process has used most of its BIOLM process quota

LOBYTQ, node process has used most of its BYTLM job quota

LODIOQ, node process has used most of its DIOLM process quota

LOENQU, node process has used most of its ENQLM job quota

LOFILQ, node process has used most of its FILLM job quota

LOPGFQ, node process has used most of its PGFLQUOTA job quota

LOPRCQ, node process has used most of its PRCLM process quota

LOTQEQ, node process has used most of its TQELM job quota

LOWEXT, node process working set extent is too small

LOWSQU, node process working set quota is too small

PRBIOR, node process buffered I/O rate is high

PRBIOW, node process waiting for buffered I/O

PRCCOM, node process waiting in COM or COMO

PRCCUR, node process has high CPU rate

PRCMUT, node process waiting for a mutex

PRCPUL, node process has used most of its CPUTIME process quota

PRCPWT, node process waiting in COLPG, PFW, or FPG

PRCQUO, node process waiting for a quota

PRCRWA, node process waiting in RWAST

PRCRWC, node process waiting in RWCAP

PRCRWM, node process waiting in RWMBX

PRCRWP, node process waiting in RWPAG, PWNPG, RWMPE, or RWMPB

PRCRWS, node process waiting in RWSCS, RWCLU, or RWCSV

PRCUNK, node process waiting for a system resource

PRDIOR, node process direct I/O rate is high

PRDIOW, node process waiting for direct I/O

PRLCKW, node process waiting for a lock

PRPGFL, node process high page fault rate


3–24


3.11 Lock Contention Summary Window


The Lock Contention Summary window shown in Figure 3–12 determines which resources are under contention. It displays all the OpenVMS Lock Manager resources that have potential lock contention situations. The Lock Contention

Summary window is available only for groups; attempting to open a Lock

Contention Summary for a node opens the node’s group window.

Figure 3–12 Lock Contention Summary Window

EVMS Lock Contention Summary


Resource Name Master Node Parent Resource

DECW$SERVER_2680009D_0066_0

DECW$SERVER_268000A7_0069_0

DECW$CLIENT_268000A6_0071_0

DECW$CLEINT_268000A7_0069_0

DECW$CLIENT_2680009B_003A_0

DECW$CLIENT_2680009E_0067_0

DECW$CLIENT_2680009B_003C_0

LCKPAG

LCKPAG

LCKPAG

LCKPAG

LCKPAG

LCKPAG

LCKPAG

Help

Duration Status

0 00:09:10 VALID

0 00:09:10

0 00:09:10

0 00:09:10

VALID

VALID

VALID

0 00:09:10

0 00:09:10

0 00:09:10

VALID

VALID

VALID

ZK−7956A−GE

Locks are written to AMDS$LOCK_LOG.LOG; see Section B.3 for more information. To interpret the information displayed in the Lock Contention

Summary window, you should have an understanding of OpenVMS lock management services. For more information, see the OpenVMS System Services

Reference Manual.

Note

Lock contention data is accurate only if every node in an OpenVMS

Cluster environment is in the same group. Multiple clusters can share a group, but clusters cannot be divided into different groups without losing accuracy.

You can open a Lock Contention Summary window from the Event Log or System

Overview windows, as follows:

• In the Event Log window, click MB3 on any lock contention-related event and choose Display from the menu.

• In the System Overview window:

1.

Click MB3 on any node or group line, and choose Display from the menu.

2.

Choose Lock Contention Summary from the submenu.

3–25



Table 3–12 describes the Lock Contention Summary window data fields.

Table 3–12 Lock Contention Summary Window Data Fields

Field Displays

Resource Name

Master Node

The resource name associated with the $ENQ system service call.

The node on which the resource is mastered.

Parent Resource The name of the parent resource. If no name is displayed, the resource listed is the parent resource.

Duration

Status

The amount of time elapsed since DECamds first detected the contention situation.

The status of the lock. See the $ENQ(W) description in the OpenVMS

System Services Reference Manual.

You can open a Single Lock Summary window from the Lock Contention

Summary window. See Section 3.12 for more information.

Figure 3–13 shows how to determine which filters can or cannot be displayed.

To filter specific locks from the display, choose Filter Data... from the Customize menu on the Lock Contention Summary window. A filter dialog box appears with a list of locks currently being filtered from the display.

To add a filter, use either of the following methods:

• Type the name of a filter in the Input Lock Name to Filter field and click on the Add button. You can use the asterisk ( * ) wildcard character to specify a range of filters. For example, $DSA*$WAITER will filter all locks beginning with $DSA and ending with $WAITER and anything in between.

• Click on a lock in the Lock Contention Summary window. The name of the lock will appear in the Input Lock Name to Filter field (as shown in

Figure 3–13). You must click on the Add button to add the filter.

3–26



Figure 3–13 Filtering Lock Events

EVMS Lock Contention Summary


Resource Name Master Node

$DSA0064_COPIER LOADQ

MOU$_DAD44: ZAPNOT

Status

0 00:25:35 INVALID

0 00:24:30 VALID

Lock Contention Summary Filtering

Filtered Resources Name List

ALIAS$

DQS$

NSCHED_

CACHE$

AUDRSV$

Input Lock Name to Filter:

MOU$_DAD44:

Help

OK

Add

Remove Cancel

ZK−7957A−GE

You can remove a lock from the filter list by selecting a lock and clicking on the Remove button. Any lock contentions affected by the removed filter will be displayed.

DECamds detects the following lock contention-related events and displays them in the Event Log window. Italicized words are replaced with actual values.

LCKCNT, node possible contention for resource resource

LRGHSH, node lock hash table too large n entries

RESDNS, node resource hash table dense percentage full n resources, hash table size n

RESPRS, node resource hash table sparse, only percentage full n resources, table size n

3–27


3.12 Single Lock Summary Window


The Single Lock Summary window shown in Figure 3–14 displays data about a blocking lock and all locks in the granted, conversion, and waiting queues. You can use it to display detailed information about a lock contention situation. The lock name is specified in the title bar. All locks are written to AMDS$LOCK_

LOG.LOG; see Section B.3 for more information.

Figure 3–14 Single Lock Summary Window

LCKPAG Single Lock Summary for DECW$SERVER_268000A7_D069_0


Granted Lock

Node:

LKID:

LCKPAG

050007F3

Process Name: DECW$SERVER_0

Lock Type: Local Copy

Resource Name: DECW$SERVER_268000A7_0069_0

Parent Resource:

Help

Granted Queue

Node

........

LCKPAG

Process Name

DECW$SERVER_0

LKID

........

GR Mode Duration

.....

Flags

............................................................

050007F3 EX 0 00:00:26 NOQUEUE SYNCSTS SYSTEM NODL CKW

Conversion Queue

Node Process Name LKID GR Mode Duration RQ Mode Flags

3–28

Waiting Queue

Node

........

Process Name

LCKPAG DECW$TE_00A7

LKID

........

RQ Mode Duration

.....

Flags

............................................................

010007F5 EX 0 00:00:26 SYSTEM NODL CKW NODL CKBLK

ZK−7966A−GE

In a Single Lock Summary window, if DECamds cannot determine the node name for the group, it uses the cluster system ID (CSID) value, which the OpenVMS

Cluster software uses to uniquely identify cluster members.



To open a Single Lock Summary window, do one of the following:

• In the Lock Contention Summary window, double-click on any field. You can also click MB3 on any field, and choose Display Lock from the menu.

• In the View menu of the Lock Contention Summary window, choose Display

Lock.

• In the Event Log window, click MB3 on any lock blocking-related or lock waiting-related event, and choose Display from the menu.

Table 3–13 describes the Single Lock Summary window data fields.

Node

Process

Name

LKID

GR Mode

RQ Mode

Duration

Flags

Table 3–13 Single Lock Summary Window Data Fields

Field Displays

Node

LKID

Process

Name

Lock Type

Resource

Name

Parent

Resource

Granted Lock

The node name on which the lock is granted.

The lock ID value (which is useful with SDA).

The name of the process owning the blocking lock.

One of the following: Local Copy, Process Copy, or Master Copy.

The name of the resource.

The name of the parent resource (if any).

Granted, Conversion, and Waiting Queue

The node on which the lock block resides.

The process name of the process owning the lock.

The lock ID value (which is useful with SDA).

One of the following modes at which the lock is granted: EX, CW, CR, PW,

PR, NL.

One of the following modes at which the lock is requested: EX, CW, CR, PW,

PR, NL.

The length of time the lock has been in the current queue (since the console application found the lock).

The flags specified with the $ENQ(W) request.

You can open a window about a specific process in the Single Lock Summary window by double-clicking on the process name.

Note

Processes that are labeled unknown are associated with system locks.

They cannot be opened.

DECamds detects the following single lock-related events and displays them in the Event Log window. Node is replaced by the name of the node to which the event is related. Process is replaced by the name of the process to which the event is related.

3–29



LCKBLK, node process blocking resource resource

LCKWAT, node process waiting for resource resource granted to process on node node

3.13 Cluster Transition/Overview Summary Window

The Cluster Transition/Overview Summary window shown in Figure 3–15 displays information about each node in an OpenVMS Cluster. This window is very similar to the System Overview window; however, the Cluster Transition window lists only one cluster for each set of nodes in a cluster, while the System

Overview window lists all the nodes and the user-defined groups the nodes are in.

The window displays summary information as well as information about individual nodes: System Communication Services (SCS) name, SCS ID, Cluster

System ID, Votes, Lock Directory Weight value, cluster status, and last transition time.

The data items shown in the window correspond to data that the Show Cluster utility displays for the SYSTEM and MEMBERS classes. A status field display of "unknown" usually indicates that DECamds is not communicating with the node.

3–30



Figure 3–15 Cluster Transition/Overview Summary Window

Cluster Transition/Overview Summary


Summary

Formed

Last Trans:

Votes:

Expected Votes:

Failover Step:

29−APR−1996

1−AUG−1996

12

15

55

11:47 Members In:

10:46 Members Out:

Quorum:

QD Votes:

Failover ID:

29

2

8

65535

072

Help

...................................................................................................................................................................................

Cluster Members

SCS Name

SAREKS

WEEKS

MACHU

RUMAD

DFODIL

AZSUN

CLAWS

CALPAL

VAX5

CRNPOP

LOADQ

GNRS

PITMOD

4X4TRK

VMSRMS

ALTOS

FARKLE

TSAVO

ETOSHA

CLAIR

MILADY

CHOBE

ZOON

ZAPNOT

2BOYS

ORNDT

BARNEY

ARUSHA

SUB4

GLOBBO

SCS Id CSID

4CC7

4CBB

FDAA

4CA7

FFA2

4CA1

FE94

4C93

4D12

4EF0

FD77

4C60

FF60

4D56

4C39

4C34

4C32

FD32

100E5

100DD

20002

100F7

100FC

20006

100EB

100C8

100EA

20003

4C31

FC2B

FE29

FF26

FD24

4DOF

100A2

100F9

20008

100FF

100FA

100DB

FE03

4CFE

20001

100CF

4CF3 100CE 1

0

1

4CDF

4ED8

4CD6

100F4

100F6

20009

0

0

1

0

0

1

1

0

0

20004

100ED

100C5

100CO

100FE

100FO

100D9

100FD

1

0

1

0

0

1

0

0

1

1

0

0

1

1

0

0

0

0

Votes Expect

1

15

3

13

3

13

13

15

15

13

3

13

13

13

3

13

3

15

13

13

15

13

15

3

15

5

13

3

13

13

Quorum

8

2

7

2

7

7

8

8

7

7

8

2

7

2

7

2

2

7

7

8

7

7

8

2

7

7

1

8

3

7

Lck:DirWt

0

1

0

1

1

0

1

0

0

0

1

0

0

0

1

0

1

0

1

0

1

1

0

0

0

0

0

0

0

0

Status Transition Time

UNKNOWN

UNKNOWN

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

MEMBER

16−JUL−1996

10−JUL−1996

28−JUL−1996

28−JUL−1996

28−JUL−1996

30−JUL−1996

16−JUL−1996

21−JUN−1996

1−AUG−1996

28−JUL−1996

8−JUN−1996

28−JUL−1996

31−JUL−1996

26−JUL−1996

28−JUL−1996

6−JUL−1996

28−JUL−1996

2−JUL−1996

29−JUN−1996

24−JUL−1996

28−JUL−1996

1−AUG−1996

30−JUL−1996

17−JUL−1996

15−JUN−1996

12−JUN−1996

28−JUL−1996

20−JUL−1996

4−JUL−1996

26−JUL−1996

08:47

11:54

11:08

12:01

11:55

06:41

12:01

06:56

08:19

09:10

11:51

10:46

15:47

14:25

09:43

12:06

11:52

11:55

21:43

16:22

06:48

10:46

12:11

13:28

13:36

13:51

11:56

06:43

16:37

11:55

ZK−8545A−GE

To open the Cluster Transition/Overview Summary window, do either of the following:

• In the System Overview window, click MB3 on a node line. Choose Display from the menu displayed and Cluster Transition Summary from the submenu.

The system displays the Cluster Transition/Overview Summary window.

• In the Event Log window, Click MB3 on a cluster-related event. Choose

Display from the menu displayed and Cluster Transition Summary from the list displayed.

3–31



Note: The Cluster Transition Summary menu option is not available for nodes that are not in the cluster; it is not available from lines that display groups.

3.13.1 Data Displayed

The Cluster Transition/Overview window has two panel displays:

• The Summary (top) panel displays cluster summary information.

• The Cluster Members (bottom) panel lists each node in the cluster.

Table 3–14 describes the Summary panel data fields.

Table 3–14 Data Items in the Summary Panel of the Cluster Transition/Overview

Summary Window

Data Item Description

Formed

Last Trans

Votes

Expected Votes

Failover Step

Members In

Members Out

Quorum

QD Votes

Failover ID

Date and time the cluster was formed.

Date and time of the most recent cluster state transition.

Total number of quorum votes being contributed by all cluster members and quorum disk.

Number of votes expected to be contributed by all members of the cluster as determined by the connection manager. This value is based on the maximum of the EXPECTED_VOTES system parameter and the maximized value of the VOTES system parameter.

Current failover step index; shows which step in the sequence of failover steps the failover is currently executing.

Number of members of the cluster DECamds has a connection to.

Number of members of the cluster DECamds either has no connection to or has lost connection to.

Number of votes required to keep cluster above quorum.

Number of votes given to Quorum Disk. A value of 65535 means there is no Quorum Disk.

Failover Instance Identification: unique ID of a failover sequence; indicates to system managers whether a failover has occurred since the last time they checked.

Table 3–15 describes the Cluster Members panel data fields.

Table 3–15 Data Items in the Cluster Members Panel of the Cluster

Transition/Overview Summary Window


SCS Name

SCS id

CSID

Votes

Expect

System Communication Services name for the node (system parameter

SCSNODE)

System Communication Services identification for the node (system parameter SCSYSTEMID)

Cluster System Identification

Number of votes the member contributes

Expected votes to be contributed as set by the EXPECTED_VOTES system parameter


3–32



Table 3–15 (Cont.) Data Items in the Cluster Members Panel of the Cluster

Transition/Overview Summary Window


Quorum

Lck:DirWt

Status

Recommended quorum value derived from the expected votes

Lock Manager distributed directory weight as determined by the

LCKDIRWT system parameter

Current cluster member status: MEMBER, UNKNOWN, or BRK_NON

(break nonmember)

Transition Time Time cluster member had last transition

3.13.2 Notes About the Display

Following are notes about the display of data in the window:

• No highlighting conventions are used in the window; all data items are displayed in normal mode.

• You cannot filter out any data.

• The data items in the window are sorted on an "as-found" basis. You cannot change the sort criteria.

• When you click on an item, DECamds temporarily stops updating the window for 15 seconds or until you choose an item from a menu.

• DECamds signals the LOVOTE event when the difference between the cluster’s quorum and votes values is less than the threshold for the event:

LOVOTE, ’node’ VOTES count is close to or below QUORUM

The default threshold for LOVOTE is 1.

• You can change collection intervals.

3.14 System Communications Architecture Summary Window

The System Communications Architecture Summary (SCA Summary) window shown in Figure 3–16 displays information about a selected node’s virtual circuits and connections to other nodes in a cluster. (The display represents the view one node has of other nodes in the cluster.) More than one type of virtual circuit indicates that more than one path to the remote node exists.

3–33



Figure 3–16 SCA Summary Window

DFODIL System Communication Architecture Summary


MACHU

RUNAD

DFODIL

AZSUN

CLAWS

CALPAL

VAX5

CRNPOP

LOADQ

GNRS

PITMOD

4X4TRK

VMSRMS

ALTOS

FARKLE

TSAVO

ETOSHA

CLAIR

MILADY

CHOBE

ZOON

ZAPNOT

2BOYS

ORNOT

BARNEY

ARUSHA

SUB4

NodeName

Local SysApp

VC (Type)

Remote SysApp State

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

(LAN) OPEN

(LAN) OPEN

PEA0: (LAN) OPEN

PEA0: (LAN) OPEN

PEAO:

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

PEA0:

PEA0:

PEA0:

PEA0:

PEA0:

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

(LAN) OPEN

........

Messages

........

Rcvd

0.00

0.00

0.00

0.01

0.00

0.00

0.04

0.00

0.00

0.00

0.04

0.00

0.00

0.00

0.00

0.04

0.04

0.00

0.01

0.00

0.04

0.00

0.04

0.01

0.00

0.04

0.00

0.00

0.00

0.00

0.04

0.04

0.00

0.00

0.04

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.04

0.00

0.00

0.00

0.04

0.00

0.04

0.00

0.00

0.04

0.00

KB Mapped

......

Block Data (KB)

Sent Rcvd

........

Block Transfer

Sent

........

Reqd

**Use MB3 to switch between raw or rate display**

Datagrams

........

Sent Rcvd Credit Wait CDT

0

13

0

0

0

4

0

2

18

0

4

8

0

0

0

15

0

0

0

0

15

3

0

0

31

0

0

46

16

3025

47

32964

16

224

16

258

16

168

148

16

16

24

16

0

16

65

916

6416

17

16059

15

16

16

26

15

7

0

7

22

148

432

7

116

7

7

7

8

226

7

156

135

7

7

38

7

171

7

268

7

191

7

0

0

0

0

14

12

22

0

35

0

0

0

0

21

0

12

17

0

0

4

0

14

0

6

0

29

0

50

26

0

26

39

561

2772

26

447

25

25

25

29

888

25

610

505

25

25

154

25

595

25

927

25

698

25

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

256

0

358

1

0

0

0

0

0

0

0

0

110

0

0

0

0

0

0

0

341

39

0

0

0

ZK−8546A−GE

Each line in the window shows either a summary of all system applications

(SysApps) using the virtual circuit communication or the communication on the connection between a local and a remote SysApp. The data displayed in the window is similar to the information that the Show Cluster utility displays for the CIRCUITS, CONNECTIONS, and COUNTERS classes. Unlike Show Cluster, however, this display shows only SCA connections to other OpenVMS nodes; it does not show SCA connections to the Disk Storage Architecture (DSA) or to devices such as FDDI or DSSI disk controllers.

By clicking MB3 on a node name and choosing View SysApps from the pop-up menu, you can display the system applications that are using virtual circuits.

This option expands the list below a virtual circuit to show all the system applications that contribute to that virtual circuit. (The SysApp lines are dimmed and right-justified.)

To hide the display of system applications, click MB3 and choose Hide SysApps from the pop-up menu.

To display a menu that allows you to toggle between Raw and Rate data, click

MB3 on the data to the right of the State field. (For messages, the default is the display of rate data; raw data is the default for all other types of data.)

To open an SCA Summary window, follow these steps:

1.

In the Cluster Transition/Overview Summary window, click MB3 on an SCS name.


2.

Choose Display SCA Summary.

3–34



The system displays the System Communication Architecture (SCA)

Summary window.

Table 3–16 describes the SCA Summary window data fields.

Table 3–16 Data Items in the SCA Summary Window


NodeName

VC(Type)

State

Messages

Block Transfer

KB Mapped

SCS name of the remotely connected node.

The virtual circuit being used and its type.

The state of the virtual circuit connection.

Relatively small data packets sent and received between nodes for control information.

Fields listing the count of the number of block data transfers and requests initiated.

Field listing the number of kilobytes mapped for block data transfer.

Note: This field is available in Raw data format only.

Block Data (KB) Fields listing in kilobytes the data transferred via block data transfer.

Datagrams Number of unacknowledged messages sent between virtual circuits.

Credit Wait

CDT Wait

Local SysApp

Number of times the connection had to wait for a send credit.

Number of times the connection had to wait for a buffer descriptor.

Name of the local system application using the virtual circuit.

Remote SysApp Name of the remote system application being communicated to.



• The window does not follow highlighting conventions: virtual circuit lines are displayed normally and are left-aligned; SysApp lines are dimmed and are indented by a column.

• You cannot filter out any data.

• The data items in the window are sorted on an "as-found" basis. You cannot change sort criteria at this time.

• DECamds signals the LOSTVC event when a virtual circuit between two nodes has been lost. This loss might be due either to a cluster node crashing or to cluster problems that caused the virtual circuit to close.

LOSTVC, <node> lost virtual circuit (<string>) to node <node>


3–35


3.15 NISCA Summary Window


The Network Interconnect System Communication Architecture (NISCA) is the transport protocol responsible for carrying messages such as disk I/Os and lock messages across Ethernet and FDDI LANs to other nodes in the cluster. More detailed information about the protocol is in the OpenVMS Cluster Systems manual.

The NISCA Summary window shown in Figure 3–17 displays detailed information about the LAN (Ethernet or FDDI) connection between two nodes. DECamds displays one window per virtual circuit provided the virtual circuit is running over a PEA0: device.

The purpose of this window is to view statistics in real time and to troubleshoot problems found in the NISCA protocol. The window is intended primarily as an aid to diagnosing LAN-related problems. The OpenVMS Cluster Systems manual describes the parameters shown in this window and tells how to use them to diagnose LAN-related cluster problems.

The window provides the same information as the OpenVMS System Dump

Analyzer (SDA) command SHOW PORTS/VC=VC_nodex. (VC refers to virtual circuit; nodex is a node in the cluster. The system defines VC-nodex after a

SHOW PORTS command is issued from SDA.)

3–36

Figure 3–17 NISCA Summary Window



DFODIL NISCA Connection to MACHU


Transmit

Item

Packets

Unsequenced (DG)

Sequenced

Lone ACK

ReXmt Count

ReXmt Timeout

ReXmt Ratio

Bytes

Raw Rate

37246

3

18752

18486

5

14

N/A

1817582

0.02

0.00

0.00

0.00

0.00

0.00

0.0000

1.04

Receive

Item

Packets

Unsequenced (DG)

Sequenced

Lone ACK

Duplicate

Out of Order

Illegal ACK

Bytes

Raw

19908

3

18786

1115

5

0

0

1646125

Rate

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.92

Help

Congestion Control

Item

Transmit Window Current

Transmit Window Grow

Transmit Window Max

Transmit Window Reached

Roundtrip uSec

Roundtrip Deviation uSec

Retransmit Timeout uSec

UnAcked Messages

CMD Queue Length

CMD Queue Max

Value

9

5

16

16

11230

0

1

1312

31729

0

VC Closures

Item

SeqMsg TMO

CC DFQ Empty

Topology Change

NPAGEDYN Low

Count

0

0

0

0

Channel Selection

Item Value

Buffer Size

Channel Count

Channel Selections

Protocol

Local Device

Local LAN Address

Remote Device

Remote LAN Address

1412

1

15

1.4.0

ES_LANCE

AA−00−D4−00−6D−FF

E2_TGEC

AA−00−D4−00−39−4C

Packets Discarded

Item

No Xmt Chan

Ill Seq Msg

TR DFQ Empty

CC MFQ Empty

Count

0

0

0

0

Item

Rcv Short Msg

Bad Checksum

TR MFQ Empty

Cache Miss

Count

0

0

0

0

ZK−8547A−GE

To open an NISCA Summary window, do one of the following:

• In the SCA Summary window, click MB3 on a row with the PEA0: Virtual

Circuit. Choose View SysApps from the popup menu, click MB3 on a SysApps node, and Choose Display NISCA. The system displays the NISCA Summary window.

Note: If the Display NISCA option is dimmed, the NISCA protocol is not running for that system application.

• Double-click MB1 on a row with a PEA0: to display an expanded list below the node name.

• Double-click MB1 on a SysApps node to display the NISCA Summary window.

3–37



3.15.1 Data Displayed

Panels in the NISCA Summary window contain the data described in the following tables.

Table 3–17 lists data items displayed in the Transmit Panel, which contains data packet transmission information.

Table 3–17 Data Items in the Transmit Panel


Packets

Unsequenced (DG)

Sequenced

Lone ACK

ReXmt Count

ReXmt Timeout

ReXmt Ratio

Bytes

Number of packets transmitted through the virtual circuit to the remote node, including both sequenced and unsequenced (channel control) messages, and lone acknowledgments.

Count and rate of the number of unsequenced datagram packages transmitted.

Count and rate of the number of sequenced packages transmitted.

Sequenced messages are used for application data.

Count and rate of the number of lone acknowledgments.

Number of packets retransmitted. Retransmission occurs when the local node does not receive an acknowledgment for a transmitted packet within a predetermined timeout interval.

Number of retransmission timeouts that have occurred.

Ratio of ReXmt Count current and past to the current and past number of sequenced messages sent.

Count and rate of the number of bytes transmitted through the virtual circuit.

Table 3–18 describes data items displayed in the Receive Panel, which contains data packet reception information.

Table 3–18 Data Items in the Receive Panel


Packets

Unsequenced (DG)

Sequenced

Lone ACK

Duplicate

Out of Order

Illegal Ack

Bytes

Number of packets transmitted through the virtual circuit to the remote node, including both sequenced and unsequenced (channel control) messages, and lone acknowledgments.

Count and rate of the number of unsequenced packages received.

Count and rate of the number of sequenced packages received.

Sequenced messages are used for application data.

Count and rate of the number of lone acknowledgments.

Number of redundant packets received by this system.

Number of packets received out of order by this system.

Number of illegal acknowledgments received.

Count and rate of the number of bytes received through the virtual circuit.

Table 3–19 describes data items displayed in the Congestion Control Panel, which contains transmit congestion control information.

3–38



The values in the panel list the number of messages that can be sent to the remote node before receiving an acknowledgment and the retransmission timeout.

Table 3–19 Data Items in the Congestion Control Panel


Transmit Window

Current

Transmit Window

Grow

Transmit Window

Max

Transmit Window

Reached

Current value of the pipe quota (transmit window). After a timeout, the pipe quota is reset to 1 to decrease congestion and is allowed to increase quickly as acknowledgments are received.

The slow growth threshold: size at which the rate of increase is slowed to avoid congestion on the network again.

Maximum value of pipe quota currently allowed for the virtual circuit based on channel limitations.

Number of times the entire transmit window was full. If this number is small as compared with the number of sequenced messages transmitted, the local node is not sending large bursts of data to the remote node.

Roundtrip uSec

Roundtrip

Deviation uSec

Retransmit

Timeout uSec

Average roundtrip time for a packet to be sent and acknowledged.

The value is displayed in microseconds.

Average deviation of the roundtrip time. The value is displayed in microseconds.

Value used to determine packet retransmission timeout. If a packet does not receive either an acknowledging or a responding packet, the packet is assumed to be lost and will be resent.

UnAcked Messages Number of unacknowledged messages.

CMD Queue

Length

Current length of all command queues.

CMD Queue Max Maximum number of commands in queues so far.

Table 3–20 describes data items displayed in the Channel Selection Panel, which contains channel selection information.

Table 3–20 Data Items in the Channel Selection Panel


Buffer Size Maximum PPC data buffer size for this virtual circuit.

Channel Count Number of channels connected to this virtual circuit.

Channel Selections Number of channel selections performed.

Protocol NISCA Protocol version.

Local Device Name of the local device that the channel uses to send and receive packets.

Local LAN Address Address of the local LAN device that performs sends and receives.

Remote Device

Remote LAN

Address

Name of the remote device that the channel uses to send and receive packets.

Address of the remote LAN device performing the sends and receives.

Table 3–21 describes data items displayed in the VC Closures panel, which contains information about the number of times a virtual circuit has closed for a particular reason.

3–39



Table 3–21 Data Items in the VC Closures Panel


SeqMsg TMO

CC DFQ Empty

Topology Change

NPAGEDYN Low

Number of sequence transmit timeouts.

Number of times the channel control DFQ was empty.

Number of times PEDRIVER performed a failover from FDDI to

Ethernet, necessitating the closing and reopening of the virtual circuit.

Number of times the virtual circuit was lost because of a pool allocation failure on the local node.

Table 3–22 lists data items displayed in the Packets Discarded Panel, which contains information about the number of times packets were discarded for a particular reason.

Table 3–22 Data Items in the Packets Discarded Panel


No Xmt Chan

Ill Seq Msg

TR DFQ Empty

CC MFQ Empty

Rcv Short Msg

Bad Checksum

TR MFQ Empty

Cache Miss

Number of times there was no transmit channel.

Number of times an illegal sequenced message was received.

Number of times the Transmit DFQ was empty.

Number of times the Control Channel MFQ was empty.

Number of times a short transport message was received.

Number of times there was a checksum failure.

Number of times the Transmit MFQ was empty.

Number of messages that could not be placed in the cache.



• No highlighting conventions are used in the NISCA Summary window.

• You cannot sort or filter the data displayed in this window.


3–40

4

Performing Fixes

You can perform fixes to resolve resource availability problems and improve system availability.

This chapter covers the following topics:

• Understanding fixes

• Performing fixes

• Typical fix examples

Caution

Performing certain actions to fix a problem can have serious repercussions on a system, including possibly causing a system failure. Therefore, only experienced system managers should perform fixes.

4.1 Understanding Fixes

When DECamds detects a resource availability problem, it analyzes the problem and proposes one or more fixes to improve the situation. Most fixes correspond to an OpenVMS system service call.

The following fixes are available from DECamds:

Fix Category Possible Fixes

Memory usage fixes

Process fixes

Adjust Process Quota

Limit fix

Process state fixes

Adjust working set

Purge working set

Delete a process

Exit an image

Change limits for AST, BIO, DIO,

ENQ, FIL, PRC, and TQE process quota limits

Resume a process

Suspend a process

Process priority fixes Lower or raise a process priority

Quorum fix Adjust cluster quorum

System fix Crash node

System Service Call

$ADJWSL

$PURGWS

$DELPRC

$FORCEX

None

$RESUME

$SUSPND

$SETPRI

None

None

4–1


4.1 Understanding Fixes

Before you perform a fix, you should understand the following information:

• Fixes are optional.

• You must have write access to perform a fix. (See Section 1.3 for more information about DECamds security.)

• You cannot undo many fixes. (After using the crash node fix, for example, the node must be rebooted.)

• The exit image, delete process, and suspend process fixes should not be applied to system processes. Doing so can require rebooting the node.

• Whenever you exit an image, you cannot return to that image.

• Processes that have exceeded their job or process quota cannot be deleted.

• DECamds ignores fixes applied to the SWAPPER process.

4.2 Performing Fixes

Standard OpenVMS privileges restrict write access of users. When you run the Data Analyzer, you must have the CMKRNL privilege to send a write (fix) instruction to a node with a problem.

To initiate a fix, perform one of the following actions:

• From any of the data windows, double-click on a process, and then choose an action from the Fix menu.

• Click MB3 on an event, and choose Fix from the menu.

DECamds displays a dialog box listing the fixes you can perform for the selected event. The recommended choice is highlighted. When you click on OK or Apply,

DECamds performs one of the following actions:

• If the event you selected is not specific to a certain process, DECamds automatically performs the fix. Some fixes are performed automatically when

‘‘(automatic)’’ is displayed next to the selection.

• If the event is specific to a process, DECamds displays another dialog box in which you can specify the fix parameters. For example, for the Adjust

Working Set Size fix, you specify a new working set size for the process.

DECamds performs the highlighted fix as long as the event still exists. If the event you are fixing has changed, the dialog box disappears when you click on

OK, Apply, or Cancel, and the fix is not performed.

Table 4–1 summarizes all fixes alphabetically and specifies the windows from which they are available.

4–2



Table 4–1 Summary of DECamds Fixes

Problem to be Solved Fix

Process quota has reached its limit and has entered

RWAIT state

Cluster hung

Adjust Process Quota

Limit

Adjust Quorum

Working set too high or low

Runaway process

Node resource hanging cluster

Process looping, intruder

Endlessly process loop in same PC range

Node or process low memory

Process previously suspended

Runaway process, unwelcome intruder

Adjust Working Set

Change Process

Priority

Crash Node

Delete Process

Exit Image

Purge Working Set

Resume Process

Suspend Process

Available From

Single Process Summary

Event Log

Effects

Process receives greater limit.

Node Summary

Cluster

Transition/Overview

Summary

Memory Summary


Event Log

CPU Summary


Event Log

System Overview

Node Summary

Single Lock Summary

Any process window

Any process window

Quorum for cluster is adjusted.

Removes unused pages from working set; page faulting might occur.

Priority stays at selected setting.

Node crashes with operator requested shutdown.

Process no longer exists.

Exit from current image.

Event Log

Memory Summary


Event Log

Memory Summary

CPU Summary

Process I/O Summary


Event Log

Memory Summary

CPU Summary

Process I/O Summary


Frees memory; page faulting might occur.

Process starts from point it was suspended.

Process gets no computes.

The following sections provide reference information about each DECamds fix.

4–3



4.2.1 Adjust Quorum Fix

When you perform the Adjust Quorum fix, DECamds displays a dialog box similar to the one shown in Figure 4–1.

Figure 4–1 FIX Adjust Quorum Dialog Box

BHAK − FIX quorum node

This fix will force a cluster quorum adjustment on the entire OpenVMS Cluster upon which this fix is run. Pressing OK will adjust the quorum, while pressing

Cancel will avoid quorum adjustment.

OK

Cancel

ZK−8991A−GE

The Adjust Quorum fix forces the node to refigure the quorum value. This fix is the equivalent of the Interrupt Priority C (IPC) mechanism used at system consoles for the same purpose. The fix forces the adjustment for the entire cluster so that each node in the cluster will have the same new quorum value.

The Adjust Quorum fix is useful when the number of votes in a cluster falls below the quorum set for that cluster. This fix allows you to readjust the quorum so that it corresponds to the current number of votes in the cluster.

4–4



4.2.2 Adjust Process Quota Limit

When you perform the Adjust Process Quota Limit fix, DECamds displays a dialog box similar to the one shown in Figure 4–2.

Figure 4–2 FIX Adjust Process Quota Limit Dialog Box

CALPAL − FIX Adjust Process Quota Limit

Process Name : BATCH_1944 (7100210E)

Current Limit : 600

Select Quota to Modify and then use Slider to Adjust

AST

FIL

600

BIO

TQE

DIO

ENQ

PRC

BYT

1200

Fix Process Quota Limit Size Scale

OK Apply

Cancel

ZK−8992A−GE

If a process is waiting for a resource, you can use the Adjust Process Quota Limit fix to increase the resource limit so that the process can continue. The increased limit is only in effect for the life of the process, however; any new process will be assigned the quota set in the UAF.

To use this fix, select the resource and then use the slide bar to change the current setting. Finally, select one of the following:

• OK — to apply the fix and exit the window

• Apply — to apply the fix and not exit the window (so that you can continue to make changes)

• Cancel — not to perform the fix and exit the window

4–5



4.2.3 Adjust Working Set Fix

When you perform the Adjust Working Set fix, DECamds displays a dialog box similar to the one shown in Figure 4–3.

Figure 4–3 FIX Adjust Working Set Size Dialog Box

DELPHI − FIX Adjust Working Set Size

Process Name : NET_34934 (62A01E6B)

Ws Count : 1144

20

544

Fix Working Set Size Scale

32000

OK Apply

Cancel

ZK−7953A−GE

Adjusting the working set can give needed memory to other processes that are page faulting. In your adjustment, try to bring the working set size closer to the actual count being used by nonpage faulting processes.

Caution

If the automatic working set adjustment is enabled for the system, a fix to Adjust Working Set Size will disable the automatic adjustment for the process.

4.2.4 Change Process Priority Fix

When you perform the Change Process Priority fix, DECamds displays a dialog box similar to Figure 4–4.

4–6



Figure 4–4 FIX Process Priority Dialog Box

DELPHI − FIX Process Priority


Priority : 5/ 4

0

4

Fix Process Priority Scale

31

OK Apply

Cancel

ZK−7972A−GE

Setting a priority too high for a compute-bound process allows it to consume all the CPU cycles on the node, which can affect performance dramatically. On the other hand, setting a priority too low prevents the process from getting enough

CPU cycles to do its job, which can also affect performance.

4.2.5 Crash Node Fix

When you perform the Crash Node fix, DECamds displays a dialog box similar to

Figure 4–5.

Figure 4–5 FIX Crash Node Dialog Box

AMDS − FIX crash node

******* WARNING *******

******* IRRECOVERABLE FIX *******

Pressing OK will force a system crash on the node listed in the title!!!

Press cancel to avoid crashing the node

OK

Cancel

ZK−7954A−GE

4–7



Caution

The crash node fix is an operator-requested bugcheck from the driver. It happens immediately when you click on OK in the Fix Crash Node dialog box. After performing this fix, the node cannot be restored to its previous state. After a crash, the node must be rebooted.

Recognizing a System Failure Forced by DECamds

Because a user with suitable privileges can force a node to fail from the Data

Analyzer by using the Crash Node fix, system managers have requested a method for recognizing these particular failure footprints so that they can distinguish them from other failures. These failures all have identical footprints: they are operator-induced system failures in kernel mode at IPL 8. The top of the kernel stack is similar to the following display:

SP => Quadword system address

Quadword data

1BE0DEAD.00000000

00000000.00000000

Quadword data

Quadword data

TRAP$CRASH

SYS$RMDRIVER + offset

4.2.6 Exit Image and Delete Process Fixes

When you perform either the Exit Image or Delete Process fix, DECamds displays a dialog box similar to Figure 4–6.

Figure 4–6 FIX Process State Dialog Box — Exit Image or Delete Process

DELPHI − FIX suspend or resume process


State : PFW

Exit Image Delete Process

OK Apply

Cancel

ZK−7971A−GE

You cannot reverse the action when you delete a process that is in a resource wait state. You must reboot the node. Deleting a process on a node that is in a resource wait state might not have an effect on the process.

Exiting an image on a node can stop an application that is required by the user.

Check the single process window first to determine which image it is running.

4–8



Caution

Deleting or exiting a system process could corrupt the kernel.

4.2.7 Purge Working Set Fix

When you perform the Purge Working Set fix, DECamds displays a dialog box similar to Figure 4–7.

Figure 4–7 FIX Purge Working Set Dialog Box

DELPHI − FIX Purge Working Set


Ws Count : 482

OK Apply

Cancel

ZK−7973A−GE

Continual purging of a working set on a node could force excessive page faulting, which affects system performance.

4.2.8 Suspend Process and Resume Process Fixes

When you perform either the Suspend Process or Resume Process fix, DECamds displays a dialog box similar to the one shown in Figure 4–8.

4–9



Figure 4–8 FIX Process State Dialog Box — Suspend or Resume Process

DELPHI − FIX suspend or resume process


State : PFW

Suspend Resume

OK Apply

Cancel

ZK−7955A−GE

Suspending a process that is consuming excess CPU time can improve perceived

CPU performance by freeing the CPU for use by other processes. Conversely, resuming a process that was using excess CPU time while running might reduce perceived CPU performance.

Caution

Do not suspend system processes, especially JOB_CONTROL.

4.3 Examples for Fixing Low Memory Availability

This section describes two approaches for solving a low memory problem, which is a common resource availability problem.

The procedure in Section 4.3.1 uses DECamds default settings. The procedure in

Section 4.3.2 shows how you can use DECamds to make a more detailed analysis and investigation. Both examples begin at the Event Log window entry.

4.3.1 Performing a Fix Using Automatic Fix Settings

When a process is page faulting, for example, it may signal a problem of available memory. A low memory (LOMEMY) event is generated. To fix this problem, you should purge the working sets of inactive processes. This will free up memory for the process that is page faulting. DECamds offers a quick, direct way to fix this and similar problems by performing the following steps:

1.

Click MB3 on the event and choose Fix.

If the event is related to a specific process, DECamds displays a dialog box with fixes you can perform. If the event is not related to a specific process but may be related to more than one process, DECamds automatically performs the fix.

In the low memory example, DECamds displays a dialog box suggesting the automatic Purge Working Set fix.

2.

Click on OK or Apply to perform the fix.

4–10



The Purge Working Set fix purges the working set of the five processes that are the highest consumers of memory and are not page faulting. If this fix is not sufficient and the low memory event entry returns, repeat the fix every

15 or 20 seconds until enough working sets are purged to eliminate the event message. If two or three purges are not sufficient, then you should investigate manually.

4.3.2 Performing a Fix Using Manual Investigation

DECamds lets you manually display additional information related to an event before performing a fix. The following example uses the same low memory problem described in the previous section to investigate and select specific fixes for the problem.

For this example, manually select the processes you want to fix from the Memory

Summary window. You also may want to refer to data in the CPU Summary window.

To investigate the low memory event before fixing it, perform the following steps:

1.

Click MB3 on the event and choose Display.

DECamds displays a dialog box with a window name highlighted to indicate the recommended path for information. In the example shown in Figure 4–9, the Memory Summary window is recommended.

Figure 4–9 Sample Fix Dialog Box

DISPLAY − LOMEMY, DELPHI free memory is low

Event Display Choices

Memory Summary

Node Summary

OK

Apply Cancel

ZK−7958A−GE

2.

Click on Apply to open the Memory Summary window shown in Figure 4–10 and keep the dialog box.

4–11



Figure 4–10 DECamds Memory Summary Window

File View Fix

DELPHI Memory Summary

Customize Help

PID

Process

Name

62AQE61D

62AQE802

62AQE86B

62AQE8B4

TGOODWIN_1

TGOODWIN_2

YURYAN

BATCH_2878

62AQE611

62AQE873

. . . the Giant!

_RTA57:

62AQE88B WEINER_1

62AQE83A ABRAMSON

62AQE614

62AQE8C1

62AQE85E

DECW$TE_E614

UCX$BOOT_BG6112

CWINSOR

8542

5578

5233

2537

2469

1235

1032

1028

831

774

740

Count

Working Set

Size Extent Rate

11498 32004

9548 32004

6994

3394

32000

32000

3544

1744

1448

1444

35.44

0.00

0.00

0.00

32000 0.01

32000 0.00

32004 0.00

32000 0.00

6094 32000 0.00

1144 32000 0.80

994 32000 0.00

Page Fault

I/O

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.05

0.00

ZK−7959A−GE

3.

To determine which process consumes the most memory and is not page faulting, you can sort and examine the data in the Memory Summary window.

In this example, the process TGOODWIN_1 is consuming the most memory and is page faulting.

4.

Select the Node Summary window from the Low Memory dialog box and click on Apply to display the window. DECamds displays a window similar to

Figure 4–11.

4–12



Figure 4–11 DECamds Node Summary Window

DELPHI Node Summary


Model:

O.S.:

Uptime:

Memory:

CPUs:

DEC 7000 Model 630

OpenVMS V7.0

12 00:39:15.14

192.00 Mb

4

CPU Process State Queues

COM

WAIT

0 1 2 3 4 Curr

0.00

Peak

1.79

0.89

2.77

CPU Modes (Avg all processors)

0 25 50 75 100

U

I

K

E

S

C

M

N

0

3

78

8

1

0

4

7

Page Faults (per second)

0 16

Total

Hard

System

32

Memory (Pages in thousands)

0 49 98

Free

Used

Modified

Bad

147

48

196

64 80 96 Curr

66.50

10.80

0.00

Peak

1177.24

67.33

0.00

245 294 344 393 Curr

24609

354061

14546

0

Peak

23848

354147

15549

0

31

11

18

35

14

0

12

36

I/O (per second)

0

WIO

DIO

BIO

16 32 48 64 80 96 Curr

0.00

39.17

30.17

Peak

6.34

81.34

157.90

ZK−7962A−GE

The Node Summary window in Figure 4–11 confirms there is little free memory available. (The Node Summary window also can show other activity that is relevant in diagnosing the problem, such as a high number of page faults.)

5.

Purge the working sets. Choose which process’s working sets are to be purged by performing the following steps: a.

In the Memory Summary window, select any process, click MB3 on the count field, and choose Fix from the menu.

b.

Click on OK or Apply in the Fix dialog box.

4–13

5

Customizing DECamds

This chapter describes how to organize data collection, analysis, and display by filtering, sorting, and customizing DECamds. It also describes how some of these tasks can optimize the performance of DECamds.

5.1 Customizing DECamds Defaults

To set DECamds application values such as bar graph colors and automatic collection options, choose DECamds Customizations from the Customize menu of the Event Log or System Overview window. DECamds displays the DECamds

Application Customizations dialog box as shown in Figure 5–1.

5–1



Figure 5–1 DECamds Application Customizations Dialog Box

DECamds Application Customizations

Event Color

NoEvent Color

Current Values

Collection Interval Factor

Red

Green

1

Automatic Collection Options

Node

CPU

Disk

Volume

CluTran

Memory

I/O

Page/Swap

Lock

Application State Options

Show Nodes

Lock Event Collect

Automatic Event Investigation

Highlight Events

OK

Apply Default Cancel

ZK−7938A−GE

Table 5–1 lists the items you can customize.

To save your changes from one use to the next, choose Save DECamds

Customizations from the Customize menu of the Event Log or System Overview window. The changes are stored in the AMDS$APPLIC_CUSTOMIZE.DAT file.

Note

Subsequent installations of DECamds will not overwrite existing customization files. The installation procedure will check for the existence of each customization file. If found, the procedure will provide any new file with the .TEMPLATE file extension. The installer must check the new .TEMPLATE files for new features implemented in future releases; any changes will be stated in the online release notes in the following location:

SYS$HELP:AMDS0nn.RELEASE_NOTES

Note that nn refers to the version number of the release.

5–2



Table 5–1 DECamds Application Defaults

Field Default Function

Current Values

Event Color

NoEvent Color

Collection

Interval Factor

Red Specifies the bar graph color used for signaled events.

Green Specifies the bar graph color used for nonsignaled events.

1 This value is multiplied by a window’s collection interval definition. Used to force windows to have longer time spans between data collection. Increasing this number decreases the use of the Data Analyzer’s CPU and LAN.

Automatic Collection Options

Node

CPU

Memory

I/O

Disk

Volume

Page/Swap

On

Off

Off

Off

Off

Off

On

Lock

CluTran

On

On

Determines whether node data is collected at startup.

Determines whether CPU data is collected at startup.

Determines whether memory data is collected at startup.

Determines whether I/O data is collected at startup.

Determines whether disk data is collected at startup.

Determines whether volume data is collected on startup.

Determines whether page and swap data is collected at startup.

Determines whether lock contention data is collected at startup.

Determines whether a view of the cluster from the node on which Collect Cluster Transition Information was selected is collected.

Application State Options

Show Nodes On

Lock Event

Collect

Automatic

Event

Investigation

Highlight

Events

Off

Off

On

Determines whether the System Overview window starts up with individual node names displayed.

Determines whether DECamds automatically collects additional data about all the processes waiting for a locked resource.

Determines whether additional data is collected when

DECamds detects an event.

Determines whether event-related data is highlighted.

5.1.1 Setting Default Data Collection

By default, DECamds collects the following categories of data when started:

• Node Summary



You can change the default amount of data collected when DECamds starts by choosing DECamds Customizations from the Customize menu in the Event Log or

System Overview window. The DECamds Application Customizations dialog box appears and you can click on the Automatic Collection Options buttons to select

5–3



or disable the categories you want. To save the settings for the next time you run

DECamds, choose Save DECamds Customizations from the Customize menu.

5.1.2 Setting Automatic Event Investigation

Automatic Event Investigation enhances the speed with which you can pursue a specified event. When this option is enabled, DECamds automatically collects follow-up data on the event. When this option is disabled, you must initiate follow-up data collection when an event occurs.

To enable automatic event investigation, choose Enable Automatic Event

Investigation from the Control menu of the System Overview or Event Log window. To disable it, choose Disable Automatic Event Investigation.

You also can set Automatic Event Investigation by choosing DECamds

Customizations from the Customize menu; then click on the Automatic Event

Investigation button in the resulting DECamds Application Customizations dialog box. To save the settings for the next time you run DECamds, choose Save

DECamds Customizations from the Customize menu.

Note that enabling this option can significantly increase CPU, memory, and LAN traffic load. By default, DECamds does not automatically investigate events that might require attention. Automatic investigation applies only to events that are detected after you enable the option. It does not apply to lock-related events, which you can control using the DECamds Application Customizations dialog box.

5.1.3 Setting Automatic Lock Investigation

With Automatic Lock Investigation, the Data Analyzer automatically investigates any signaled lock contention events. Setting this option allows you to determine more quickly the blocking lock in a resource contention situation.

Note that this option sometimes uses more DECamds memory, CPU, and LAN bandwidth to investigate locks that are very transient.

To enable automatic investigation of locks, click on the Lock Event Collect button in the DECamds Application Customizations dialog box.

5.2 Filtering Data

DECamds can collect and display every event regardless of how important or unimportant an event is to you. However, you can narrow the focus so that the events that you want to see are displayed. You can use the following methods to determine which events qualify for your attention:

• Filter all events on a global severity basis. For example, you might not want to see any event that has less than a 40 severity value.

• Define specific event criteria. For example, you can refine the global filtering by also defining that DSKRWT event (high disk device Rwait count) must pass your specifications before being considered an event worth displaying or logging.

Figure 5–2 shows the process an event must pass through to qualify as important enough to be logged or displayed for your attention.

5–4



Figure 5–2 Event Qualification

Data Analyzer gets information from the Data Provider

Event Severity Check

Set in Event Log Filter dialog box.

Choose Filter Data... from the

Customize menu of the Event Log.

Test Threshold Values

Set in event customization dialog box. Choose Customize Events from Event Log Customize menu.

Double−click on an event.

Data meets or exceeds values to signal an event

?

Yes

No

Data meets or exceeds threshold values

?

Yes

Add 1 to Occurrence counter.

No

Do nothing.

Do nothing.

Test Threshold Values

Set in event customization dialog box. Choose Customize Events from Event Log Customize menu.

Double−click on an event.

Occurrence count>=set value

?

Yes

Signal event.

No

Do nothing.

Event Severity Check

Set in Event Log Filter dialog box.

Choose Filter Data... from the

Customize menu of the Event Log.

Data meets or exceeds values to display an event

?

Yes

Display event and write the event to the AMDS$LOG file,

OPCOM, or user file.

No

Write the event to the AMDS$LOG file,

OPCOM, or user file.

ZK−7949A−GE

5–5



5.2.1 Filtering Events

To specify the events to be displayed in the Event Log window, perform the following steps:

1.

Choose Filter Data... from the Customize menu. A filter dialog box appears.

Table 5–2 describes the filter options.

Table 5–2 Event Log Filters

Filter Description

Severity

Event Bell

Bell Volume

Event Highlight

Event Signal

Event Timeout

(secs)

Event Escalation

Time (secs)

Event Escalation

Severity

Controls the severity level at which events are displayed in the

Event Log menu. By default, all events are displayed. Increasing this value reduces the number of event messages in the Event Log window and can improve perceived response time.

Determines which events are marked by an audible signal by specifying a minimum event severity value. When a new event is displayed, if the severity value is the same or greater than the specified value, an audible notification is given. To disable the sound, specify a value of 101.

Controls the pitch or sound level at which the bell is rung when an event is signaled whose priority is greater than the Event Bell filter.

Determines which events are marked by a visual signal by specifying a minimum event severity value. When a new event is displayed, if the severity value is the same or greater than the specified value, an event is highlighted. To disable highlighting, specify a value of 101.

Determines the severity value at which DECamds signals an event for attention. Only events that qualify are passed on to be checked by any filters you may set for a specific event. Increasing this value reduces the number of event messages that need to be tested to see if further attention is warranted, which can improve perceived response time.

Determines how long an informational event is displayed (in seconds).

Determines how long an event must be signaled before it is sent to the operator communication manager (OPCOM). DECamds uses this value along with the Event Escalation Severity value. Both criteria must be met before the event is signaled to OPCOM.

Determines which events are sent to OPCOM. DECamds uses this value along with the Event Escalation Time (secs) value. Both criteria must be met before the event is signaled to OPCOM.

2.

Modify the settings, which will apply to the current session. To save these settings from session to session, choose Save Filter Changes from the

Customize menu in the Event Log window.

5–6



You can also filter data in the following data windows:

• CPU Summary


• Memory Summary



• Volume Summary


The modifiable options that are displayed in the filter dialog box for the window vary with the window.

Figure 5–3 shows the CPU Summary Filtering dialog box. For a process to be displayed in the CPU Summary window, it must have a Current Priority of 4 or more and be in any of the process states indicated except HIB, HIBO, or SUSPO.

No other processes are displayed.

Figure 5–3 CPU Summary Filtering Dialog Box

CPU Summary Filtering

Current Filter Values

Current Priority 4

CPU Rate 0.000

Select value, then either use arrows to change value or input new value and ‘Apply’ or ‘OK’ the change:

Process States

COLPG

MWAIT

PFW

LEF

CEF LEFO

HIB

HIBO

SUSP

SUSPO

FPG

COM

COMO

CUR

OK


ZK−7943A−GE

If the Enable Highlighting option is on, any process that signals an event is included in the display, regardless of whether it meets the filter criteria.

5–7



To change the value of a filter, turn the filter button on by clicking on it, and then click on the up or down arrow. Click on OK or Apply for the filter to take effect.

To return to system default values, click on Default.

Changing a Filter Category

Some data windows also allow you to filter data by category. For example, in the CPU Summary window, you also can filter by the Process State category to display only processes in certain states. Category buttons that are selected display the associated information.

In the CPU Summary window, to display only inactive processes, select the HIB and HIBO buttons under Process States, and deselect all other process states.

When you click on OK or Apply, only inactive processes appear in the CPU

Summary window.

5.2.2 Customizing Events

You can define criteria by which specific events are qualified for your attention.

For example, you can refine the global filtering by also defining that DSKRWT event (high disk device Rwait count) must pass your specifications before being considered an event worth displaying or logging. To define specific event criteria, perform the following steps:

1.

Choose Customize Events from the Customize menu in the Event Log window.

Figure 5–4 shows the Customize Events dialog box that appears.

5–8



Figure 5–4 Customize Events Dialog Box

Customize Events

HIHRDP,

HIMWTQ,

HINTER,

HIPWIO,

HIPWTQ,

HISYSP,

HITTLP,

HMPSYN, high hard page fault rate many processes waiting in MWAIT high interrupt mode time high paging Write I/O rate many processes waiting in Page WAIT high system page fault rate high total page fault rate high MP synchronization mode time

LCKBLK,

LCKCNT,

LCKWAT,

LOASTQ, lock blocking lock contention lock waiting process has used most of ASTLM quota

LOBIOQ,

LOBYTQ,

LODIOQ,

LOENQU, process has used most of BIOLM quota process has used most of BYTLM quota process has used most of DIOLM quota process has used most of ENQLM quota

LOFILQ,

LOMEMY,

LOPGFQ,

LOPGSP,

LOPRCQ,

LOSWSP,

LOTQEQ,

LOVLSP,

LOWEXT, process has used most of FILLM quota free memory is low process has used most of PGFLQUOTA quota low page file space process has used most of PRCLM quota low swap file space process has used most of TQELM quota low disk volume free space low process working set extent

LOWSQU, low process working set quota

LRGHSH, large hash table

NOPGFL,

NOPROC, no page file cannot find process

NOSWFL,

PRBIOR,

PRBIOW, no swap file high process Buffered I/O rate process waiting for Buffered I/O

PRCCOM, process waiting in COM or COMO

OK Select

Cancel

ZK−7944A−GE

2.

Double-click on an event that you want to customize. A dialog box appears with the event you select. The dialog box also contains an explanation of what might cause this event to occur. Figure 5–5 shows the LOWSQU Event

Customization window.

5–9



Figure 5–5 LOWSQU Event Customization Window

LOWSQU, low process working set quota

Event Format: LOWSQU, <node> <process> working set quota is too small

Signaled From: Memory or Single Process Summary

Event Class Type: Memory

Event Description

The process page fault rate exceeds the threshold and the percentage of

Working Set Size to Working Set Quota exceeds the threshold.

Event Investigation Hints

This event indicates the process needs more memory, but may not be able to get it due to either the WSQUO value in the UAF file being set too low for size of memory allocation requests or the system is memory constrained.

Event Customize Options

Severity

Occurrence

40

3

Class

Threshold 1

N/A

50

150.000

Page faults per second

Percent WSQuota over WSCount Threshold 2

Select value, then either use arrows to change value or input new value and "Apply" or "OK" the change:

Event Escalation Action Options

OPCOM USER NONE

Type in procedure to be run (e.g., amds$system:amds$event_mail_sample.com)

OK


ZK−7945A−GE

Figure 5–5 shows the values you can set in any Event Customization window.

To change the value of an option, click on an option and then use the arrow buttons to increase or decrease the value. A higher number indicates a more severe event.

3.

Modify the settings that will apply to the current session. To save these settings from session to session, choose Save Event Customizations from the

Customize menu in the Event Log window.

5–10



The following sections describe the event customization options.

Severity Option

Severity is the relative importance of an event. Events with a high severity must also exceed threshold settings before an event can be signaled for display or logging.

Occurrence Option

Each DECamds event is assigned an occurrence value, that is, the number of consecutive data samples that must exceed the event threshold before the event is signaled. By default, events have low occurrence values. However, you might find that a certain event only indicates a problem when it occurs repeatedly for an extended period. You can change the occurrence value assigned to that event so that DECamds signals it only when necessary.

For example, suppose page fault spikes are common in your environment, and

DECamds frequently signals intermittent HITTLP, total page fault rate is high events. You could change the event’s occurrence value to 3, so that the total page fault rate must exceed the threshold for three consecutive collection intervals before being signaled to the Event Log.

To avoid displaying insignificant events, you can customize an event so that

DECamds signals it only when it continuously occurs.

Automatic Event Investigation (see Section 5.1.2) uses the occurrence value to determine when to further investigate an event. When enabled, the automatic event investigation is activated when the Occurrence count is three times the

Occurrence setting value.

Class Option

You can customize certain events so that the event threshold varies depending on the class of computer system the event occurs on. This feature is particularly useful in environments with many different types and sizes of computers.

By default, DECamds uses only one default threshold for each event, regardless of the type of computer the event occurs on. However, for certain events (in particular, CPU, I/O, and memory usage events) the level at which resource use becomes a problem depends on the size and type of computer. For example, a page fault rate of 100 may be important on a VAXstation 2000 system but not on a VAX 7000 system.

DECamds provides three additional predefined classes for CPU, I/O, and Memoryrelated events. You can specify threshold values for each class in addition to the default threshold for an event. To specify an additional event threshold for each class, edit the file AMDS$THRESHOLD_DEFS.DAT located in the

AMDS$CONFIG directory.

Table 5–3 defines CPU, I/O, and Memory classes.

5–11



Table 5–3 CPU, I/O, and Memory Class Definitions

Class

1

Description

Class 1

Class 2

Class 3

Class 4

CPU Classes

All VAXft systems, VAXstation/VAXserver 4000, MicroVAX 4000

Higher VUP workstations: VAXstation/VAXserver 3100-M76, MicroVAX 3100-

M76, MicroVAX 3100-8*, VAXstation 3100-9*, MicroVAX 3100-9*, VAXstation

4000-9*

VAX/VAXserver 6000, 7000, 9000, 10000

All Alpha systems

Class 1

Class 2

Class 3

Class 4

I/O Classes

All VAX systems, VAXft systems, VAXstation/VAXserver 4000, MicroVAX 4000

Higher VUP workstations: VAXstation/VAXserver 3100-M76, MicroVAX 3100-

M76, MicroVAX 3100-8*, VAXstation 3100-9*, MicroVAX 3100-9*, VAXstation

4000-9*

VAX/VAXserver 6000, 7000, 9000, 10000

All Alpha systems

Systems with less than or equal to 24 MB of memory

Systems with more than 24 MB and less than or equal to 64 MB of memory

Systems with more than 64 MB of memory

All Alpha systems

Memory Classes

Class 1

Class 2

Class 3

Class 4

1

If no class is defined, DECamds uses the default threshold value.

You can specify class-based thresholds only for the following events:

• CPU-related events:

HINTER, node interrupt mode time is high

HICOMQ, node many processes waiting for CPU

HMPSYN, node MP synchronization mode time is high

HIPWTQ, node many processes waiting in COLPG, PFW, or FPG

HIMWTQ, node many processes waiting in MWAIT

• I/O-related events:

HIBIOR, node buffered I/O rate is high

HIDIOR, node direct I/O rate is high

HIPWIO, node paging write I/O rate is high

• Memory-related events:

LOMEMY, node free memory is low

HIHRDP, node hard page fault rate is high

HISYSP, node high system page fault rate

HITTLP, node total page fault rate is high

RESPRS, node resource hash table sparse

RESDNS, node resource hash table dense

5–12



As an example of setting a class-based threshold, the HITTLP, total page fault

rate is high event is a memory-related event, so the thresholds are based on the memory class definitions shown in Table 5–3. The default threshold for this event is 20 page faults per second. A page fault rate of 20 may be important on a VAXstation 2000 system, but it is not important on a VAX 7000 system.

To account for this, you can specify the following additional thresholds for the

HITTLP, total page fault rate is high event:

Class

1 (systems with less than or equal to 64 MB of memory)

2 (systems with 24 MB to 64

MB of memory)

3 (systems with more than 64

MB of memory)

4 (Alpha systems)

Threshold

20

40

100

100

Description

Event is triggered at the default threshold of 20 page faults per second.

Event is triggered at 40 page faults per second.

Event is triggered at 100 page faults per second.

Event is triggered at 100 page faults per second

Threshold Options

Threshold values are compared to an event’s description to determine whether an event meets the criteria for display or log. Threshold values are used in conjunction with the occurrence and severity values. Increasing event threshold values can reduce CPU use and improve perceived response time because more instances must occur for the threshold to be crossed, so fewer thresholds are crossed and fewer events are triggered.

Note

Setting a threshold too high could mask a serious problem.

You can read a description of an event by choosing Customize Events from the

Customize menu in the Event Log window, then double-clicking on the event. The

Event Customization dialog box displays an Event Description field.

Most events are checked against only one threshold; however, some have dual thresholds, where the event is triggered if either one is true. For example, for the

LOVLSP, node disk volume free space is low event, DECamds checks both of the following thresholds:

• Number of blocks remaining (LowDiskFreeSpace.BlkRem)

• Percentage of total blocks remaining (LowDiskFreeSpace.Percent)

Note

Events with both high severity and threshold values are signaled to the operator communication manager (OPCOM). For more information about signaling events to OPCOM, see Section 2.3.3.

5–13


5.3 Sorting Data


Choose Sort Data... from the Customize menu to change the order of the information displayed in a window. A dialog box appears in which you can specify sort criteria. All sort criteria must be met for a process to be displayed.

You can sort data in the following windows:

• CPU Summary


• Volume Summary

• Event Log


• Memory Summary



Figure 5–6 shows a sample Memory Summary Sorting dialog box.

5–14

Figure 5–6 Memory Summary Sorting Dialog Box

Memory Summary Sorting

Sort Order

Ascending Descending

Sort Field

Process PID

Process Name

Working Set Count

Working Set Size

Working Set Extent

Page Fault Rate

Paging I/O Rate

Unsorted



OK


ZK−7961A−GE

Sorting is based on two variables: the sort order and the sort field. You can choose only one sort criterion for each variable—one for the sort order, and one for the sort field. To sort Memory Summary data to list the processes with the highest page fault rates first, for example perform the following steps:

1.

Choose Sort Data... from the Customize menu on the Memory Summary window. The Memory Summary Sorting dialog box appears; current sort field settings are displayed. (By default, DECamds sorts Memory Summary data on the Working Set Count field in descending order.)

2.

Change sort settings by choosing Page Fault Rate and Ascending order.

3.

Click on OK or Apply.

4.

To save sort settings, choose Save Sort Changes on the Customize menu.

5–15


5.4 Setting Collection Intervals


A collection interval is the time the Data Analyzer waits before requesting more information from Data Provider nodes. Changing the collection interval helps you control the performance of DECamds and its consumption of system resources.

The frequency of polling remote nodes for data (collection intervals) can affect perceived response time. You want to find a balance between collecting data often enough to detect potential resource availability problems before a node or cluster experiences a severe problem, and seldom enough to optimize perceived response time. Increasing the collection interval factor decreases CPU consumption and

LAN load, but response time might appear slower because the intervals are longer.

Collection intervals do not affect memory use.

To change a collection interval, choose Collection Interval from the Customize menu. Figure 5–7 shows a sample Memory Summary Collection Interval dialog box.

Figure 5–7 Memory Summary Collection Interval Dialog Box

Memory Summary Collection Interval

Current Collection Interval: 3.00

Based on Collection Interval Factor: 1

Display Interval (sec)

Event Interval (sec)

NoEvent Interval (sec)

3.00

5.00

30.00

Select value, then either use arrows to change value or input new value and ‘Apply’ or ‘OK’ the change:

OK


ZK−7939A−GE

Table 5–4 describes the fields on the Memory Summary Collection Interval dialog box.

5–16



Table 5–4 Memory Summary Collection Interval Fields

Current Collection

Interval

Based on Collection

Interval Factor

Displays the number of seconds between requests for data. You can change the value for all collection intervals for all windows by choosing DECamds Customizations from the Customize menu of the Event Log or System Overview window. The DECamds

Application Customizations dialog box appears and you can increase or decrease the collection interval factor.

Displays the number with which the collection interval is multiplied.

Display Interval (sec)

Event Interval (sec)

Displays the collection interval for displaying data in a window.

You can change the interval by clicking on the up or down arrows in the dialog box.

Displays the collection interval used when events are found.

This value is used by default when you start background collection. You can change the interval by clicking on the up or down arrows in the dialog box.

NoEvent Interval (sec) Displays the collection interval when no events are found. You can change the interval by clicking on the up or down arrows in the dialog box.

To apply the changes, click on OK or Apply. To save collection interval changes, choose Save Collection Interval Changes from the Customize menu.

To change back to DECamds default values for the window, click on Default. To exit without making any changes, click on Cancel.

Table 5–5 lists the default window collection interval values (in seconds) provided with DECamds for each window type.

Table 5–5 Default Window Collection Intervals

Window

CPU Modes Summary

CPU Summary

Disk Status Summary

Volume Summary

Lock Contention

Memory Summary

Node Summary

Page/Swap File Summary

Process Identification Manager

2

Process I/O Summary

Single Lock Summary


Display

1

5.0

5.0

30.0

15.0

10.0

5.0

5.0

30.0

60.0

10.0

10.0

5.0

Event

1

5.0

10.0

15.0

15.0

20.0

10.0

5.0

30.0

60.0

10.0

10.0

5.0

No Event

5.0

30.0

60.0

120.0

60.0

30.0

10.0

2400.0

240.0

30.0

20.0

20.0

1

1

All times are in seconds and cannot be less than .5 second.

2

Process Identification Manager supports the CPU, Memory, Process I/O, and Single Lock Summary window sampling.

5–17


5.5 Optimizing Performance with System Settings


DECamds is a compute-intensive and LAN traffic-intensive application. At times, routine data collection, display activities, and corrective actions can cause a delay in perceived response time.

This section explains how to optimize perceived response time based on actual measurements of CPU utilization rates (throughput). Performance improvements can be made in the following areas:

Area

DECamds software

System settings

Hardware configuration

Discussed in...

Section 5.5.1

Section 5.5.2

Section 5.5.3

Site configurations vary widely, and no rules apply to all situations. However, the information in this section can help you make informed choices about improving your system performance.

The following factors affect perceived response time:

• Load on monitored nodes including applications and peripherals (especially number of disks)

• Number of monitored nodes and users

• Size of operating system tables and lists on monitored nodes (process and lock)

• Version of operating system running on monitored nodes

• LAN traffic, cluster communications, nodes booting, and network-based applications and tools

5.5.1 Optimizing DECamds Software

When DECamds starts, it polls the LAN to locate all nodes running the DECamds

Data Provider, creates a communications link, and collects data from each

Data Provider node on the LAN. (See Section 1.1 for more information about establishing a communications link between nodes.)

The initial polling process creates a short-term high load of CPU and LAN activity. After establishing a communications link with other nodes, DECamds reduces polling frequency, thereby reducing the CPU and LAN load.

Note

Each request to collect a new category of data increases memory and LAN requirements. Memory requirements vary with the number of categories collected and the number of nodes being polled.

Polling frequency does not affect memory because polling only changes how frequently existing data is replaced with updated data.

The following sections describe system settings that you can change to improve performance and the ability of DECamds to handle data collection demands.

5–18



5.5.1.1 Setting Process Quotas

To improve the performance of DECamds, you might need to change process quotas. The quotas used extensively by DECamds are ASTLM, TQELM, BIOLM,

BYTLM, and WSEXTENT. The values listed in Section A.2 are suggestions for a

50-node cluster.

The following process quotas are recommended:

Quota

ASTLM

TQELM

BIOLM

WSEXTENT

BYTLM

Recommended Value

1

4 times the node count





1

node count is the number of nodes a Data Analyzer monitors simultaneously.

Perform the following steps to change process quotas:

1.

Increase the process quotas assigned to the process initiating DECamds in the system’s user authorization file (UAF).

2.

Log out, log back in, and restart DECamds.

5.5.1.2 Setting LAN Load

The maximum size for data packets is 1500 bytes. When the amount of data is greater than 1500 bytes, DECamds must send multiple requests to complete the data collection request.

Table 5–6 shows the LAN load for various levels of collection intervals and data collection. You can modify a data collection window’s collection intervals (as explained in Section 5.4) or reduce the scope of data collection (as explained in

Section 5.1.1) to reduce LAN activity.

Table 5–6 LAN Load

Data

Outgoing Packet

Size (in bytes) on

Alpha Systems

Configuration data

CPU Modes

CPU Summary

Disk Status

Summary

Fix

Hello Message

Lock Contention

Memory

Summary

129

201

178

473

24

N/A

240

275

Outgoing Packet


VAX Systems

285

Return Packet Size (in bytes)

88

129

171

473

24

N/A

240

275

48 + (64* no. of processors)

16 per active process

56 per disk

12

32

76 per resource



5–19



Table 5–6 (Cont.) LAN Load

Data

Outgoing Packet


Alpha Systems

Node Summary

Page/Swap File

Process I/O

Summary

Single Lock

(Waiting)

Single Process

Summary

Volume

Summary

319

208

236

272

491

430

Outgoing Packet


VAX Systems

241

208

229

Return Packet Size (in bytes)

48 + (64 * no. of processors)

46 per page/swap file


272

471

430

32 per waiter

00

28 per disk

5.5.1.3 Setting Window Customizations

The Sort, Filter, and collection interval settings at the data window level affect performance. Follow these guidelines to balance customization with performance:

• Filter out data to improve CPU performance. Reducing the collection criteria increases performance. See Section 5.2 for information on filtering data.

• Use unsorted windows to improve performance. Sorting requires extra computations. See Section 5.3 for information on sorting data.

• Increase collection interval values to improve performance. See Section 5.4

for information on changing collection intervals.

5.5.2 Optimizing System Settings

Changing several system settings might improve the performance of DECamds on your system. The following sections discuss these settings and how to change them.

5.5.2.1 Setting Data Link Read Operations

Increase read operations to the data link by changing the logical name

AMDS$COMM_READS in the AMDS$CONFIG:AMDS$LOGICALS.COM

command procedure. The AMDS$COMM_READS logical name controls the number of requests for data (read operations) queued to the data link.

If you increase data collection, increase the number of requests that can be queued. Compaq recommends two requests for each node being monitored. Each read operation queued requires 1500 bytes of BYTLM quota.

5.5.2.2 Setting the Communications Buffer

Increase the communications buffer by changing the logical name AMDS$COMM_

BUFFER_SIZE in the AMDS$CONFIG:AMDS$LOGICALS.COM command procedure. The buffer controls the size of the global section used for communication between the provider node and the communications process.

When DECamds cannot keep up, it displays the following warning message:

AMDS$_COMMBUFOVF---communications buffer overflow.

Increase the buffer by 25 percent.

In addition to increasing the value of the AMDS$COMM_BUFFER_SIZE logical name, set the system parameter GBLPAGFIL on the provider node to cover the increase. This adds to the amount of data collection that DECamds can perform.

5–20



The value of the GBLPAGFIL system parameter must always be higher than the number of FREE_GBLPAGES. To determine the value of FREE_GBLPAGES, enter the following commands:

$ A = F$GETSYI("FREE_GBLPAGES")

$ SHO SYM A

The value of A must conform to the following formula:

2 * ( (buffer_size / 512) + 512)

5.5.3 Optimizing Performance with Hardware

Table 5–7 provides an approximate guideline for the number of nodes you can monitor when running DECamds on certain computer types.

Table 5–7 Monitoring Nodes

Monitoring Computer Type

VAXstation 3100

VAXstation 4000 Model 60

VAX 6000

1

VAX 4000 Model 90

DEC 3000 Model 400

DEC 4000 Model 620

DEC 7000 Model 720

Number of Nodes Monitored

Observation Only Observe and Fix

0-30

20-60

75-150

75-150

0-50

0-70

Any number

1

With DECwindows display directed to a workstation

0-20

20-50

65-130

65-130

0-50

0-70

Any number

Follow these suggestions when choosing and configuring a console:

• Use fast hardware.

Because DECamds is compute and memory-intensive, which is compounded by its real-time DECwindows-based display medium, faster CPUs will improve throughput and perceived response time.

• Use multiprocessors.

DECamds runs two processes: one handling calculations and display; one handling communications between the monitoring node and the remote nodes.

A multiprocessor reduces the DECwindows server process competition for

CPU time. On single processor systems, the processes must compete.

• Run the monitoring portion of DECamds on a standalone system.

If a cluster is experiencing system resource problems, you can still use

DECamds.

5–21

A

Installing the DECamds Data Analyzer

This appendix explains how to install the DECamds Data Analyzer software on

OpenVMS Alpha and OpenVMS VAX Version 6.2 and later systems.

Beginning with OpenVMS Version 7.2, the Data Provider ships as part of the

OpenVMS installation. Installing or upgrading to OpenVMS Version 7.2 or later automatically installs the Data Provider on your system. You can run the Data

Provider on any VAX or Alpha Version 6.2 or later system.

Note

The Compaq Availability Manager web site might refer you to a more recent version of the Data Provider than the one on the current OpenVMS

VAX or Alpha operating system CD-ROM. Compaq recommends that you install the DECamds Data Provider software using the version indicated at the following URL: http://www.openvms.compaq.com/openvms/products/availman/

Section A.7 explains how to start using the Data Provider.

This chapter contains the following sections:

• General installation information

• Data Analyzer installation requirements

• Downloading the Data Analyzer software

• Installing Data Analyzer software from a PCSI kit

• Postinstallation tasks on Data Provider nodes

• Postinstallation tasks on the Data Analyzer node

• Starting to use the Data Provider

• Determining and reporting problems

• Running the Installation Verification Procedure (IVP) separately

A.1 General Installation Information

DECamds provides online release notes. Compaq strongly recommends that you read the release notes before proceeding with the installation. You can print the text file of the release notes from the following location:

SYS$HELP:AMDS072-1B.RELEASE_NOTES

A–1


A.1 General Installation Information

DECamds consists of client and server software:

• The client software, the Data Analyzer, provides the graphical user interface to display DECamds information to users.

• The server software, the DECamds Data Provider (RMDRIVER), collects the data that DECamds analyzes and displays.

In earlier versions of OpenVMS, you needed to install both both the Data

Analyzer and Data Provider software on your system from the latest DECamds kit. Beginning with OpenVMS Version 7.2, you need to install only the Data

Analyzer software on the system where you run the client, or graphical user interface. You need to do this to obtain the new library for DECamds Version 7.2

and later.

A.2 Data Analyzer Installation Requirements

This section provides a checklist of hardware and software requirements for installing the DECamds Data Analyzer. A typical installation takes approximately 5 to 10 minutes per node, depending on your type of media and system configuration.

•

Hardware requirements

– A workstation monitor. For any hardware configuration without a

DECwindows Motif display device, use the DECwindows server to direct the display to a workstation or an X terminal.

– 16 MB of memory for VAX systems and 32 MB for Alpha systems, for the

Data Analyzer portion of DECamds.

You should use a more powerful system as the number of nodes and the amount of collected data rises. Table A–1 shows general guidelines for the default Data Analyzer node. Note that the following table does not preclude

DECamds from running on a less powerful system than listed for the number of nodes being monitored.

Table A–1 Recommended System Requirements

Number of

Monitored Nodes Recommended Alpha Hardware

1-30

20-50

40-90

91 or more

DEC 3000 Model 400, 32 MB

DEC 3000 Model 400, 64 MB

DEC 3000 Model 500

DEC 4000 Model 620

Recommended VAX

Hardware

VAXstation 3100, 16 MB



VAX 6000-420

•

Operating system version

At least one of the following:

OpenVMS VAX Version 6.2 or higher

OpenVMS Alpha Version 6.2 or higher

•

Display software

DECwindows Motif for OpenVMS Version 1.1 or higher installed on the

Data Analyzer node system.

A–2



•

Privileges

Operation

Monitor only (read-only access)

Implement fixes (write access)

Stop, start, reload, or restart the Data

Provider node. Includes changing security or group name.

Privileges Needed

OPER

OPER, CMKRNL

OPER, CMKRNL, LOG_IO, SYSNAM,

SYSPRV

Note

For OpenVMS Version 6.2 and later, if the Data Provider is running on the same node as the Data Analyzer node, you must also have either

SYSPRV privilege or ACL access to the RMA0: device.

•

Disk space

– 3500 blocks on VAX systems.

– 4000 blocks on Alpha systems.

To determine the number of free disk blocks on the current system disk, enter the following command at the OpenVMS DCL prompt:

$ SHOW DEVICE SYS$SYSDEVICE

•

System parameter settings

These settings are the same as those required for operating system installation. The Installation Verification Procedure (IVP) requires additional space as follows:

GLBPAGFIL

WSMAX

1200

16384

You can modify WSMAX and GLBPAGFIL using the System Management utility (SYSMAN). See the OpenVMS System Manager’s Manual for more information.

• Process account quotas (minimum)

ASTLM

BIOLM

BYTLM

FILLM

PRCLM

PGFLQUO

TQELM

WSEXTENT

150

51

75000

20

3

25600

100

16384

Note that the AMDS$COMM_READS logical determines the default value.

If you are reinstalling the Data Analyzer, or have changed AMDS$COMM_

READS, then the following formulas are used to determine the default value:

ASTLM >= (AMDS$COMM_READS*3)

BIOLM >= (AMDS$COMM_READS+1)

BYTLM >= (AMDS$COMM_READS*1500)

A–3



TQELM >= (AMDS$COMM_READS*2)

User account quotas are stored in the file SYSUAF.DAT. Use the OpenVMS

Authorize utility (AUTHORIZE) to verify and change user account quotas.

For more information on modifying account quotas, see the description of the

Authorize utility in the OpenVMS system management documentation.

Note

On both Alpha and VAX systems, Compaq recommends that you perform a system disk backup before installing any software. Use the backup procedures that are established at your site. For details about performing a system disk backup, see the OpenVMS Backup utility documentation.

A.3 Obtaining the Data Analyzer Software

The Data Analyzer software is available on the OpenVMS operating system layered product CD-ROM or from the Compaq Availability Manager web site.

Follow these steps to download the software from the web:

1.

From the Availability Manager home page, click Software Download. The

Availability Manager home page is at the following URL: http://www.openvms.compaq.com/openvms/products/availman/

2.

Complete the user survey, which allows you to proceed to the Download web page.

3.

Click one or both of the DECamds executables:

DECamds - Alpha: decamds0721b.pcsi-dcx_axpexe

DECamds - VAX: decamds0721b.pcsi-dcx_vaxexe

4.

Save the executable to a device and directory of your choice.

5.

Run the executable and accept the default file name. The result will be:

DECamds: DEC-VMS-AMDSV0702-1B-1.PCSI

The next section provides installation instructions for the Data Analyzer.

A.4 Installing Data Analyzer Software from a PCSI Kit

This section describes the installation procedure on an OpenVMS Version 6.2 or later systems from a POLYCENTER Software Installation (PCSI) kit.

•

Starting the installation

Use the following procedure to start the installation:

Enter the OpenVMS DCL command PRODUCT, the name of the task to be performed, and the name of one or more products. For example, to install

DECamds Version 7.2, enter the following command:

$ PRODUCT INSTALL AMDS/SOURCE=device directory/HELP where:

device directory refers to the device and the directory where the kit is located.

A–4



For a description of the features you can request with the PRODUCT

INSTALL command when starting an installation such as running the

IVP, purging files, and configuring the installation, see the POLYCENTER

Software Installation Utility User’s Guide.

As an installation procedure progresses, the system displays a percentage message to indicate how much of the installation is done. For example:

Percent Done: 15%

...30%

...46%

...62%

...76%

...92%

%PCSI-I-SUCCESS, operation completed successfully

If you started the installation using the /LOG qualifier, the system displays details of the installation.

•

Stopping and restarting the installation

Use the following procedure to stop and restart the installation:

1.

To stop the procedure at any time, press Ctrl/Y.

2.

Enter the PRODUCT REMOVE command to reverse any changes to the system that occurred during the partial installation. This deletes all files created up to that point and causes the installation procedure to exit.

3.

Go back to the beginning of the installation procedure to restart the installation.

•

Recovering from errors

If the installation procedure fails for any reason, the following message is displayed:

%POLYCENTER Software Installation utility

%INSTAL-E-INSFAIL, The installation of DECamds 7.2-1B has failed.

An error during the installation can occur if one or more of the following conditions exist:

• The operating system version is incorrect.

• The prerequisite software version is incorrect.

• Quotas necessary for successful installation are inadequate.

• Process quotas required by the POLYCENTER Software Installation utility are inadequate.

• The OpenVMS Help library is currently in use.

If you receive any error message beginning with

%PCSI-E-INSTAL

, refer to the OpenVMS DCL HELP/MESSAGE utility for POLYCENTER Software

Installation information and a possible solution to the problem.

If the installation fails, you must restart the installation procedure. If the installation fails due to an IVP failure, contact a Compaq support representative.

A–5



Sample Installation on an OpenVMS Alpha System

Example A–1 shows a sample installation on an OpenVMS Alpha system. This sample was run on a system that had no previous version of DECamds installed.

Depending on which layered products you have on your system, you might see additional messages and questions when you perform your installation.

Example A–1 Sample OpenVMS Alpha Installation

$ product install amds

The following product has been selected:

DEC VMS AMDS V7.2-1B

Do you want to continue? [YES]

Layered Product

Configuration phase starting ...

You will be asked to choose options, if any, for each selected product and for any products that may be installed to satisfy software dependency requirements.

DEC VMS AMDS V7.2-1B: DECamds (Availability Manager for Distributed

Systems) V7.2-1B

COPYRIGHT © © 1994, 1995, 1999 -- All rights reserved

Compaq Computer Corporation

License and Product Authorization Key (PAK) Information

Do you want the defaults for all options? [YES]

DECamds Data Provider Installation Verification Procedure

DECamds Startup File

DECamds Logicals Customization File

DECamds Data Provider Security Access File

DECamds Data Analyzer Security Access File

DECamds Data Analyzer Installation Verification Procedure (IVP)

IVP may fail due to the following PQL values being too low:

PQL_MASTLM, PQL_MBIOLM, PQL_MTQELM, or PQL_MBYTLM

See the file AMDS$SYSTEM:AMDS$PCSI_IVP_OUTPUT.LOG for help on failure.

Do you want to review the options? [NO]

Execution phase starting ...

The following product will be installed to destination:

DEC VMS AMDS V7.2-1B DISK$ALPHA_V72:[VMS$COMMON.]

Portion done: 0%...20%...30%...40%...50%...60%...70%...80%...90%...100%

The following product has been installed:

DEC VMS AMDS V7.2-1B Layered Product

%PCSI-I-IVPEXECUTE, executing test procedure for DEC VMS AMDS V7.2-1B ...

%PCSI-I-IVPSUCCESS, test procedure completed successfully

DEC VMS AMDS V7.2-1B: DECamds (Availability Manager for Distributed

Systems) V7.2-1B


A–6



Example A–1 (Cont.) Sample OpenVMS Alpha Installation

This product requires the following SYSGEN parameters:

GBLPAGES add 1172

A.5 Postinstallation Tasks on Data Provider Nodes

Perform the following tasks after installing DECamds on Data Provider nodes:

1.

If you have not read the release notes, please read them.

2.

Modify user accounts.

Users who maintain the security or group name files or load new versions of the driver need privileges associated with the driver startup procedure.

3.

Add AMDS$STARTUP.COM to the node’s startup and shutdown procedures to provide for automatic startup and shutdown of the Data Provider driver when a node is booted or shut down.

Add the following command line to SYS$MANAGER:SYSTARTUP_

VMS.COM:

$ @SYS$STARTUP:AMDS$STARTUP.COM STOP

Also, edit SYSHUTDWN.COM to add the following line:

$ @SYS$STARTUP:AMDS$STARTUP.COM STOP

4.

Modify default security files.

To implement fixes, which require write access, the security files must bemodified. The Data Provider security file contains a list of three-part codes representing Data Analyzer nodes that have read or write access to that node.Refer to Section 1.3 for complete instructions about designing security files.

5.

Assign a node to a group.

See Section 1.3.2.1.

6.

Start DECamds (the Data Provider).

Even though the IVP starts and stops the driver, you must start the Data

Provider drivers by entering the following command on each node:

$ @SYS$STARTUP:AMDS$STARTUP.COM START

Note

Starting, stopping, or reloading DECamds (the AMDS$STARTUP.COM

procedure) requires at least TMPMBX, NETMBX, SYSNAM, LOG_IO, and

CMKRNL privileges. Use the OpenVMS Authorize utility (AUTHORIZE) to determine whether users have the required privileges and then make adjustments as needed.

A–7


A.5 Postinstallation Tasks on Data Provider Nodes

A.5.1 Starting, Stopping, and Reloading DECamds

To start and stop the Data Provider driver, enter the following command. (Use this command if a node will be used to both provide and collect system data.)

$ @SYS$STARTUP:AMDS$STARTUP.COM [parameter] where the optional parameter is one of the following:

NODRIVER

START

STOP

RELOAD

Defines the default input and output logicals on the Data Analyzer node driver. Use this parameter on the Data Analyzer node where the Data

Provider driver is not running. It is the default.

Starts the Data Provider driver.

Stops the Data Provider driver.

Loads a new Data Provider driver. Use this parameter when installing a new version of DECamds.

Note

If you use the OpenVMS Snapshot Facility, stop the DECamds Data

Analyzer and Data Provider node drivers before taking a system snapshot.

A.6 Postinstallation Tasks on a Data Analyzer Node

Perform the following tasks after installing the DECamds Data Analyzer:

1.

If you were previously running an earlier version of DECamds, check the differences between the .DAT or .COM files on your system and the associated

.TEMPLATE files provided with the new kit. Change your existing files as necessary.

Note

The new .TEMPLATE files may contain important changes. However, to avoid altering your customizations, the upgrade procedure does not modify your existing customized versions of these files. Check the new

.TEMPLATE versions of these files provided with the kit, and make the appropriate change to your files.

2.

Modify default DECamds security files on each Data Analyzer node.

The security files must be modified to implement fixes (fixes require write access). Refer to Section 1.3 for complete instructions about designing security files.

3.

Define the system directory logical name AMDS$SYSTEM.

To define the logical name AMDS$SYSTEM on systems running the Data

Analyzer but not the Data Provider, enter the following command:

$ @SYS$STARTUP:AMDS$STARTUP.COM NODRIVER

This command requires SYSNAM privilege. The NODRIVER parameter specifies that the procedure is to define the input and output logical names in

AMDS$LOGICALS.COM.

4.

Modify user accounts as needed.

A–8


A.6 Postinstallation Tasks on a Data Analyzer Node

To use DECamds, user accounts require certain privileges and quotas:

• Using the Data Analyzer node for data collection (read access) requires

TMPMBX, NETMBX, and OPER privileges.

• Performing fixes (write access) requires the CMKRNL privilege in addition to TMPMBX, NETMBX, and OPER.

• Using the AMDS$STARTUP.COM to start, stop, or reload the Data

Provider requires at least TMPMBX, NETMBX, SYSNAM, LOG_IO, and

CMKRNL privileges.

5.

Start the application.

For example, the following command starts DECamds with all input files read from AMDS$SYSTEM and all output files written to the current default directory. Only data from group A nodes and group B nodes is displayed.

$ AVAIL /CONFIGURE=AMDS$SYSTEM /LOG_DIRECTORY=SYS$LOGIN-

_$/GROUP=(GROUP_A, GROUP_B)

See Chapter 2 for startup options.

A.7 Starting to Use the Data Provider

Before starting to use the Data Provider, you need to move and remove several files to make the Data Provider RMDRIVER part of OpenVMS.

Move these Files

Move the following files:

File

AMDS$DRIVER_ACCESS.DAT

AMDS$LOGICALS.COM

Old Directory Location New Directory Location

SYS$COMMON:[AMDS] SYS$COMMON:[SYSMGR]

SYS$COMMON:[AMDS] SYS$COMMON:[SYSMGR]

These new directory locations should not affect previous copies of

AMDS$DRIVER_ACCESS.DAT that are in the AMDS$SYSTEM directory because the AMDS$SYSTEM logical is now a search list for

SYS$COMMON:[AMDS] and SYS$COMMON:[SYSMGR]. Previous copies of the files will still be valid; however, new copies of the files will be placed in the new locations.

Delete this File

Also, because the installation replaces the following file, remove it from your system:

SYS$COMMON:[AMDS]AMDS$RMCP.EXE

Data Provider Commands

To start to use the Data Provider, perform either of the following tasks:

• Run the SYS$STARTUP:AMDS$STARTUP START command procedure at the OpenVMS DCL prompt ($).

• Add the @SYS$STARTUP:AMDS$STARTUP START command to the

SYSTARTUP_VMS.COM command file in the SYS$MANAGER directory.

A–9


A.8 Determining and Reporting Problems

A.8 Determining and Reporting Problems

If you encounter a problem while using DECamds, report the problem to Compaq.

Depending on the nature of the problem and the type of support you have, take one of these actions:

• If your software contract or warranty agreement entitles you to telephone support, contact a Compaq support representative.

• If the problem is related to the DECamds documentation, see the Preface of this manual for instructions.

A.9 Running the Installation Verification Procedure Separately

Usually the Installation Verification Procedure (IVP) runs during installation.

Should system problems occur after you install DECamds, check the integrity of installed files by executing the following command procedure:

$ @SYS$TEST:AMDS$IVP.COM

The IVP leaves the Data Provider in the same state in which it was found. For example, if the Data Provider is running, the IVP stops and starts it.

A–10

B

DECamds Files and Logical Names

The DECamds Data Analyzer installation procedure installs files and defines logical names to customize the environment.

The installation procedure defines all logical names in executive mode in the system table (with the /SYSTEM /EXECUTIVE qualifiers). However, you can define logical names in /JOB or /GROUP tables, preceding the system definitions.

Table B–1 and Table B–2 explain the files installed and logical names defined with the Data Analyzer.

Table B–3 and Table B–4 explain the files installed and logicals defined on each node running the Data Provider.

Logical names are added to the logical name table when the

AMDS$LOGICALS.COM procedure is invoked by AMDS$STARTUP.COM.

Note

Logical names can be a search list of other logicals.

The logical names in Table B–2 and Table B–4 must be defined in the job, group, or system table. If you change the name, define the new logical in the job, group, or system table.

B.1 Files and Logical Names for the Data Analyzer Node

Table B–1 and Table B–2 contain the names of all files created on a Data

Analyzer node when DECamds is installed.

Table B–1 Files on the Data Analyzer Node

Directory-Logical:File-Name

AMDS$HELP:AMDS$HELP.HLB

AMDS$CONFIG:AMDS$*.DAT

AMDS$SYSTEM:AMDS073.RELEASE_NOTES

AMDS$CONFIG:AMDS$COMM.EXE

AMDS$SYSTEM:AMDS$CONSOLE.EXE

AMDS$CONFIG:AMDS$CONSOLE.UID

AMDS$CONFIG:AMDS$CONSOLE_

ACCESS.DAT

1

Function

Help library

Customization files

Product Release Notes

Communication image

Data Analyzer image

User interface description file

Data Analyzer security file

1

Can be provided as a TEMPLATE file, depending on whether the file was found during installation.


B–1


B.1 Files and Logical Names for the Data Analyzer Node

Table B–1 (Cont.) Files on the Data Analyzer Node

Directory-Logical:File-Name

SYS$MANAGER:AMDS$LOGICALS.COM

1

AMDS$SYSTEM:AMDS$VMS*-*.LIB

AMDS$TEST:AMDS$IVP.COM

SYS$STARTUP:AMDS$STARTUP.COM

Function

Logical name definition file

DECamds version-specific libraries

Installation verification procedure

DECamds startup file

1

Can be provided as a TEMPLATE file, depending on whether the file was found during installation.

Table B–2 Logical Names Defined for the Data Analyzer

Logical Name Definition

AMDS$COMM_BUFFER_SIZE This value is the size (in bytes) of the communications buffer between the

AMDS$CONSOLE process and the

AMDS$COMM process.

AMDS$COMM_READS

AMDS$COMM_PKT_RETRY

AMDS$COMM_PKT_TMOUT

This value is the number of read aheads posted by the DECamds communications process

(AMDS$COMM) to handle the delivery of remote response packets from the Data Provider to the Data Analyzer node.

Specifies the number of retries before quitting and issuing a "delivery path lost message."

Specifies the timeout period (in seconds) for packet retry for the Data Analyzer.

AMDS$CONFIG The device and directory location for the following DECamds input files:

AMDS$APPLIC_CUSTOMIZE.DAT

AMDS$COMM.EXE

AMDS$CONSOLE.UID

AMDS$CONSOLE_ACCESS.DAT

AMDS$VMS*-*.LIB

All customization files AMDS$*_DEFS.DAT

Default

300000 bytes

50 read aheads

4

10

AMDS$SYSTEM

AMDS$DPI

AMDS$LOG

This value specifies the DPI value of your display device.

The device and directory location for the following DECamds output files:

AMDS$EVENT_LOG.LOG

AMDS$LOCK_LOG.LOG

75 or 100

AMDS$SYSTEM

B.2 Files and Logical Names for Data Provider Nodes

Table B–3 and Table B–4 contain the names of all files created on a node when a

Data Provider is installed.

B–2



Table B–3 Files on Nodes Running the Data Provider

Directory-Logical:File-Name Function

SYS$MANAGER:AMDS$DRIVER_ACCESS.DAT

1

SYS$MANAGER:AMDS$LOGICALS.COM

1

AMDS$SYSTEM:RMCP.EXE

SYS$HELP:AMDS072-1B.RELEASE_NOTES

SYS$HELP:AMDS$HELP.HLB

SYS$LOADABLE_IMAGES:RMDRIVER.EXE,

SYS$LOADABLE_IMAGES:RMDRIVER.STB

2

SYS$LOADABLE_IMAGES:SYS$RMDRIVER.EXE,

SYS$LOADABLE_IMAGES:SYS$RMDRIVER.STB

3

SYS$STARTUP:AMDS$STARTUP.COM

SYS$TEST:AMDS$IVP.COM

Data Provider security file

Logical name definition file

Management interface to the

Data Provider

Product Release Notes

Help library

Data Provider (VAX systems)

Data Provider (Alpha systems)

DECamds startup file

Installation verification procedure

1

Can be provided as a .TEMPLATE file, depending on whether the file was found during installation.

2

VAX specific

3

Alpha specific

Table B–4 Logical Names Defined on Nodes Running the Data Provider

Logical Name Definition Default

AMDS$CONFIG

AMDS$DEVICE

AMDS$GROUP_NAME

AMDS$NUM_DL_READS

The device and directory location for the

DECamds input file AMDS$DRIVER_

ACCESS.DAT.

This logical is translated as the first LAN device to which the Data Provider or Data Analyzer node attempts to connect. The attempts are made in this order: AMDS$DEVICE, FXA0,

XEA0, XQA0, EFA0, ETA0, ESA0, EXA0, EZA0,

FCA0, ECA0.

If your LAN line is not in this list, use

AMDS$DEVICE.

If the Data Analyzer node and Data Provider run on the same node, *RMA0 is used.

The group to which the node is assigned. Choose an alphanumeric string of up to 15 characters.

The group name is defined on the node running the Data Provider and is used by the Data

Analyzer node to display nodes in the System

Overview window.

The number of data link reads to be posted by the

Data Provider as read-ahead buffers. Generally between 4 and 8 should be sufficient to allow the

Data Provider to process without having to wait for a data link buffer to be cleared.

AMDS$SYSTEM

Undefined

DECAMDS

5 data link reads

1

1

Each read request requires 1500 bytes of BYTCNT quota used for the starting process.


B–3



Table B–4 (Cont.) Logical Names Defined on Nodes Running the Data Provider

Logical Name Definition Default

AMDS$RM_DEFAULT_

INTERVAL

The number of seconds between multicast hello messages from the Data Provider to the Data

Analyzer node when the Data Provider is inactive or is only minimally active.

The minimum value is 15. The maximum value is 300.

AMDS$RM_OPCOM_READ This logical defined as TRUE allows OPCOM messages for read failures from the Data

Provider. Defined as FALSE, the message facility is disabled.

AMDS$RM_OPCOM_

WRITE

This logical defined as TRUE allows OPCOM messages for write (fix) successes and failures from the Data Provider. Defined as FALSE, the message facility is disabled.

AMDS$RM_SECONDARY_

INTERVAL

The number of seconds between multicast hello messages from the Data Provider to the Data

Analyzer node when the Data Provider is active.

The minimum value is 15. The maximum value is 1800.

30

TRUE

TRUE

90

B.3 Log Files

The DECamds Data Analyzer records two log files:

• An events log file named AMDS$EVENT_LOG.LOG. This ASCII text file records all event messages displayed in the Event Log window.

• A lock contention log file named AMDS$LOCK_LOG.LOG. This ASCII text file records all lock contention information displayed in the Lock Contention window.

Both log files are created when the DECamds application is started. Either file can be edited while the application is running.

Event Log File and Lock Log File Enhancements

Prior to Version 7.2, the Event Log File and Lock Log File were created with a default creation size of 1 block and a default extension size of 1 block. This sometimes resulted in a very fragmented log file (and disk) when DECamds was allowed to run for a long period of time.

Two new logicals in the AMDS$LOGICALS.COM file allow users to define additional sizes in log files. The following table describes these logicals and their default values.

Logical

AMDS$EVTLOG_ALLOC_SIZE

AMDS$EVTLOG_EXTNT_SIZE

Description

Sets the initial size of the log files.

Sets the extension size of a file when it needs to grow.

Default Value

100 blocks

0 blocks

The default value for AMDS$EVTLOG_EXTNT_SIZE causes DECamds to use the system defaults for extent size.

B–4


B.4 Event Log File

B.4 Event Log File

The event log file keeps a record of the events detected by DECamds. You can review it without a DECwindows terminal. Every 30 minutes, DECamds writes a message to the file, noting the date and time.

Example B–1 is an example of AMDS$LOG:AMDS$EVENT_LOG.LOG.

Example B–1 Sample Event Log File

Time Sev Event

Opening DECamds Event Log on date/time: 11:16:07.00

11:16:07.98 0 CFGDON, PROD12 configuration done


11:16:09.65 0 CFGDON, AXPND1 configuration done


11:16:11.89 0 CFGDON, VAXND1 configuration done



11:16:14.57 60 HIDIOR, PROD12 direct I/O rate is high

11:16:14.57 70 HITTLP, PROD12 total page fault rate is high

11:16:14.57 80 LOMEMY, PROD12 free memory is low



11:16:15.32 70 HITTLP, AXPND1 total page fault rate is high

11:16:25.33 60 HIBIOR, PROD09 buffered I/O rate is high

11:16:35.46 60 HIBIOR, AXPND1 buffered I/O rate is high

11:16:40.62 95 LOSWSP, AXPND1 DISK$ALPHAVMS015:[SYS0.SYSEXE]SWAPFILE.SYS swap file space is low

















11:17:39.84 60 HINTER, PROD07 interrupt mode time is high







11:18:15.68 60 HIBIOR, AXPND1 buffered I/O rate is high


11:18:40.57 75 HIHRDP, PROD10 hard page fault rate is high


11:18:55.91 60 HINTER, PROD07 interrupt mode time is high



11:19:09.67 75 HIHRDP, PROD09 hard page fault rate is high


B.5 Lock Contention Log File

Example B–2 is an example of a Lock Contention Log File.

B–5


B.5 Lock Contention Log File

Example B–2 Sample Lock Contention Log File

***********************************************

Time: 9-JUL-2000 14:23:46.68

Master Node: AXPND1

Resource Name: QMAN$JBC_ALIVE_01

Parent Resource Name: QMAN$MSR_$10$DKA300.....ñ.....

RSB Address: 805B1400, GGMODE: EX, CGMODE: EX

Hex Representation

514D414E 244A4243 (Bytes 0 - 7)

5F414C49 56455F30 (Bytes 8 - 15)

31000000 00000000 (Bytes 16 - 23)

00000000 000000C0 (Bytes 24 - 31)

Status: VALID

***********************************************

Time: 9-JUL-2000 14:28:42.44

Resource Name: QMAN$JBC_ALIVE_01

Parent Resource Name: QMAN$MSR_$10$DKA300.....ñ.....

Blocking Lock Data

Node: AXPND1, PID: 2020008C, Name: JOB_CONTROL

LKID: 0200015E, GR Mode: EX

Flags: NOQUEUE

Local Copy

Blocked Lock on WAITING queue

Node: AXPND1, PID: 2020008D, Name: QUEUE_MANAGER

LKID: 2000013B, RQ Mode: CR

Flags: NODLCKW

Local Copy

***********************************************

B.6 OPCOM Log

The following examples show some of the OPCOM messages that appear in the operator log file from the Data Provider:

%%%%%%%%%%% OPCOM 2-JAN-2000 08:16:21.92

%%%%%%%%%%%

Message from user RMDRIVER

RMA0: - No privilege to access from node 2.2

This message means that the node does not have the privilege to perform a read operation.

%%%%%%%%%%% OPCOM 2-JAN-2000 10:10:45.08

%%%%%%%%%%%


RMA0: - No privilege to write from node 2.2

This message means that the Data Provider does not have the privilege to perform a write operation.

%%%%%%%%%%% OPCOM 2-JAN-2000 12:35:05.28

%%%%%%%%%%%


RMA0: - Process 2390003c modified from node 2.2

This message means that the Data Provider has successfully performed a write operation on the node.

B–6

Glossary

Following is an alphabetical listing of terms used in this manual and their definitions.

automatic data collection

Data collection that begins automatically when the Data Analyzer runs and recognizes a Data Provider. By default, this feature is enabled.

The default data windows for which automatic collection is enabled are:

Node Summary


Lock Contention Summary

Cluster Transition Summary

Automatic Event Investigation

Enhances the speed with which you can pursue a specified event. When this option is enabled, DECamds automatically collects follow-up data on the event.

When this option is disabled, you must initiate follow-up data collection when an event occurs.

To enable automatic event investigation, choose Enable Automatic Event

Investigation from the Control menu of the System Overview or Event Log window. To disable it, choose the Disable Automatic Event Investigation menu item.

This feature does not apply to any lock contention events. To enable automatic lock contention detection, use the DECamds Application Customizations dialog box, as explained in Section 5.1.

collection interval

The frequency at which the Data Analyzer will send requests to a Data Provider to collect data.

See also Data Analyzer, Data Provider.

Data Analyzer

The portion of DECamds that collects and displays system data from Data

Provider nodes. You can also perform fixes with the Data Analyzer.

See also Data Provider, fix.

Data Provider

The portion of DECamds that is installed to provide system data when requested by authorized Data Analyzers. A Data Provider node uses the OpenVMS LAN drivers to receive and send data across the network.

See also Data Analyzer.

Glossary–1

Glossary–2 data window

A Data Analyzer window that displays additional data. A number of different data windows are available as follows (see also Chapter 3):

CPU Modes Summary

CPU Summary

Disk Status Summary

Volume Summary

Single Disk Summary

Lock Contention Summary

Memory Summary

Node Summary


Process I/O Summary

Single Lock Summary


Cluster Transition/Overview Summary

System Communication Architecture Summary

NISCA Summary

event

A description of a potential resource availability problem, based on rules defined by the Data Analyzer and customized thresholds. Events trigger display changes in data windows such as color and item highlighting.

See also Data Analyzer, data window.

Event Log window

One of two primary Data Analyzer windows that displays events as they occur.

For each event, you can display more detailed information to investigate the underlying problem by double-clicking on the event. You can also perform fixes for some events from this window.

See also System Overview window.

fix

A corrective action made to a Data Provider node but initiated from the Data

Analyzer node.

group

A set of remote Data Provider nodes with similar attributes; for example, all the members of an OpenVMS Cluster can be in the same group. The group that a node belongs to is determined by the translation of the AMDS$GROUP_NAME logical on each Data Analyzer.

occurrence value

The number of consecutive data samples that must exceed the event threshold before an event is signaled.

page

A unit used by the operating system to section memory. On VAX systems, a page is 512 bytes. On Alpha systems, a page can be 8 kilobytes (8192 bytes), 16 KB,

32 KB, or 64 KB.

pagelet

A unit used by the OpenVMS Alpha operating system to break down the page into smaller addressable units. One pagelet is the same as a VAX page: 512 bytes.

security triplet

A three-part access code located in the AMDS$DRIVER_ACCESS.DAT and

AMDS$CONSOLE_ACCESS.DAT files that enables communications to be established between the Data Analyzer and Data Provider.

System Overview window

One of two primary Data Analyzer windows that graphically displays groups and the nodes that belong to each group. The System Overview window provides summary data about CPU, Memory, Process I/O usage, Number of Processes in

CPU Queues, Operating System Version, and Hardware Model for the nodes being monitored.

Glossary–3

Index

A

Access for Data Analyzers, 1–4 for Data Providers, 1–4 read-only, 1–9

Account field, 3–20

Adjust Working Set fix, 4–6

AMDS$*_DEFS.DAT files, B–1

AMDS$APPLIC_CUSTOMIZE.DAT file, 5–2

AMDS$COMM.EXE file, B–1

AMDS$COMMBUFOVF logical name, 5–20

AMDS$COMM_BUFFER_SIZE logical name,

5–20, B–2

AMDS$COMM_READS logical name, 5–20, A–3,

B–2

AMDS$CONFIG logical name, B–2, B–3

AMDS$CONSOLE.EXE file, B–1

AMDS$CONSOLE.UID file, B–1

AMDS$CONSOLE_ACCESS.DAT file, 1–4, B–1

AMDS$DEVICE logical name, B–3

AMDS$DRIVER_ACCESS.DAT file, 1–4, B–3 example, 1–8

AMDS$EVENT_LOG.LOG file, 2–7, B–4

AMDS$EVTLOG_ALLOC_SIZE

DECamds logical, B–4

AMDS$EVTLOG_EXTNT_SIZE

DECamds logical, B–4

AMDS$GROUP_NAME logical name, 1–8, B–3

AMDS$IVP.COM file, B–2, B–3

AMDS$LOCK_LOG.LOG file, B–4

AMDS$LOGICALS.COM file, 1–8, B–1, B–3

AMDS$LOG logical name, B–2

AMDS$NUM_DL_READS logical name, B–3

AMDS$RM_DEFAULT_INTERVAL logical name,

B–3

AMDS$RM_OPCOM_READ logical name, B–4

AMDS$RM_OPCOM_WRITE logical name, B–4

AMDS$RM_SECONDARY_INTERVAL logical name, B–4

AMDS$STARTUP.COM file, 1–8, A–8, B–2

AMDS$THRESHOLD_DEFS.DAT file, 5–11

AMDS$VMS*-*.LIB files, B–2

AMDS073.RELEASE_NOTES, B–1, B–3

Assigning a node to a group, 1–8

ASTLM (AST limit) quota current count, 3–23

Automatic data collection, 5–3 event investigation, 5–3 fix, 4–10 investigation for events, 5–4 investigation for locks, 5–4

Availability messages setting broadcast intervals, 1–10

AVAIL command, 2–1

B

Bell filter, 5–6

Bell volume, 5–6

Blocks number free on a volume, 3–6 number used on a volume, 3–6 percentage number used on a volume, 3–6

Broadcast intervals, B–3 setting for node availability messages, 1–10

Buffered I/O, 3–22 byte limit (BYTLM), 3–24 field, 3–13 limit (BIOLM), 3–23

Limit Remaining field, 3–13 rate, 3–13 rate display, 2–4 wait state, 3–23

Byte limit remaining for process I/O, 3–13

C

Change process priority fix, 4–6

Choosing data categories, 2–6

Classes

CPU, 5–12 customizing, 5–11

I/O, 5–12 memory, 5–12 thresholds, 5–12

Cluster hung, 4–3

Cluster Transition/Overview Summary window,

3–30 to 3–33

Index–1

Collecting data automatic at startup, 5–3 by category, 2–6 by event, 2–6 choosing a data category, 2–6 default, 5–3 for events, 5–3 for lock events, 5–3 options, 5–3 recommendations to handle heavy workloads,

5–18 single node or group, 2–6 stopping, 2–6

Collection intervals changing at window level, 5–16, 5–20 globally, 5–16 collection factor, 5–17 default values, 5–17 display factor, 5–17

Event, 5–17 factor setting, 5–3

NoEvent, 5–17

Collect menu, 2–6

Command qualifiers, 2–1

Communications buffer, 5–20

Compaq DECamds

See DECamds

Compute wait state, 3–23

/CONFIGURE qualifier, 2–1

Control wait state, 3–23

Conversion locks, 3–29

Corrective action, 2–9

CPU Modes Summary window, 3–14 to 3–15

CPU queues display of number of processes in, 2–4

CPUs (central processing units) capability, 3–15 classes, 5–12

CPU identifier (ID), 3–15

CPU Modes field, 3–11

CPU Process State Queues field, 3–11 default data collection, 5–3 execution rates, 3–22 filter categories, 5–8 filtering data, 5–7 improving performance by suspending, 4–10 response, 5–13 load in gathering data, 2–6 modes, 3–11, 3–15 number active on a node, 3–11 peak usage, 3–15 percentage used, 3–15 process identifier (PID), 3–15, 3–16 name in, 3–15

CPUs (central processing units) process (cont’d) priority, 3–16 state queues, 3–11 time, 3–16 program counter, 3–20 setting process priorities, 4–7 state, 3–15, 3–16 time rate, 3–15, 3–16 time limit for single process modes, 3–23 usage, 3–14, 3–15 display of, 2–4 wait state, 3–16

CPU Summary window, 3–15 to 3–16

Crash Node fix, 4–7

Customizing automatic investigation when events detected,

5–3 collection interval factor, 5–3, 5–20 default settings, 5–1 effect on performance, 5–20 hiding node names on startup, 5–3 highlighting event data, 5–3 security files, 1–7 template customization files, 5–2

D

Data display default, 2–5 event details, 2–9 link, 5–20 transfer security, 1–4

Data Analyzer, 1–2 access for Data Providers, 1–4

AMDS$CONSOLE_ACCESS.DAT file, 1–4 data exchange with Data Provider, 1–8 files used for, B–1 log files, B–4 security file, A–8 setting up after installing, A–8 starting, 2–1 to 2–2, A–9 system directory definition, A–8 typical setup, 1–2 user account privileges and quotas, A–8

Data collection

See Collecting data

Data Provider, 1–2 access for Data Analyzers, 1–4

AMDS$DRIVER_ACCESS.DAT file, 1–4 data exchange with Data Analyzer, 1–8 files, B–3 restarting, 1–8 restrictions, 1–2 security file, A–7 setting up after installation, A–7

Index–2

Data Provider (cont’d) starting, A–8

Data windows

See also specific window names

Cluster Transition/Overview Summary, 3–30

CPU Modes Summary, 3–14

CPU Summary, 3–15

Disk Status Summary, 3–2 hierarchy, 3–1

Lock Contention Summary, 3–25

Memory Summary, 3–17

NISCA Summary, 3–36

Node Summary, 3–10 overview, 3–1

Page/Swap File Summary, 3–8

Process I/O Summary, 3–12

Single Disk Summary, 3–6

Single Lock Summary, 3–28

Single Process Summary, 3–19

System Communications Architecture (SCA)

Summary, 3–33

Volume Summary, 3–5

DECamds

Data Analyzer system requirements, A–2 installation requirements, 1–3 log file enhancements, B–4 overview, 1–1 processing model, 1–2 security features, 1–3 starting, 2–1 to 2–2, A–8

DECamds Data Analyzer installation requirements, A–2 preparing for installation, A–1 system requirements, A–2

DECwindows Motif used on DECamds, 1–2

Defaults automatic data collection, 5–3 collection intervals, 5–17 customizing, 5–1 data display, 2–5 event color, 5–3 event highlighting, 5–3 event investigation, 5–3 lock event collect state, 5–3 options, 2–5 setting default data collection, 5–3

Deleting events, 2–10

Device name field, 3–3, 3–6

DIO

See Direct I/O

DIOLM (Direct I/O limit), 3–23

Direct I/O

DIO rate field, 3–13

Limit Remaining field, 3–13 use display, 2–4 wait state, 3–23

Disk error messages, 3–4

Disk space required for installation, A–3

Disk Status Summary window, 3–2 to 3–4

Disk volumes error messages, 3–6

Displaying default data, 2–5 event data, 2–9 options, 2–5

Display software installation requirements, A–2

DSKERR error message, 3–4

DSKINV error message, 3–4

DSKMNV error message, 3–4

DSKOFF error message, 3–4

DSKQLN error message, 3–6

DSKRWT error message, 3–4

DSKUNA error message, 3–4

DSKWRV error message, 3–4

Duration field, 3–26, 3–29

E

ENQLM (enqueue limit) job quotas in use, 3–23

Error messages

CPU, 3–16 disk status, 3–4 disk volume, 3–6 lock contention, 3–27 memory, 3–18 node, 3–11 page/swap file, 3–9 process I/O, 3–13 single lock, 3–29 single process, 3–24

Errors field, 3–4

Escalation severity filter, 5–6 time filter, 5–6

Event Log window, 2–7 deleting events from, 2–10 display fields, 2–7 filters, 5–6 menus, 2–8 using, 2–7

Events

See also Event Log window automatic investigation, 5–4 bell filter, 5–6 bell volume, 5–6 changing default highlighting, 5–3 severity filter, 5–7, 5–11 corrective action, 2–9 creating thresholds for different computer classes, 5–11

Index–3

Events (cont’d) customizing based on frequency of occurrence,

5–11 deleting, 2–10 displaying more information, 2–9 escalation severity filter, 5–6 time filter, 5–6 filtering, 5–6 highlight filter, 5–6 highlighting color, 5–3 lock contention investigation, 5–4 log files, B–4 removing from the Event Log window, 2–6 sending messages to OPCOM, 2–10 severity filter, 5–6 severity values, 2–7 signal filter, 5–6 temporary freeze, 2–10 timeout, 2–8 filter, 5–6

Exiting Image and Deleting Process fix, 4–8

Explicit wait state, 3–23

F

Fault rate for pages, 3–18

File Name field, 3–9

File protection for security, 1–4

Files, B–1 to B–6

FILLM (file limit) job quota in use, 3–23

Filtering data at window level, 5–20 changing a filter category, 5–8 event qualification, 5–5 methods, 5–4

Filtering events, 5–6

Filters bell, 5–6 bell volume, 5–6 changing severity values for events, 5–6, 5–7,

5–11 escalation severity, 5–6 escalation time, 5–6

Event Log, 5–6 highlight, 5–6 signal (display), 5–6 timeout, 5–6

Fixes, 4–1 adjust working set, 4–6 automatic, 4–10 changing process priority, 4–6 working set size, 4–6 cluster hung, 4–3 crashing a node, 4–7 deleting a process, 4–8 examples, 4–10 to 4–13

Fixes (cont’d) exiting an image, 4–8

Fix menu, 2–9 intruder, 4–3 list of available, 4–1 manual, 4–11 memory too low, 4–3 memory usage, 4–1 options, 2–9 performing, 4–2 to 4–10 process, 4–1, 4–3 purging a working set, 4–9 quorum, 4–1 resuming a process, 4–9 runaway process, 4–3 summary, 4–3 suspending a process, 4–9 system, 4–1 understanding, 4–2 working set too high or too low, 4–3

Flags field, 3–29

Free field, 3–6

G

Getting Started, 2–1

Granted locks, 3–29

GR mode field, 3–29

/GROUP qualifier, 2–1

Groups, 2–4 collapsing information, 2–5 collecting data, 2–6 expanding information, 2–5 how to assign a node, 1–8 of nodes creating, 1–7

H

Hanging cluster fix, 4–3

Hardware

DECamds installation requirements, A–2 model field, 3–11 security triplet address, 1–6

Hardware model display of for node, 2–4

Hello message broadcasts

See Broadcast intervals

Help, 2–2

HIBIOR error message, 3–11

HICOMQ error message, 3–11

Hide Nodes, 2–5 changing default behavior, 5–3

HIDIOR error message, 3–11

Highlighting changing default behavior, 5–3 customizing color, 5–3 filter, 5–6

Index–4

HIHRDP error message, 3–11

HIMWTQ error message, 3–11

HINTER error message, 3–11

HIPWIO error message, 3–11

HIPWTQ error message, 3–11

HISYSP error message, 3–11

HITTLP error message, 3–11

HMPSYN error message, 3–11

I

I/O (input/output) average number of operations pending for a

BIO volume, 3–6 limit remaining, 3–13 buffered limit (BIOLM), 3–23 buffered wait state, 3–23 byte limit remaining, 3–13 classes, 5–12 default data collection, 5–3

DIO limit remaining, 3–13 direct limit (DIOLM), 3–23 direct wait state, 3–23

Fault rate for pages, 3–18 open files, 3–13 paging, 3–22

PIO, 3–13 process identifier (PID), 3–13 summary, 3–12 summary for node, 3–11

I/O field, 3–11

IEEE DECamds protocol, 1–4

Images Activated field, 3–22

Installation requirements

DECamds, 1–3

Installing DECamds post-installation tasks, A–7

Installing software

DECamds Data Analyzer, A–1

Installing the DECamds Data Analyzer, A–4 requirements, A–2

Interrupt priority level

See IPL

Intruder fix, 4–3

Investigating events, 4–11

IPID (internal PID), 3–15

IPL (interrupt priority level) on DECamds, 1–2

J

Job quotas in use, 3–23 to 3–24

JOB_CONTROL process, 4–10

L

LAN polling, 5–18 setting load, 5–19

LCKBLK error message, 3–29

LCKCNT error message, 3–27

LCKWAT error message, 3–29

LIB files, B–2

LKID field, 3–29

LOASTQ error message, 3–24

LOBIOQ error message, 3–13, 3–24

LOBYTQ error message, 3–13, 3–24

Lock Contention Summary window, 3–25 to 3–27 default data collection, 5–3 detailed data, 3–28 logging information, B–4

Locks automatic investigation, 5–4 types of, 3–29

Lock Type field, 3–29

LODIOQ error message, 3–13, 3–24

LOENQU error message, 3–24

LOFILQ error message, 3–13, 3–24

Log files, B–4 lock contention, B–5

Logging event messages, B–4

Logical names

AMDS$COMMBUFOVF, 5–20

AMDS$COMM_BUFFER_SIZE, 5–20, B–2

AMDS$COMM_READS, 5–20, B–2

AMDS$CONFIG, B–2, B–3

AMDS$DEVICE, B–3

AMDS$GROUP_NAME, B–3

AMDS$LOG, 2–7, B–2

AMDS$NUM_DL_READS, B–3

AMDS$RM_DEFAULT_INTERVAL, B–3

AMDS$RM_OPCOM_READ, B–4

AMDS$RM_OPCOM_WRITE, B–4

AMDS$RM_SECONDARY_INTERVAL, B–4

Data Analyzer node, B–1

Data Provider node, B–3 requirements, B–1 sending messages to OPCOM, 1–9

/LOG_DIRECTORY qualifier, 2–1

LOMEMY error message, 3–11

Looping process fix, 4–3

LOPGFQ error message, 3–24

LOPGSP error message, 3–9

Index–5

LOPRCQ error message, 3–24

LOSWSP error message, 3–9

LOTQEQ error message, 3–24

LOVLSP error message, 3–6

LOWEXT error message, 3–18, 3–24

LOWSQU error message, 3–18, 3–24

LRGHSH error message, 3–27

M

Manual fix, 4–11

Master Node field, 3–26

Memory classes, 5–12 default data collection, 5–3 distribution on a node, 3–11 example fix for low, 4–10 fixes, 4–1 fix for low, 4–3 investigating low memory, 4–11

Memory field, 3–11 process identifier (PID), 3–18 rate display, 2–4 requirements, 5–18 sorting data, 5–14 total for a node, 3–11 wait state, 3–23

Memory field, 3–11

Memory Summary window, 3–17 to 3–18

Messages sending to OPCOM, 1–9

Mount field, 3–4

Multicast messages customizing interval, B–3

Mutexes held, 3–22

N

Network address security triplet, 1–5

Network Interconnect System Communication

Architecture

See NISCA

NISCA Summary window, 3–36 to 3–40

Node Name field, 3–9

Nodes assigning to a group, 1–8 availability messages setting broadcast intervals, 1–10 crash fix, 4–7 default data collection, 5–3 error messages, 3–11 field, 3–29 hardware model of, 2–4 shutdown procedure, A–7

Node Summary window, 3–10 to 3–12

NOPGFL error message, 3–9

NOPROC error message, 3–11

NOSWFL error message, 3–9

O

Occurrence value, 5–11

Online release notes, A–1

OPCOM (Operator Communication Manager) filtered event messages sent to, 5–6 sending events to, 2–10 sending messages, 1–9 using on DECamds, 1–4

Open files field, 3–13 for process I/O, 3–13 limit remaining, 3–13

Limit Remaining (Files) field, 3–13

OpenVMS Clusters including in groups, 1–7

Operating system field, 3–11 version displayed, 2–4 version requirements for installing the Data

Analyzer, A–2

Operations count rate, 3–6

OpRate field, 3–6

Owner ID field, 3–20

P

Page/Swap file error messages, 3–9

Page/Swap File Summary window, 3–8 to 3–9

Page faults, 3–22 adjust working set, 4–6 field, 3–11

I/O Rate field, 3–18 purging working sets, 4–9

Rate field, 3–18

Page files default data collection, 5–3

Pagelets number used, 3–9 percentage number used, 3–9 reserving for use, 3–9 total available in page file, 3–9

Pages number used, 3–9 percentage number used, 3–9 reserving for use, 3–9 total available in page file, 3–9

Paging I/O, 3–22

Parameter (system) settings required for installation, A–3

Parent Resource field, 3–26, 3–29

Index–6

Partitioning groups, 1–7

Path field, 3–3, 3–6

PC (program counter) field, 3–20

Performance improving, 5–1 with hardware, 5–21 optimizing by customizing data collection, 5–18

PGFLQUO (page file quota) job quotas in use,

3–23

PID (process identifier), 3–16

CPU, 3–15 field, 3–13, 3–18, 3–20

I/O, 3–13 memory, 3–18 single process, 3–20

PIO (paging I/O) field, 3–13 rate, 3–13

Polling

LAN, 5–18

PRBIOR error message, 3–13, 3–24

PRBIOW error message, 3–24

PRCCOM error message, 3–16, 3–24

PRCCUR error message, 3–24

PRCCVR error message, 3–16

PRCLM process limit job quotas in use, 3–24

PRCMUT error message, 3–24

PRCMWT error message, 3–16

PRCPUL error message, 3–24

PRCPWT error message, 3–16, 3–24

PRCQUO error message, 3–24

PRCRWA error message, 3–24

PRCRWC error message, 3–24

PRCRWM error message, 3–24

PRCRWP error message, 3–24

PRCRWS error message, 3–24

PRCUNK error message, 3–24

PRDIOR error message, 3–13, 3–24

PRDIOW error message, 3–24

Priority field, 3–20 process, 3–20 process fix, 4–1

Private LAN transport security, 1–3

Privileges installation requirements, A–3 to run Data Analyzer node, A–8 to run Data Provider node, A–8 to start DECamds, A–7

PRLCKW error message, 3–24

Problems, reporting, A–10

Processes account quotas required for installation, A–3 displaying number of in CPU queues, 2–4 fixes, 4–1 looping fix, 4–3 name in CPU, 3–15

Processes (cont’d) privileges, 1–4 quotas recommended, 5–19 setting, 5–19 states, 3–21 queues on CPUs, 3–11

Process I/O Summary window, 3–12 to 3–13

Process Name field, 3–13, 3–18, 3–20, 3–29

PRPGFL error message, 3–18, 3–24

PRPIOR error message, 3–13, 3–18, 3–24

PSLs (processor status longwords), 3–20

Purge Working Set fix, 4–9

Q

Qualifiers

See Command qualifiers

Queue field, 3–6

Quorum fix, 4–1

Quotas in use for jobs, 3–23 limit fix, 4–3 process account quota requirements, A–3 process mode, 3–23 recommended for processes, 5–19 to run Data Analyzer node, A–8 wait state, 3–23

R

Read-only access, 1–9

Recording event messages, B–4 lock contention information, B–4

Release Notes, 5–2, A–1

Reporting problems, A–10

RESDNS error message, 3–27

Reservable field, 3–9

Resource Name field, 3–29

Response time external factors, 5–18 optimizing by customizing data collection, 5–18 system hardware, 5–21

RESPRS error message, 3–27

Restarting the Data Provider, 1–8

Restrictions on the Data Provider, 1–2

Resume Process fix, 4–9

RMCP.EXE file, B–3

RMDRIVER.EXE file, B–3

RQ Mode field, 3–29

Runaway process fix, 4–3

Rwait field, 3–4

Index–7

S

SCA (System Communications Architecture)

Summary window, 3–33 to 3–35

Security customizing files, 1–7

Data Analyzer file, A–8

Data Provider file, A–7 data transfe, 1–4 file protection, 1–4 logging issues with OPCOM, 1–4 private LAN transport, 1–3 process privileges, 1–4 read-only access, 1–9 steps after installing

Data Analyzer, A–8

Data Provider, A–7

Security triplets access verification code, 1–5 format, 1–5 how they work, 1–6 network address, 1–5 password, 1–5 verifying, 1–7 wildcard address, 1–6

Setting broadcast intervals, 1–10

Severity events, 5–6

Show Nodes, 2–5 changing default behavior, 5–3

Shutdown procedure, node, A–7

Signal filter, 5–6

Single Disk Summary window, 3–6

Single Lock Summary window error messages, 3–29

Granted, Conversion, and Waiting Queue lock,

3–29

Granted lock fields, 3–29

Single node collecting data, 2–6

Single Process Summary window error messages, 3–24 execution rates, 3–22 quotas, 3–23

SMP (symmetric multiprocessing), 2–4

Sorting data, 5–14, 5–15 for memory data, 5–14

Starting DECamds, 2–1 to 2–2

Starting the Data Analyzer, 2–1 to 2–2, A–9

Starting the Data Provider, A–8

State field, 3–20

Status field, 3–4, 3–26

Stopping data collection, 2–6

Suspended process fix, 4–3, 4–9

Swap files default data collection, 5–3

SWAPPER process fixes ignored, 4–2

SYS$HELP.HLB file, B–1, B–3

SYS$RMDRIVER.EXE file, B–3

SYS$STARTUP.COM file, B–3

System fix, 4–1 load recommendations for hardware, 5–21 parameter settings required for installation,

A–3 processes, 4–10

System Communications Architecture

See SCA

System Overview window, 2–2, 2–3 defining groups, 1–7 hiding node name on startup with customization file, 5–3 menus, 2–4

T

Thresholds customizing for events or computer classes,

5–11

Timeout filter, 5–6

Total field, 3–9

TQELM process limit job quotas in use, 3–23

Trans (transaction) field, 3–4

Transactions number of operations for disk, 3–4

Triplets

See Security triplets

U

UICs (user identification codes) single process summary, 3–20

UIC field, 3–20

Uptime field, 3–11

Used field, 3–6, 3–9

% Used field, 3–6, 3–9

User accounts

Data Analyzer node privileges and quotas, A–8

Data Provider node privileges, A–3

Username field, 3–20

V

Version display of operating system, 2–4

Viewing groups, 2–5

View menu, 2–5

Volume default data collection, 5–3

I/O operations, 3–6 number of free blocks, 3–6

Index–8

Volume (cont’d) number of used blocks, 3–6 operations count rate, 3–6 percentage number of blocks used, 3–6

Volume Name field, 3–3, 3–6

Volume Summary window, 3–5 to 3–6

W

Waiting queue locks, 3–29

Wait states buffered I/O, 3–23 compute, 3–23 control, 3–23

CPU, 3–16 direct I/O, 3–23 disk status, 3–4 explicit, 3–23 memory, 3–23 quota, 3–23

RWAIT, 3–4

Wildcard address security triplet, 1–6

Working set count, 3–18 default, 3–22 extent, 3–18, 3–22 global pages, 3–21 private pages, 3–22 purging, 4–9 quota, 3–22 size, 3–18, 3–22 size fix, 4–6 too high or too low, 4–3 total pages, 3–22

WSdef field, 3–22

WSextent field, 3–22

WSquo field, 3–22

Index–9

Compaq Version 7.3 DECamds, Data Analyzer, Data Provider User’s Guide

Assistant Bot

DECamds User’s Guide

Order Number: AA–Q3JSE–TE

April 2001

This guide explains how to use DECamds software to detect and fix system availability problems. It also explains how to install DECamds.

Revision/Update Information:

This guide supersedes the DECamds

User’s Guide, Version 7.1.

Operating System and Version:

Data Analyzer:OpenVMS Alpha and

VAX Version 7.2 or later

Data Provider: OpenVMS Alpha and

VAX Version 6.2 or later

Software Version:

Compaq DECamds Version 7.3

Compaq Computer Corporation

Houston, Texas

Contents

Preface

Intended Audience

Document Structure

Related Documents

Reader’s Comments

How to Order Additional Documentation

Conventions

Monospace text

Overview of DECamds

1.1 How Does DECamds Work?

1.2 Where to Install the DECamds Data Analyzer

1.3 DECamds Security Features

Getting Started

2.1 Starting DECamds

2.2 Using the System Overview Window

2.3 Using the Event Log Window

Managing DECamds Data Windows

3.1 Disk Status Summary Window

3.2 Volume Summary Window

3.3 Single Disk Summary Window

3.4 Page/Swap File Summary Window

3.5 Node Summary Window

3.6 Process I/O Summary Window

3.7 CPU Modes Summary Window

3.8 CPU Summary Window

3.9 Memory Summary Window

3.10 Single Process Summary Window

3.11 Lock Contention Summary Window

3.12 Single Lock Summary Window

3.13 Cluster Transition/Overview Summary Window

3.14 System Communications Architecture Summary Window

3.15 NISCA Summary Window

Performing Fixes

4.1 Understanding Fixes

4.2 Performing Fixes

4.3 Examples for Fixing Low Memory Availability

Customizing DECamds

5.1 Customizing DECamds Defaults

5.2 Filtering Data

5.3 Sorting Data

5.4 Setting Collection Intervals

5.5 Optimizing Performance with System Settings

Installing the DECamds Data Analyzer

A.1 General Installation Information

A.2 Data Analyzer Installation Requirements

A.3 Obtaining the Data Analyzer Software

A.4 Installing Data Analyzer Software from a PCSI Kit

%PCSI-E-INSTAL

A.5 Postinstallation Tasks on Data Provider Nodes

A.6 Postinstallation Tasks on a Data Analyzer Node

A.7 Starting to Use the Data Provider

A.8 Determining and Reporting Problems

A.9 Running the Installation Verification Procedure Separately

DECamds Files and Logical Names

B.1 Files and Logical Names for the Data Analyzer Node

B.2 Files and Logical Names for Data Provider Nodes

B.3 Log Files

B.4 Event Log File

B.5 Lock Contention Log File

B.6 OPCOM Log

Glossary