NetWare 5.1: Server Operating System Administration Guide

NetWare 5.1: Server Operating System Administration Guide
NetWare Server Operating System
December 17, 1999
doc_tpl.fm
Rev 99a
28 October 99
Legal Notices
Novell, Inc. makes no representations or warranties with respect to the contents or use of this documentation, and
specifically disclaims any express or implied warranties of merchantability or fitness for any particular purpose.
Further, Novell, Inc. reserves the right to revise this publication and to make changes to its content, at any time,
without obligation to notify any person or entity of such revisions or changes.
Further, Novell, Inc. makes no representations or warranties with respect to any software, and specifically disclaims
any express or implied warranties of merchantability or fitness for any particular purpose. Further, Novell, Inc.
reserves the right to make changes to any and all parts of Novell software, at any time, without any obligation to
notify any person or entity of such changes.
This product may require export authorization from the U.S. Department of Commerce prior to exporting from the
U.S. or Canada.
Copyright © 1993-2000 Novell, Inc. All rights reserved. No part of this publication may be reproduced,
photocopied, stored on a retrieval system, or transmitted without the express written consent of the publisher.
U.S. Patent Nos. 4,555,775; 5,157,663; 5,349,642; 5,455,932; 5,553,139; 5,553,143; 5,594,863; 5,608,903;
5,633,931; 5,652,854; 5,671,414; 5,677,851; 5,692,129; 5,758,069; 5,758,344; 5,761,499; 5,781,724; 5,781,733;
5,784,560; 5,787,439; 5,818,936; 5,828,882; 5,832,275; 5,832,483; 5,832,487; 5,859,978; 5,870,739; 5,873,079;
5,878,415; 5,884,304; 5,893,118; 5,903,650; 5,905,860; 5,913,025; 5,915,253; 5,925,108; 5,933,503; 5,933,826;
5,946,467; 5,956,718; 5,974,474. U.S. and Foreign Patents Pending.
Novell, Inc.
122 East 1700 South
Provo, UT 84606
U.S.A.
www.novell.com
NetWare Server Operating System Administration Guide
January 2000
104-001241-001
Online Documentation: To access the online documentation for this and other Novell products, and to get
updates, see www.novell.com/documentation.
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Novell Trademarks
For a list of Novell trademarks, see the final appendix of this book.
Third-Party Trademarks
All third-party trademarks are the property of their respective owners.
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
Rev 99a
28 October 99
doc_tpl.fm
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
Rev 99a
28 October 99
doc_tpl.fm
Rev 99a
28 October 99
17
Contents
Server Operating System .
Understanding .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
New NetWare 5 Kernel. . . . . . . . . . . . . . . . . . . . . .
The Run Queue. . . . . . . . . . . . . . . . . . . . . . . .
Load Balancing . . . . . . . . . . . . . . . . . . . . . . . .
Concurrency and Parallelism . . . . . . . . . . . . . . . . .
Preemption . . . . . . . . . . . . . . . . . . . . . . . . . .
Platform Support Module . . . . . . . . . . . . . . . . . . .
The Server Console . . . . . . . . . . . . . . . . . . . . . . .
Network Connections . . . . . . . . . . . . . . . . . . . . . .
NetWare Is Self-Tuning . . . . . . . . . . . . . . . . . . . . .
Java-Based Components of NetWare 5 . . . . . . . . . . . . .
The Novell JVM for NetWare and the Just In Time Compiler
The Novell Object Request Broker (ORB) . . . . . . . . . .
Setting Up .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
14
14
15
17
18
18
20
21
23
24
24
25
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Changing the Server’s Software Configuration . . . . . .
Copying Additional NetWare Files after Installation . .
Installing, Uninstalling, and Configuring a Server Product .
Changing the Language NetWare Uses . . . . . . . . . .
Code Pages . . . . . . . . . . . . . . . . . . . . . . . .
Using Unicode . . . . . . . . . . . . . . . . . . . . .
Changing Server Message Files . . . . . . . . . . . . . .
Prerequisites . . . . . . . . . . . . . . . . . . . . . .
Procedure. . . . . . . . . . . . . . . . . . . . . . . .
Changing NLM Message Files . . . . . . . . . . . . . . .
Changing the Server Keyboard Type . . . . . . . . . . .
Changing Code Pages . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Contents
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
28
29
30
30
31
32
33
33
34
35
36
36
doc_tpl.fm
Optimizing
Rev 99a
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Applying Patches . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Matching Performance Capabilities . . . . . . . . . . . . . . . . . . . .
Setting the Load Balancing Threshold . . . . . . . . . . . . . . . . . .
Setting Server Parameter Values . . . . . . . . . . . . . . . . . . . . .
Understanding Server Defaults . . . . . . . . . . . . . . . . . . . .
Finding Current Parameter Values from MONITOR . . . . . . . . . .
Printing Current Parameter Values to a File . . . . . . . . . . . . . .
Finding Current Parameter Values with SET . . . . . . . . . . . . .
Checking Current Parameter Values with DISPLAY ENVIRONMENT
When to Modify Defaults . . . . . . . . . . . . . . . . . . . . . . . .
When Not to Modify Defaults . . . . . . . . . . . . . . . . . . . . .
Restoring Default Values with RESET ENVIRONMENT . . . . . . .
Monitoring Dynamic Statistics . . . . . . . . . . . . . . . . . . . . .
Using Optional Commands in Startup Files . . . . . . . . . . . . . .
Using SET Commands in Startup Files . . . . . . . . . . . . . . . .
Minimizing Replication Traffic . . . . . . . . . . . . . . . . . . . . . . .
Service Advertising . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Managing
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Starting and Stopping the Server . . . . . . . . . . . . . . . . . . .
Starting the Server . . . . . . . . . . . . . . . . . . . . . . . . .
Bringing Down the Server . . . . . . . . . . . . . . . . . . . . .
Starting and Stopping Processors . . . . . . . . . . . . . . . . . . .
Starting Processors . . . . . . . . . . . . . . . . . . . . . . . .
Displaying Processor Information . . . . . . . . . . . . . . . . .
Stopping Processors . . . . . . . . . . . . . . . . . . . . . . . .
Using Server Batch Files . . . . . . . . . . . . . . . . . . . . . . .
Using EDIT to Create and Edit Text and Batch Files . . . . . . .
Using NWCONFIG to Create and Edit Startup Batch Files . . . .
Making a Command Optional in a Server Batch (.ncf) File . . . .
Changing Server Name and IPX Internal Address in Autoexec.ncf
Using Server Utilities and Console Commands . . . . . . . . . . . .
How to Access Help for Utilities and Console Commands. . . . .
Using NetWare Loadable Modules . . . . . . . . . . . . . . . .
Loading and Unloading NetWare Loadable Modules . . . . . . .
Monitoring the NetWare Server . . . . . . . . . . . . . . . . . . . .
Accessing MONITOR . . . . . . . . . . . . . . . . . . . . . . .
What to Monitor and Why . . . . . . . . . . . . . . . . . . . . .
Displaying Server Alerts . . . . . . . . . . . . . . . . . . . . . . . .
28 October 99
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
39
40
40
43
44
45
46
46
47
47
48
48
48
49
49
50
50
50
51
52
52
53
55
55
56
56
57
58
60
61
62
63
63
64
65
66
66
67
72
doc_tpl.fm
Managing Workstation Connections . . . . . . . . . . . . . . .
Using MONITOR for Connection Management Tasks . . . .
Monitoring Workstation Connections to the Server. . . . . .
Sending Console Messages to Workstations. . . . . . . . .
Clearing a Workstation Connection. . . . . . . . . . . . . .
Disabling and Enabling Logins . . . . . . . . . . . . . . . .
Clearing a Connection that Uses Auto Reconnect . . . . . .
Controlling Resource Allocation with Locks . . . . . . . . .
Configuring Watchdog Packets. . . . . . . . . . . . . . . .
Enabling Console Message when Watchdog Logout Occurs
Using Java on the Server . . . . . . . . . . . . . . . . . . . .
Environment Variables for Java . . . . . . . . . . . . . . . . .
Viewing and Setting Variable Values . . . . . . . . . . . . .
Using the Novell ORB . . . . . . . . . . . . . . . . . . . . . .
Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . .
Procedure. . . . . . . . . . . . . . . . . . . . . . . . . . .
Common Maintenance Tasks . . . . . . . . . . . . . . . . . .
Backing Up . . . . . . . . . . . . . . . . . . . . . . . . . .
Checking Server Error Logs . . . . . . . . . . . . . . . . .
Checking Disk Drives . . . . . . . . . . . . . . . . . . . . .
Reviewing Server Cache Statistics . . . . . . . . . . . . . .
Checking Free Space on Server Disks and Volumes . . . .
Reviewing User Accounts . . . . . . . . . . . . . . . . . .
Testing the Uninterruptible Power Supply . . . . . . . . . .
Documenting the Network . . . . . . . . . . . . . . . . . .
Preventive Maintenance Tasks . . . . . . . . . . . . . . . . .
Preventing Memory Problems . . . . . . . . . . . . . . . .
Preventing Power Supply Errors . . . . . . . . . . . . . . .
Preventing Static Electricity Problems . . . . . . . . . . . .
Using Accounting. . . . . . . . . . . . . . . . . . . . . . . . .
Setting Up Accounting . . . . . . . . . . . . . . . . . . . .
Calculating Charge Rates for Accounting . . . . . . . . . .
Viewing Accounting Totals . . . . . . . . . . . . . . . . . .
Rev 99a
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
28 October 99
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Contents
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
73
74
74
77
77
78
79
80
82
82
83
83
84
85
85
86
87
87
87
87
88
88
88
89
89
91
91
91
92
93
93
94
95
doc_tpl.fm
Troubleshooting .
Rev 99a
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A Troubleshooting Methodology. . . . . . . . . . . . . . . . . . . . . . . . . .
Keep Your Servers Patched . . . . . . . . . . . . . . . . . . . . . . . . . .
Eliminate Obvious Causes . . . . . . . . . . . . . . . . . . . . . . . . . . .
Gather Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Simplify the System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Use Research Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Develop Hypotheses and Perform Tests . . . . . . . . . . . . . . . . . . .
Look for Additional Help . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Document the Solution. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Creating a Core Dump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Performing a Core Dump to a Local Drive . . . . . . . . . . . . . . . . . . .
Sending the Core Dump to Novell . . . . . . . . . . . . . . . . . . . . . . .
Slow Network, Performance Problems . . . . . . . . . . . . . . . . . . . . . .
Resolving Slow Server Response . . . . . . . . . . . . . . . . . . . . . . .
Insufficient Packet Receive Buffers, No ECB Available Count Errors . . . . . . .
High Utilization Statistics in MONITOR . . . . . . . . . . . . . . . . . . . . . .
What Is Normal? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
What Are the Most Common Solutions? . . . . . . . . . . . . . . . . . . . .
Suspected Multiprocessing Problems . . . . . . . . . . . . . . . . . . . . . . .
Unable to Communicate with Server, -625 Errors . . . . . . . . . . . . . . . . .
Transaction Tracking Errors, -621 Errors, FD93 Errors . . . . . . . . . . . . . .
Workstations Can’t Communicate with Server . . . . . . . . . . . . . . . . . .
Errors Reading from and Writing to Network, Server Cannot Be Found Errors
Connections Dropped Periodically . . . . . . . . . . . . . . . . . . . . . . .
Tips for Resolving General Communication Problems . . . . . . . . . . . .
Admin Account Has Been Deleted or Admin Password Is Unknown . . . . . . .
Memory Error Messages, Memory Problems . . . . . . . . . . . . . . . . . . .
NetWare Doesn’t Recognize All the Memory in the Server . . . . . . . . . .
Server Displays Memory Error Messages . . . . . . . . . . . . . . . . . . .
Resolving Server Memory Leaks . . . . . . . . . . . . . . . . . . . . . . .
Resolving Server Memory Problems. . . . . . . . . . . . . . . . . . . . . .
Disk Errors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
General Disk I/O Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Resolving Server Disk Space Problems . . . . . . . . . . . . . . . . . . . .
Mirrored Partitions Do Not Remirror Automatically . . . . . . . . . . . . . .
Mirroring Takes a Long Time . . . . . . . . . . . . . . . . . . . . . . . . .
Mirroring Stops Just Before It Is Finished . . . . . . . . . . . . . . . . . . .
Server Console Hangs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Server Abends. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
28 October 99
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
97
98
98
99
100
101
102
103
104
105
105
106
107
107
107
109
109
109
110
116
117
120
122
122
125
125
127
127
127
128
130
131
132
132
134
135
136
137
137
139
139
doc_tpl.fm
Types of Abends . . . . . . . . . . . . . . . . .
Processor Exception Abends . . . . . . . . .
Software Exceptions . . . . . . . . . . . . .
Hardware Abends . . . . . . . . . . . . . .
CPU Hog Abends. . . . . . . . . . . . . . .
Resolving Abends . . . . . . . . . . . . . . . .
Preventive Steps . . . . . . . . . . . . . . .
General Troubleshooting Tips . . . . . . . .
Resolving Abends Using the Abend.log File .
Managing the Abend Auto Recovery Mechanism
Responding to the Abend Manually . . . . .
Responding to the Abend Automatically . . .
Preventive Maintenance Tasks . . . . . . . . .
$
Novell Trademarks .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Rev 99a
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
28 October 99
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
140
140
142
142
143
144
144
145
146
150
150
153
154
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
155
Contents
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
Rev 99a
28 October 99
doc_tpl.fm
Rev 99a
17
28 October 99
Server Operating System
Although it is frequently more convenient to refer to the computer where
NetWare® resides as the server, the NetWare 5 server is the network operating
system that is installed on a server class computer. NetWare 5 turns the
computer into a high performance network platform that provides
connectivity between workstations, printers, mass storage devices, and other
networks such as the Internet. NetWare 5 also provides network users with a
server platform for hosting Java applications.
The NetWare server manages the transmission of data among the network
devices, stores and retrieves data from hard disks, manages one or more file
systems, ensures data integrity, manages printers and printing, and allocates
and manages memory.The NetWare server is designed to operate on either a
uniprocessing or multiprocessing computer. For further information on the
server, its console, and its performance, click any of the links below or click
Understanding for more options.
If you are ready to begin installing the server, see NetWare 5.1 Installation
Guide. If you are planning to use a multiprocessing computer as your server,
see “New NetWare 5 Kernel” on page 14 for information on the higher level
of support for multiprocessing.
NetWare 5 uses pure IP as its default communication protocol, but also
supports IPX. If you choose IP, you can also choose to have passive IPX in
Compatibility Mode. To decide which protocol to use, see NetWare Server
Communications.
The NDS® database and the NDS tree are the foundation for convenient
network administration. See Contents > Network Services Documentation >
Directory Services > NDS Concepts and Planning in the online
documentation.
Server Operating System
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
NetWare offers a choice of file systems—Novell Storage Services (NSS)
advanced file system and NetWare’s traditional file system. For more
information, see Novell Storage Services (NSS) and Traditional File Services.
For information about security services for your network, see Auditing the
Network.
For information about NetWare’s use of memory, see NetWare Server
Memory.
For information about NetWare’s use of disks and other storage devices, see
NetWare Server Disks and Storage Devices.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
17
28 October 99
Understanding
Server.exe, the principal executable for NetWare®, includes bound-in
modules, such as those for Media Manager, NetWare Peripheral
ArchitectureTM (NWPA), and the Link Support LayerTM (LSL). When
server.exe is executed, the NetWare server is loaded into memory, takes
control of the hardware resources, and begins to load its component modules.
127(NetWare uses DOS as a cold boot loader. It does not run on top of DOS.
Because the system is modular, individual modules can be updated without
replacing server.exe. A new module simply replaces the previous version.
127(The directory in which server.exe is stored is called the startup directory or the
boot directory. By default, server.exe is stored in the c:\nwserver directory. Updated
modules may be copied to the startup directory where they will overwrite previous
versions. Server.nlm, however, requires patches.
NetWare 5 is based on an integrated kernel. The new multiprocessing kernel
(MPK) is completely multithreaded, supports preemption, and—as an
integrated kernel—runs on both multiprocessor and uniprocessor systems.
NetWare 5 has also been architected with an eye to the future—it supports as
many as 32 processors.
Although NetWare 5 has minimum hardware requirements (see System
Requirements), we recommend that you provide a high-performance
hardware platform so you can get the full benefit of the power and scalability
that NetWare 5 offers.
To ensure that no component of your system hampers the performance of any
other, see also “Matching Performance Capabilities” on page 40.
Understanding
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
For information about
See
NetWare 5 integrated kernel
“New NetWare 5 Kernel” on page 14
Server console security and interfaces
“The Server Console” on page 20
28 October 99
Connections to server, licensing and authentication “Network Connections” on page 21
How the NetWare 5 server optimizes its own
performance
“NetWare Is Self-Tuning” on page 23
Java-based components of NetWare 5
“Java-Based Components of NetWare 5” on page
24
New NetWare 5 Kernel
The core of the NetWare 5® operating system is the new integrated kernel.
MPK (multiprocessing kernel) manages threads, schedules processor
resources, handles interrupts and exceptions, and manages access to memory
and the I/O subsystems.
For explanations and descriptions, click:
Š “The Run Queue” on page 14
Š “Load Balancing” on page 15
Š “Concurrency and Parallelism” on page 17
Š “Preemption” on page 18
Š “Platform Support Module” on page 18
The Run Queue
A thread is a stream of control that can execute its instructions independently.
A simpler definition is that a thread is a unit of execution. It is not the code
itself.
+,17For more information on threads, see Bil Lewis and Daniel J. Berg’s Threads
Primer: A Guide to Multithreaded Programming © Sun Microsystems, Inc. Their
definition of `thread’ has been quoted above.
The kernel maintains a data structure called the run queue which contains
threads that are in a state of readiness. In a uniprocessor system, there is only
one run queue from which the processor can pick up threads for execution.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
In a multiprocessing system where more than one processor is available, there
is more than one possible solution. The distribution of threads to multiple
processors could be handled by a global run queue (all processors sharing a
single run queue) or by per-processor run queues (known also as local run
queues or distributed run queues). Or by some combination of both.
To compare the two approaches:
Š Global run queue. This approach to distributing threads has the
advantage of automatic load balancing. The reason is that no processor
remains idle as long as the run queue has threads ready. However, the
global solution has the drawback of becoming a bottleneck as the number
of processors increases in a system, although under certain scheduling
policies (such as a real time scheduling policy), a global queue may be
necessary.
Š Per-processor run queue. This approach has the advantage of being
able to exploit cache affinity—where threads are preferentially scheduled
on the processor on which they last ran. In addition, this approach does
not have the bottleneck problem associated with the global run queue
approach.
With local queues, however, it becomes necessary to ensure that the load
on the processors—the number of threads in the queue—does not become
severely imbalanced. A load balancing mechanism is required to handle
load imbalances so that threads do not pile up at one processor while
another processor remains idle.
The NetWare 5 kernel uses the per-processor run queue. As implemented, a
processor can pick up threads for execution only from its local run queue. This
makes the NetWare 5 scheduler highly scalable compared to an
implementation using a global run queue. To address load imbalance,
NetWare 5 uses a sophisticated load balancing algorithm.
Load Balancing
Two important requirements of any load balancing scheme are
Š Stability. This means not overreacting to small load imbalances.
Š Ability to distribute the processing load quickly.
The NetWare 5 scheduler handles the stability requirement by using a
threshold. The threshold determines how much load imbalance is permitted in
the system before the load balancing mechanism kicks in.
Understanding
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Because a low threshold value would have less tolerance for load imbalances
than a higher value, the lower value could lead to excessive thread movement
due to frequent load balancing. A higher value would have more tolerance for
load imbalances, with the result that the load balancing mechanism would be
triggered less often. However, an optimum value of the threshold would
prevent excessive thread migration while addressing load imbalance only as
needed.
The NetWare 5 scheduler periodically calculates the system-wide load and the
mean load and uses the latter to compare loads and to determine whether an
individual processor is overloaded or underloaded.
The load balancing threshold and the calculated mean load are then used to
determine the high and low trigger loads.
Š A processor is overloaded when its load exceeds the high trigger load.
Š A processor is underloaded when it is below the low trigger load.
In this situation, the scheduler then moves threads from the overloaded
processor to the underloaded processor with the result that the loads are
balanced. See the following figure for an illustration of the relationship
between the mean load, the load balancing threshold, and the high and low
trigger loads.
Figure 1
Mean Load, Threshold, and High and Low Trigger Loads
High Trigger Load
(determined by Mean and Threshold)
Increasing
Processor
Workload
Threshold
= value
Mean
Low Trigger Load
(determined by Mean and Threshold)
Without the margin provided by the threshold, threads would constantly move
from one processor to another, thereby compromising the productivity of the
system.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
:$51,1*Although the threshold is configurable, we strongly recommend that you
retain the preset optimum value. If after careful consideration, you decide to modify the
threshold, test in an isolated system before modifying the value for a production
environment. If you modify the threshold, remember that you can always reset it at the
optimum value.
To modify the threshold, access MONITOR’s Available Options > Server
Parameters > Multiprocessor > System Threshold. For the procedure, see
“Setting the Load Balancing Threshold” on page 43.
Concurrency and Parallelism
NetWare® has always had threads (typically referred to as `processes’) but has
not exploited the potential for parallelism in multithreaded code.
Multithreading enables multiple paths of parallel execution through the code
path. A software developer identifies tasks that can be performed
concurrently, that are not dependent on being performed in a fixed sequence,
and provides the mechanisms for assigning tasks to multiple threads and for
appropriate synchronization to protect data shared by the threads.
In a uniprocessor environment, multithreaded code allows threads to run
concurrently. This means that one or more threads are active on the same
processor. The threads appear to run at the same time—although they do not
actually do so. One thread, for example, can be blocking while another thread
is executing code. They are perceived as executing simultaneously because
processors are very fast and time quanta are very small.
On the other hand, it is the availability of hardware systems with multiple
processors that makes it possible to have multiple threads actually running at
exactly the same time on different processors. When threads execute
simultaneously on multiple processors, they are running in parallel.
Multithreaded code allows more efficient processor utilization by exploiting
parallelism.
NetWare has supported concurrency but not parallelism—until now. With
NetWare 5, applications can be written to exploit the parallelism available in
multiprocessor (MP) hardware and the support for parallelism in the server
operating system, and your system will benefit from the performance gains
and scaling that server applications such as Oracle* Database Server and
GroupWise® provide.
127(A single binary supports both uniprocessor and multiprocessor systems
precisely because the NetWare 5 kernel is multithreaded.
Understanding
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Preemption
NetWare 5 allows for preemption of threads, within constraints. New NetWare
modules may be written to be preemptible.
127(Earlier versions of NetWare implemented a nonpreemptible round-robin (FIFO)
scheduling policy where threads were scheduled to run in the order that they entered
the run queue. On a uniprocessor system, NetWare is fast and very efficient.
For an application to exploit preemption, the code must be explicitly written
to be preemptible. Critical sections boundaries may be marked by calling
scheduler API functions that signal a critical code section. Critical sectioning
is used to keep program data in a consistent state and to prevent code that
doesn’t lend itself to concurrent execution from executing concurrently. A
thread cannot be preempted when it is in a critical section.
By default, if an application thread is running, it will not be preempted until
the following conditions are met.
Š The code where the thread is running must be preemptible. This is
indicated by a flag set in the module’s NLM file format. When the code
is loaded into memory, the memory pages are flagged as preemptible.
Š The thread cannot be in a critical section of the code.
Š The thread has run long enough to qualify for preemption. The scheduler
checks the elapsed time with every tick.
Support for preemption provides
Š An execution environment that allows simplified application
development. Developers can rely on the scheduler to handle preemption.
Š A mechanism that prevents ill-behaved modules from monopolizing the
processor.
The kernel itself is not preemptible.
Platform Support Module
Besides NetWare 5, all that is necessary to enable multiprocessing on a
multiprocessor computer is the Platform Support Module (PSM) for your
specific hardware platform and NetWare 5. No other modules are required.
127(Use the PSMs that come in the NetWare 5 red box and that load during
installation. The PSMs for NetWare 4.11 smp.nlm do not work for NetWare 5.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
The PSM is an interrupt abstraction for multiple processors. As a device driver
for the processors, it shields NetWare from hardware-dependent or platformspecific details. It enables secondary processors to be brought online and taken
offline.
During installation, NetWare 5 detects multiple processors by reading the MP
configuration table in BIOS and then determines which of the available
NetWare 5 Platform Support Modules (PSMs) matches the MP hardware
platform.
The network administrator then has the option to load the PSM or to run
NetWare on Processor 0 only. The installation program will modify the
startup.ncf file to load the PSM whenever the server is started.
In the NetWare 5 red box, Novell® provides mps14.psm which supports any
hardware platform that complies with the Intel* Multiprocessor Specification
1.1 and 1.4. Compaq* also provides a PSM for its system requirements.
Contact Tricord* for information on their PSM for NetWare 5.
In addition to scalability, NetWare 5 multiprocessing offers these benefits:
Š Backward compatibility for applications written for NetWare 4.11
smp.nlm. In fact, NetWare 5 supports everything written to CLIB in
previous releases. Older application code is simply funneled to processor
0.
Š An integrated multiprocessing kernel (MPK) that also supports
uniprocessing. One binary runs on both uniprocessor and multiprocessor
hardware platforms.
Š Kernel support for preemption. (Applications must be explicitly enabled
by their developers to take advantage of the kernel’s preemption
support.).
Š Platform support for MPK in a single platform support module.
Understanding
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
The Server Console
The server console includes the monitor and keyboard where you view and
control NetWare® server activity. From the server console you can
Š Shut down and restart the server
Š Load and unload NLM programs
Š Monitor server performance
Š View network traffic
Š Set configuration parameters
Š Edit configuration and other batch files
Š Send messages
You can use the Novell® remote console utility, rconsole.exe, to set up a
workstation as a server console. Or you can use the Java-based utility,
RConsoleJ. See both Using DOS-Based Remote Console and Using JavaBased Remote Console (RConsoleJ).
The server console also supports more than one interface:
Š Command line. Commands can be entered at the server console prompt,
with or without options and parameters. The prompt is indicated by a
colon (:). LOAD is an example of a server console command.
Š Character-based menuing. Utilities use menus, lists, and portals. The
key utility for monitoring the server—MONITOR—is an example of this
interface.
Š Java-based GUI . The new Servertop application ConsoleOne uses a
graphical user interface.
When NetWare 5 is installed, the server console defaults to the ConsoleOne
screen. However, you can toggle between screens when you need access to the
command line or if you choose to keep MONITOR loaded. For the procedure
for toggling between console screens, see “Accessing MONITOR” on page
66.
+,17For information on bypassing the server splash screen or replacing it with a
screen of your own choice, see SERVER in Utilities Reference.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Be aware of the following console security issues:
Š Unauthorized access of the server console. Control keyboard access by
requiring the use of a password. (See NetWare Server Security.)
Use the SECURE CONSOLE command to secure your console against
breaches of security. (See SECURE CONSOLE in Utilities Reference.)
You can also use the screen saver and console-locking features in
SCRSAVER. (See SCRSAVER in Utilities Reference.)
Š Software tampering. An expert could use the built-in debugger to
disable or bypass the security system. To prevent this, use SECURE
CONSOLE.
Š Hardware tampering. Keep your server in a secure location. An
intruder could disable the power-on password or remove hard disks to
access data.
Network Connections
NetWare® licenses client connections in large multiples, such as 100, 250, or
1000 licenses per server.
A network login passes through three connection phases until a license is
consumed:
Š Not Logged-In . Unlicensed—the connection does not count against
connection licenses—and is not authenticated to NDS®. This can be the
initial attachment of the client before authentication or licensing occurs
or it could also indicate users who have logged out, leaving a Not LoggedIn connection to the server on which a default drive was located.
Unlicensed connections are indicated by an asterisk (*) in MONITOR’s
Active Connection list.
Understanding
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Authenticated . Unlicensed and authenticated only to NDS. This
connection cannot—in its current state—access other resources on the
server. This type of connection occurs most often when users are
browsing the NDS tree to view objects located in an NDS replica held on
another server. This type of connection can also be from other NetWare
servers carrying out NDS synchronization or time synchronization
activities.
As unlicensed connections, they do not consume licenses, and as such,
they are indicated by an asterisk (*) in MONITOR’s Active Connection
list.
Š Authenticated/Licensed . Both authenticated and licensed. Also called a
Normal connection. The authenticated state comes about when the user
enters a valid username and password. From this information, NDS
establishes the user’s identity and authenticates the user via the security
methods found in its encryption algorithms. However, a user’s connection
doesn’t actually become authenticated and licensed until the user requests
access to the server’s file and print services. Typically, such requests are
initially performed through the container, profile, or user login script.
Once the user is authenticated during the initial login process,
background authentication establishes connections to any other servers
that are specified in the login scripts. An authenticated/licensed
connection consumes one of the connection licenses. Most connections
are of this type.
All connections described as licensed or unlicensed, authenticated or Not
Logged-In, are considered active connections and are tracked by the server.
See “Using MONITOR for Connection Management Tasks” on page 74.
NetWare 5 connection licenses are managed through Novell Licensing
Services. You can increase your NetWare 5 connections by purchasing
additional licenses and installing them using NLS Manager or Netware
Administrator. For more information, contact your Novell Authorized
Reseller.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
NetWare Is Self-Tuning
In many ways, NetWare® is a self-tuning system. It allocates resources
according to need and availability.
Typically, new resources are not immediately allocated when a request is
received. The operating system waits a specified amount of time to see if
existing resources become available to service the demand. If resources
become available, no new resources are allocated. If they do not become
available within the time limit, then new resources are allocated.
The time limit ensures that sudden, infrequent peaks of server activity do not
result in permanently allocating unneeded resources.
For example, when the server is started, all free memory is assigned to file
caching. However, as demand increases for other resources (such as directory
cache buffers), the number of available file cache buffers decreases.
The following parameters are dynamically configured by the operating
system:
Š Directory cache buffers
Š File locks
Š Kernel processes
Š Kernel semaphores
Š Load balancing for multiple processors
Š Maximum number of open files
Š Memory for NLMTM programs
Š Router/server advertising
Š Routing buffers
Š Service processes
Š TTS transactions
Š Turbo FAT index tables
In addition to the parameters configured by NetWare, you can adjust the value
of many server parameters to optimize the server for your network. See
“Setting Server Parameter Values” on page 44 or refer to MONITOR and SET
in Utilities Reference.
Understanding
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Java-Based Components of NetWare 5
NetWare 5 includes the following Java-based components:
Š “The Novell JVM for NetWare and the Just In Time Compiler” on page
24, along with the Java Naming Directory Interface (JNDI) and Java
Class Libraries (NJCL) for Novell, enable you to run Java-based applets
and applications on the server.
Š “The Novell Object Request Broker (ORB)” on page 25 enables you to
run CORBA*-compliant Java-based applications on the server.
The Java-based components are installed automatically when you install
NetWare 5. To use Java-based components, you must use TCP/IP. Instructions
for using Java-based components are found in “Managing” on page 51.
The following NetWare 5 components may also be of interest:
Š Java-based ConsoleOneTM server management utility. See Additional
Products > ConsoleOne or ConsoleOne in Utilities Reference. These are
located in the NetWare 5 online documentation.
Š NetWare Enterprise Web Server. See Contents > Web Services in the
NetWare 5 online documentation.
The Novell JVM for NetWare and the Just In Time Compiler
The Novell JVM for NetWare is comprised of several NLM programs that let
the NetWare server run Java-based applications and applets, including multithreaded applications and applications that use a graphical interface.
The JVM can run any application certified as 100% Pure Java, as described by
JavaSoft*. However, applications that are not 100% Pure Java must be ported
to NetWare.
NetWare 5 includes the Symantec* Just In Time Compiler (JIT) to improve the
performance of Java-based applications.
NetWare 5 also includes drivers for popular video cards and a utility that
automatically recognizes and configures your video card during NetWare 5
installation.
To run the Novell JVM, you need a minimum of 48 MB RAM and at least 64
MB to run graphical applications. The more memory your server has, the more
Java-based applications it can run.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
For information about Java Class Libraries for Novell and the Java Naming
Directory Interface (JNDI), download the Novell Developer Kit (NDK) from
the Novell DeveloperNet Web site (http://developer.novell.com/ndk/
netware.htm#java).
The Novell Object Request Broker (ORB)
The Common Object Request Broker Architecture (CORBA) is a
specification adopted by the Object Management Group (OMG), a consortium
of platform vendors, ISVs, and end users. The CORBA specification describes
a software bus, called an Object Request Broker (ORB), that provides an
infrastructure for distributed object computing.
NetWare 5 includes the server component of the Novell ORB, based on
VisiBroker for NetWare. The Novell ORB server component supports the
execution and management of distributed objects on the NetWare 5 server.
The Novell ORB requires the TCP/IP protocol and DNS naming services.
For information about loading the Novell ORB on the server, see “Using the
Novell ORB” on page 85.
To load and run CORBA-compliant applications on the network or to provide
a development environment for CORBA applications, you will need the
Novell Developers Kit (NDK).
The NDK contains the client component of the Novell ORB, development
tools, and online documentation for developing and administering VisiBroker
for NetWare applications.
You can download a copy of the NDK free of charge from the Novell
DeveloperNet Web site (http://www.developer.novell.com/ndk) or contact
your Novell Authorized Reseller.
Understanding
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
Rev 99a
28 October 99
doc_tpl.fm
Rev 99a
17
28 October 99
Setting Up
For information about installing a NetWare® server, including hardware and
software requirements and installation instructions, see NetWare 5.1
Installation Guide.
For information about migrating from IPXTM to TCP/IP, see Migrating IPX to
IP in NetWare Server Communications.
After installation, you may want to use these setup procedures:
Š “Changing the Server’s Software Configuration” on page 28
Š “Installing, Uninstalling, and Configuring a Server Product” on page 30
Š “Changing the Language NetWare Uses” on page 30
The server’s default language is English, but the server and the clients can
use different languages. For information about the various localized
components and procedures for changing them, see
“Code Pages” on page 31
“Changing Server Message Files” on page 33
“Changing NLM Message Files” on page 35
“Changing the Server Keyboard Type” on page 36
Setting Up
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Changing the Server’s Software Configuration
After installing NetWare®, you can make changes, including
Š “Copying Additional NetWare Files after Installation” on page 29
Š “Setting Server Parameter Values” on page 44
Š Changing time synchronization. See Network Time Management.
Š Adding or changing volumes. See Traditional File Services.
Š Modifying the Directory tree. See Directory Services.
Before changing the server’s configuration, view the current configuration by
entering the CONFIG command at the server console prompt:
CONFIG
The CONFIG command displays the following information:
Š Server name, internal network number, server uptime.
Š Hardware settings, node (station) addresses, external network number of
the cabling scheme, board name, frame type, and communication
protocol for each network board.
Š The server’s NDS® tree and bindery context.
Š On some systems, CONFIG might display information about NetWare
Loadable ModulesTM (NLMTM programs). The kind and amount of NLM
information depends on the system.
See CONFIG in Utilities Reference.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Copying Additional NetWare Files after Installation
If necessary, you can copy files from the NetWare CD-ROM or another server
after installation. Use the Copy Files Options of NWCONFIG. Follow this
procedure.
Prerequisites
‰ NetWare installation CD-ROM
Procedure
At the server console prompt, enter
[LOAD] NWCONFIG
Select Copy Files Options.
The system displays a box containing the default source path from which
the files will be copied.
Press Enter to accept the default path, or press F3 to specify a different
path.
If you press F3, the system displays a prompt with the current source path.
Backspace to delete the path and enter the new path.
Once you have specified the source path, the system displays the prompt
Specify a server boot path:. This is the destination directory—the location
to which the files will be copied. This directory must be the boot
directory.
Enter the path to the boot directory.
The system displays a list of file groups. An X next to each group means
the group is preselected.
Delete the X next to any group(s) you do not want to copy.
To delete the X, use the arrow keys to move the cursor to the desired
group and press Enter.
To reselect a group, move the cursor to the group and press Enter again.
When finished selecting groups, press F10.
The system copies the files.
See NWCONFIG in Utilities Reference.
Setting Up
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Installing, Uninstalling, and Configuring a Server
Product
The NWCONFIG utility allows you to install, uninstall, and configure other
products such as name space modules and NLMTM programs.
If you have questions during the installation or configuration of a product,
refer to the documentation that came with the product.
Some products might not have configuration or uninstall options. A message
explaining this is displayed if you try to perform unsupported operations.
At the server console prompt, enter
[LOAD] NWCONFIG
Select Product Options > View/Configure/Remove an Installed Product.
The list of Currently Installed Products appears. If no products are
currently installed, the list is empty.
Do one of the following:
Š
To install a product, press Insert and follow the prompts.
Š
To uninstall a product, select the product from the list and press
Delete.
Š
To set configuration options for a product, select the product from the
list and follow the prompts.
For more information, see NWCONFIG in Utilities Reference.
Changing the Language NetWare Uses
The default language for NetWare® is English, but other languages are
available. To change the language of your NetWare server and its NLMTM
programs, see the following sections:
Š “Code Pages” on page 31
Š “Changing Server Message Files” on page 33
Š “Changing NLM Message Files” on page 35
Š “Changing the Server Keyboard Type” on page 36
Š “Changing Code Pages” on page 36.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Code Pages
A code page is a table storing a character set supporting one or more language
scripts. When you press a key on a keyboard (a letter, symbol, or number), the
computer receives a numeric code that represents that keystroke. Code pages
store these numeric codes. Many personal computer operating systems
support multiple code pages and allow you to switch among them.
For example, DOS uses code page 437 for several languages that use Roman
alphabetical characters, including English, French, and German, but requires
code page 850 for Portuguese. DOS code page 850 (Portuguese) removes the
symbol for f (franc) and inserts an (O acute). Different computer operating
systems use different code pages for the same language. For example, DOS
uses code page 437 for English, but Windows 95 uses code page 1252.
In a single-byte code page, up to 256 codes are available to represent lower
and upper-case letters, numbers, punctuation marks, and all the mathematical
symbols on your keyboard.
However, 256 codes are not sufficient to represent all the letters and characters
used in the writing systems of every language. Some non-alphabetic writing
systems, such as Chinese, Japanese, and Korean, contain thousands of
characters and require a double-byte code page.
Differences between single-byte and double-byte code pages usually cause
display and readability problems. For example, a document created with
Windows 95 in Japan is probably created with code page 932. The same
document will not look the same when displayed on a Windows 95 computer
using code page 1252 in the United States . Unrecognized characters will be
replaced with a symbol such as a heart. In the past, these substituted characters
might cause a database such as NDSTM to fail to recognize objects.
To help resolve these problems, a convention, called Unicode*, has been
adopted.
Setting Up
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Using Unicode
Unicode is a 16-bit character representation, defined by the Unicode
Consortium, that supports up to 65,536 unique characters. Unicode allows the
characters for multiple languages to be represented using a single Unicode
representation.
Any character that your code page does not understand is substituted in your
display by the 4-digit hexadecimal value of the Unicode character, surrounded
by square brackets, for example: [00FF]
Because NDS supports Unicode, substituted characters do not prevent NDS
from recognizing an object. For example, your company’s European office
might create an Organizational Unit object to represent Finance in western
Europe. They might use DOS code page 852 to make the generic currency
symbol a part of the object name (OU=¤W-Euro).
When this object is accessed in the United States, using DOS code page 437
or Windows 95 code page 1252, the currency symbol (¤) is replaced by square
brackets surrounding the Unicode number for the currency symbol, [00A4].
NDS recognizes the Unicode number, so the object can still be opened and
accessed.
However, the object name (containing the square brackets and unicode
number) will be difficult for users to understand. If the name is too difficult to
interpret, the only solution is to determine which code page was used to create
the object, then view the object using that code page. Changing code pages can
be troublesome; see “Changing Code Pages” on page 36 for guidelines.
The following table shows ranges of Unicode numbers, with a description of
each range and a list of code pages that might be used to view the character
correctly. However, switching to one of the suggested code pages does not
guarantee that you will see the correct results. For example, characters in the
range 4E00-9FFF (Han Ideographs) are used in Japan, China, and Korea. But
switching to code page 932 (Japanese) does not display the character correctly
if the character is used only in China.
The most reliable way to determine the character is to refer to the Unicode
Standard, Version 2.0. Access the Unicode World Wide Web site for more
information: http://www.unicode.org. The web site also includes charts of
Unicode characters.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Table 1
Rev 99a
28 October 99
Unicode Ranges, Descriptions, and Code Pages
Unicode Range
Description
Geographical Region Windows Code
Pages
DOS Code Pages
0080 - 00FF
Extended Latin
Western Europe
437, 850,860, 863,
865
0100 - 01FF
Extended Latin
Central Europe
1250, 1257
852, 775
0300 - 03FF
Greek
Greece
1253
737
0400 - 04FF
Cyrillic
Russia
1251
855, 866
0590 - 05FF
Hebrew
Israel
1255
862
0600 - 06FF
Arabic
Middle East
1256
864
2500 - 26FF
Line Drawing and
Graphics
N/A
N/A
Most DOS code
pages
4E00 - 95FF
Han Ideographs
Far East
932, 936, 949, 950
932, 936, 949, 950
AC00 - D7FF
Hangul Syllables
Korea
949
949
FE70 - FEFF
Arabic
Presentation
Forms
Middle East
N/A
864
FF00 - FFEF
Full and Half Width
Variants
Far East
932, 936, 949, 950
932, 936, 949, 950
Changing Server Message Files
English is the default language for both the NetWare® operating system and
its NLMTM programs and utilities.
You specified a language for the server when you installed it. To change the
server language later, you must first copy new language files to the server, then
set the server to the new language and restart the server.
Prerequisites
‰ Installation CD-ROM or access to another NetWare server containing the
language files
Setting Up
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Procedure
127(This procedure requires you to restart the server because executing server.exe
sets the server language.
If you are copying the files from the NetWare installation CD-ROM,
insert the CD.
At the server console prompt, enter
[LOAD] NWCONFIG
Select Product Options > Choose an Item or Product Listed Above >
Install an Additional Server Language.
The screen displays a box containing the path from which you last
installed files. Make sure this path is to the language files. The language
files on the NetWare CD are located at:
Drive or volume:\NW50\INSTALL\current_language
If the path is correct, press Enter; if the path is incorrect, press F3 and type
in the correct path to the language files.
If you are installing the files from another server, type the path to that
server; the system might prompt you for a user name and password.
Follow the prompts.
The system displays a window listing groups of language files.
Use the arrow keys to move the cursor to the language you want to install
and press Enter.
The system marks the selected language with an X. You can specify
several languages, if desired.
To deselect a language, move the cursor to the language and press Enter
again.
When you have finished selecting languages, press F10.
The system copies the selected language files.
Press Esc until the Other Installation Items/Products menu is displayed.
Select Change Server Language.
The screen displays a box containing the path from which you last
installed files.
Press Enter to accept the path.
The system displays a menu listing the available languages.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Select the desired language from the menu.
If this is the first time you have changed the language, the system displays
a prompt specifying the default destination where the language files are
to be placed. The location is the default boot directory, c:\nwserver.
The language files must be placed in your boot directory.
Accept the default or type the path to your boot directory.
Once you specify the boot directory, the system loads the language files.
Restart the server for the change to take effect.
See NWCONFIG in Utilities Reference.
Changing NLM Message Files
Use the LANGUAGE command to change the language for subsequently
loaded NLMTM programs. The LANGUAGE command does not change the
language for currently loaded modules.
,03257$17If you change the NLM language without changing the server language,
the NLM still displays some character strings and screen titles in the server language.
For consistent language display, you should change the server language as explained
in the section, “Changing Server Message Files” on page 33.
At the server console prompt, enter
LANGUAGE LIST
The server displays a list of languages and their ID numbers. Not all
languages in the list are currently supported.
Note the name or number of the language you want to use.
At the server console prompt, enter
LANGUAGE Language_name |number
Any subsequently loaded NLM programs will be displayed in the new
language.
See LANGUAGE in Utilities Reference.
Setting Up
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Changing the Server Keyboard Type
NetWare® allows you to use keyboard types other than U.S. English by
loading the KEYB utility. The language you specify with KEYB must match
the language of your keyboard, not the server language. The keyboard type
can be changed without restarting the server.
,03257$17Changing the keyboard type causes some of the keys on your keyboard
to represent different characters. Do not use this command unless you have the
appropriate keyboard for the language you are specifying, and you are familiar with the
keyboard’s use. Otherwise, you might not know which keys to use to change the
keyboard back to its original language.
At the server console prompt, enter
[LOAD] KEYB
From the list of keyboard types, find the keyboard type that matches the
one you are using.
To change the keyboard type to match your keyboard, enter
[LOAD] KEYB <keyboard_type>
For example,
[LOAD] KEYB GERMANY
See KEYB in Utilities Reference.
Changing Code Pages
Changing code pages on the NetWare® server or on DOS can usually be done
while the computer is operating. However, changing code pages on a
Windows 3.1x or a Windows 95 workstation might require installing
additional software and restarting Windows. For background information
about Unicode and other code pages, see “Code Pages” on page 31.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
To change the code page on . . .
Do the following . . .
NetWare Server
At the server console prompt, type
28 October 99
CHARSET CPxxx
where xxx is the number of the code page you want
to load.
The following code pages are supported: 437, 737,
850, 852, 855, 857, 860, 861, 863, 865, 866, 869.
DOS
Follow instructions that come with the DOS
operating system.
Windows 3.1x, Windows 95, Windows NT
Follow instructions that come with the operating
system.
You might be required to install a new set of
Unicode translation tables, found on the NetWare
product CD. You will need the following:
Š UNI_newpage#.platform: Translates the new
code page to Unicode and Unicode to the new
code page
Š UNI_MON.platform: Converts characters to
lower case to facilitate comparisons
Š UNI_UPR.platform: Handles upper case
characters
Platform extensions include:
.W16 (Windows 3.1x)
.W32 (Windows 95, Windows NT)
.DOS (DOS)
See CHARSET in Utilities Reference.
Setting Up
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
Rev 99a
28 October 99
doc_tpl.fm
Rev 99a
17
28 October 99
Optimizing
Optimizing the NetWare® server requires staying up-to-date with Novell®
Technical Support at http://support.novell.com. (If you have an internet
connection, click on the Novell logo on the documentation browser to access
http://support.novell.com.)
Make sure that you install the latest patches, updated modules and drivers for
NetWare 5 and for third-party components. See “Applying Patches” on page
40 for information on where and how to obtain the most recent operating
system updates. Then patch your server before you optimize.
Optimizing the server also requires isolating and resolving bottlenecks.
(Bottlenecks are never really eliminated—they just move from one
component to another.) Because your system is constrained by its worstperforming components, be sure that your hardware components are well
matched for power and potential performance. Upgrade poorly performing
components as soon as possible. See “Matching Performance Capabilities” on
page 40.
Although NetWare® is largely self-tuning and provides excellent performance
off the shelf, some parameters can be tweaked to improve performance. When
your server is patched, the modules and drivers updated, and your hardware
components are matched and upgraded, see “Setting Server Parameter
Values” on page 44 for information on server defaults and settable parameters.
To identify where you can adjust the value of a server parameter and tune your
server for its particular processing load, see “Monitoring the NetWare Server”
on page 66.
Protocols can also create bottlenecks. See how the tips in “Minimizing
Replication Traffic” on page 50 and “Service Advertising” on page 50 can
help reduce network traffic.
Optimizing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Applying Patches
Novell® makes every attempt to solve problems in the NetWare® operating
system software by periodically releasing updates you can install on your
server. The latest operating system update contains all previous updates, so be
sure to install the most current file. It will keep your operating system current.
The latest operating system updates can be found at the following locations:
Š In the patch lists on Novell’s technical support World Wide Web site
http://support.novell.com. If you have an internet connection, click on the
Novell logo on the documentation browser to access http://
support.novell.com.
Š By File Transfer Protocol (FTP): Anonymous FTP to ftp.novell.com.
Š By Gopher: gopher.novell.com.
Š On the Novell Support EncyclopediaTM (NSEProTM): Released monthly
on CD-ROM and sent to subscribers.
Review the NetWare README file for information about installing NetWare
patches.
Matching Performance Capabilities
Each component’s performance characteristics can have an effect on the
system’s performance as a whole. Bottlenecks can be created by a great
number of subcomponents and component interactions. Protocols create
bottlenecks. Specific software functionality creates bottlenecks. In fact, every
part and process of a system is a potential bottleneck. However, it’s important
to keep in mind that only one bottleneck—the system’s weakest point—can
exist at one time.
Another point to remember is that all systems have a weak point somewhere.
Your goal is to match the performance capabilities of your system’s
components in such a way that no single component hampers the performance
of any other component. The following figure shows a system with matched
components.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Figure 2
Rev 99a
28 October 99
A Well-Designed System
Lower
Performance
Higher
Performance
a b c de
a,b,c,d,e represent the performance capabilities
of each system component
Note that all components in a well-designed system are within the same range
of performance. But in this case, a less than optimum component is cost
effective. The components are so closely matched that upgrading the lowerend component a would not increase performance enough to make it worth the
cost.
The following figure shows a poorly designed system with an inefficient mix
of low- and high-performance components.
Figure 3
A Poorly Designed System
Lower
Performance
f
Higher
Performance
g
hi j
f,g,h,i,j represent the performance capabilities
of each system component
In this system, components f and g have performance characteristics that are
considerably inferior to components h, i, and j. With component f as the
bottleneck, components h, i, j, and even g are constrained by f so that they run
much more slowly than their designs would otherwise permit. In this system,
the bottleneck is significantly hampering the system’s performance.
This is a poorly designed system because the high-performance components
aren’t able to perform anywhere near their full potential. In fact, because of
the high-performance characteristics of h, i, and j, this system may have cost
more than the system in Figure 3. But it cannot perform as well because of the
greater disparity between its low- and high-performance components.
Optimizing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
The unbalanced system in the following figure would benefit noticeably if you
were to upgrade component f to the performance level of g. But if you were to
upgrade f to a performance level beyond g—as shown in the following
figure—all additional performance characteristics in f will be lost because the
bottleneck moves to component g.
Figure 4
Moving the Bottleneck
Lower
Performance
f
Higher
Performance
g
hi j
Upgrade f (partial return on investment)
Thus you derive only a partial return on your investment in the upgrade. To
accomplish more than moving the bottleneck, both components must be
upgraded to the level of h before a full return on investment in h, i, and j can
be reached.
It is useless to test any single component of a system to its fullest capacity
without allowing for the possibility that another component is actually
bottlenecking system performance. You can upgrade system components, but
you won’t see any significant improvement until you isolate the system
bottleneck, assess its relationship to the performance characteristics of the
entire system, and take appropriate action.
Bottleneck isolation is critical if you’re building high-performance systems.
You need to know how and where your system is bottlenecking. Moreover,
you need to know how tightly that bottleneck is coupled with the performance
characteristics of the rest of the system. If the poorly performing component
is in a different performance category than the other components, you may
obtain a greater return on investment by making the necessary upgrades to
move the bottleneck to the highest possible performance level.
This discussion is taken from Performance Analysis: Isolating the Real
Bottleneck in a System, in the January 1996 AppNotesTM. See http://
developer.novell.com/research/for online publications.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Setting the Load Balancing Threshold
The load balancing threshold applies only to multiprocessor servers. Through
extensive testing, Novell has determined the optimum threshold for load
balancing across multiple processors—it addresses serious load imbalance
while avoiding excessive thread migration. Without that play in the load
balancing mechanism, threads would constantly move from one processor to
another.
The optimum threshold has been preset by Novell and, in most cases, it
provides the most efficient processor utilization. Although we strongly
recommend that you do not change the threshold, this value is configurable.
:$51,1*Adjusting this threshold should be done only with considerable research
and consultation. Experimenting with different threshold levels can best be done in an
isolated test system. Do not experiment in a production environment. Remember that
you can always return to the optimum setting.
Use this procedure to set the load balancing threshold.
In MONITOR’s Available Options, select Server Parameters >
Multiprocessor.
The multiprocessor parameters appear in the upper window.
Highlight the System Threshold field.
The pop-up window displays the current setting and the range of valid
values. The optimum value is 1536.
To change the threshold setting, press Enter and type a new value in the
field.
To exit MONITOR, press Alt + F10.
See MONITOR and SET Multiprocessor Parameters in Utilities Reference.
Optimizing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Setting Server Parameter Values
Default values for NetWare® server parameters provide maximum
performance for most systems. You should seldom need to modify these
values. But when you need to modify the default values, we recommend that
you read the topics listed here.
Š “Understanding Server Defaults” on page 45
Š “Accessing MONITOR” on page 66
Š “Finding Current Parameter Values from MONITOR” on page 46
Š “Printing Current Parameter Values to a File” on page 46
Š “Finding Current Parameter Values with SET” on page 47
Š “Checking Current Parameter Values with DISPLAY ENVIRONMENT”
on page 47
Š “When to Modify Defaults” on page 48
Š “When Not to Modify Defaults” on page 48
Š “Restoring Default Values with RESET ENVIRONMENT” on page 48
Š “Monitoring Dynamic Statistics” on page 49
Š “Using Optional Commands in Startup Files” on page 49
Š “Using SET Commands in Startup Files” on page 50
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Understanding Server Defaults
NetWare 5TM has been tuned for the specific duties of a network server. It has
been thoroughly tested with software tools designed to identify inefficiencies
so they can be eliminated. NetWare gives outstanding off-the-shelf
performance.
NetWare is also self-tuning: the following parameters are dynamically
configured by the operating system:
Packet receive buffers
File cache buffers
Directory cache buffers
Memory for NLMTM programs
Service processes
Kernel processes
Kernel semaphores
Load balancing for multiple processors
File locks
Maximum number of open files
Turbo FAT index tables
TTS transactions
Router/server advertising
To tune the server for its particular processing load, two utilities are available
to check or set server parameter values:
Š SET is a command line utility executed at the server console prompt. If
you are an experienced NetWare administrator, you might find that using
SET is faster for routine tuning as soon as you become familiar with the
names of the server parameters. See SET in Utilities Reference.
Š MONITOR is a character-based menuing utility, the key NetWare tool for
performance monitoring at the server level. MONITOR also allows you
to configure SET parameters from its Server Parameters option. This
means you can set server parameter values from within MONITOR
without memorizing parameter names and their valid values. See
MONITOR in Utilities Reference.
Parameter values are persistent and are stored in a hidden configuration file.
A NetWare 5 server retains any settings that have been modified from the
default values—even when it is taken down and then rebooted. The server will
come up again with all the settings you have made to tune the server.
Optimizing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Finding Current Parameter Values from MONITOR
Use this procedure to explore parameter settings in MONITOR. For more
information, see MONITOR in Utilities Reference.
At the server console prompt, enter
MONITOR
Select Available Options > Server Parameters.
Press F1 for descriptions of the server parameter categories.
Press Esc to return to the menu of parameter categories.
Select a parameter category.
The parameters are displayed in the upper window. As you highlight the
field for each parameter, a popup window displays a description of the
parameter, its current setting, and its range of valid values.
When you have finished, press Esc until you return to the Available
Options menu.
Printing Current Parameter Values to a File
Settings for NetWare 5 server parameters are now persistent, whether you set
them using MONITOR or whether you use the SET command. Nevertheless,
you should record your optimum server configuration as part of documenting
your network.
Use this procedure to write current values of settable parameters to the
sys:\system\setcmds.cp file.
At the server console prompt, enter
MONITOR
Select Available Options > Server Parameters.
Press F3 to write the current parameter values to sys:\system\setcmds.cp,
the default file for recording the value of settable parameters.
From a workstation, open the file with a text editor and print it.
Put the printout in the notebook that you use for documenting your
network.
For more information, see MONITOR in Utilities Reference.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Finding Current Parameter Values with SET
Use this procedure to explore parameter settings in SET.
At the server console prompt, enter
SET
A numbered list of server parameter categories is displayed.
Enter the number associated with one of the categories.
A list of parameter names is displayed. The name of each parameter
appears highlighted in white with the current setting highlighted in
yellow. The information also includes the limits (or range of valid values),
the .ncf file the parameter can be set in, and a brief description.
If necessary, press any key to display additional parameters in the same
category. Press Escape to exit the category.
Run SET again to select another category.
For a complete description of SET parameters and guidelines for their use, see
SET in Utilities Reference.
Checking Current Parameter Values with DISPLAY ENVIRONMENT
You can also check parameter values by using two new console commands:
DISPLAY ENVIRONMENT and DISPLAY MODIFIED ENVIRONMENT.
Š The DISPLAY ENVIRONMENT command displays the current search
paths and the current values of the settable server parameters. (Hidden
parameters are not displayed.)
At the server console prompt, enter
DISPLAY ENVIRONMENT
The name of each parameter appears highlighted in white. The current
value appears in yellow. The limits (or range of valid values) and the
default setting (if applicable) also appear.
Š In comparison, the DISPLAY MODIFIED ENVIRONMENT displays
only the server parameters that have values modified from the defaults.
At the server console prompt, enter
DISPLAY MODIFIED ENVIRONMENT
The display is similar to that of DISPLAY ENVIRONMENT. The
difference is that only modified parameters values are shown.
Optimizing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
For more information, see DISPLAY ENVIRONMENT and DISPLAY
MODIFIED ENVIRONMENT in Utilities Reference.
When to Modify Defaults
NetWare 5 defaults were chosen after thorough testing to provide both
excellent performance and system balance.
Before modifying defaults, use the MONITOR utility to track and log
performance statistics. See “Accessing MONITOR” on page 66. You will also
find the server log files useful in assessing the server’s performance. See
“Checking Server Error Logs” on page 87.
You can modify default values to
Š Compensate temporarily for a poorly performing subsystem, allowing the
server to remain in service.
Š Resolve inefficiencies.
Š Optimize performance and tune the server for its processing load.
For example, if your server has a slow hard disk or insufficient RAM, you can
compensate temporarily by modifying defaults. The long-term solution is to
replace the disk or to add RAM. See “Matching Performance Capabilities” on
page 40.
When Not to Modify Defaults
If you were to modify some defaults, you could sacrifice the system’s balance
and performance.
NetWare 5 defaults have been chosen after thorough testing. We recommend,
for example, that you do not modify the default settings of the Load Balancing
Threshold or the Transaction Tracking parameters.
Restoring Default Values with RESET ENVIRONMENT
If you want to restore the default values of the settable server parameters, you
can use the RESET ENVIRONMENT command.
When you run RESET ENVIRONMENT , you are prompted at each changed
server parameter value.
At the server console prompt, enter
RESET ENVIRONMENT
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
When the first parameter with modified values appears, you can choose to
reset all values to their defaults, or just reset the value of the parameter that is
displayed. Or you can also skip to the next modified parameter.
For more information, see RESET ENVIRONMENT in Utilities Reference.
Monitoring Dynamic Statistics
MONITOR reports many useful statistics and updates most of them every
second with the return values from direct calls to the operating system.
However, not all statistics are of equal value to the network administrator. The
key dynamic statistics are displayed in MONITOR’s General Information
screen, which appears as soon as MONITOR is accessed. For a description of
these statistics, see “What to Monitor and Why” on page 67.
ConsoleOne also allows you to monitor server performance. See Procedures
> Setting Up > ConsoleOne for information on accessing the servertop.
Using Optional Commands in Startup Files
During initialization, NetWare executes two server boot files that are created
by the install process and that reside in the System directory:
Š startup.ncf . This file executes first and loads the server’s disk drivers,
name spaces, and some SET parameters.
Š autoexec.ncf . This file stores the server name and IPXTM internal
network number, loads the LAN drivers and settings for the network
boards, binds the protocol to the installed drivers, and loads other NLMTM
programs.
Š
,03257$17Even if you use IP, the internal network number stored in this file
serves as a unique server ID. IP uses this number to provide passive IPX support
called Compatibility Mode.
An .ncf file executes all the commands within it; the commands are not
optional. However, you can make a command optional by placing a question
mark (?) in front of it. For the procedure, see “Making a Command Optional
in a Server Batch (.ncf) File” on page 61.
Optimizing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Using SET Commands in Startup Files
Remember that some SET commands can be executed only from the
startup.ncf file. To find out whether a SET command must be executed in
startup.ncf, refer to the explanations of the SET command parameters in SET
in Utilities Reference.
When you install the server operating system, the installation process adds
SET commands to the autoexec.ncf file—commands that control time zone
and daylight savings time information, bindery context, and time server type.
You can add more SET commands to the autoexec.ncf file, if desired. Use
NWCONFIG to modify server batch files. When the server is rebooted, any
parameter values set by commands in the autoexec.ncf file override previous
values for those parameters.
Minimizing Replication Traffic
Careful choice and placement of tree replicas can minimize replication traffic
on the network.
Š Make sure that the replica containing the user object is on the server
nearest the user.
Š Do not create more partition replicas than you need. The more replicas,
the more traffic required to synchronize them all.
For more information about optimizing NDS®, see the online documentation:
Contents > Network Services Documentation > Directory Services > NDS
Database Management > Working with Replicas.
Service Advertising
For IPX internetworks, NetWare Link Services Protocol (NLSP) is a faster
way to advertise network services than Service Advertising Protocol (SAP).
Replace SAP/RIP segments of your network with NLSP segments. For more
information about NLSP, see Novell Guide to NLSP Migration.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
17
28 October 99
Managing
Managing the NetWare® server operating system involves such tasks as
Š “Starting and Stopping the Server” on page 52
Explains how to start the server, how to bring down the server through an
orderly shutdown procedure that prevents data corruption, and how to
restart the server after a brief shutdown.
Š “Starting and Stopping Processors” on page 55
Explains how to start and stop individual processors in a multiprocessing
server. Only secondary processors can be handled this way.
Š “Using Server Batch Files” on page 57
Explains how to use .ncf batch files to load modules, execute console
commands, and configure subsystems.
Š “Using Server Utilities and Console Commands” on page 63
Explains how to access help for server utilities and console commands,
discusses types of NLM programs, and provides instructions for loading
and unloading an NLM.
Š “Monitoring the NetWare Server” on page 66
Explains why and what to monitor and how to use the MONITOR utility.
Š “Managing Workstation Connections” on page 73
Explains how to use utilities and commands for disabling login, clearing
connections, controlling resource allocation, sending messages, and
working with watchdog packets.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š “Using Java on the Server” on page 83
Explains how to set environment variables for Java, load and unload
java.nlm, run Java-based applets and applications, and load and unload
the Novell ORB.
Š “Common Maintenance Tasks” on page 87
Explains checking error logs, drives, cache statistics, free space, user
accounts, the UPS, and includes documenting the network.
Š “Preventive Maintenance Tasks” on page 91
Explains how to prevent memory problems and power supply errors.
Š “Using Accounting” on page 93
Explains how to enable accounting on the server, set up charge rates, and
view accounting statistics.
Starting and Stopping the Server
For information and procedures on starting the server, bringing it down using
an orderly shutdown, and restarting the server without exiting to DOS, see
Š “Starting the Server” on page 52
Š “Bringing Down the Server” on page 53
Starting the Server
To start the server, change to the startup directory (usually C:\nwserver) and
enter the following at the DOS prompt:
SERVER
+,17You may also place the command to execute server.exe in the DOS
autoexec.bat file. Then, whenever the server computer is powered up, the server starts
automatically.
When you execute server.exe, the Novell® logo is displayed. To prevent the
Novell® logo (server splash screen) from displaying, you can load SERVER
with the -nl (no logo) command option.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
If the splash screen is active and you want to see a list of modules as they load,
you can still display the system console screen by using the procedure below
to toggle between server console screens.
Press Alt + Esc, tapping Esc until the system console screen is displayed.
A list of modules scrolls up the screen. If you have a color monitor, the
color in which each module name is displayed indicates how the module
is loaded.
Cyan (light blue) indicates a module is loaded by server.exe
Red indicates the module is loaded from the startup directory
White indicates a module loaded from the autoexec.ncf file
Purple indicates a module is loaded by other modules
After the server is running, you can display a list of all loaded modules by
entering the following at the server console prompt:
MODULES
The module names are displayed in colors that represent the functional
group they belong to, as described above.
For more information, see SERVER and MODULES in Utilities Reference.
Bringing Down the Server
To ensure data integrity, use the DOWN command at the server console before
turning off power to the server. DOWN writes all cache buffers to disk, closes
all files, updates the appropriate Directory Entry Tables and File Allocation
Tables, and exits the network.
If you want to bring the server down and then restart the server immediately,
without exiting to DOS, use the RESTART SERVER command instead of
DOWN.
127(Bringing down or restarting a server during a remote console session
terminates the session.
Use MONITOR to inform all users that they should close all files and log
out.
D From Available Options, select Connections.
E Highlight each connection you want to inform and press F5 to mark
it.
F Press F4 to display a Message Text window.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
G Type the message and press Enter to send the message to the
designated connections.
For example, you might send a message like this:
Please close files and log out from <servername>
Going down in 5 minutes
Execute one of the following commands at the server console prompt:
D To bring down the server and exit to the local operating system, enter
DOWN
You can now shut off power to the system, if necessary.
E To reboot the server without exiting to DOS, enter
RESTART SERVER
Although most modules, including those in protected address spaces, unload
automatically when you down or restart the server, some modules might
require that you unload them individually first. You can unload them manually
with the UNLOAD command. Or you can create a batch file called
shutdown.ncf and place the unload commands in it. If you create
shutdown.ncf, it is executed automatically whenever you down or restart the
server.
For more information about creating batch files, see “Using EDIT to Create
and Edit Text and Batch Files” on page 58. For more information about the
UNLOAD command, see UNLOAD in Utilities Reference. Also see DOWN
and RESTART SERVER in Utilities Reference.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Starting and Stopping Processors
When the Platform Support Module (PSM) is loaded, NetWare 5 can start the
secondary processors from the SET Auto Start Processors = ON command in
the startup.ncf file.
+,17Server console commands for processors affect only secondary processors.
Processor 0 cannot be taken offline while the server is running.
You can start or stop secondary processors at any time while the server is
running by using the console commands, START PROCESSORS and STOP
PROCESSORS. (If you choose to start secondary processors manually, you
can change the value to SET Auto Start Processors = OFF in the startup.ncf
file.) When any secondary processor is stopped, the associated threads are
automatically switched to another processor.
The procedures are
Š “Starting Processors” on page 55
Š “Displaying Processor Information” on page 56
Š “Stopping Processors” on page 56
Starting Processors
To bring a specific secondary processor online, enter at the console prompt:
START PROCESSORS n
To specify two or more processors to be brought online (but not all secondary
processors), enter at the console prompt:
START PROCESSORS n n
To bring all secondary processors online, enter at the console prompt:
START PROCESSORS
For more information, see START PROCESSORS in Utilities Reference.
If you choose to start secondary processors automatically when the server
boots, make sure the startup.ncf file contains the command, SET Auto Start
Processors = ON.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Displaying Processor Information
To determine which processors are currently online, run the DISPLAY
PROCESSORS command.
You can also obtain generic processor information from the console
command, CPUCHECK. Per-processor statistics are also available from
MONITOR’s Available Options > Kernel > Processors.
Displaying Processor Status
Before stopping a secondary processor, determine whether it is online by
entering at the console prompt:
DISPLAY PROCESSORS
Displaying Processor Information
To determine processor speed and other generic information, enter at the
console prompt:
CPUCHECK
For example, with a single processor, the output for CPUCHECK might be
similar to the following:
Processor Speed 199 MHZ, CPUID 52C:1BF
Family 5, Model 2, Stepping C
For more information, see DISPLAY PROCESSORS, CPUCHECK, and
MONITOR in Utilities Reference.
Stopping Processors
Processor 0 is the boot processor and cannot be taken offline.
If your server platform has multiple processors, use DISPLAY
PROCESSORS to identify which processors are currently online and to obtain
their ID numbers. See “Displaying Processor Information” on page 56.
To take a secondary processor offline, specify the processor number when you
enter the command at the console prompt:
STOP PROCESSORS n
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
To specify more than one of the secondary processors to take offline, enter at
the console prompt:
STOP PROCESSORS n n
To take all secondary processors offline, enter at the console prompt:
STOP PROCESSORS When a secondary processor is taken offline, the
threads running on that processor are switched to another processor
automatically.
For more information, see DISPLAY PROCESSORS and STOP
PROCESSORS in Utilities Reference.
Using Server Batch Files
Information on server batch files and procedures includes the following:
Š “Using EDIT to Create and Edit Text and Batch Files” on page 58
Š “Using NWCONFIG to Create and Edit Startup Batch Files” on page 60
Š “Making a Command Optional in a Server Batch (.ncf) File” on page 61
Š “Changing Server Name and IPX Internal Address in Autoexec.ncf” on
page 62
A server batch file is an executable file containing server console commands.
You execute the file by entering its name at the server console prompt.
NetWare then executes all the commands in the file in order.
By default the NetWare installation process creates two batch files that are
always executed whenever the server is booted:
Š startup.ncf —This file is executed first. It loads the server’s disk drivers
and name spaces, and can execute certain server parameters that cannot
be executed anywhere else.
Š autoexec.ncf —This file is executed after startup.ncf. It sets time
synchronization parameters and the bindery context, stores the server
name, loads the communication protocol, loads the LAN drivers and
settings for the network boards, binds the protocol to the installed drivers,
loads other NLM programs, and executes server parameters.
In addition, you can create a batch file called shutdown.ncf to execute
commands whenever you down the server. The DOWN command executes
shutdown.ncf automatically whenever the server is downed.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
You can create any other batch files you need. Simply create a text file with a
.ncf extension. Place NetWare console commands into the batch file in the
order in which you want them executed.
Using EDIT to Create and Edit Text and Batch Files
127(Each time you save the file, you can increase the file size by up to 4 KB.
At the server console prompt, enter
[LOAD] EDIT [path]
If you specify a path, include both the directory and file name. The file
must be located on a mounted volume or on a local drive of the server.
If you do not specify a path, the system displays a prompt at which you
can either enter the filename or display a browse window.
Š
To enter the filename, type either the complete file path (volume,
directory, and file name) or just the filename, and press Enter.
If you enter only the filename, the system assumes a default location
of sys:system.
If the file does not exist, EDIT displays a prompt that asks whether
the file should be created. Choose Yes to create the new file or No to
redisplay the file name prompt. If you choose Yes, EDIT displays a
blank screen so you can enter text into the file.
Š
To display the browse window, press Insert.
To see the contents of a volume or directory, highlight the volume or
directory name and press Enter. Use the arrow keys to scroll up and
down the list of files and directories. To display a file, highlight the
file name and press Enter.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Create or edit the file as needed.
The keys used to edit text are described in the following table.
Table 2
Editing Keys Available in the EDIT NLM
Key
Function
Arrow keys
Move up/down one line or left/right one character.
Backspace
Delete character to left of cursor.
<Ctrl>+<PageUp>
Move to beginning of document.
<Ctrl>+<PageDown>
Move to end of document.
<Delete>
Delete the character at the cursor location or delete
blocked text.
<End>
Move to end of line.
<Esc>
Exit the file. You are prompted to save any changes.
<F5>
Highlight (block) text for copying or deleting.
Toggles highlighting on and off.
<F6>
Copy highlighted (blocked) text to a buffer.
<Home>
Move cursor to beginning of line.
<Insert>
Insert text from buffer.
<PageUp>
Move up one page
<PageDown>
Move down one page.
<Tab>
Move cursor four spaces to the right.
When finished, press Esc. Select Yes to save the file or No to exit the file
without saving changes.
The system redisplays the path name prompt.
To edit or create another file, enter the file path name. To redisplay the
console prompt, press Esc.
See DOWN in Utilities Reference.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Using NWCONFIG to Create and Edit Startup Batch Files
The following procedure explains how to use NWCONFIG to create or edit
the startup.ncf and autoexec.ncf files.
At the server console prompt, enter
[LOAD] NWCONFIG
Choose NCF Files Options.
The system displays a menu containing options to create or edit both the
startup.ncf and autoexec.ncf files and to convert a NetWare 3.1x
autoexec.ncf file to NetWare 5.
Highlight the desired option and press Enter.
If you select the option to upgrade the NetWare 3.1x file, follow the
screen prompts. NWCONFIG compares the file with the current NetWare
configuration and makes changes to the file as necessary. You can edit the
file before saving it.
If you select an option to edit one of the files, the selected file appears in
a window.
If you select an option to create one of the files, the system displays the
existing file, if there is one, in one window and a new file with default
entries in another window, so you can compare the two.
The default entries include only essential contents of the file and reflect
the current system setup. For example, the default entries for autoexec.ncf
might include time zone information and load and bind commands for
currently installed LAN drivers, but would not include custom
configuration settings you have in the original file.
The cursor is active in the new file. Press Tab to toggle between the two
files.
Edit the new or existing file as necessary.
Editing and navigation keystrokes are listed at the bottom of the screen.
For additional help, press F1.
When finished, save the file by pressing F10 and selecting Yes when
prompted. To exit without saving, press Alt + F10.
,03257$17Each time you edit the autoexec.ncf file or the startup.ncf file,
NetWare saves the previous version as autoncf.old or startncf.old.
127(Changes to the file take effect once you reboot the server.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
See NWCONFIG in Utilities Reference.
Making a Command Optional in a Server Batch (.ncf) File
By default, an .ncf file executes all the commands within it. However, to make
a command optional, you can place a question mark (?) in front of it.
The question mark causes the .ncf file to pause at the command and prompt
you to execute it. You can respond either Yes or No. You can also specify a
default response.
For example, the following line in the autoexec.ncf file causes the file to pause
and ask whether to execute the LOAD MONITOR command:
? LOAD MONITOR
When the autoexec.ncf file is executed, it displays the following prompt:
LOAD MONITOR? y
The default response is Yes. If you do not respond to the prompt, the default
response is executed after 10 seconds.
Both the default response and the time period are configurable.
Setting the Default Response
There are two ways to configure the default response to the command prompt:
Š Specify the default on the command line in the .ncf file. Place the default
response (a Y or N) immediately following the question mark. For
example,
?N LOAD MONITOR
In this example, the default response is N. If the user does not respond to
the prompt, MONITOR is not loaded.
Š Set the Command Line Prompt Default Choice parameter to either ON or
OFF. ON means the default response is Y; OFF means the default
response is N. The parameter is in MONITOR > Available Options >
Server Parameters > Miscellaneous.
The parameter setting is overridden by a default specified on the
command line.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Setting the Time Period
To configure the time period before the default response is executed, set the
Command Line Prompt Time Out parameter to a time in seconds. The
parameter is in MONITOR > Available Options > Server Parameters >
Miscellaneous. The default value is 10 seconds.
Changing Server Name and IPX Internal Address in Autoexec.ncf
The server’s name is stored in the autoexec.ncf file. You can change the server
name by editing the file. However, changing the server name in the
autoexec.ncf file does not affect NDS® server objects. You must use NetWare
Administrator to rename NetWare Server objects.
,03257$17Be sure to use NetWare Administrator to update any references to the
former server name in the Directory tree and in the timesync.cfg file.
The server’s IPX internal network number is also stored in the autoexec.ncf
file. Even if you use Novell’s Native IP protocol, the internal network number
is used as a server ID and will still be stored in the autoexec.ncf file. Do not
delete the number.
To change either the server name or the IPX internal network number in the
autoexec.ncf file, use NWCONFIG or EDIT at the server console, and then
reboot the server.
To use NWCONFIG to change the settings, see “Using NWCONFIG to Create
and Edit Startup Batch Files” on page 60. To use EDIT, see “Using EDIT to
Create and Edit Text and Batch Files” on page 58.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Using Server Utilities and Console Commands
Server utilities are NetWare Loadable ModulesTM (NLMTM programs). Some
NLM programs are loaded whenever you start the server, some are loaded by
other NLM programs when they are needed, and some must be loaded
manually.
Server console commands are executed by entering the command at the server
console prompt.
+,17When you enter a command at the server console prompt, the command is
stored in the command queue. To repeat a command, press the Up-arrow key until the
desired command is displayed. You can make changes to the command or press Enter
to execute the command. Up to 16 commands can be saved in the queue.
For more information, see the following topics:
Š “How to Access Help for Utilities and Console Commands” on page 63
Š “Accessing MONITOR” on page 66
Š “Using NetWare Loadable Modules” on page 64
Š “Loading and Unloading NetWare Loadable Modules” on page 65
How to Access Help for Utilities and Console Commands
Š For a complete list of all utilities and console commands, with
descriptions, syntax, and examples, see Utilities Reference.
Š To display a list of console commands on the server console, enter the
following at the server console prompt:
HELP
Š To see online help for a console command, enter either of the following
at the server console prompt:
HELP command_name
or
HELP ALL
Š To see online help for an NLM that displays a menu, such as MONITOR,
load the NLM, then press F1. You can also press F1 at subsequent menus,
lists, and information windows.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Using NetWare Loadable Modules
When loaded, an NLM program is dynamically linked to the operating system,
and the NetWare server allocates a portion of memory to it.
The amount of memory an NLM program uses depends on the task. Some
tasks make calls that cause the operating system to allocate more memory. The
NLM uses the memory to perform a task, and then returns control of the
memory to the operating system when the NLM is unloaded. When an NLM
is unloaded, all allocated resources are returned to the operating system.
NetWare has five types of NLM programs:
Š NWPA drivers (.cdm and .ham extensions) control communication
between the operating system and host adapters.
Š LAN drivers (.lan extension) control communication between the
operating system and the network boards.
You can load and unload LAN drivers while the server is running and
while users are logged in.
Š PSM modules (.psm extension) provide support for multiple processors
in server class computers.
Š Management utilities and server applications modules (.nlm
extension) allow you to monitor and change configuration options.
For example, you can run VREPAIR on a dismounted volume or use
MONITOR to view server statistics.
Once you finish your tasks, you can unload the utility and free memory
for other server functions.
Š Name space support (.nam extension) allows non-DOS naming
conventions to be stored in the directory and file naming system.
Some NLM programs, such as utilities, can be loaded, used, and then
unloaded. Other NLM programs, such as LAN driver and disk driver NLMs,
must be loaded every time the server is started.
NCF files (startup.ncf and autoexec.ncf) allow you to store commands that
load NLM programs every time the NetWare server is started.
Most NLM programs included with NetWare are copied to sys:system during
installation. As you acquire additional NLM programs, decide where you
want to copy them. The operating system must be able to find the NLM
programs when a LOAD command is issued.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
NLM programs can be copied to any of the following areas:
Š Sys:system directory
Š Any network directory on the NetWare server
Š A local drive of the NetWare server
For more information about using NLM programs, see LOAD, UNLOAD,
and SEARCH in Utilities Reference.
Loading and Unloading NetWare Loadable Modules
To load a NetWare Loadable Module (NLM), enter the following at the server
console prompt:
[LOAD] [path]NLM_name
Š If the NLM is stored in the sys:system directory or if you have a server
search path to the directory where the NLM resides, enter just the NLM
name (instead of the complete path). For information about server search
paths, see SEARCH in Utilities Reference.
Š If you want the NLM to be loaded whenever the server boots, add the
command to autoexec.ncf.
Š If you want the NLM to be displayed in a language other than English, see
“Changing NLM Message Files” on page 35.
127(In most cases you can load an NLM by entering just the name of the NLM at the
console prompt . The exception is if there is a .ncf file of the same name as the NLM. In
this case, you must use the LOAD command. Using the LOAD command tells the server
to load the NLM instead of executing the file of the same name. For information about
the LOAD command, see LOAD in Utilities Reference.
To unload a NetWare Loadable Module (NLM), enter the following at the
server console prompt:
UNLOAD NLM_name
To unload an NLM from a protected address space, enter the following at the
server console prompt:
UNLOAD ADDRESS SPACE = space_name module_name
For more information about loading and unloading modules, see LOAD and
UNLOAD in Utilities Reference. For more information about unloading
modules from protected address spaces, see Unloading Protected Address
Spaces.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Monitoring the NetWare Server
Monitoring the server allows you to
Š Isolate performance bottlenecks
Š Optimize server performance
Š Troubleshoot and resolve server problems
MONITOR is the key NetWare® tool for performance monitoring at the server
level. It runs in the NetWare kernel so that it can make calls directly to the
operating system. Most statistics are updated every second.
You can also use the Java-based ConsoleOne for monitoring. See Contents >
Server Operating System > ConsoleOne.
Accessing MONITOR
Š To load MONITOR at the console prompt, enter
[LOAD] [path] MONITOR
The LOAD command is necessary only if you have a batch file with the
same name as the NLM. (Using the LOAD command indicates that
MONITOR is a loadable module.)
The path is specified only if you moved the utility from its default
directory sys:system to a directory that does not have a search path.
Š If MONITOR is running but is not the active screen at the console, there
are two ways to toggle to the MONITOR screen:
Press Alt + Esc. To cycle through other console screens, continue to hold
Alt down and tap Esc until the current MONITOR screen reappears. Use
the bar at the top of the screen for navigation.
Or press Ctrl + Esc. A numbered list of available console screens appears.
Enter the number for the NetWare 5 Console Monitor.
Š To exit MONITOR, use either of the following procedures:
Repeatedly press Esc until the Exit confirmation box appears. Then press
Enter to return to the console prompt.
Or to bring up the Exit confirmation box immediately, press Alt + F10.
Then press Enter to return to the console prompt.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
What to Monitor and Why
A monitoring utilitycan be overwhelming in the volume of information and
statistics it provides. To simplify your access to key parameters,
“MONITOR’s General Information” on page 67 includes most of them on one
screen, including performance indicators that pertain to memory utilization
and disk I/O.
See “Monitoring Allocated Services” on page 70 for an explanation of the
interaction of three types of configurable parameters that control dynamically
allocated services. The guidelines allow you to be proactive in maintaining the
level of service your users require.
“Other Information and Statistics to Monitor” on page 71 includes error logs,
available free disk space, and users accounts. To be prepared for power supply
interruptions, test your UPS (uninterruptable power system) periodically.
Becoming familiar with your server’s day-to-day performance and its
characteristic response to its unique load will help you to better interpret the
information and statistics the monitoring utilities provide.
MONITOR’s General Information
Most of the important performance indicators that can be used to track,
diagnose, and resolve server problems appear in the Figure 5 on page 68. This
window is displayed when you load MONITOR.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Figure 5
Rev 99a
28 October 99
General Information Screen in MONITOR
The onscreen performance indicators are described following.
Š Utilization. Average of the server’s total processing capacity that was
used during the last second, expressed as a percentage. The remainder of
the time is spent in an idle loop. Monitor the patterns of utilization.
Does utilization go to 100 percent? Does it reach a plateau or does it peak
for a few seconds and then drop? What time and how frequently does it
peak?
Š Server Up Time. This is the amount of time that has elapsed since the
server was most recently started. If your server has Auto Restart enabled,
use this field to determine whether your server has abended and restarted.
You can also use this information to detect power failures or intruders
who bring down the server.
Š Online Processors. The number of enabled and online processors. The
Platform Support Module (PSM), loaded from startup.ncf, enables
NetWare to use secondary processors in a multiprocessing platform. The
default is to autostart all processors.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Total Cache Buffers. This number shows how many cache buffers are
currently available for file caching. This number decreases as you load
NLM programs.
Because file caching has a dramatic impact on server performance, you
want this number to be as high as possible.
Š Dirty Cache Buffers. This number shows how many buffers contain
data which has been changed but not yet written to disk.
Monitor this number to see whether it is increasing. If it is, you might
have a bottleneck in the file system.
Š Long Term Cache Hits. This number shows the cumulative percentage
of requests for disk blocks that were already in cache. (When the
requested data is already in memory, disk reads don’t need to be made.)
Use this percentage to assess overall disk cache utilization. If this value
falls below 90%, the disk thrashes and server performance degrades. The
solution is to install more physical memory (RAM).
For another MONITOR statistic that is useful in assessing RAM, see
Understanding LRU Sitting Time.
Š Current Disk Requests. This number shows how many read requests
are pending.
Monitor this number to see whether it is increasing. If it is, a slow hard
disk might be your bottleneck.
Š Packet Receive Buffers. This number shows how much memory was
allocated for holding client requests while the server processes them. The
server allocates more packet receive buffers as needed (within minimum
and maximum values). Buffer size depends on the network adapter.
This number is the result of the interaction of three configurable
parameters. Monitor the current value to ensure that it does not reach the
maximum value.
Š Maximum Service Processes. This number shows how many task
handlers that the server can allocate to service client requests. As the
number of client requests increases, the server creates more service
processes. When no more service processes can be allocated, server
performance is adversely affected.
Monitor the current service processes to ensure that it does not reach this
value.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Current Service Processes. This number shows how many task handlers
are currently being used to service client requests.
Use this number to determine whether the server has enough free task
handlers to service client requests. If the number of current service
processes reaches the maximum, the value of the Maximum Service
Processes parameter must be increased. (See MONITOR’s Available
Options > Server Parameters > Communications Parameters.)
Š Current Connections. The number of active connections, whether
licensed or unlicensed, authenticated or Not Logged-In.
Š Open Files. The number of files currently being accessed by the server
and by other clients. Certain files are always open, such as the hidden files
that support NDS.
Monitoring Allocated Services
There are limits to resources in every computing environment. NetWare
allocates resources according to need and availability. When a request is
received, the operating system doesn’t immediately allocate new resources. It
waits a specified amount of time to see if existing resources become available
to service the demand.
Dynamically allocated services are controlled by the interaction of three types
of settable parameters. If you understand how they work together, you can
interpret the statistics that you monitor and know how to configure the
system’s response to demand. The three types of parameters that interact to
control allocation are
Š Maximum limits control the amount of server resources the operating
system can allocate for a particular service.
Š Minimum limits allow the operating system to allocate a minimum
amount of resources as soon as a request is received.
Low minimum limits will slow the growth of a particular service. High
minimum limits allow rapid growth.
Š Wait time limits control how rapidly the operating system can allocate
a new resource.
If resources become available, no new resources are allocated. If they
don’t become available within the time limit, new resources are allocated.
The time limit ensures that sudden, infrequent peaks of server activity
don’t permanently allocate unneeded resources.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Consider, for example, the allocation of directory cache buffers. If the
minimum number of directory cache buffers is set to 20, the system allocates
another buffer resource as soon as a request is made—until 20 cache buffers
have been allocated.
When 20 directory cache buffers are allocated, the system waits 2.2 seconds
(default) when a request comes in, and then allocates another buffer if the
request is still active.
However, if the minimum number of directory cache buffers is set to 40, the
system allocates 40 directory cache buffers before it starts slowing the growth
by waiting 2.2 seconds after each request.
Other Information and Statistics to Monitor
Common maintenance tasks also include monitoring:
Š “Checking Server Error Logs” on page 87.
Š “Checking Disk Drives” on page 87.
Š “Reviewing Server Cache Statistics” on page 88.
Š “Checking Free Space on Server Disks and Volumes” on page 88.
Š “Reviewing User Accounts” on page 88.
Š “Testing the Uninterruptible Power Supply” on page 89.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Displaying Server Alerts
You can enable the server to display alert messages under various conditions,
such as when NLM programs fail to relinquish control of a processor. Many
alerts are controlled by server parameters that set a threshold when the alert
will be generated.
To hear a bell whenever an alert is received, set the following server parameter
(MONITOR > Available Options > Server Parameters > Miscellaneous):
Sound Bell for Alerts = On
Available alert parameters are described in the following table.
Condition that Generates Alert
Relevant Parameters
Spurious interrupts are detected
Display Spurious Interrupt Alerts
Spurious Interrupt Alert Threshold
(See SET Miscellaneous Parameters in Utilities
Reference.)
Lost interrupts are detected
Display Lost Interrupt Alerts
Lost Interrupt Alert Threshold
(See SET Miscellaneous Parameters in Utilities
Reference.)
NLMs fail to relinquish control of processor
Display Relinquish Control Alerts
CPU Hog Timeout Amount
(See SET Miscellaneous Parameters in Utilities
Reference.)
Incomplete IPX packets are received
Display Incomplete Packet Alerts
(See SET Miscellaneous Parameters in Utilities
Reference.)
Available file cache buffers are nearing minimum
Minimum File Cache Report Threshold
(See the SET File Caching Parameters in Utilities
Reference.)
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
Condition that Generates Alert
Relevant Parameters
A volume is becoming full
Volume Low Warn All Users
28 October 99
Volume Low Warning Threshold
Volume Low Warning Reset Threshold
(See the SET File System Parameters in Utilities
Reference.)
Many other alerts can be generated by the system, but are not controlled by
server parameters. For example, if the address of a server on the network is
changed, an alert will be broadcast on other server consoles. You can control
these alerts with the ALERT command. See ALERT in Utilities Reference.
Managing Workstation Connections
You can perform management tasks for workstation connections to the
NetWare® server from the Connections option in MONITOR’s Available
Options menu. Console commands that serve the same purpose are also listed
for some tasks.
Š “Using MONITOR for Connection Management Tasks” on page 74
Š “Monitoring Workstation Connections to the Server” on page 74
Š “Sending Console Messages to Workstations” on page 77
Š “Clearing a Workstation Connection” on page 77
Š “Disabling and Enabling Logins” on page 78
Š “Controlling Resource Allocation with Locks” on page 80
Š “Configuring Watchdog Packets” on page 82
Š “Enabling Console Message when Watchdog Logout Occurs” on page 82
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Using MONITOR for Connection Management Tasks
The tasks for which you can use MONITOR include
Š View a list of active workstation connections on the server
Š Determine which users are logged in
Š View information about a connection
Š Configure a watchdog to monitor connections for activity
Š View a list of files that the connection has open
Š Terminate one or more workstation connections
You can also use CLEAR STATION n or CLEAR STATION ALL at the
console prompt. See CLEAR STATION in Utilities Reference.
Š Send a message to one or more workstations
Monitoring Workstation Connections to the Server
The following table contains tasks and procedures.
Table 3
Monitoring Server Connections
To
Do
View a list of active server connections
1. Select Available Options > Connections. The list
of Active Connections appears.
Determine which users are logged in
2. (Optional) Press F3 to re-sort the connection list.
List items without an asterisk (*) represent loggedin users.
List items with an asterisk (*) are unlicensed
connections (this means they do not count against
a server’s total licenses). These connections are
authenticated to NDS but are not logged in.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
To
Do
View information about a connection
1. Select Available Options > Connections to
display the list of Active Connections.
Š Connection number
Š Connection name
2. Highlight a connection from the list.
Š Elapsed connection time
3. Press Tab to activate the Connection Information
screen in the upper window.
Š Network node address
4. For an explanation of the information fields and
statistics, press F1.
Š Read/write requests
Š Authentication and license status
Š Files in use
Š Locks and semaphores in use
Š Supervisor rights
Configure a watchdog to monitor connections for
activity
See “Configuring Watchdog Packets” on page 82
and “Enabling Console Message when Watchdog
Logout Occurs” on page 82.
View a list of files that the connection has open.
1. Select Available Options > Connections.
Use this procedure when you want to determine
whether a crashed workstation has open files
remaining on the server.
2. Select the connection from the list.
3. After determining whether files are open on the
server, press Esc to return to the Active
Connections list.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
To
Do
Terminate one or more connections
1. Select Available Options > Connections.
Use this procedure to clear the connection when a
workstation has crashed and left open files on the
server.
2. Highlight the connection (or for multiple
connections, highlight each one and press F5 to
mark it).
Note: Transaction monitoring (TTS) provides
automatic backout for database records or files
stored on the server if a workstation has crashed
and left those files open on the server. Also a file is
protected if the Transaction (T) file attribute is set
for it.
3. Press Delete.
4. At the Clear Connection? prompt, select Yes.
Deleted connections are no longer listed.
Press F6 to clear all unlicensed connections,
indicated in the connection list by an asterisk (*).
These connections are authenticated but are not
logged in and therefore do not count against a
server’s total licenses.
Note: If you know the connection number, you can
also terminate a connection at the server console
prompt with CLEAR STATION n. Use CLEAR
STATION ALL to terminate all connections. See
CLEAR STATION in Utilities Reference.
Send a message to one or more workstations
1. Select Available Options > Connections.
Use this procedure when you need to warn users to
log out before you down the server.
2. Highlight the connection (for multiple
connections, highlight each connection and press
F5 to mark it and then press F4 to confirm your
selections).
3. In the text-entry box that appears, enter the
message.
4. At the confirmation prompt, select Yes to send
the message or select No to redisplay the text-entry
window.
Note: In the Active Connections list, items without
an asterisk (*) represent logged-in users.
For an explanation of NDS connection types, see “Network Connections” on
page 21.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Sending Console Messages to Workstations
To send a message from the server console to all workstations or to a single
user, access MONITOR.
Choose MONITOR’s Available Options > Connections.
Press F8 to display more options on the task bar.
On the list of Active Connections, highlight the connection for the user
you want to send a message to. Or use F5 to mark additional recipients.
To display the message composition box, press F4.
Type your message in the message box and press Enter to send it.
A confirmation box appears.
You can either choose to send the message immediately or you can press
Esc to discard it.
For more information, see MONITOR in Utilities Reference. For information
about receiving and clearing messages at the workstation, see the
documentation for your NetWare client.
Clearing a Workstation Connection
You can use either MONITOR or CLEAR STATION to clear a client
connection when the workstation has crashed and left open files on the server.
127(If you clear the connection while the workstation is in the middle of a transaction
or a file update, the files might be saved with incorrect data.
Using MONITOR to Clear the Connection
Select Available Options > Connections.
A list of active connections appears.
(Optional) To view a workstation’s open files, select the workstation’s
connection.
After viewing files, press Esc to return to the Active Connections list.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Clear one or more connections, as follows:
Š
To clear one connection, highlight the connection and press Delete.
At the Clear Connection? prompt, select Yes.
Š
To delete multiple connections, highlight each connection and press
F5 to mark it. Then press Delete. At the Clear Connection? prompt,
select Yes.
Š
To clear all unused (not logged in) connections, press F6.
Deleted connections are no longer listed on the Active Connections list.
Using CLEAR STATION to Clear the Connection
To determine the workstation’s connection number, from MONITOR’s
Available Options, select Connections.
Highlight the appropriate connection in the Active Connections list.
The connection information, which includes the connection number,
appears in the upper window.
Note the connection number.
Press Alt + Esc and then tap Esc until you cycle to the server console
prompt.
At the server console prompt, enter either
CLEAR STATION n
or
CLEAR STATION ALL
Replace n with the connection number of the workstation, or use the ALL
parameter to clear all connections.
For more information, see CLEAR STATION or MONITOR in Utilities
Reference.
Disabling and Enabling Logins
If you need to make repairs to the server, you can prevent users from logging
in by entering the following at the server console prompt:
DISABLE LOGIN
This does not affect users who are already logged in to the network. However,
a user who logs out cannot log in again until you execute ENABLE LOGIN.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
To allow users to log in again, enter the following at the server console
prompt:
ENABLE LOGIN
You can also use ENABLE LOGIN to enable a SUPERVISOR account when
it has been locked by the intruder detection function.
See DISABLE LOGIN and ENABLE LOGIN in Utilities Reference.
Clearing a Connection that Uses Auto Reconnect
Client 32 has automatic reconnection capabilities that make possible a
complete restoration of the user environment after a network failure. Because
auto-reconnect is enabled by default, you need to prevent workstations from
reconnecting while you are trying to perform some function on the server.
Client 32 also supports file caching on the client side, which allows files to be
cached as they are read from the server and as they are written to the server.
With the default settings for the Auto Reconnect Level and the File Cache
Level parameters, Client 32 will restore all files to their actual state before the
workstation connection was cleared (or the server went down). The
workstation monitor displays a message:
Client32: Please wait while Application retries
request to ServerName. To stop waiting, press Enter.
NetWare will then return an error and try to
reconnect in the background.
You can combine the use of console commands to prevent clients from
reconnecting for long enough to complete your work on the server.
Š To warn users, see “Sending Console Messages to Workstations” on page
77.
Š To prevent auto-reconnect users from reattaching to the server while you
are working on it, enter DISABLE LOGIN at the console prompt before
you clear the connections. See “Disabling and Enabling Logins” on page
78.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Then use CLEAR STATION to terminate the connection as in “Clearing
a Workstation Connection” on page 77. With login disabled, a
workstation with auto-reconnect enabled will still try to reestablish its
connection automatically, but the necessary services on the login server
won’t be available. (After the connections are cleared, workstation
attachments appear as *NOT-LOGGED-IN entries in MONITOR’s
Active Connections window.)
Š When you have completed your server tasks, enter ENABLE LOGIN to
allow users to log in again.
+,17Note that if you don’t clear a workstation’s connection and a communication
failure occurs (such as a broken or unplugged network cable), the workstation will
be able to auto-reconnect when the problem is fixed.
For a more complete discussion, refer to Disconnecting NetWare Client that
Have Automatic Reconnection Enable, in the February 1997 AppNotesTM
Controlling Resource Allocation with Locks
A lock prevents a file or record from being updated by more than one user at
a time. By controlling the number of file and record locks available to a
workstation or a server, you control access to files and records.
You can increase the number of locks if clients cannot open files. Or you can
limit the number of locks to prevent overuse of file resources.
Preventing Workstations from Overusing File Resources
If you are running out of file resources, use MONITOR to limit the number of
locks for the server and for each connection.
Before limiting file and record locks, make sure the workstations have an
adequate number of locks to access the files and records they need. Limiting
needed locks can cause applications to generate errors.
From MONITOR’s Available Options, select Server Parameters > Locks.
127(You can also set locks parameter values with SET. See SET Locks
Parameters in Utilities Reference.
To edit a parameter value, highlight it and enter the new value.
For the server, limit the total number of file and record locks allowed by
decreasing the values of the following parameters:
Maximum Record Locks
Maximum File Locks
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
The changes to the parameter values take effect immediately and are
persistent.
For each client, limit the total number of file and record locks allowed by
decreasing the value of the following parameters:
Maximum Record Locks Per Connection
Maximum File Locks Per Connection
Press Esc until you return to Available Options.
Increasing File Record Locks
Use MONITOR to provide more file record locks if applications fail either
because clients cannot open enough files or because not enough record locks
are available.
From MONITOR’s Available Options, select Server Parameters > Locks.
127(You can also set locks parameters with SET. See SET Locks Parameters
in Utilities Reference.
To edit a parameter value, highlight it and enter the new value.
Increase the values of the following parameters to increase the number of
file and record locks allowed for the server:
Maximum Record Locks
Maximum File Locks
The changes to the parameter values take effect immediately and are
persistent.
Increase the values of the following parameters to increase the number of
file and record locks allowed for each workstation:
Maximum Record Locks Per Connection
Maximum File Locks Per Connection
The changes to the parameter values take effect immediately and are
persistent.
Press Esc until you return to Available Options.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Configuring Watchdog Packets
The server sends watchdog packets to a workstation to determine if the
workstation connection is still active. If there is no response from the
workstation within a set amount of time, the server terminates the connection.
To provide an early warning that a connection has failed, set the values of the
watchdog parameters so that the server sends twice the default number of
packets in the shortest amount of time allowed.
If you enable an alert, a console message displays when the server terminates
a connection. For this procedure, see “Enabling Console Message when
Watchdog Logout Occurs” on page 82.
From MONITOR’s Available Options, select Server Parameters >
Communications.
Change the following watchdog parameters to the following values:
Number of Watchdog Packets=20
Delay Between Watchdog Packets=10
Delay Before First Watchdog Packet=16
Console Display Watchdog Logouts=ON
Press Esc until you return to Available Options.
Enabling Console Message when Watchdog Logout Occurs
If you enable an alert, a console message displays when the server terminates
a connection.
From MONITOR’s Available Options, select Server Parameters >
Communications.
Change the following parameter to the indicated value by highlighting the
parameter and pressing Enter:
Console Display Watchdog Logout=ON
The changes to the parameter values take effect immediately and are
persistent.
Press Esc until you return to Available Options.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Using Java on the Server
Most of the information you need about JVM for NetWare can be located at
VisiBroker for NetWare documentation site (http://
www.developer.novell.com/ndk/doc_orb.htm).
See the following sections for information about using and managing Javabased components on the NetWare 5 server:
Š “Environment Variables for Java” on page 83
Š “Using the Novell ORB” on page 85
Environment Variables for Java
The Novell JVM for NetWare is automatically installed when you install
NetWare 5. The installation process sets default values for key environment
variables. Setting environment variables for Java is necessary to run Javabased applications and applets.
The installation process places some of the variables into the sys:\etc\java.cfg
file. The java.cfg file is read whenever you load Java. You can modify or add
to the java.cfg file.
The following table describes some key environment variables. Other key
environment variables for Java are available in the Novell Developer Kit
(NDK) doc.
Environment Variable
Description
OSA_HOME
Specifies the location of libraries. The default value
is sys:\java.
JAVA_HOME
Specifies the location of the JVM.
XLOCALEDIR
Specifies the location of files needed to support
other languages, including keyboard and font files.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Viewing and Setting Variable Values
To view current variable values and to set values, follow these steps:
To load java.nlm at the server console prompt, enter:
java
To view current environment variables and variable values, enter the
following at the server console prompt:
envset
To set a value for the current session, enter the envset command as
follows:
envset variable=value;value
Make sure there are no spaces on either side of the equal sign or between
values. To set multiple values for variables such as CLASSPATH,
separate the values with semicolons.
The value of a variable set with envset lasts until Java is unloaded.
To append a value to the CLASSPATH variable, use $CLASSPATH to
represent the existing values, for example:
envset CLASSPATH=$CLASSPATH;sys:myclasses
To make lasting changes to variable values, enter the envset command
line in the sys:\etc\java.cfg file. The java.cfg file is read whenever Java is
loaded.
To edit the file, use EDIT at the server console prompt, or use a word
processing program at a workstation.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Using the Novell ORB
The following ORB runtime components are installed automatically when you
install NetWare® 5:
Š ORB .jar files
The .jar files make up the Novell ORB.
Š orbcmd.nlm
The ORBCMD makes ORB commands available at the server console.
Once ORBCMD is loaded, the syntax for issuing ORB commands at the
server console is identical to the VisiBroker command syntax. See the
VisiBroker for NetWare online documentation included in the Novell
Developer Kit (NDK) available at the VisiBroker for NetWare
Documentation site (http://www.developer.novell.com/ndk/
doc_orb.htm).
Š osagent.nlm
The OSAGENT is a directory of CORBA objects and services. It enables
the ORB to find objects and services when requested by a client.
,03257$17To run CORBA applications on the network or to provide a development
environment for CORBA applications, you will need the client component of the Novell
ORB, development tools, and the VisiBroker online documentation for developing and
administering VisiBroker applications.
These components and tools are available on the Novell Developer Kit (NDK). You can
download a copy of the NDK free of charge from the Novell DeveloperNet Web site
(http://www.developer.novell.com) or contact your Novell Authorized Reseller.
The following procedure explains how to load and unload the server component of the
Novell ORB and make ORB commands available in the NetWare environment when
you install and run a distributed application.
Prerequisites
‰ 64 MB RAM on the server (the more memory your server has, the more
CORBA applications it can run).
‰ 85 MB available disk space in the Sys volume.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
‰ TCP/IP installed and configured, DNS running, and a sys:\etc\hosts file
on the server containing an entry for host name and IP address. If you did
not select TCP/IP and enable DNS during NetWare 5 installation, you
must do so before trying to run CORBA applications. See NetWare
Server Communications and Novell DNS/DHCP Services
Administrator’s Guide.
‰ The paths to the following Java archive files in the CLASSPATH
environment variable: sys:\java\lib\vbjapp.jar, sys:\java\lib\vbjorb.jar,
and sys:\java\lib\vbjtools.jar. See “Viewing and Setting Variable Values”
on page 84.
Procedure
Install the files for a CORBA application in the same volume where the
Novell ORB resides.
Enter the application pathnames into the CLASSPATH environment
variable.
See “Viewing and Setting Variable Values” on page 84.
Load ORBCMD at the server console prompt.
orbcmd
ORBCMD automatically loads other NLM programs, including the
Novell JVM, and enables the use of CORBA commands such as
osagent at the server console. You must load orbcmd.nlm before you
run any other ORB-based application clients, servers, or tools from the
NetWare server console.
Start the osagent at the server console prompt:
osagent
The osagent command starts the Smart Agent, which provides Novell
ORB’s directory service and failure detection services.
Install client components and set up the distributed environment.
Refer to the NDK, available free of charge from the Novell DevelopNet
Web site (http://www.developer.novell.com/ndk).
To unload the Novell ORB, enter the following at the server console
prompt:
unload osagent
unload orbcmd
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Common Maintenance Tasks
Backing Up
Maintaining a daily backup schedule is crucial to the security of your data.
For information about the backup software included with NetWare, see
Backup and Restore.
Checking Server Error Logs
Check server error logs at least once a week, even if you don’t believe there
have been errors. Problems caught early are easier to solve. View the error
logs with EDIT at the server console or with a text editor from a workstation.
See EDIT in Utilities Reference.
When needed, NetWare creates the following error logs, the first four in
sys:system:
Š boot$log.err for errors generated during the process of starting the server
Š sys$log.err for server errors, including system messages, alerts, and
security violations
Š vol$log.err for volume errors
Š tts$log.err for data backed out by the Transaction Tracking System
Š abend.log for information generated by the abend process. This file is
moved to sys:system as soon as the server is started after the abend.
To keep error log files from using too much disk space, modify the values of
the relevant Error Handling server parameters. See SET Error Handling
Parameters in Utilities Reference.
Checking Disk Drives
Once a week check the Storage Devices statistics in MONITOR to make sure
that the operating status of each disk is active.
If you rarely use the diskette drive on your server, you might want to test the
drive about twice a year to make sure it’s running properly.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Reviewing Server Cache Statistics
For more information about monitoring and optimizing server memory, see
Optimizing in Server Memory.
Review file server cache statistics weekly.
These statistics are found on the General Information screen of
MONITOR.
Divide the value of Total Cache Buffers by the value of Original Cache
Buffers to determine the percentage of available cache.
Ideally, file server cache should be 70% or more of the total cache.
If the percentage falls below 50%, add more memory to the server.
Checking Free Space on Server Disks and Volumes
Check server disks and volumes weekly to make sure they have enough
free space.
To determine the amount of free space, load MONITOR and select
Volumes. The Mounted Volumes screen lists all mounted volumes, their
capacity, and the percentage of the volume that is full.
Keep at least 15% to 20% free space on any disk.
Keep at least 20% free space on the Sys volume and at least 10% free
space within other volumes. If the Sys volume fills up, your server hangs.
If the free space is less than the recommended values, increase the size of
the volume or remove unneeded files and directories.
You can use the VREPAIR utility to repair volumes. See VREPAIR in Utilities
Reference.
Reviewing User Accounts
Review all users and objects on the network once a month to make sure
that there are no intruders, obsolete accounts, or unauthorized accounts.
The network administrator should know all objects and accounts on the
network.
Make sure accounts are closed for all users who have left the system.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Testing the Uninterruptible Power Supply
If your network has an Uninterruptible Power Supply (UPS), check it twice a
year to make sure it’s running properly. Choose a time when no users are
logged into the network.
Down the server and exit to DOS, but do not turn off the computer,
While the computer is still running, unplug it.
If your UPS system is working properly, the computer should continue to
run.
If the server powers off, your UPS batteries are probably low or dead.
Check and replace them.
Documenting the Network
It is important to document your network. Keeping records of network layout,
hardware and software inventory, configuration, repairs, and backup
schedules will save you time and work if you need to rebuild or replace parts
of the network, justify new equipment, or restore the network after a disaster.
You will need much of this information if you ever need to call Novell
Technical Support. You might want to keep documentation in a three-ring
binder or online in a database.
:$51,1*Whether you store your network documentation in a binder or in a
database, keep multiple copies and update them regularly. Always store one or more
copies offsite—along with your taped backups. Documentation does no good if it’s
destroyed with the network itself in a major disaster.
Information you should keep in your network documentation includes the
following:
Š Inventory of all hardware and software.
Version numbers, serial numbers, warranty information, date purchased,
license information, etc.
Š Configuration information for workstations .
Copies of autoexec.bat, net.cfg (for DOS and Windows 3.1x clients) or
the registry (for Windows 95 and Windows NT).
Any standard directory structures.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Configuration information for server.
Copies of the autoexec.ncf, startup.ncf, and autoexec.bat files
Contents of the DOS partition
Record of hardware interrupts, port addresses, frame types and other
settings (type CONFIG at the console prompt to see the settings)
Services provided by each server, such as file and print, databases, etc.
General directory structure of each volume.
Settings of all server parameters, especially if you have modified the
defaults to optimize your servers (use MONITOR to copy all the
parameter settings to a file. See “Printing Current Parameter Values to a
File” on page 46 . )
Š Configuration information for printers.
For queue-based printers: map of printers, print queues, and print servers,
the version of PSERVER, and the dates and versions of print drivers.
Š NDSTM tree structure.
Print out the NDS structure from NetWare Administrator. Document
replicas and partitions.
Š Procedures.
Login processes and backup procedures. Most important, procedures for
restoring the system from backups.
Š Backup information.
Rotation schedule of backup media, labeling system, names and phone
numbers of backup operators, and locations of backup disks and tapes.
Š User information.
Login scripts, special menus.
Š Software and hardware repairs and service calls.
Repair records and receipts, service calls log, solutions in case the same
problem occurs again.
Š Diagram of network and cable plant .
Locations of servers, workstations, printers, routers, hubs, etc.; name of
each element; the kinds of cables and connections.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Changes to the system.
Record any updates to the network or its components.
Preventive Maintenance Tasks
The following sections provide tips for avoiding common server problems.
Preventing Memory Problems
Š Make sure you have adequate memory installed on the system
Š Unload unneeded TSR (terminate-and-stay-resident) programs (DOS
clients only).
Š Modify the net.cfg file to load only necessary drivers.
Š Modify the config.sys file to load only necessary drivers.
Š Check the autoexec.bat and config.sys files to make sure no memory
managers, such as himem.sys, or emm386.exe, are being loaded, and that
there is no DOS=HIGH statement in either file. Make sure no DOS device
drivers are being loaded.
For information about freeing up memory on a server, see “Resolving Server
Memory Problems” on page 131.
Preventing Power Supply Errors
An inconsistent power source is the most common cause of hardware
problems. It also produces the most devastating results.
Power outages cause workstations, network servers, print servers, and backup
devices to reboot. When this happens, all information stored in RAM is lost,
and sometimes hardware is damaged.
Power spikes and brownouts can also cause a variety of hardware errors.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
You can have reliable network performance only if you plan for power outages
and fluctuations and protect against them. The following tips can help you do
this:
Š Add a dedicated power feed and ground line from your breaker box to
critical equipment. Make sure the ground line connects to earth ground.
Š Install an uninterruptible power supply (UPS) or a standby power system
(SPS) to provide power to critical equipment for 15 minutes after a power
outage.
The capacity of such power supplies is limited, so you may not want to
plug nonvital hardware (such as monitors or printers) into the UPS or SPS
line.
Š Train users to save data and log out of the network when the lights go out.
Then you can bring down the NetWare® server in an orderly fashion
without forcing users off the system while the server is running on UPS
or SPS power.
Š Install a surge suppressor or power conditioner on all power lines that are
used by computers. Many UPS and SPS devices already have this feature.
Š Do not allow anyone to plug fans, printers, copy machines, vacuum
cleaners, or other motor-driven appliances into the dedicated line or into
any power line with computers.
Insert dummy plugs into open outlets to prevent people from plugging
such appliances into computer power sources.
Preventing Static Electricity Problems
An electrostatic discharge (ESD) must equal about 3000 volts before you can
feel it, but sensitive electronic components such as microchips and circuit
boards can be damaged by ESDs of as little as 20 or 30 volts. These small
discharges do not cause a component to fail immediately, but can cause the
component to degrade over time and fail at a later date.
Novell recommends you take the following precautions to protect electronic
components from electrostatic discharge:
Š Ground yourself and your equipment with a proper wrist strap and mat
before working on computers and computer boards. Test grounds daily to
make sure they have not become loose or intermittent.
:$51,1*DO NOT use a wrist strap when working on monitors! They can carry
a large voltage that can reach you through the strap.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Never touch components or integrated chips by their electrical leads.
Š Do not allow anyone to touch you when you are working on boards that
contain integrated circuits. They may cause a static charge.
Š Transport and store boards and integrated circuits in static-shielding bags,
usually grey-silver in color. The bags must be in perfect condition
because tiny pin holes will defeat their purpose. Anti-static bags (blue or
pink) do not shield their contents from external static fields and should
not be used.
Š Keep nonconductors, such as polyester clothing and plastic, away from
open computers and components. Nonconductors are a source of static
charges.
Š Never place components on any conductive surface, such as metal.
Š Keep the humidity of any area where you have open computers at 70 to
90 percent. Static electricity problems are more likely to occur in low
humidity.
Using Accounting
Accounting allows you to monitor ®NetWare server usage and charge users
for server resources.
Š “Setting Up Accounting” on page 93
Š “Calculating Charge Rates for Accounting” on page 94
Š “Viewing Accounting Totals” on page 95
Setting Up Accounting
The following procedure explains how to set up accounting:
Prerequisites
‰ A workstation running Windows 3.1x or Windows 95, and NetWare
Administrator.
‰ 6 MB of available memory on the Windows workstation.
‰ Supervisor object right to the Server object where you are setting up
accounting.
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Procedure
Start NetWare Administrator.
NetWare Administrator is found in sys:\public or sys:\public\win95.
Select the Server object you want to set up accounting for.
The Object menu appears.
Select Details > Accounting.
Confirm that you want to install accounting on this server.
Accounting page buttons appear on the right side of the dialog box with
the other page buttons.
Use the accounting buttons to edit the accounting pages.
Click OK.
Calculating Charge Rates for Accounting
Charge rates for using server resources are specified as multipliers and
divisors. To set the charge amount to No Charge, enter 0 as the multiplier. Use
the following formula to calculate a charge rate:
CHARGE
(charge rate multiplier)
=
CHARGE RATE
ESTIMATED USAGE
(charge rate divider)
For example, if you want to charge $200 per month for server disk storage, and
there are usually 100,000 blocks used each month on your hard drive:
Š 20,000 would be the multiplier (number of cents in $200)
Š 100,000 would be the divisor (disk block usage per month)
Š Therefore, users would be charged two-tenths of one cent per disk block
used.
20,000 cents
=
1 cent / 5 blocks read
100,000 blocks read
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Viewing Accounting Totals
The ATOTAL utility provides a total for the accounting services used on your
network. ATOTAL compiles information from the system accounting records
and lists the following:
Š Total connect time in minutes
Š Total service requests
Š Total blocks read
Š Total blocks written
Š Total disk storage in blocks per day
‰ A workstation running DOS 3.30 or later, Windows 95.
‰ Accounting set up on the NetWare server. (See “Setting Up Accounting”
on page 93.)
‰ Supervisor object right to the Server object you want to see accounting
information for.
Change to the sys:system directory.
To view daily and weekly totals for accounting services, enter
ATOTAL
To redirect ATOTAL data to a file, enter
ATOTAL > filename
Managing
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
Rev 99a
28 October 99
doc_tpl.fm
Rev 99a
17
28 October 99
Troubleshooting
See “A Troubleshooting Methodology” on page 98 for the steps to follow
when you troubleshoot any server problem.
See “Preventive Maintenance Tasks” on page 154 for links to information that
will help you prevent network problems.
See “Creating a Core Dump” on page 105 for information on creating full core
dumps and cacheless core dumps.
The following symptoms indicate common error conditions. Each section
includes procedures to troubleshoot the problems.
Š “Slow Network, Performance Problems” on page 107
Š “Insufficient Packet Receive Buffers, No ECB Available Count Errors”
on page 109
Š “High Utilization Statistics in MONITOR” on page 109
Š “Suspected Multiprocessing Problems” on page 116
Š “Unable to Communicate with Server, -625 Errors” on page 117
Š “Transaction Tracking Errors, -621 Errors, FD93 Errors” on page 120
Š “Workstations Can’t Communicate with Server” on page 122
Š “Admin Account Has Been Deleted or Admin Password Is Unknown” on
page 127
Š “Memory Error Messages, Memory Problems” on page 127
Š “Disk Errors” on page 132
Š “Server Console Hangs” on page 137
Š “Server Abends” on page 139
Troubleshooting
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
A Troubleshooting Methodology
Novell’s Technical Support Representatives recommend that you follow the
same basic steps in all troubleshooting efforts. Following a standard procedure
helps you solve technical problems faster and in an orderly way.
The following tasks are the ones that a Novell Support Representative would
ask you to perform if you called Technical Support. By following these steps
on your own, you may be able to resolve problems without assistance. If you
do need to call Technical Support, you will already have the information
prepared to help support representatives solve your technical problems more
quickly.
Š “Keep Your Servers Patched” on page 98
Š “Eliminate Obvious Causes” on page 99
Š “Gather Information” on page 100
Š “Simplify the System” on page 101
Š “Use Research Tools” on page 102
Š “Develop Hypotheses and Perform Tests” on page 103
Š “Look for Additional Help” on page 104
Š “Document the Solution” on page 105
Keep Your Servers Patched
Many problems you might experience with your server will have already been
solved by Novell and provided in the form of a patch file. It is essential that
you load all server and client patches. See “Applying Patches” on page 40.
Use the File Finder at http://support.novell.com to locate patches for
server.nlm and updates for other modules. Copy updated modules to the
server’s startup directory where they will overwrite the previous versions.
In addition, be sure you use up-to-date drivers for NetWare 5. Use drivers that
ship in the Novell red box, drivers that are tested, certified, and made available
on Novell’s Support Connection (http://support.novell.com), or certified
drivers released by hardware vendors. Check product certification bulletins on
DeveloperNet (http://developer.novell.com) to learn which versions are most
recent.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
ODI (Open Data-Link Interface)TM HSM Specification 3.31 supports NBI
(NetWare Bus Interface), HIN (Hardware Instance Number), instance unload,
and industry-standard PCI Hot Plug technology. Do not mix use of a driver
written to the 3.3x/1.1x ODI Specification with 3.2 ODI Specification support
modules. The LAN driver certification bulletin will list the specification a
LAN driver is written to. In some cases, you will need to contact the IHV
(Independent Hardware Vendor) who wrote the driver.
NetWare 5 drivers for storage devices and host bus adapters are written to the
NWPA (NetWare Peripheral ArchitectureTM) Specification. The .dsk drivers
are no longer supported.
Eliminate Obvious Causes
Before taking other steps, be sure to eliminate any obvious causes for network
problems—such as checking to be sure everything is plugged in and the power
is on.
For example, consider the following:
Š Check Novell’s Knowledgebase for known problems with software and
hardware.
+,17If you are reading this documentation from your Netscape browser and you
have access to the internet, click the Home icon in the toolbar at the lower part of
the navigation column. Then click either the Novell Links button in the the
navigation column or Novell. (in red) at the top of that column. In the pop-up
window, click Novell Technical Support.
Š Inventory the hardware and cabling systems to make sure everything is
plugged in, connected, and terminated properly.
Š Check for user errors.
Š Make sure all hardware is Novell certified. Check Novell’s World Wide
Web site (http://developer.novell.com) for a list of hardware and software
products certified to run with NetWare.
+,17Again, if you are reading this documentation from your Netscape browser
and you have access to the internet, click the Home icon in the toolbar at the lower
part of the navigation column. Then click either the Novell Links button in the the
navigation column or Novell. (in red) at the top of that column. In the pop-up
window, click Novell Technical Support.
Š Make sure there are no rights problems. Insufficient rights prevent you or
your users from accessing volumes, directories, and files and from
performing file operations.
Troubleshooting
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Gather Information
Once you have updated your server with current patches, modules, and drivers
and have eliminated any obvious causes, start gathering information about the
context of the problem.
Ask yourself the following questions:
Š When does the problem happen? For example, a problem that occurs
regularly at midnight could be related to file compression. A problem
occurring in the morning could be related to spikes in network usage as
users log in.
Š Can you duplicate the problem? Or is it constant or intermittent? If
intermittent, at what intervals does it happen ?
Š Can the problem be associated with other activities happening on the
network at the same time or with specific hardware and software? For
example, does the problem occur only when the network activity is
highest? Does it occur only on certain workstation platforms?
Š Did anything change shortly before the problem began to occur? For
example, was new software installed?
Š What components of the network are involved? the server? some or all of
the workstations? printers? other peripheral devices?
Š What error or console messages are displayed? What module is the source
of the message? Check server error logs.
Š What do configuration and log files reveal? If you think you will need to
call Novell Technical Support, gather information about your server and
network, including log files such as abend.log, sys$log.err, and
vol$log.err. Also have your hardware and software configuration and
version numbers available.
You can download a Server Configuration Information tool from the
Minimum Patch List on http://support.novell.com. Extract the utility by
running confgx.exe (where x represents the current version number). It
allows you to create a text file that lists all modules loaded on the server;
the contents of the startup.ncf, autoexe.ncf, config.sys, and autoexec.bat
files ; a directory of sys:system and your local drive; and other
configuration information.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š If you suspect NDS® problems, gather the following information:
Number of servers in the tree
Structure of the tree
Number of objects per partition
Error messages
Names of servers holding master replicas
Versions of DS on pertinent servers
Š Compare the affected machines to similar machines. Do some have the
problem, while others do not? If so, what is different between the affected
and unaffected machines.
Simplify the System
If there is any danger of losing data, back up the system before you do any
further troubleshooting.
Simplify the system and eliminate as many variables as possible to isolate the
source of the problem:
Š Try running the server without the autoexec.ncf file or startup.ncf file.
You can either comment out the commands in the .ncf files or you can
execute SERVER with the -na or -ns options. (To comment out a
command, type REM and a space at the beginning of the command line.)
See SERVER in Utilities Reference for an explanation of these load
options.
Š Unload all NLM programs. Remove all nonessential products, including
virus scanners and server metering modules, until the server software is
in bare bones form.
Š Remove or disconnect peripheral devices.
Š Remove extra network boards or LAN adapters.
When you have simplified the server and taken it to bare bones, note whether
the problem happens again. To determine whether the OS is the source of the
problem, restore each component of the system, one-by-one, checking to see
whether the addition of the component generates the problem.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Use Research Tools
Novell and other vendors provide research tools you can use to resolve
problems, for example:
Š The Novell Knowledgebase at http://support.novell.com for Technical
Information Documents (TIDs) and patches. You can browse by product
or search using partial filenames and wildcards. You might look for the
Top 20 Downloads or the most recently released files.
Š Support forums at http://support.novell.com where you can post technical
questions and receive a response from a Novell Certified System
Operator (SysOp). You can also browse responses that have already been
posted.
Š Novell’s development World Wide Web site, containing information
about third-party products certified to run with NetWare, at
http://developer.novell.com. You can use a form-based search facility to
located tested and approved bulletins by category.
Š NetWare user groups, located in most cities and countries. NetWare Users
International (NUI) offers NetWare support forums on the World Wide
Web.
Š Vendor World Wide Web sites provide information and updated software
to download.
Š NetWare utilities, such as MONITOR. This utility can provide important
clues to the source of server or network performance problems.
Š ManageWise®, Novell’s network management application. ManageWise
provides an integrated set of management services for managing and
controlling servers and Simple Network Management Protocol (SNMP)
devices. It lets you monitor and analyze network traffic trends on
ethernet, token ring, and FDDI networks.
Š NetWork analyzers. Analyzers are good tools for gathering
troubleshooting information about packet traffic on the physical network.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Develop Hypotheses and Perform Tests
Once you have gathered information, eliminated obvious causes, or simplified
your system, it’s time to determine what you think is the most likely cause of
the problem. You might develop several hypotheses. If so, decide which is the
most likely, the next mostly likely, and so on. Base your hypotheses on the data
you’ve gathered and your own experience with the system.
Once you have developed your hypotheses, test them, keeping these tips in
mind:
Š Decide which hypotheses seem the most likely and test these first. Test
the least likely hypotheses last. The only exception might be a hypothesis
that is very quick and easy to test; in which case, you might want to
eliminate it first.
Š Break hypotheses down into the smallest possible components and test
each one separately. For example, if you suspect the problem is with the
physical network, try to isolate which cable segments, cards, or
connectors might be the source of the problem and then test each one
individually. Test only one thing at a time.
Š Try swapping components. If you suspect a faulty component is causing
the problem, replace it with a component that you know is good. Be sure
to swap only one component at a time.
Š Simplify the system. If you haven’t already done so, try programs
removing components from the system. If the problem still occurs, you
have a smaller set of components left that can be the source of the
problem. If the problem goes away, start reloading components one at a
time. After each addition, note whether the problem occurs. See
“Simplify the System” on page 101.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Look for Additional Help
Be prepared, when asked, to tell the support representative what
troubleshooting steps you have already taken.
Using Novell Authorized Resellers
Locate Service Partners by locality or area. Check Novell’s World Wide Web
site (http://novell.com or http://support.novell.com) for a list of resellers and
distributors worldwide.
Opening a Support Incident
An incident is defined as the resolution of one Novell issue, problem or
question, regardless of the number of communications required to achieve
resolution.
Novell customers may purchase technical support on a per-incident basis. Perincident support can be charged to a major credit card, or can be pre-purchased
in any quantity with a personal check, credit card, or purchase order. To
purchase single incidents as needed, call your regional Novell Support Center.
To pre-purchase technical support incidents in any quantity, call 1-888-3214272 in the United States or Canada, or call your regional Novell Support
Center.
Current telephone numbers for Novell Support Centers for the Americas,
EMEA (Europe, Middle East, and Africa), and Asia-Pacific are available
under Novell Telephone Support at http://support.novell.com. For the
Americas, the Technical Support telephone numbers are 1-800-858-4000 and
+1-801-861-4000.
If the problem is a NetWare bug and no patches were previously available,
then there will be no charge.
Depending on the type of prepaid service agreement your company has in
place, you may be able to open an electronic incident, update an already open
incident, or check the status or history of your incident.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Document the Solution
Once you resolve a server or network problem, document the solution
thoroughly. The information will be helpful if similar problems occur in the
future. Be sure to document changes to the server or network configuration,
software and hardware updates or additions, new version numbers, workarounds, etc.
Creating a Core Dump
Novell may request a core dump when a server experiences a lockup or abend
and other troubleshooting has failed to resolve the problem. A core dump is a
copy of a NetWare server’s memory at the time the server abended. Core
dumps can be analyzed by Novell engineers and are often the key to finding
software bugs. However, now that servers typically use large amounts of
RAM, creating a core dump is less likely to be a routine troubleshooting
procedure. Your abend.log file may be equally valuable for diagnosis.
,03257$17Novell will expect you to install all necessary patches before requesting
a core dump. To send a core dump file to Novell, you must first open a support incident,
which will be billable. You will not be charged until the incident is resolved or closed. If
the problem is a NetWare bug and no patches were previously available, then there will
be no charge.
The procedure is explained following, but your support representative can
walk you through it when you call.
You can perform the following types of core dumps:
Š Full core dump. Copies all server memory to a local drive or device.
Š Full core dump minus cache. Copies all server memory except file
cache (disk cache) to a local drive or device. The cacheless core dump is
smaller, but it provides as much useful information as a full core dump.
A core dump can be started two ways:
Š By responding to the core dump choices displayed by NetWare after an
abend has occurred.
The prompt—Type?—allows you to specify a full core dump or a
cacheless core dump, as explained above.
The prompt— Device?—allows you to specify a local drive or a DOSwritable device.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
127(Other types of core dumps may become possible in the future. Your Novell
Technical Support Representative will inform you if that is the case.
Š By forcing a core dump by typing .c while in the NetWare debugger.
If the server is not completely frozen, you can enter the debugger by
pressing these keys: Shift + Shift + Alt + Escape.
+,17Press the left Shift key with the left hand, and press both the right Shift key
and Alt with the right hand. Then press Escape with the most convenient finger of
the left hand.
Then type .c. The server then displays core dump choices.
When the core dump is finished, enter g to exit the debugger and to
redisplay the server console prompt. Or else enter q to exit to DOS.
Performing a Core Dump to a Local Drive
After you start the core dump, you are asked to specify the DOS drive letter
and file path to which the memory image file will be written. The size of the
full image file will be approximately equal to the total RAM installed in the
server. For a cacheless core dump, the size of the image file will be
approximately equal to total RAM minus the amount of file cache (disk cache
memory).
The drive can be any writable DOS device that contains enough storage space.
But remember that it must be set up, not only before the server abends—but
before the server is booted.
When the image is written to a local hard drive on the server, the default name
of the image file is c:\coredump.img. Once the file is on the hard disk, it can
be compressed, copied to diskettes, backed up to tape, or sent by FTP to
ftp.novell.com (if you have opened a support incident).
The image file can also be written to a NetWare drive later, after the server is
up and running. This can be done using an NLM called imgcopy.nlm or any
other third-party NLM that provides this functionality. Use File Finder to
locate imgcopy.nlm on Novell’s World Wide Web site,
http://support.novell.com.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Sending the Core Dump to Novell
Before sending the memory image to Novell, contact a support representative
to open a support incident. You will be assigned to a Technical Support
Engineer who will help you analyze the memory image file. The Technical
Support Engineer will make arrangements to receive the image either in the
regular mail or through the internet and will advise you on the best media
format to use.
Slow Network, Performance Problems
Resolving Slow Server Response
To diagnose slow server response problems, identify whether the following
conditions exist:
Š The workstation network board is slow or faulty.
Š Network cabling is faulty.
Š The server network board is slow or faulty.
Š Too many users are using the network.
Š The server speed is not set to the highest speed.
Š The server hard disk is slow or faulty.
Š The server is low on memory.
Š The volume has too many deleted files that have not been purged.
Š Network traffic is extremely high.
Š The cabling system is experiencing too much interference.
Š A hard disk has failed or is failing.
Š Insufficient directory buffers, cache buffers, or packet receive buffers
have been allocated.
Š An EISA controller board needs to be configured to use interrupts.
To resolve slow server response problems, perform the following actions.
Š Check the computer’s documentation for switch information. Set the
processor speed to its highest setting. Use SPEED to verify that the
processor is running at the appropriate speed.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š If a workstation or the server seems slow, insert a new network board into
the slow computer to check performance. If the speed is still below
normal, reinstall the original network board and then replace the cable
attaching the workstation or server to the network.
Š Load MONITOR to check the status of packet receive buffers and service
processes. Compare their values to the maximum allowable value.
Use MONITOR > Available Options > Server Parameters to increase the
values for the following parameters if your system has reached the
maximum value: Maximum Service Processes and Maximum Packet
Receive Buffers. (You can also use SET at the command line.) Also see
Assessing Server RAM.
Š Use MONITOR > Available Options > Storage Devices and check the
Hot FixTM status of all hard disks that use the Traditional NetWare file
system. Verify that all mirrored disks are still mirrored.
Š Run the NetWare Administrator utility to purge deleted files. Or you can
set the Purge attribute on files you want to be purged immediately after
being deleted.
Š Use MONITOR > Available Options > LAN/WAN Drivers and check the
LAN driver statistics. If you have more than one network board, compare
the boards’ Total Packets Transmitted statistics. If one board is receiving
most of the traffic, recable the network so that the boards have equal
loads.
Š If you are on a multiserver network or an internetwork, recable the system
with a backbone to reduce network traffic.
Š Check the cabling for interference from fluorescent lights, microwaves,
radar, X rays, and copy machines. Either move the cable or shield it from
the source of interference.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Insufficient Packet Receive Buffers, No ECB Available
Count Errors
The ECB (Event Control Block) counter is incremented when a device sends
a packet to your NetWare server, but no packet receive buffer is available.
The server allocates more packet receive buffers after each incident until it
reaches its maximum limit (configured with a server parameter).
If you are using an EISA busmaster board (such as the NE3200TM board), you
will probably need to increase both the minimum and maximum number of
packet receive buffers.
See Minimum Packet Receive Buffers and Maximum Packet Receive Buffers
in SET Communications Parameters in Utilities Reference.
No ECB Available Count messages can also indicate that the driver is not
configured correctly or that the Topology Specific Module (TSM) and the
Hardware Specific Module are incompatible.
This value is maintained by the TSM module.
High Utilization Statistics in MONITOR
Network performance is a key concern for network administrators and for
Novell as well. Unfortunately, there is sometimes confusion about
performance indicators and what their statistics mean.
For example, the idea that processor utilization is the key performance
indicator for NetWare is much too simple. Some network administrators are
concerned when the Utilization value in MONITOR’s General Information
screen approaches 100%—on the assumption that the higher the percentage,
the worse NetWare’s performance is. This is entirely false.
What Is Normal?
Consider first what the Utilization value represents: the average of the server’s
total processing capacity that was used during the last second (update
interval). The remainder of the capacity was spent in the idle loop process. In
other words, it is an indication of how much of that time the processor had
something to do. A high utilization value means that NetWare is using that
percentage of the processor’s capacity and wastes less time doing nothing.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Some processes make efficient use of the processor and as a result may cause
100% utilization. This type of utilization is entirely appropriate. Most of the
time when utilization moves up to 100%, it means that the thread is using the
processor efficiently. It may stay at 100% for a couple of minutes. This is
normal.
It is not normal, however, when the utilization is at 100% for 15 to 20 minutes,
when connections are dropped, and server performance deteriorates
noticeably. High utilization with these conditions indicates a problem. If
you’re not seeing these conditions, your utilization may be normal, even when
it’s at 100%.
How do you know what is normal for your server? You will recognize
problems if you baseline your server. Know what is normal and know the
difference between a cosmetic problem and a true performance problem. (You
can test for a cosmetic problem by loading or unloading any NLM. This will
cause the processor information in MONITOR to be recalculated.)
What Are the Most Common Solutions?
Before troubleshooting high utilization problems, make sure that you have
followed the steps in “A Troubleshooting Methodology” on page 98. Check
the Novell Support Connection at http://support.novell.com for NetWare 5
patches or updated modules. Available patches will contain fixes for reported
high utilization problems related to the actual code of the operating system and
NDS.
However, there are also a number of high utilization conditions that can still
result from problems with configuration, levels of NLMs, and with tuning
issues. You can use the following list to help you resolve problems.
The items in the list are categorized, but are otherwise in no particular order.
The list represents the collective experience of Novell Technical Support
Representatives . We recommend that you review each item, using each to
carefully analyze your system. Except for problems new to NetWare 5, you
will be able to resolve the problem on your own in almost every case.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Operating System Issues
Š Server resources. Server resources can be divided into LAN, disk, and
processor resources. LAN and disk resources are the number of buffers
available. Processor resources are the number of service processes
available. When a network packet arrives or when a disk read or write is
performed, a buffer should be available to hold the data until a service
process is available to service it. Running low on either buffers or service
processes for even a brief period can cause high utilization on the server
and delay for the users.
To understand how server parameters work together to control the
allocation of server resources, see “Monitoring Allocated Services” on
page 70.
For the procedure to configure the allocation of packet receive buffers
(communication buffers, see Increasing Maximum and Minimum Packet
Receive Buffers and SET Communications Parameters for parameter
descriptions.
For information on controlling the allocation of directory cache, see
Tuning Directory Cache and SET Directory Caching Parameters.
For parameter descriptions, see also the service processes parameters in
SET Miscellaneous Parameters.
Novell suggests that the following server parameters be set as indicated
to avoid running low on server resources. This is especially important if
you have 100 or more users. As a guideline, 2 to 3 per connection is
recommended for the minimum values. You can use SET or MONITOR
> Available Options >Server Parameters.
Minimum Packet Receive Buffers=value
Maximum Packet Receive Buffers=4000
Minimum Directory Cache Buffers=value
Maximum Directory Cache Buffers=4000
Maximum Service Processes=1000
New Service Process Wait Time=0.3
Š Processor speed. Use the SPEED command at the server console. In a few
cases, a system board drops the processor speed after the server boots. A
processor speed rating drastically lower than what is expected may
indicate a configuration problem. See SPEED in Utilities Reference.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Processor hog. See MONITOR > Available Options > Kernel >
Processors to determine whether one thread is hogging the processor. You
can set the CPU Hog Timeout Amount to a value lower than the default.
Specify the amount of time in seconds to wait before terminating a thread
that has not relinquished control of the processor. See SET Miscellaneous
Parameters.
Storage Devices and Adapter Issues
Š Drivers. See “Keep Your Servers Patched” on page 98. For a brief
explanation of NetWare Peripheral Architecture (NWPA), see Drivers for
Host Adapters and Storage Devices.
Š Free blocks. Adequate Free Blocks are essential. A Free Block is a disk
block that has no salvageable files stored there. A file that has been both
deleted and purged becomes free space. Maintain a minimum of 1000 free
blocks on each NetWare volume that has suballocation enabled.
Suballocation, normally a low priority process, uses free blocks to
perform its function. When free blocks are low, suballocation can go into
an “aggressive” mode. Maintaining over 1000 free blocks will usually
avoid this problem.
To check how many Free Blocks you have on a volume, use MONITOR
> Volumes > Volume Information > Free Blocks. If there are not at least
1000 free blocks on the volume, run (at a workstation) PURGE /A from
the root of the volume. This removes all deleted files from the directories
and subdirectories and increases your free disk space. See PURGE in
Utilities Reference.
If you have applications that create large numbers of temporary files you
may want to set the P (Purge) flag on the directories where these files are
stored. Every temporary file that is created will be put on the deleted file
list. These files are kept on the disk until PURGE is run. You could also
run SET Immediate Purge of Deleted Files=ON at the server console
prompt.
Š Suballocation. Use a disk block size of 64 KB. It is very important to
monitor the disk space. Novell Technical Support recommends keeping
10% to 20% of the volume space free to avoid suballocation problems.
Suballocation does not have any server parameters to adjust. See
Checking Available Disk Space.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Compression. When using compression with the Traditional File System,
it is essential to have the server patched. Because it takes processor cycles
to compress and decompress files, compression needs to be monitored
carefully to avoid utilization problems. The default SET File System
Parameters for compression take this into consideration.
For example, file compression is set by default to occur during off-hours
or periods of low server usage. Make sure that change have not been made
that are causing high utilization problems.
File decompression occurs on the fly. However, a very large file, such as
100 MB, can take a noticeable amount of time even with a Pentium*
processor.
The SET Deleted Files Compression Option= 2 causes the immediate
compression of files that have been deleted. This can cause high
utilization because the processor immediately compresses files upon their
deletion.
Directories that are flagged IC (Immediate Compression) can cause
excessive decompression during work hours. Users with disk space
restrictions may flag their home directories to IC in an attempt to save
disk space. This can affect server performance.
To ensure that frequently used files are not compressed, you can specify
an appropriate value for the SET Days Untouched Before Compression
parameter
To proactively eliminate compression as a possible problem, use SET
Enable Files Compression=OFF. This will cause files to be queued for
compression but the files will not be compressed. However, accessing
compressed files causes them to be decompressed.
Memory Issues
Š Insufficient RAM. Memory resources are critical in a server environment.
The best indicator for assessing server RAM is displayed in MONITOR
> Available Options > Disk Cache Utilization > LRU Sitting Time. The
value of this field should be a minimum of 15 to 20 minutes. A number
lower than this indicates that the server’s memory resources may be low.
See Assessing Server RAM. Another good indicator, Long Term Cache
Hits, is displayed on MONITOR’s General Information screen.
Š See also “Memory Error Messages, Memory Problems” on page 127.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Directory Services Issues
Š Directory Services partitions. Novell recommends having no more than
three replicas of each partition in the tree. Efficient tree design is essential
to avoid utilization problems. Because NDS needs to maintain
synchronization among all servers in the replica ring, the more replicas
there are of any partition, the more traffic will be on the wire. Three
replicas are enough to provide fault tolerance and to allow for NDS
recovery if a database were to become corrupt.
Š Directory Services synchronization. To test the possibility that NDS
synchronization is the problem, turn off inbound and outbound
synchronization for an hour by setting dstrace=!D60. See MONITOR >
Available Options > Server Parameters > Directory Services. Also see
SET Directory Services Parameters in Utilities Reference.
After turning synchronization off, wait 5 to 15 minutes to allow the server
to catch up with queued work. Cancel the test by setting dstrace=!E.
Š Directory Services errors. To test for other errors, run SET dstrace=ON.
See MONITOR > Available Options > Server Parameters > Directory
Services. Check for persistent synchronization errors on the Also see SET
Directory Services Parameters in Utilities Reference.
Š Bindery emulation. Only one thread is created for all bindery
connections; consequently printers using bindery emulation can
monopolize a processor.
Printing Issues
Š Bindery emulation. Eliminate printers using bindery emulation or reduce
them to a minimum. Only one thread is created for all bindery
connections; consequently bindery connections can monopolize a
processor. NDPS (Novell Distributed Print Services) is the preferred print
service for NetWare 5 and is fully compatible with existing queue-based
printing resources. This means that you can make the transition from your
legacy printing setup gradually.
Š Number of printers. No more than 40 printers can be serviced from one
server. More than that will degrade server performance. However, this
number depends on the type of printing that is being done. Printing CAD/
CAM designs take much more processor time than text documents. If you
are concerned about utilization problems, we recommend that you set
your print devices to do the processing instead of the server. This will
slow down printer output but will relieve the utilization on the server.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Client Issues
Š Hung connections. Check for them and clear them. See “Clearing a
Workstation Connection” on page 77.
Š Client software. You can identify client problems by changing SET NCP
Parameters. See also MONITOR > Available Options > Server
Parameters > NCP.
By setting these server parameters to ON, the server displays a warning
each time a bad packet is received at the server. The warning message also
includes the MAC (Media Access Control) addresses of devices where
the packets originated.
Display NCP Bad Component Warnings=ON
Reject NCP Packets With Bad Components=ON
Display NCP Bad Length Warnings=ON
Reject NCP Packets With Bad Length=ON
Update client software.
Š Cabling. If you have installed high speed 100 MB network boards, you
may need to upgrade your cabling, too.
Š RConsole. If RConsole is run from a DOS window on Microsoft*
Windows* and you leave it running while you open another DOS window
or toggle to it, server utilization will exceed its limits (max out).
In addition, REMOTE and RSPX, the two server modules that enable a
workstation to access the server console remotely, can aggravate high
utilization, can aggravate its effects.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com) for high utilization problems
involving particular software or hardware products; and contact a Novell
Authorized Reseller or Novell Technical Support.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Suspected Multiprocessing Problems
Perform the following steps if you suspect a multiprocessing problem. You
will be able to eliminate most probable causes.
1. Are the secondary processors online?
If the server is already up, run DISPLAY PROCESSORS at the server
prompt. Online processors appear in green, and offline processors are in
red. (Processor 0 cannot bot be taken offline.) Run STOP PROCESSORS.
If the server is down, add the command SET Auto Start Processors=ON
to startup.ncf. Then bring up the server and then run STOP
PROCESSORS.
If the problem still occurs, go to Step 2. If not, go to Step 4.
2. To verify that this is not an interrupt issue, remove the PSM. The
mps14.psm, cpqmpk.psm, or *.psm in startup.ncf is the interrupt
abstraction layer and communicates with the APIC (Advanced
Programmable Interrupt Controller) rather than the PIC for interrupt
handing in the machine.
Then bring up the server and run STOP PROCESSORS again.
If the problem still occurs, it not a multiprocessing issue. If it no longer
occurs, go to Step 3.
3. Check with the hardware vendor and make sure you have the latest
machine updates (current ROM BIOS, etc.), If you confirm that your
hardware is current and the problem still occurs, it’s possible that you are
using a new version that wasn’t available at the time Netware 5 was
released. Whatever is different about the hardware requires a newer PSM
to work with the machine.
127(Problems that occur only in a multiprocessing environments are not
necessarily MP problems. More likely, it is an issue with a piece that has been
loaded in an MP environment and that does not function properly.
4. Gather information as explained in “A Troubleshooting Methodology” on
page 98. Use startup.ncf and autoexec.ncf to determine what software
pieces are running. Check with the software vendors and developers and
tell them what steps you have taken and give them enough information to
duplicate the problem.
If the problem still occurs, review the Novell Knowledgebase
(http://support.novell.com) for recent TIDs that might resolve your problem;
and contact Novell Technical Support.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Unable to Communicate with Server, -625 Errors
Communication problems among servers can generate messages such as
Unable to communicate with server or messages that a Directory partition
operation can’t be completed. If DSTRACE is set to on, -625 errors appear in
the DSTRACE output because the Directory cannot be synchronized among
the servers.
Follow these steps to troubleshoot server-to-server communication problems.
1. Is the server down? Has it abended?
Run DISPLAY SERVERS at the server console prompt to see if the server
is visible on the network.
If the server is down, restart it. If the server has abended, troubleshoot the
abend. See “Server Abends” on page 139.
2. Is time synchronized?
Enter TIME at the server console to see if server time is synchronized to
the network.
If time is not synchronized, troubleshoot the time synchronization
problem. For general information about time synchronization, see
Network Time Management.
3. Is the IPX internal network number of the server unique? Are IP
addresses unique? Do all replicas have the correct address for each
server?
If all the servers in a replica ring cannot communicate with one of the
servers in the ring, there is probably an address problem.
Make sure all addresses are unique. Make sure all replicas have the
correct address for each server. Load DSREPAIR, select Advanced
Options > Replica and Partition Operations > (select a replica) > Display
Replica Information.
4. Do all replicas have the correct server name?
Incorrect or duplicate server names in Directory replicas make it
impossible to communicate with the server. Make sure all replicas have
the correct name for each server. Load DSREPAIR, select Advanced
Options > Replica and Partition Operations > (select a replica) > Display
Replica Information.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
If names are not correct, contact Novell Technical Support or an
Authorized Novell Reseller for assistance in renaming servers.
5. If the servers share the same cabling system, are addresses unique for
each cable segment?
If addresses are not unique, change addresses as necessary.
6. Are LAN drivers loaded and protocols bound? Are LAN adapters and
drivers the latest versions?
Load and bind LAN drivers, if necessary. If drivers are not the latest
versions, update the drivers. In addition, review Novell’s Knowledgebase
for known problems with specific adapters. Replace adapters if necessary.
7. Is the LAN functioning?
Ping the servers to see if the connection is active. Run TRACK ON at the
server console to monitor traffic into and out of the server. View LAN
statistics in MONITOR to make sure packets are being sent and received.
(MONITOR > LAN/WAN Drivers; highlight a driver and press Tab to
expand the information window.) Review statistics such as packets
received and sent, collisions, and retries.
If the LAN connection is malfunctioning, troubleshoot the physical link
between the servers. See “Tips for Resolving General Communication
Problems” on page 125.
8. Have servers just been upgraded from NetWare 3.x?
Some routers don’t have enough memory to buffer the increase in packets
from NetWare 4.x or 5.x servers. Make sure the routers have enough
memory.
9. If the servers are using IPX, are routers filtering SAP types 004, 26b, 278?
If yes, remove the filters.
10. Have any other changes been made to routers before the problem started?
If yes, try using the RESET ROUTER command to reset the server’s
internal router. Troubleshoot other routers.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
11. Are servers separated by 56 KB WAN links?
If yes, this could be the source of the problem. Consider the following:
Is there a guaranteed transfer rate or does the rate decline when network
traffic is heavy? If the rate declines, the link might not provide the
performance you need to maintain consistent communication.
Are you trying to transfer large files or an entire replica across the link?
You might be able to synchronize replicas across a 56 KB link, but not
transfer an entire replica.
Do the servers have to traverse more than one 56 KB link?
To solve the problem, you might need to provide faster links.
12. Do the affected servers have utilization at or close to 100%?
If yes, take steps to reduce server utilization. See “High Utilization
Statistics in MONITOR” on page 109.
13. If using IPX, are IPX connections set to the recommended value (2000)?
If not, enter SPXCONFIG I = 2000 at the server console.
14. Is the database locked?
If DSREPAIR is running, the database will be locked. Other applications
might also lock the database while writing to it.
If the database is locked, wait for DSREPAIR or other module to finish
using the database or shut down the utility or module.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Transaction Tracking Errors, -621 Errors, FD93 Errors
Any error that contains -621 or FD93 means that the Transaction Tracking
System (TTS) has shut down. The server operating system cannot track and
write the transactions at the maximum transaction level set for the server.
The problem might be temporary, because a partition operation generated a
large number of transactions, or it might be recurrent, because the server can’t
handle normal peak transaction requests.
To enable TTS after it has been disabled, enter the following at the server
console prompt:
ENABLE TTS
To troubleshoot TTS shutdowns, follow these steps.
1. Have you attempted NDS partition operations, such as removing a
replica? Has the server displayed messages such as Growing TTS
memory tables?
Certain NDS partition operations require the server to track and write
more transactions than it can handle. When this happens, the server
displays error messages such as Growing TTS memory tables.
Reduce the value of the Maximum Transactions server parameter so the
server can track and write transactions without becoming overburdened.
You might also need to reduce the value of the TTS UnWritten Cache
Wait Time parameter.
To modify the parameters, load MONITOR, select Available Options >
Server Parameters > Transaction Tracking and set parameter values as
follows:
Š Reduce the value of the Maximum Transactions parameter to 5,000.
Š If TTS shutdown continues to occur, reduce the value of the
Maximum Transactions parameter to 1,000.
Š If TTS shutdown continues to occur, reduce the value of the TTS
UnWritten Cache Wait Time parameter to 11 seconds.
Š Before removing a replica, set the Maximum Transactions parameter
to 1,000 and the TTS UnWritten Cache Wait Time to 11 seconds.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
2. Has the server abended with a message such as Abend: CPU Hog
Detected by Timer that lists Transaction Tracking as the running process?
If the server abended because TTS was not relinquishing the CPU, TTS
might simply need more time to finish backing out a large transaction.
If your server experiences this abend, set the CPU Hog Timeout Amount
parameter to 0 to shut off the CPU monitoring process (MONITOR >
Available Options > Server Parameters > Miscellaneous). Restart the
server and allow TTS to finish backing out its transactions. When
finished, set the CPU Hog Timeout Amount parameter back to its original
value.
3. Has the server experienced memory errors, such as Memory Allocator out
of available memory, along with the TTS shutdown?
NDS operations, such as migrating objects, merging trees, or creating
objects, sometimes create a large number of TTS transactions in a short
amount of time and cause TTS to use up available memory.
To prevent the problem, follow these suggestions:
Š Use the latest version of NDS.
Š Reduce maximum transactions on the affected server to 100.
Š Add more memory to the affected server.
4. Has the Sys volume become full?
TTS backs out its transactions to the Sys volume, so if the volume
becomes full TTS must shut down.
Remove files from the Sys volume or increase the volume size.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Workstations Can’t Communicate with Server
See the following sections to troubleshoot different kinds of workstation to
server problems:
Š “Errors Reading from and Writing to Network, Server Cannot Be Found
Errors” on page 122.
Š “Connections Dropped Periodically” on page 125.
Š “Tips for Resolving General Communication Problems” on page 125.
Errors Reading from and Writing to Network, Server Cannot Be
Found Errors
When workstations can’t communicate with the server, users lose their
connections to the server or cannot log in at all. Users might see Server Cannot
Be Found messages or errors about reading from or writing to the network.
Follow these steps to troubleshoot the problem.
1. Can other clients connect to the server?
If other clients can connect to the server, then the problem is with this
client, not with the server or network. Consider the following:
Š If the user cannot log in, did the user power off without logging out?
If the user powers off without logging out, the client connection
remains open and the user will not be able to log in again until the
connection is cleared. The server’s watchdog process will retry the
connection a designated number of times, then the connection will
time out and the server will clear the connection. Once the
connection is cleared, the user will be able to log in again.
Watchdog processes are controlled by SET parameters accessed
through MONITOR (MONITOR > Available Options > Server
Parameters > Communications).
Š Has the user’s password expired?
Sometimes users don’t understand that they must change their
passwords before they use up their grace logins.
Š Are there other restrictions on the user account, such as time periods
when the user is not allowed to log in? Check the user’s properties in
NetWare Administrator.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Has intruder lockout been activated?
Š Is the workstation LAN cable functioning?
2. Has the server abended?
If the server abended, the server will display an abend message on the
console screen. If the abend auto recovery process restarted the server, the
message might have scrolled off the screen. Check the abend.log file in
Sys:system to read the message.
To troubleshoot a server abend, see “Resolving Abends” on page 144. For
general information about types of server abends, see “Types of Abends”
on page 140. To manage the abend auto recovery process, see “Managing
the Abend Auto Recovery Mechanism” on page 150.
3. Is the server console hung?
If you cannot enter commands at the server console prompt, then the
console is hung. In some cases, you won’t be able to toggle among server
screens.
Try pressing Ctrl + Alt + Esc. This should display a prompt to down the
file server and exit to DOS. If the prompt is not displayed, try entering the
debugger by pressing Shift + Shift + Alt + Esc. If you can enter the
debugger, type q and press Enter to exit to DOS. Reboot the server. See
“Server Console Hangs” on page 137 to troubleshoot the problem.
If neither of the commands just described works, the only way to exit the
server is to force a non-maskable interrupt (NMI) error. Contact Novell
Technical Support, a Novell Authorized Reseller, or your computer
hardware vendor for assistance.
If you do not have time to obtain the assistance to generate an NMI, you
can power off the server, but this can corrupt operating system files. Try
this measure at your own risk.
4. Is the Sys volume mounted?
Type VOLUMES at the server console to display a list of mounted
volumes. If volume Sys is not mounted, mount it. The server does not
broadcast to the network until volume Sys is mounted.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
5. Are there LAN problems?
At the server, execute TRACK ON to make sure that traffic is flowing
into and out of the server.
View LAN driver statistics to see if packets are being lost or if there are
excessive collisions or retries. To see LAN statistics, load MONITOR and
select LAN/WAN Drivers, then press Tab to expand the information
window, which contains the statistics.
For tips on troubleshooting communications problems, see “Tips for
Resolving General Communication Problems” on page 125
If necessary, use LANalyzer® or a similar product to monitor network
traffic and isolate the problem.
6. Is server utilization high?
Check utilization statistics in MONITOR. For help in responding to high
utilization, see “High Utilization Statistics in MONITOR” on page 109.
If you suspect heavy LAN traffic is causing the problem, try unplugging
the LAN cable. If heavy LAN traffic was the problem, utilization should
drop.
7. Is the server receiving No ECB Available Count messages or messages
about insufficient packet receive buffers?
These messages could indicate that the server is running out of resources.
For help in resolving these errors, see “Insufficient Packet Receive
Buffers, No ECB Available Count Errors” on page 109.
8. Is the NDS database closed or locked by another utility, such as
DSREPAIR?
If the NDS database is not functioning, the workstation clients will not
display the login screen.
9. Are servers receiving -625 or other communication errors?
See “Unable to Communicate with Server, -625 Errors” on page 117 or
see the online documentation: Reference > Error Codes and System
Messages > Novell Error Codes > NDS Error Codes.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Connections Dropped Periodically
1. Is server utilization high?
Check server utilization statistics in MONITOR. See “High Utilization
Statistics in MONITOR” on page 109.
2. Are users using old NetWare clients?
If so, upgrade to the newest version of the appropriate NetWare client.
3. Do any of the workstations have the same node number?
Node numbers must be unique for each workstation. If there are duplicate
node numbers, change numbers so that all workstations have unique
addresses.
4. Are any network boards malfunctioning in the server or workstation? Are
there old boards or drivers?
Check Novell’s Knowledgebase (http://support.novell.com) for any
known problems with your drivers and boards; update drivers; replace
boards if necessary.
5. Are cables terminated correctly?
Check all cables to make sure they are are firmly connected, terminated
correctly, and not running near any equipment that might cause
interference.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Tips for Resolving General Communication Problems
To resolve general network communication problems, perform the following
actions:
Š Check all hubs and repeaters to make sure they are on. Check
concentrators and look for flashing yellow lights that indicate collisions.
Š Make sure all routers are functioning properly and have enough memory.
A router with insufficient memory becomes the weak link in the chain and
can slow down an entire LAN.
Š Check the cables for proper termination. Make sure cables are not running
close to sources of interference.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Make sure the server network adapter is seated properly and is
initializing.
Š Run CONFIG at the server console to make sure a protocol is bound to
each board and to check the configuration of the network boards. Then
bring down the server, turn off the power, and check the actual network
board settings:
Š Make sure all settings agree with the settings used to load the driver.
If you load the driver with an interrupt that conflicts with the board’s
setting, the network board cannot broadcast on the network.
Š Make sure the node address on the board is a legal address.
(Addresses 0 and FFFFFFFFFFFF are reserved; do not use them.)
Š Make sure that the memory range for the I/O ports and memory
addresses do not overlap. If there are conflicts, reconfigure the
equipment so that no conflicts exist.
Š Make sure that adapters do not conflict with other boards, the monitor
board, or ports in the server. Make sure no two pieces of hardware are
using the same I/O port, interrupt, or memory address. Avoid interrupts 2,
9, and 15.
Š Make sure the network adapter works correctly; replace it if it doesn’t.
Check the lists of certified products at http://developer.novell.com to
make sure the adapter works with NetWare. Make sure drivers are the
latest versions.
Š Run VOLUMES at the server console. If volume Sys is not mounted,
mount it. The server does not broadcast to the network until volume Sys
is mounted.
Š Check workstation addresses and server node numbers:
Each node address on the network should be unique. Sometimes you can
find the problem by turning off all workstations and turning them on one
at a time
Š Make sure workstations are using the correct frame type.
Š Have a user log in to the network from a workstation. Load MONITOR
and select Connections.
If the Active Connections list shows the user’s login name, the server is
receiving and responding to the workstation’s request.
If the workstation receives a Server cannot be found error, make sure the
server and workstation are using the same frame type.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
If using 10baseT, try using a cross-cable to connect the server directly to
the workstation. If this solves the communication problem between the
server and workstation, then something on the network is causing the
problem.
Admin Account Has Been Deleted or Admin Password
Is Unknown
There is only one solution when the Admin account has been deleted or the
Admin password has been lost: Contact Novell Technical Support for
assistance.
To avoid losing the Admin account, it is recommended that you create one or
more additional user accounts with the same privileges as the Admin account.
,03257$17Do not assign privileges to the additional accounts by setting up security
equivalence with Admin. The accounts will lose security equivalence if Admin is deleted.
Memory Error Messages, Memory Problems
See the following sections to troubleshoot different kinds of memory
problems, to resolve memory leaks, and to resolve memory problems by
freeing memory:
Š “NetWare Doesn’t Recognize All the Memory in the Server” on page 127
Š “Server Displays Memory Error Messages” on page 128
Š “Resolving Server Memory Leaks” on page 130
Š “Resolving Server Memory Problems” on page 131
NetWare Doesn’t Recognize All the Memory in the Server
1. Does config.sys or autoexec.bat include a DOS=HIGH statement, or
commands to load memory managers or DOS device drivers? For
example, is there a command to load himem.sys or emm386.exe? Both
are memory managers.
Comment out these statements from config.sys or delete config.sys
altogether. Comment out these statements from autoexec.bat. (To
comment out a command, type REM and a space at the beginning of the
command line.)
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
If there is a memory manager in the server, NetWare relies upon the
memory manager to determine the amount of available memory, instead
of registering the memory itself. Some memory managers cannot
recognize more than 64 MB of memory. DOS device drivers take memory
away from NetWare’s memory pool.
Make sure Windows 95 is not being used to boot the server. Windows 95
autoloads memory managers.
2. Is the server BIOS out of date?
An out of date BIOS might be reporting the wrong amount of memory.
Update the BIOS if a newer version is available.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Server Displays Memory Error Messages
Typical memory error messages include, Cache memory allocator out of
available memory, Short term memory allocator is out of memory, and Cache
memory allocator exceeded the minimum cache buffer limit.
1. Is the server loading a memory manager or a DOS device driver?
Check the autoexec.bat and config.sys files to make sure no memory
managers, such as himem.sys, or emm386.exe, are being loaded, and that
there is no DOS=HIGH statement in either file. Make sure no DOS device
drivers are being loaded.
Comment out these statements from config.sys or delete config.sys
altogether. Comment out these statements from autoexec.bat. (To
comment out a command, type REM and a space at the beginning of the
command line.)
If there is a memory manager in the server, NetWare relies upon the
memory manager to determine the amount of available memory, instead
of registering the memory itself. Some memory managers cannot
recognize more than 64 MB of memory. DOS device drivers take memory
away from NetWare’s memory pool. Make sure Windows 95 is not being
used to boot the server. Windows 95 autoloads memory managers.
2. Is the server BIOS out of date?
An out of date BIOS might be reporting the wrong amount of memory.
Update the BIOS if a newer version is available.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
3. Is the Reserved Buffers Below 16 MB parameter value too low, for
example under 200?
If so, use MONITOR to increase the value (MONITOR > Available
Options > Server Parameters > Memory). Increase the value to 200 or
higher, especially if there is a CD-ROM or tape device that needs memory
below 16 MB.
4. Is memory being registered manually?
Manually registering memory can cause memory fragmentation. Some
old system boards might require you to register memory manually, but the
better solution is to upgrade to a newer board so that NetWare will register
the memory automatically.
If memory has been registered manually, reboot the server to free memory
and do not manually register memory again. Upgrade the system board if
necessary.
5. Do memory errors occur when a volume is mounting?
If yes, the server might be low on memory.
To free memory temporarily, see “Resolving Server Memory Problems”
on page 131. To solve the problem, add more RAM.
6. Does the LRU sitting time, as shown in MONITOR, average less than 15
minutes during peak work hours?
If yes, the server might be low on memory.
To free memory temporarily, see “Resolving Server Memory Problems”
on page 131. To use the LRU Sitting Time to tune memory, see Tuning
File Cache. To solve the problem, add more RAM.
7. Do the LRU Sitting Time and Long Term Cache Hits gradually decline
over time, even when network traffic has not increased and no new
applications have been installed on the server?
If yes, the server might have a memory leak. See “Resolving Server
Memory Leaks” on page 130.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Resolving Server Memory Leaks
The following information is available in more detail from Novell’s
Knowledgebase at http://support.novell.com.
A memory leak means that an NLM or set of NLM programs has requested
memory from the server, but has not returned the memory when finished with
it. Over time, the amount of available memory decreases until eventually the
server generates memory error messages. The memory leak may be slow or
fast depending on the amount of memory requested each time.
If you reboot the server, the memory is returned to the memory pool, and the
low memory error messages stop temporarily, until the memory leak ties up
enough memory to generate the error messages again.
To see if the server has a memory leak, restart the server and monitor memory
statistics over time. Note whether the LRU Sitting Time and the Long Term
Cache Hits statistics gradually decline, when traffic hasn’t increased and no
new applications are installed on the server. If they do, follow the steps below
to find the source of the problem.
1. Load all the latest patches on the server.
Server patches are available from Novell’s web site and other locations.
See “Applying Patches” on page 40 for a list of sources.
2. Restart the server to free memory and establish a baseline for memory
use.
3. Load MONITOR and select Loaded Modules. Highlight a module name
you suspect might be the source of the leak and press Tab to view the
memory statistics for the module.
Write down the Bytes in Use value for the module. Use this number as the
baseline as you monitor the module’s memory use over time.
4. Repeat Step 3 for each NLM you suspect might be the source of a memory
leak.
5. When the memory error messages occur again, repeat Step 3 to view the
memory statistics for each suspected NLM. Note whether memory use
increased substantially for any of the modules.
If there is a memory leak, one or more modules will show a large increase
in the Bytes in Use value.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
6. When you discover the source of the memory leak, contact the module
vendor to tell them about the problem. If possible, update the module or
remove the module from the server.
Resolving Server Memory Problems
To free server memory temporarily (until you can add more memory to the
server), do one or more of the following:
Š Delete unused files and directories and purge them. Use the NetWare
Administrator utility to purge deleted files.
Š Use MONITOR to lower the value of the Minimum File Delete Wait
Time parameter to shorten the time period before deleted files are purged.
Or change the Immediate Purge of Deleted Files parameter to On so that
files are purged immediately after they are deleted. Both parameters are
found in MONITOR > Available Options > Server Parameters > File
System.
Š Unload any NLM, such as NWCONFIG or MONITOR, that is not
currently needed.
Š Dismount volumes that are not being used and reduce the size or number
of volumes that the server supports.
Š Streamline the directory structure. Every subdirectory in the traditional
file system takes at least one directory block (by default, a 4 KB block of
memory). Thus, subdirectories with only one file require as much
memory as directories with 32 files.
If you combine directories so that most directories have about 32 files,
and then purge the deleted subdirectories and files, you will free up
memory.
Š Try converting volumes from the traditional file system to the NSS file
system. It requires less memory to mount NSS volumes.
Š If you are using the traditional file system, remove name space support.
(The NSS file system does not use name space support.)
:$51,1*This is a destructive step that destroys all the extended file
information. Before taking this step, try to free up enough memory so that the
volume mounts and you can back up the data.
Have all users log out, and then unload all modules except the volume’s
disk drivers. Dismount any mounted volumes.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
To remove the name space, load VREPAIR, select Set VRepair Options
and then select two options: Write All Directory and FAT Entries Out to
Disk and Remove Name Space Support From The Volume. Then run
VREPAIR > Repair a Volume on the volume that would not mount.
Š As a last resort, back up all files in the volume, bring down the server, and
use NWCONFIG to reinitialize the volume. Specify a block size of 64 KB
and turn the Block Suballocation option to Off.
(This setting uses a lot of disk space but increases the amount of memory
available.)
Š Check the autoexec.bat and config.sys files to make sure no memory
managers, such as himem.sys, or emm386.exe, are being loaded, and that
there is no DOS=HIGH statement in either file. Make sure no DOS device
drivers are being loaded
Disk Errors
See the following sections to resolve disk I/O, disk space, and mirroring
problems.
Š “General Disk I/O Errors” on page 132
Š “Resolving Server Disk Space Problems” on page 134
Š “Mirrored Partitions Do Not Remirror Automatically” on page 135
Š “Mirroring Takes a Long Time” on page 136
Š “Mirroring Stops Just Before It Is Finished” on page 137
General Disk I/O Errors
General disk I/O problems can produce error messages such as Device name
deactivated by driver due to device failure, Device_name deactivated for
unknown reason, Could not redirect block number on device number, and
Some or all volume segments are missing.
Follow these steps to troubleshoot the problem.
1. Is all hardware installed and connected correctly?
For example, are cables terminated correctly, are the connections firm, are
disks installed correctly, and is the power on? If not, correct the problem
and note whether the errors still occur. If they do, continue with the
following steps.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
2. Does NetWare recognize all devices?
At the server console prompt, execute LIST DEVICES to make sure all
devices are recognized.
If devices are not recognized, run SCAN FOR NEW DEVICES and
SCAN ALL to register the devices with the server.
Execute LIST DEVICES again to make sure the devices are now
recognized. If they are not, determine whether the correct CDMs are
loaded for the devices not being recognized.
If the device is corrupted, you might need to use a third-party application
to salvage lost or corrupted volumes.
3. If this is a SCSI device, does the controller recognize devices when you
start the server?
If not, troubleshoot the hardware. Refer to your hardware documentation
or contact the hardware vendor.
4. Are .ham drivers loaded for all disk channels and are they the latest
versions?
Some .ham drivers might have to be loaded reentrantly. Load the
appropriate and most recent .ham for each disk channel.
5. Are .cdm drivers loaded for all devices and are they the latest versions?
If not, load the appropriate and most recent .cdm for each device.
6. Are .hams and .cdms configured correctly?
Are interrupt, port, slot, and custom settings all correct for the hardware?
Are interrupts unique for each driver? To see port, interrupt, and slot
settings, load MONITOR, select Storage Devices, and highlight a .ham.
Press Tab to expand the information window to see the settings for the
selected module. Configure the drivers, if necessary.
7. Is there a valid partition on the device?
Load MONITOR and select Storage Devices to see if partitions are
visible.
Third-party software can be purchased to salvage lost or corrupted
volumes. If you plan to use such a product, do not recreate the partition.
If partitions have been lost and you will not attempt to recover the data on
them, you can use NWCONFIG to recreate partitions. Restore data from
a backup.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
8. Has the Hot Fix redirection area been used or has it been filled?
To see Hot Fix information, load MONITOR and select Storage Devices.
Scroll to highlight a Hot Fixed partition, then press Tab to expand the
information window.
If the Hot Fix area is full or if it has grown since the last time you checked,
the disk is probably failing. Troubleshoot or replace the disk.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Resolving Server Disk Space Problems
When a disk fills up, you will see error messages and alerts about insufficient
disk space.
Follow these steps to troubleshoot the problem:
1. Does the volume have space available?
If yes, check the disk drives and disk channel to see if a failure has
occurred.
If no, continue with the following steps.
2. Is there free (unassigned) space on the disk?
If yes, add the free space to the volume that is filling up. Load
NWCONFIG, select Standard Disk Options > NetWare Volume Options;
press Ins to add or modify volume segments.
3. Have deleted files been purged?
When a file is deleted, it remains on the disk until it is purged. To purge
all deleted files in a directory, start NetWare Administrator, select the
directory containing the files, click Tools > Salvage > Purge. You can also
purge selected files.
Set the Immediate Purge attribute on files or directories so they are purged
as soon as they are deleted.
4. Are there many unnecessary files, such as .tmp files and mail files or
unneeded NLM programs?
If yes, delete and purge the unneeded files and programs. If the Sys
volume is full, users can still log on and delete files.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
5. Are users restricted to a certain amount of disk space?
If users are reporting that they are out of disk space, there might be
restrictions on the amount of disk space they can use.
If user restrictions are causing the problem, assign them more disk space
or ask them to delete and purge unneeded files.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Mirrored Partitions Do Not Remirror Automatically
When mirrored partitions become unsynchronized, they should resynchronize
automatically. If partitions do not resynchronize, follow these steps.
1. Load NWCONFIG and select Standard Disk Options > Mirror/Unmirror
Disk Partitions. Highlight the out of sync partition and press F3 to
manually start remirroring.
2. If the partitions still do not resynchronize, you must recreate the mirrored
set. Follow these steps.
Determine which disk partition has the data you want to save and mirror.
Delete the other disk partition(s).
Recreate new partitions in place of the ones you deleted.
Mirror the partition containing data to the new partitions.
For information about mirroring, see Mirroring and Duplexing Partitions.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Mirroring Takes a Long Time
If volumes are very large, mirroring can sometimes take several hours to
complete; this is normal. The following steps might help to speed the
mirroring process.
1. Are you using the latest disk drivers?
If not, update drivers to the latest versions.
2. Are there any problems related to the disk drive or drivers?
Refer to Novell’s Knowledgebase (http://support.novell.com) for known
problems with specific hard disks or drivers. Contact the drive vendor to
ask if there are ways to speed drive mirroring. For example, some drivers
can be loaded with parameters that speed the mirroring process.
3. Is read-after-write-verify turned on in MONITOR?
Most disk drives provide their own form of read-after-write verification;
therefore, NetWare’s read-after-write-verify feature is not needed and can
increase the time required to mirror partitions.
If the drive provides its own read-after-write verification, turn off readafter-write verify in MONITOR. See Contents > Server Disks and
Storage Devices > Optimizing > Turning Off Read-After-Write
Verification.
4. Is the Maximum Concurrent Disk Cache Writes parameter set to its
default value or lower?
If yes, try increasing the value of the parameter (MONITOR > Available
Options > Server Parameters > File Caching). However, increasing the
value too much can cause some disk controllers or drivers to fail. You
must experiment to see if a larger value speeds the mirroring process.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Mirroring Stops Just Before It Is Finished
Sometimes, the mirroring process proceeds without error, but stops at 99%
completion. Follow these steps to troubleshoot the problem.
1. Are there any faulty disk blocks?
Review Hot Fix statistics in MONITOR to see if any redirection blocks
have been used. Load MONITOR, then select Storage Devices > Hot
Fixed Partition; press Tab to expand the information window.
If there are faulty disk blocks, troubleshoot and replace the disk if
necessary.
2. Are the latest disk drivers loaded?
If not, update the drivers to the latest versions.
3. Are there any known problems with the brand or version of the hard disk?
Check Novell’s Knowledgebase (http://support.novell.com) for
information about known problems or contact the disk vendor.
4. Is the value of the Maximum Concurrent Disk Cache Writes parameter set
to its highest value?
If yes, reduce the value (MONITOR > Available Options > Server
Parameters > File Caching).
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
Server Console Hangs
If the server console locks up so that you cannot enter commands, but there is
no abend message on the screen, follow these steps to troubleshoot the
problem. If there is an abend message on the screen, see “Server Abends” on
page 139.
1. Can you toggle among console screens?
If yes, the problem might be caused by high server utilization. See “High
Utilization Statistics in MONITOR” on page 109.
If no, continue with the following steps.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
2. Can you break into the debugger by pressing Shift + Shift + Alt + Esc?
If yes, enter the debugger, then type q and press Enter to quit to DOS.
Reboot the server. Continue with the following steps to help find the
source of the problem.
If the console is locked, you can’t toggle among screens or exit to DOS,
and you can’t enter the debugger, contact Novell Technical Support or
your computer vendor to learn how to generate a non-maskable interrupt
to down the server.
3. Did the console hang as you unloaded an NLM?
If yes, the NLM is probably the source of the problem. Contact the NLM
vendor.
4. Are disk and LAN drivers the latest versions?
If not, update disk and LAN drivers. See “Keep Your Servers Patched” on
page 98 for information on NetWare 5 drivers.
5. Did the console hang after you mounted the last volume?
If yes, the LAN adapter might not be seated correctly or might not be
configured correctly. Check the board and its configuration and correct
any problems.
If the problem still occurs, follow the troubleshooting steps in “A
Troubleshooting Methodology” on page 98; review the Novell
Knowledgebase (http://support.novell.com); and contact a Novell Authorized
Reseller or Novell Technical Support.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Server Abends
See the following sections for a general overview of abends, and for
information about types of abends, resolving abends using the abend.log file,
and helping your server respond appropriately to abends:
Š “Overview” on page 139
Š “Types of Abends” on page 140
Š “Resolving Abends” on page 144
Š “Managing the Abend Auto Recovery Mechanism” on page 150
Overview
The NetWare operating system is very resilient, but errors can arise. Serious
problems are usually accompanied by abend (abnormal end) messages. When
an abend message appears, either NetWare or the CPU has detected a critical
error condition and started NetWare’s fault handler. NetWare abruptly halts
the server to protect it from data corruption. NetWare uses abends to ensure
the integrity of operating system data.
When a server abends, users can’t log into the server, workstations can’t read
from or write to the server, and an abend message usually appears on the
server console. If NetWare’s auto recovery mechanism is enabled, NetWare
might restart the server automatically or suspend the offending process,
depending on the nature of the abend. If NetWare suspends threads and restarts
the server, a number in brackets appears as part of the server console prompt
to indicate how many threads have been suspended.
If there is no abend message on the console, no abend.log file in sys:system,
and no number in brackets within the server console prompt, but users still
can’t access the server, see “Workstations Can’t Communicate with Server”
on page 122. If there is no abend message, but the console is frozen so that you
cannot enter commands, see “Server Console Hangs” on page 137.
When the server abends, it displays a message similar to the following:
Abend: SERVER-5.00-message_number message_string
ADDITIONAL INFORMATION: message
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
The Additional Information section states the probable cause of the abend. It
indicates where the problem occurred and gives the name of any NLM
associated with the abend. This information helps you determine how to
resolve the abend.
The abend message, along with additional information, is saved in the
abend.log file on the C: drive. As soon as the server is restarted, the abend.log
file is moved to sys:system.
Types of Abends
See the following sections for a discussion of various kinds of abends.
Š If the abend message includes processor exception in the text, then the
abend was detected by the CPU. See “Processor Exception Abends” on
page 140.
Š If the abend message does not include the words processor exception,
then the abend was detected by NetWare. See “Software Exceptions” on
page 142.
Š If the abend message mentions NMI parity error or machine check, the
abend was caused by a hardware problem. See “Hardware Abends” on
page 142.
Š If the abend message includes the text CPU Hog Detected by Timer, the
abend occurred because a process would not relinquish the CPU in the
time specified by the CPU Hog Timeout Amount server parameter. See
“CPU Hog Abends” on page 143.
If more than one kind of abend message appears, troubleshoot hardware
abends first.
Processor Exception Abends
When the abend is detected by the CPU, it is called a processor exception.
Processor exceptions always include the words processor exception in the
abend message, for example, Abend: Page Fault Processor Exception. A
processor exception occurs when the processor encounters an invalid address
or machine instruction.
The abend message displays additional information about the abend,
including the process running at the time of the abend. If the running process
is something other than the Server process, then knowing the process name is
very helpful. It is almost always related to the abend in some fashion.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
If the running process is an Interrupt Service Routine (ISR), then the problem
is usually caused by a LAN or disk driver. Sometimes it can be caused by a
hardware problem.
Remember these tips when resolving processor exception abends:
Š Troubleshoot LAN drivers and devices first. LAN drivers are easy to
replace and they are more likely to be the source of the problem than disk
drivers and disks.
Š Update drivers before devices. If the problem isn’t solved, then replace
devices.
Š Make sure you use up-to-date drivers for NetWare 5. Use drivers that ship
in the Novell red box, drivers that are tested, certified, and made available
on Novell’s Support Connection (http://support.novell.com), and
certified drivers released by hardware vendors. Check product
certification bulletins on DeveloperNet (http://developer.novell.com) to
learn which versions are most recent.
Do not mix the use of a driver written to the 3.3x/1.1x ODI Specification
with 3.2 ODI Specification support modules. The LAN driver
certification bulletin will list the specification a LAN driver is written to.
In some cases, you will need to contact the IHV (Independent Hardware
Vendor) who supplied the driver.
NetWare 5 drivers for storage devices and host bus adapters are written to
the NWPA (NetWare Peripheral ArchitectureTM) Specification. The .dsk
drivers are no longer supported.
Š Simplify the server by removing software. If the server is stable when
software has been removed, then start reloading software one module at
a time until the problem recurs.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Software Exceptions
Abends detected by NetWare are called software exceptions. NetWare
continually monitors the status of various server activities to ensure proper
operation. If NetWare detects a condition that threatens the integrity of its data
(such as an invalid parameter being passed in a function call or certain
hardware errors), it abruptly halts the active process and displays an Abend
message on the screen. (See Resolving Critical Server Issues, AppNotesTM,
February 1995, at http://developer.novell.com/research.)
Thus, abends are NetWare’s way to ensure the integrity of its data and protect
against data corruption.
To troubleshoot a software exception, simplify the server by removing
software. If the server is stable when software has been removed, then start
reloading software one module at a time until the problem recurs
Hardware Abends
Hardware abends are either non-maskable interrupt (NMI) errors or machine
check errors.
There are two kinds of NMI error messages:
Š Abend: NMI parity error generated by IO check. The source of an IO
check abend could be anywhere in the server.
Š Abend: NMI parity error generated by system board. The source of a
system board abend could be anywhere in the system, but is most likely
to be in the system board or its memory.
If your server experiences a hardware abend, run any diagnostic programs
your hardware provides to help locate the source of the problem.
Try swapping hardware components to see whether the problem disappears
when a particular component is replaced. Swap components in this order:
1. Memory
2. Power supply
3. System board
The following hardware problems have been found to generate errors:
Š Faulty RAM
Š Faulty system board
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Faulty I/O cards, especially cards with on-board memory
Š Low or fluctuating power at the power source; power source or
Uninterruptible Power Supply going bad
Š Faulty memory extension boards
Š System board memory that is mismatched in speed or brand
Š Conflicting interrupts
Š Cards, cables, and memory modules that are dirty or not seated properly
Š Incompatibility among hardware pieces
Š Static electric discharge
Š Occasionally, a hard drive or a printer
CPU Hog Abends
CPU Hog abends occur when a process does not relinquish control of the CPU
in the time specified by the CPU Hog Timeout Amount server parameter.
When this happens, the CPU Hog Timeout process initiates the abend auto
recovery mechanism to suspend the thread. Depending on how the auto
recovery parameters are set, the server might be restarted, thus freeing the
CPU.
Review the abend message or the abend.log file to determine the running
process at the time of the abend. When a CPU Hog abend happens, the running
process shown in the abend messages indicates exactly where in the code the
problem occurred. Try removing the NLM indicated by the running process.
If this prevents the abend from recurring, then contact the vendor of the NLM.
In some cases, the process controlling CPU cycles simply needs more time to
finish. In this case, set the CPU Hog Timeout Amount parameter to 0 to
disable it and thus allow the process to finish running (MONITOR Available
Options > Server Parameters > Miscellaneous). For example, CPU Hog
abends sometimes occur when a particularly large volume is mounted. When
the process has finished running, set the CPU Hog Timeout Amount parameter
back to its original value.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Resolving Abends
Because abends can be caused by so many different conditions, there is no one
way to pinpoint and resolve the problem. Follow the general guidelines in this
section, and use tools such as the abend.log file to help you diagnose the
problem.
See Novell’s Knowledgebase (http://support.novell.com) for more detailed
troubleshooting information and for information about specific abends that
have been reported to Novell.
In some cases, the best way to solve an abend is to have Novell Technical
Support analyze a copy of the server’s memory, called a core dump. Before
sending a core dump to Novell you must contact Novell Technical Support and
open a support incident. A support engineer will tell you whether a core dump
is necessary, the kind of core dump to perform, and how to send the core dump
to Novell. For information on creating core dumps, see “Creating a Core
Dump” on page 105.
Preventive Steps
You can prevent many abends by taking the following steps:
Š Always keep the server patched. Many problems your server might
experience have already been solved by Novell and provided in the form
of a patch file. See“Applying Patches” on page 40.
Š Update LAN and disk drivers to their newest versions. Many abends can
be avoided simply by keeping all drivers up to date.
Š Update outdated NLM programs. The most recent versions of Novell
modules can be downloaded from Novell’s technical support web site,
http://support.novell.com, or from other locations where you download
patches. Check with other vendors to make sure you have the most recent
versions of third-party NLM programs.
Š Virus scan the DOS partition on the server.
Š Clean and reseat cables and adapters, make sure cables are terminated
correctly, and SCSI IDs are correct. Make sure all fans are operating.
Š Take precautions to prevent static electric discharge. Static discharge can
cause computer systems to degrade over time. See “Preventing Static
Electricity Problems” on page 92 for more information.
Š Review Novell’s Knowledgebase on the World Wide Web at http://
support.novell.com for known problems.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
General Troubleshooting Tips
Consider the following as you diagnose the problem:
Š What is the physical environment of the server?
Is the air conditioning off when the computer abends? Is the environment
dry, dusty, or hot? Excessive heat can cause server problems and increase
static electricity.
Is there any new electro magnetic force (EMF) near the server or cabling?
Large motors, transmitters, fluorescent lights, and even vacuum cleaners
can cause problems.
Has the computer been handled with appropriate static protection?
Excessive static can cause computer hardware to degrade over time. See
“Preventing Static Electricity Problems” on page 92.
Is the power source consistent? If the power source is filtered, is the
filtering still functioning?
Š What happens at the same time as the abend?
What resources are being used when the server abends, such as printing,
backup software, databases?
Look for consistency: Are there certain activities, applications, or server
processes that are always present when the server abends? If these
elements are removed, does the server function normally? Does the abend
always happen at the same time?
Š What has changed since the server started to abend?
Has traffic increased? Has new software or hardware been installed?
Have there been power outages? Have there been changes to the LAN?
Š What other errors or problems are occurring at the same time?
Are users dropping connections? Are files corrupted? Have drives
deactivated? Are there printing problems?
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š Do server statistics indicate a problem?
Review the sys$log.err and abend.log files. See “Resolving Abends
Using the Abend.log File” on page 146.
Load MONITOR and look for high server utilization, packet receive
buffers that have reached their maximum, no ECB available counts, cache
buffers lower than 60 - 70%, LRU sitting time less than 15 minutes during
peak traffic, LAN errors, and service processes that have reached their
maximum.
If the source of the abend proves to be a third-party product, please contact the
product vendor, even if the product is one that ships with NetWare, such as a
driver. Only the product vendor can supply the patch or replacement you need.
Resolving Abends Using the Abend.log File
The following information is excerpted from Troubleshooting Server
Problems Using the Abend.log File and Memory Images (Core Dumps) in
AppNotes, October 1997 (http://developer.novell.com/research). This
information will help you pinpoint the source of an bend.
When an abend or lockup occurs, NetWare writes a summary of the state of
the server to a file called abend.log on the C: drive. After the server is
restarted, the abend.log information is appended to a file with the same name
in the sys:system directory.
The abend.log file contains the following information:
Š File server name
Š Date and time of the abend
Š Text of the abend message
Š Contents of the registers on the server
Š Name of the module or process that was suspended
Š Name of the process that was running at the time of the abend
Š Stack pointer and limit
Š Stack trace
Š Modules list
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Server Name
The server name is the first piece of data saved in the abend.log file. The name
of the server can be useful if you have multiple servers. It is important to know
which servers experience the abend and which do not. Many times a simple
comparison between a server that is experiencing an abend and one that is not
provides valuable clues to the cause of the abend.
For example, suppose you have two identical servers, one of which abends
while the other one does not. The only difference between the servers is that
the one that abends is used for backups. In this case, the logical component to
focus your troubleshooting efforts on is the backup software and hardware.
Date and Time of Abend
Because of the auto recovery process in NetWare, administrators might not
notice that a server has experienced an abend, even when the error has
occurred several times. But notice the server console prompt. If a number in
brackets appears as part of the prompt, then the server has abended and faulty
threads have been suspended. The number indicates how many threads have
been suspended.
Looking at the abend.log file might be the only way to tell what kinds of
abends happened to the server . Many times a pattern in the date or time of the
abend will provide an important clue to the cause of the abend. For example,
suppose you notice that a particular server has experienced the same abend
five times, and every time the abend occurred at midnight. The logical starting
place would be processes that run at midnight, such as compression or a
backup.
Abend Message
Earlier versions of NetWare produced generic messages such as Page Fault
Processor Exception. This describes an entire class of errors that can be caused
by many different things. The abend.log file, on the other hand, contains more
specific abend messages, such as Free detected modified memory beyond the
end of the cell being returned. This message indicates that the running process
tried to free memory that had been overwritten.
Often a combination of the abend message, the running process, and the stack
trace will be sufficient to enable you to find a solution from the Novell Support
Connection web site (http://support.novell.com) or CD-ROM.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Registers
The registers are most helpful when you can establish some sort of pattern
from the contents of the registers in several abends. If the contents of the
registers (especially the EIP register) are the same in each abend, it usually
means the operating system followed the same code path to arrive at the
abend. This indicates a software bug or possibly a corrupt file. Many
resolutions to such problems are available from the Novell Support
Connection or CD-ROM. Search for the running process and the NLM that
triggered the abend.
ABENDed NLM
The ABENDed NLM is the module that owns the code that actually triggered
the abend. However, this does not necessarily mean that this module caused
the abend. A function in the named NLM might have been passed a bad value
from another module. Knowing what module was active when the server
experienced the abend is useful to identify which NLM or third-party product
to troubleshoot. It is also very helpful in looking up known resolutions from
the Novell Support Connection. The description of the resolutions usually
includes the name of the module and often the running process, the abend
message, register values, and functions from the stack trace.
Running Process
If the running process is the Server process, the abend could literally be caused
by any application. Any module can make a request to the NetWare operating
system, and the result of that request will show up as a Server process. If the
running process is something other than the Server process, knowing the name
of the process is usually quite helpful. Often the process belongs to the module
that is causing the abend. Almost always it is related to the abend in some
fashion. For example, suppose your server abends, and the running process is
identified as TCP/IP. A logical troubleshooting step would be to try to obtain
a protocol analyzer trace during such an abend and examine the TCP/IP
packets that are being sent to the server.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Stack Limit and Stack Pointer
The stack limit and stack pointer are used to determine if server memory
corruption has taken place.
The stack is an area of memory set aside for the temporary storage of values.
The stack limit is simply the smallest value the stack of the running process
can reach. If the stack pointer is at a value lower than the stack limit, or if it is
greater than the stack limit plus the stack size (3000 for server processes, other
processes can have different stack sizes), then the current stack pointer (ESP)
is incorrect, and the server has experienced memory corruption.
Stack Trace
The stack trace is a printout of the contents of the stack, one value at a time. If
the value is an address located inside a module, the module name, function (if
available), and offset are printed to the right side of the address. The stack
trace values that are of concern are those that fall inside a module.
Once again, a pattern in the stack trace of several abends will provide clues to
the possible cause of the abend. If the pattern shows that the stack trace of
several abends is exactly the same, there is likely a code path to the abend, and
there is a good chance that a fix is already available. Search the Novell
Support Connection (http://support.novell.com) under the abended NLM
name, the running process, and the function to find possible resolutions.
Modules List
Many problems that initially existed in the NetWare operating system and in
the Novell® products that run on NetWare have already been fixed by Novell.
These fixes are readily available from Novell.
The first thing to do if your server experiences abends is to look at the modules
list and compare it to a list of the latest released versions from Novell. If you
see modules that are outdated, update them, especially if one of them is the
abended NLM. The modules list is complete with version numbers and dates
to make it easy to tell which revisions were running on the server when it
experienced the abend.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Managing the Abend Auto Recovery Mechanism
When a server abends, you can respond to the abend manually or have the
server respond automatically.
When you respond manually, the server determines the nature of the abend
and displays the appropriate response option on the screen, along with
additional options for bringing down the server or executing a core dump. You
must execute an option to respond to the abend.
When the server responds automatically, it executes the appropriate response
without intervention.
Three server parameters control the Abend Auto Recovery Mechanism. How
to use these parameters is described in the following sections. You can find
these parameters in MONITOR, as follows:
Š Auto Restart After Abend: MONITOR > Available Options > Server
Parameters > Error Handling
Š Auto Restart After Abend Delay Time: MONITOR > Available Options
> Server Parameters > Error Handling
Š Developer Option: MONITOR > Available Options > Server Parameters
> Miscellaneous
,03257$17Sometimes an abend (or a faulty NLM) can cause the server console to
hang (stop functioning). In this case, the abend message is not displayed and you
cannot enter commands at the console prompt.
If this happens, press <Ctrl>+<Alt>+<Esc>. The server displays a menu; select the first
option, Down the File Server and Exit to DOS. The DOS prompt is displayed.
Responding to the Abend Manually
For information about automatic response to abends, see “Responding to the
Abend Automatically” on page 153.
To respond manually to abends, change either of the following SET
parameters to the values shown :
AUTO RESTART AFTER ABEND = 0
DEVELOPER OPTION = ON
Set Auto Restart After Abend in MONITOR > Available Options > Server
Parameters > Error Handling; set Developer Option in MONITOR > Available
Options > Server Parameters > Miscellaneous.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
When an abend occurs, the server displays a short list of options appropriate
to the nature of the abend. To respond to the abend, you must execute one of
the options by typing the first letter of the option.
The following options might be displayed. Note that several of the options
have the same first letter (such as R, S, or X). In a given abend situation, the
option list will include only one option for any given first letter.
Š S=Suspend the running process, update abend.log,
and attempt to down the server.
This option appears if the abend was software-detected, that is, detected
by NetWare. It is important to save files, shut down the server, and try to
solve the problem that caused the abend. Review the abend.log file to help
determine the source of the problem.
When you execute this option, the server stops the running process,
updates the abend.log file, and sends a message every two minutes to
users advising them to save their files and log out. The server then
attempts to shut down and restart the computer.
The amount of time before the server shuts down and restarts is
determined by the parameter Auto Restart After Abend Delay Time
(MONITOR > Available Options > Server Parameters > Error Handling).
You can set this value from 2 to 60 minutes.
Š R=Resume the running process, update abend.log,
and attempt to down the server
This option appears if the abend was an NMI (non-maskable interrupt) or
machine check error, indicating a parity error or a machine check
processor exception. It is important to save files, shut down the server,
and solve the problem causing the abend. Review the abend.log file to
help determine the source of the problem.
When you execute this option, the server sends a message every two
minutes to users advising them to save their files and log out. The server
then resumes the running process, updates the abend.log file, and
attempts to shut down and restart the computer.
The amount of time before the server shuts down and restarts is
determined by the parameter Auto Restart After Abend Delay Time
(MONITOR > Available Options > Server Parameters > Error Handling).
You can set this value from 2 to 60 minutes.
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š S=Suspend the running process and update
abend.log
This option appears if the abend was hardware-detected, that is, detected
by the processor. All hardware-detected abends have the words processor
exception in the abend message.
Common abends include page faults, general protection faults, and
invalid op codes, but there can be many others. When this option is
available, the server has determined that it cannot return the process to a
safe state, but that it does not need to shut down the computer
immediately to resolve the problem. You might still need to shut down the
computer and restart it at a later time.
When you execute this option, the server suspends the currently running
process, updates the abend.log file, but does not shut down the computer.
Server performance may be poor, because a loaded NLM is probably
malfunctioning.
Read the Additional Information part of the abend message to learn which
NLM might be causing the problem. Wait until a convenient time, then
shut down the server and restart it. Examine the abend.log file for more
information about the source of the problem.
Š S=Return the running process to a safe state and
update the abend.log file
Like the previous option, this option appears if the abend was hardwaredetected, that is, detected by the processor. All hardware-detected abends
have the words processor exception in the abend message.
These abends include page faults, protection faults, and invalid op codes.
When this option is listed, the server has determined that it can return the
process to a safe state.
When you execute this option, the server returns the running process to a
safe state and updates the abend.log file, but it does not shut down the
computer. In most cases, the machine completely recovers and no further
action is necessary. This option resolves most page fault abends.
Š Y=Copy diagnostic image to disk
Execute this option to perform a core dump that can be examined to
determine the cause of an abend.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Š X=Restart
This option appears only if DOS has been removed. Execute this option
if you want to restart the server.
127(If DOS has been removed, the server will not create or update an
abend.log file.
Š X=Update abend.log and then exit
Execute this option if you want to bring down the server and exit to DOS.
If you power off the server without first executing one of the S or R
options to resolve the abend, the server will not update the abend.log file.
Š Otherwise, power off and back on to restart
If the console has been secured, you must power off and then back on to
restart the server. If you power off the server without first executing one
of the S or R options to resolve the abend, the server will not update the
abend.log file.
127(When the server restarts, it moves the abend.log file from the DOS partition to
sys:system.
Responding to the Abend Automatically
You can require the server to respond automatically to abends. Two automatic
responses are possible.
Š When both of the following parameters are set to the values shown, the
server responds to the abend automatically by evaluating the source of the
abend and executing the appropriate S or R option described in the
previous section.
AUTO RESTART AFTER ABEND = 1
DEVELOPER OPTION = OFF
Š When both of the following parameters are set to the value shown, the
server responds to the abend by attempting to recover. If the server cannot
recover, the auto recovery process waits for the restart time to elapse, and
then downs the server and attempts to restart it.
AUTO RESTART AFTER ABEND = 2
DEVELOPER OPTION = OFF
Troubleshooting NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Use the following parameter to specify how long the server waits after an
abend before attempting to shut down and restart the computer:
AUTO RESTART AFTER ABEND DELAY TIME = minutes
Set Auto Restart After Abend and Auto Restart After Abend Delay Time in
MONITOR > Available Options > Server Parameters > Error Handling.
Set Developer Option in MONITOR > Available Options > Server Parameters
> Miscellaneous.
All parameters can be saved in the startup.ncf file.
,03257$17Because the server responds to the abend automatically, you might not
be aware that an abend occurred. Periodically check the abend.log file or the Server Up
Time statistic on the General Information screen of MONITOR to determine if there have
been any abends.
Preventive Maintenance Tasks
The following sections provide tips for preventing server problems:
Š “Monitoring the NetWare Server” on page 66
Š “Matching Performance Capabilities” on page 40
Š “Preventing Memory Problems” on page 91
Š “Preventing Power Supply Errors” on page 91
Š “Preventing Static Electricity Problems” on page 92
Š Preventing Cabling Problems
Š “Documenting the Network” on page 89
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
$
Rev 99a
17
28 October 99
Novell Trademarks
Access Manager is a registered trademark of Novell, Inc. in the United States
and other countries.
Advanced NetWare is a trademark of Novell, Inc.
AlarmPro is a registered trademark of Novell, Inc. in the United States and
other countries.
AppNotes is a registered service mark of Novell, Inc. in the United States and
other countries.
AppNotes is a registered service mark of Novell, Inc. in the United States and
other countries.
AppTester is a registered service mark of Novell, Inc. in the United States and
other countries.
BrainShare is a registered service mark of Novell, Inc. in the United States and
other countries.
C-Worthy is a trademark of Novell, Inc.
C3PO is a trademark of Novell, Inc.
CBASIC is a registered trademark of Novell, Inc. in the United States and
other countries.
Certified NetWare Administrator in Japanese and CNA-J are service marks of
Novell, Inc.
Certified NetWare Engineer in Japanese and CNE-J are service marks of
Novell, Inc.
Certified NetWare Instructor in Japanese and CNI-J are service marks of
Novell, Inc.
Certified Novell Administrator and CNA are service marks of Novell, Inc.
Certified Novell Engineer is a trademark and CNE is a registered service mark
of Novell, Inc. in the United States and other countries.
Certified Novell Salesperson is a trademark of Novell, Inc.
Novell Trademarks NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Client 32 is a trademark of Novell, Inc.
ConnectView is a registered trademark of Novell, Inc. in the United States and
other countries.
Connectware is a registered trademark of Novell, Inc. in the United States and
other countries.
Corsair is a registered trademark of Novell, Inc. in the United States and other
countries.
CP/Net is a registered trademark of Novell, Inc. in the United States and other
countries.
Custom 3rd-Party Object and C3PO are trademarks of Novell, Inc.
DeveloperNet is a registered trademark of Novell, Inc. in the United States and
other countries.
Documenter’s Workbench is a registered trademark of Novell, Inc. in the
United States and other countries.
ElectroText is a trademark of Novell, Inc.
Enterprise Certified Novell Engineer and ECNE are service marks of Novell,
Inc.
Envoy is a registered trademark of Novell, Inc. in the United States and other
countries.
EtherPort is a registered trademark of Novell, Inc. in the United States and
other countries.
EXOS is a trademark of Novell, Inc.
Global MHS is a trademark of Novell, Inc.
Global Network Operations Center and GNOC are service marks of Novell,
Inc.
Graphics Environment Manager and GEM are registered trademarks of
Novell, Inc. in the United States and other countries.
GroupWise is a registered trademark of Novell, Inc. in the United States and
other countries.
GroupWise XTD is a trademark of Novell, Inc.
Hardware Specific Module is a trademark of Novell, Inc.
Hot Fix is a trademark of Novell, Inc.
InForms is a trademark of Novell, Inc.
Instructional Workbench is a registered trademark of Novell, Inc. in the
United States and other countries.
Internetwork Packet Exchange and IPX are trademarks of Novell, Inc.
IPX/SPX is a trademark of Novell, Inc.
IPXODI is a trademark of Novell, Inc.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
IPXWAN is a trademark of Novell, Inc.
LAN WorkGroup is a trademark of Novell, Inc.
LAN WorkPlace is a registered trademark of Novell, Inc. in the United States
and other countries.
LAN WorkShop is a trademark of Novell, Inc.
LANalyzer is a registered trademark of Novell, Inc. in the United States and
other countries.
LANalyzer Agent is a trademark of Novell, Inc.
Link Support Layer and LSL are trademarks of Novell, Inc.
MacIPX is a registered trademark of Novell, Inc. in the United States and
other countries.
ManageWise is a registered trademark of Novell, Inc. in the United States and
other countries.
Media Support Module and MSM are trademarks of Novell, Inc.
Mirrored Server Link and MSL are trademarks of Novell, Inc.
Mobile IPX is a trademark of Novell, Inc.
Multiple Link Interface and MLI are trademarks of Novell, Inc.
Multiple Link Interface Driver and MLID are trademarks of Novell, Inc.
My World is a registered trademark of Novell, Inc. in the United States and
other countries.
N-Design is a registered trademark of Novell, Inc. in the United States and
other countries.
Natural Language Interface for Help is a trademark of Novell, Inc.
NDS Manager is a trademark of Novell, Inc.
NE/2 is a trademark of Novell, Inc.
NE/2-32 is a trademark of Novell, Inc.
NE/2T is a trademark of Novell, Inc.
NE1000 is a trademark of Novell, Inc.
NE1500T is a trademark of Novell, Inc.
NE2000 is a trademark of Novell, Inc.
NE2000T is a trademark of Novell, Inc.
NE2100 is a trademark of Novell, Inc.
NE3200 is a trademark of Novell, Inc.
NE32HUB is a trademark of Novell, Inc.
NEST Autoroute is a trademark of Novell, Inc.
NetExplorer is a trademark of Novell, Inc.
Novell Trademarks NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
NetNotes is a registered trademark of Novell, Inc. in the United States and
other countries.
NetSync is a trademark of Novell, Inc.
NetWare is a registered trademark of Novell, Inc. in the United States and
other countries.
NetWare 3270 CUT Workstation is a trademark of Novell, Inc.
NetWare 3270 LAN Workstation is a trademark of Novell, Inc.
NetWare 386 is a trademark of Novell, Inc.
NetWare Access Server is a trademark of Novell, Inc.
NetWare Access Services is a trademark of Novell, Inc.
NetWare Application Manager is a trademark of Novell, Inc.
NetWare Application Notes is a trademark of Novell, Inc.
NetWare Asynchronous Communication Services and NACS are trademarks
of Novell, Inc.
NetWare Asynchronous Services Interface and NASI are trademarks of
Novell, Inc.
NetWare Aware is a trademark of Novell, Inc.
NetWare Basic MHS is a trademark of Novell, Inc.
NetWare BranchLink Router is a trademark of Novell, Inc.
NetWare Care is a trademark of Novell, Inc.
NetWare Communication Services Manager is a trademark of Novell, Inc.
NetWare Connect is a registered trademark of Novell, Inc. in the United
States.
NetWare Core Protocol and NCP are trademarks of Novell, Inc.
NetWare Distributed Management Services is a trademark of Novell, Inc.
NetWare Document Management Services is a trademark of Novell, Inc.
NetWare DOS Requester and NDR are trademarks of Novell, Inc.
NetWare Enterprise Router is a trademark of Novell, Inc.
NetWare Express is a registered service mark of Novell, Inc. in the United
States and other countries.
NetWare Global Messaging and NGM are trademarks of Novell, Inc.
NetWare Global MHS is a trademark of Novell, Inc.
NetWare HostPrint is a registered trademark of Novell, Inc. in the United
States.
NetWare IPX Router is a trademark of Novell, Inc.
NetWare LANalyzer Agent is a trademark of Novell, Inc.
NetWare Link Services Protocol and NLSP are trademarks of Novell, Inc.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
NetWare Link/ATM is a trademark of Novell, Inc.
NetWare Link/Frame Relay is a trademark of Novell, Inc.
NetWare Link/PPP is a trademark of Novell, Inc.
NetWare Link/X.25 is a trademark of Novell, Inc.
NetWare Loadable Module and NLM are trademarks of Novell, Inc.
NetWare LU6.2 is trademark of Novell, Inc.
NetWare Management Agent is a trademark of Novell, Inc.
NetWare Management System and NMS are trademarks of Novell, Inc.
NetWare Message Handling Service and NetWare MHS are trademarks of
Novell, Inc.
NetWare MHS Mailslots is a registered trademark of Novell, Inc. in the United
States and other countries.
NetWare Mirrored Server Link and NMSL are trademarks of Novell, Inc.
NetWare Mobile is a trademark of Novell, Inc.
NetWare Mobile IPX is a trademark of Novell, Inc.
NetWare MultiProtocol Router and NetWare MPR are trademarks of Novell,
Inc.
NetWare MultiProtocol Router Plus is a trademark of Novell, Inc.
NetWare Name Service is trademark of Novell, Inc.
NetWare Navigator is a trademark of Novell, Inc.
NetWare Peripheral Architecture is a trademark of Novell, Inc.
NetWare Print Server is a trademark of Novell, Inc.
NetWare Ready is a trademark of Novell, Inc.
NetWare Requester is a trademark of Novell, Inc.
NetWare Runtime is a trademark of Novell, Inc.
NetWare RX-Net is a trademark of Novell, Inc.
NetWare SFT is a trademark of Novell, Inc.
NetWare SFT III is a trademark of Novell, Inc.
NetWare SNA Gateway is a trademark of Novell, Inc.
NetWare SNA Links is a trademark of Novell, Inc.
NetWare SQL is a trademark of Novell, Inc.
NetWare Storage Management Services and NetWare SMS are trademarks of
Novell, Inc.
NetWare Telephony Services is a trademark of Novell, Inc.
NetWare Tools is a trademark of Novell, Inc.
NetWare UAM is a trademark of Novell, Inc.
NetWare WAN Links is a trademark of Novell, Inc.
Novell Trademarks NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
NetWare/IP is a trademark of Novell, Inc.
NetWire is a registered service mark of Novell, Inc. in the United States and
other countries.
Network Navigator is a registered trademark of Novell, Inc. in the United
States.
Network Navigator - AutoPilot is a registered trademark of Novell, Inc. in the
United States and other countries.
Network Navigator - Dispatcher is a registered trademark of Novell, Inc. in the
United States and other countries.
Network Support Encyclopedia and NSE are trademarks of Novell, Inc.
Network Support Encyclopedia Professional Volume and NSEPro are
trademarks of Novell, Inc.
NetWorld is a registered service mark of Novell, Inc. in the United States and
other countries.
Novell is a service mark and a registered trademark of Novell, Inc. in the
United States and other countries.
Novell Alliance Partners Program is a collective mark of Novell, Inc.
Novell Application Launcher is a trademark of Novell, Inc.
Novell Authorized CNE is a trademark and service mark of Novell, Inc.
Novell Authorized Education Center and NAEC are service marks of Novell,
Inc.
Novell Authorized Partner is a service mark of Novell, Inc.
Novell Authorized Reseller is a service mark of Novell, Inc.
Novell Authorized Service Center and NASC are service marks of Novell,
Inc.
Novell BorderManager is a trademark of Novell, Inc.
Novell BorderManager FastCache is a trademark of Novell, Inc.
Novell Client is a trademark of Novell, Inc.
Novell Corporate Symbol is a trademark of Novell, Inc.
Novell Customer Connections is a registered trademark of Novell, Inc. in the
United States.
Novell Directory Services and NDS are registered trademarks of Novell, Inc.
in the United States and other countries.
Novell Distributed Print Services is a trademark and NDPS is a registered
trademark of Novell, Inc. in the United States and other countries.
Novell ElectroText is a trademark of Novell, Inc.
Novell Embedded Systems Technology is a registered trademark and NEST is
a trademark of Novell, Inc. in the United States and other countries.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Novell Gold Authorized Reseller is a service mark of Novell, Inc.
Novell Gold Partner is a service mark of Novell, Inc.
Novell Labs is a trademark of Novell, Inc.
Novell N-Design is a registered trademark of Novell, Inc. in the United States
and other countries.
Novell NE/2 is a trademark of Novell, Inc.
Novell NE/2-32 is a trademark of Novell, Inc.
Novell NE3200 is a trademark of Novell, Inc.
Novell Network Registry is a service mark of Novell, Inc.
Novell Platinum Partner is a service mark of Novell, Inc.
Novell Press is a trademark of Novell, Inc.
Novell Press Logo (teeth logo) is a registered trademark of Novell, Inc. in the
United States and other countries.
Novell Replication Services is a trademark of Novell, Inc.
Novell Research Reports is a trademark of Novell, Inc.
Novell RX-Net/2 is a trademark of Novell, Inc.
Novell Service Partner is a trademark of Novell, Inc.
Novell Storage Services is a trademark of Novell, Inc.
Novell Support Connection is a registered trademark of Novell, Inc. in the
United States and other countries.
Novell Technical Services and NTS are service marks of Novell, Inc.
Novell Technology Institute and NTI are registered service marks of Novell,
Inc. in the United States and other countries.
Novell Virtual Terminal and NVT are trademarks of Novell, Inc.
Novell Web Server is a trademark of Novell, Inc.
Novell World Wide is a trademark of Novell, Inc.
NSE Online is a service mark of Novell, Inc.
NTR2000 is a trademark of Novell, Inc.
Nutcracker is a registered trademark of Novell, Inc. in the United States and
other countries.
OnLAN/LAP is a registered trademark of Novell, Inc. in the United States and
other countries.
OnLAN/PC is a registered trademark of Novell, Inc. in the United States and
other countries.
Open Data-Link Interface and ODI are trademarks of Novell, Inc.
Open Look is a registered trademark of Novell, Inc. in the United States and
other countries.
Novell Trademarks NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Open Networking Platform is a registered trademark of Novell, Inc. in the
United States and other countries.
Open Socket is a registered trademark of Novell, Inc. in the United States.
Packet Burst is a trademark of Novell, Inc.
PartnerNet is a registered service mark of Novell, Inc. in the United States and
other countries.
PC Navigator is a trademark of Novell, Inc.
PCOX is a registered trademark of Novell, Inc. in the United States and other
countries.
Perform3 is a trademark of Novell, Inc.
Personal NetWare is a trademark of Novell, Inc.
Pervasive Computing from Novell is a registered trademark of Novell, Inc. in
the United States and other countries.
Portable NetWare is a trademark of Novell, Inc.
Presentation Master is a registered trademark of Novell, Inc. in the United
States and other countries.
Print Managing Agent is a trademark of Novell, Inc.
Printer Agent is a trademark of Novell, Inc.
QuickFinder is a trademark of Novell, Inc.
Red Box is a trademark of Novell, Inc.
Reference Software is a registered trademark of Novell, Inc. in the United
States and other countries.
Remote Console is a trademark of Novell, Inc.
Remote MHS is a trademark of Novell, Inc.
RX-Net is a trademark of Novell, Inc.
RX-Net/2 is a trademark of Novell, Inc.
ScanXpress is a registered trademark of Novell, Inc. in the United States and
other countries.
Script Director is a registered trademark of Novell, Inc. in the United States
and other countries.
Sequenced Packet Exchange and SPX are trademarks of Novell, Inc.
Service Response System is a trademark of Novell, Inc.
Serving FTP is a trademark of Novell, Inc.
SFT is a trademark of Novell, Inc.
SFT III is a trademark of Novell, Inc.
SoftSolutions is a registered trademark of SoftSolutions Technology
Corporation, a wholly owned subsidiary of Novell, Inc.
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
Rev 99a
28 October 99
Software Transformation, Inc. is a registered trademark of Software
Transformation, Inc., a wholly owned subsidiary of Novell, Inc.
SPX/IPX is a trademark of Novell, Inc.
StarLink is a registered trademark of Novell, Inc. in the United States and
other countries.
Storage Management Services and SMS are trademarks of Novell, Inc.
Technical Support Alliance and TSA are collective marks of Novell, Inc.
The Fastest Way to Find the Right Word is a registered trademark of Novell,
Inc. in the United States and other countries.
The Novell Network Symbol is a trademark of Novell, Inc.
Topology Specific Module and TSM are trademarks of Novell, Inc.
Transaction Tracking System and TTS are trademarks of Novell, Inc.
Universal Component System is a registered trademark of Novell, Inc. in the
United States and other countries.
Virtual Loadable Module and VLM are trademarks of Novell, Inc.
Writer’s Workbench is a registered trademark of Novell, Inc. in the United
States and other countries.
Yes, It Runs with NetWare (logo) is a trademark of Novell, Inc.
Yes, NetWare Tested and Approved (logo) is a trademark of Novell, Inc.
ZENworks is a trademark of Novell, Inc.
Novell Trademarks NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
doc_tpl.fm
NetWare Server Operating System Administration Guide
NetWare Server Operating System Administration Guide
104-001241-001
December 17, 1999
Novell Confidential
Rev 99a
28 October 99
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising