B PDG Master

Tivoli Tivoli Storage Manager
Version 5.5
Problem Determination Guide
SC32-0142-01
Tivoli Tivoli Storage Manager
Version 5.5
SC32-0142-01
Note! Before using this information and the product it supports, read the general information in the Notices appendix.
Edition notice This edition applies to Version 5 Release 5 of the IBM Tivoli Storage Manager Problem Determination Guide (program numbers 5608-ACS, 5608-APD, 5608-APE, 5608-APH, 5608-APR, 5608-APW, 5608-CSS, 5608-HSM, 5608-ISM, 5608-ISX, 5608-SAN, 5698-A11, 5698-A12, 5698-A13, 5698-A25, 5698-USS) and to any subsequent releases until otherwise indicated in new editions or technical newsletters. Changes since the previous edition are marked with a vertical bar (|) in the left margin. Ensure that you are using the correct edition for the level of the product. Copyright International Business Machines Corporation 1993, 2007. All rights reserved. US Government Users Restricted Rights Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.
Contents
Preface . . . . . . . . . . . . . . vii Who should read this guide . . . . . . ix Publications. . . . . . . . . . . . . xi
Tivoli Storage Manager publications . . . . . . xi
| |
Support information . . . . . . . . . xv
Getting technical training . . . . . . . . . . xv Searching knowledge bases . . . . . . . . . xv Searching the Internet . . . . . . . . . . xv Using IBM Support Assistant . . . . . . . xv Finding product fixes . . . . . . . . . . xvi Getting E-mail notification of product fixes . . xvi Contacting IBM Software Support . . . . . . xvi Determine the business impact . . . . . . xvii Describe your problem and gather background information . . . . . . . . . . . . . xvii Submit your problem to IBM Software Support xvii Accessibility features . . . . . . . . . . xviii
Chapter 1. Resolving client problems . . 1

Client diagnostic tips . . . . . . . . . . Examining error messages . . . . . . . . Examining the server activity log messages . . Identifying when and where the problem can occur . . . . . . . . . . . . . . . Reproducing the problem . . . . . . . . Collecting documentation to resolve problems with the client application . . . . . . . . . . dsmc/dsmadmc/dsmj/... will not start . . . . Client option sets . . . . . . . . . . . . Authenticating the Client . . . . . . . . . Client scheduling . . . . . . . . . . . . Including or excluding client files during backup processing . . . . . . . . . . . . . . Excluding files automatically from backup processing . . . . . . . . . . . . . Excluding Windows system files . . . . . Excluding files with the EXCLUDE.DIR statement . . . . . . . . . . . . . Using include/exclude statements in the client option set . . . . . . . . . . . . . Dictating incremental copy frequency through a server policy . . . . . . . . . . . . Compression, encryption, and subfile backup statements . . . . . . . . . . . . . Using platform-specific include/exclude statements . . . . . . . . . . . . . Resolving errors due to the incorrectly coded include/exclude list . . . . . . . . . Resolving image backup errors . . . . . . . Resolving Linux image backup errors. . . . . 1 . 1 . 1 . 1 . 2 . . . . . 2 3 4 7 9
. 12 . 12 . 12 . 13 . 14 . 15 . 15 . 15 . 16 . 16 . 16
Linux Snapshot image backup fails with error message ANS1258E . . . . . . . . . . . Resolving errors during AIX JFS2 snapshot-based backup-archive and image backup . . . . . . API problem determination . . . . . . . . . Support solutions for the Tivoli Storage Manager API . . . . . . . . . . . . . . . . Determining if data is sent to the IBM Tivoli Storage Manager storage agent rather than the server . . . . . . . . . . . . . . . Running applications that use the API as a non-root user . . . . . . . . . . . . . JBB problem determination . . . . . . . . . Determining if a backup will be journal-based . . Running the Journal Daemon in the foreground Using the journal database viewing utility . . . Using open file support and the logical volume snapshot agent . . . . . . . . . . . . . Examining the Windows system event log . . . Obtaining debug print output . . . . . . . Configuring the system for a full dump . . . . Forcing a dump for a system hang when a logical volume snapshot agent problem is suspected . . . . . . . . . . . . . . Best practices for open file support . . . . . Windows Volume Shadow Copy Services . . . . Determining VSS transient errors . . . . . . Defining Windows VSS test flags . . . . . . Windows 2003 VSS hot fixes . . . . . . . . Microsofts VSS tuning recommendations . . . VSS diagnostic information for Microsoft assistance . . . . . . . . . . . . . . Using the vsreq.exe sample program . . . . . Comparing Tivoli Storage Manager and Ntbackup.exe interaction with VSS . . . . . Show commands for the backup-archive client . . CACHE . . . . . . . . . . . . . . CLUSTER . . . . . . . . . . . . . . DOMAIN . . . . . . . . . . . . . . OPTIONS . . . . . . . . . . . . . . OPTTABLE . . . . . . . . . . . . . PLUGINS . . . . . . . . . . . . . . SESSION . . . . . . . . . . . . . . SYSTEMOBJECT . . . . . . . . . . . . SYSTEMSERVICES . . . . . . . . . . . SYSTEMSTATE . . . . . . . . . . . . TRACEFLAGS . . . . . . . . . . . . VERSION . . . . . . . . . . . . . .
18 19 20 20
22 23 25 25 26 27 28 28 28 29
29 30 30 30 30 31 31 31 32 32 33 33 33 33 33 34 34 34 34 35 35 35 35
Chapter 2. Resolving server problems

Server diagnostic tips . . . . . . . . . Checking the server activity log . . . . Checking HELP for Tivoli Storage Manager messages issued for the problem . . . . Recreating the problem . . . . . . . . . . .
37
. 37 . 37 . 37 . 37
Copyright IBM Corp. 1993, 2007
iii
Resolving errors from reading or writing to a device . . . . . . . . . . . . . . . Changing server options or settings creates errors Failing a scheduled client operation . . . . . Resolving server space issues . . . . . . . Server error messages indicate a code page conversion failure . . . . . . . . . . . Resolving a server crash . . . . . . . . . . Resolving database errors. . . . . . . . . . Resolving a hang or loop . . . . . . . . . . Using recovery log information . . . . . . . . Defining processes . . . . . . . . . . . . AUDIT VOLUME . . . . . . . . . . . BACKUP DB . . . . . . . . . . . . . BACKUP STGPOOL . . . . . . . . . . CHECKIN LIBVOLUME . . . . . . . . . CHECKOUT LIBVOLUME . . . . . . . . EXPIRATION . . . . . . . . . . . . . IMPORT . . . . . . . . . . . . . . LABEL LIBVOLUME . . . . . . . . . . MIGRATION . . . . . . . . . . . . . MOVE DATA . . . . . . . . . . . . . MOVE DRMEDIA . . . . . . . . . . . MOVE MEDIA . . . . . . . . . . . . MOVE NODEDATA . . . . . . . . . . PREPARE . . . . . . . . . . . . . . RECLAMATION . . . . . . . . . . . RESTORE STGPOOL . . . . . . . . . . RESTORE VOLUME . . . . . . . . . . Using the process messages . . . . . . . . . Using the process symptoms to resolve problems . . Resolving storage pool issues . . . . . . . .
37 38 38 38 38 39 42 44 44 46 47 47 47 48 48 48 48 49 49 49 49 50 50 50 50 51 51 51 54 57
Chapter 3. Resolving communication problems . . . . . . . . . . . . . . 61

Generating errors when connecting to the server . Resolving failed connections by clients or administrators . . . . . . . . . . . . Determining SSL errors . . . . . . . . . . 61 . 61 . 62
Conditions that can cause a warning or critical database status . . . . . . . . . . . . Conditions that can cause a warning or critical storage status . . . . . . . . . . . . . When to resynchronize the ADMIN_CENTER administrator ID password . . . . . . . . Administration Center task fails with a message . . ANRW0089E . . . . . . . . . . . . . ANRW0090E . . . . . . . . . . . . . ANRW0126E . . . . . . . . . . . . . ANRW0127E . . . . . . . . . . . . . ANRW0128E . . . . . . . . . . . . . ANRW0129E . . . . . . . . . . . . . ANRW0130E . . . . . . . . . . . . . ANRW0131E . . . . . . . . . . . . . ANRW0132E . . . . . . . . . . . . . ANRW0209E . . . . . . . . . . . . . ANRW0212E . . . . . . . . . . . . . ANRW0213E . . . . . . . . . . . . . ANRW0221E . . . . . . . . . . . . . ANRW0257E . . . . . . . . . . . . . Administration Center task fails or returns unexpected results . . . . . . . . . . . . Check the server activity log. . . . . . . . Errors caused by starting or stopping a wizard or portlet . . . . . . . . . . . . . . . Problems caused by internal errors . . . . . Administration Center messages versus Tivoli Storage Manager server messages . . . . . . . Understanding Tivoli Storage Manager messages . . Problems with the Tivoli Storage Manager server command definition file . . . . . . . . . . Diagnosing red error messages . . . . . . . . Administration Center support utility . . . . .
83 83 84 85 86 86 86 87 87 88 88 88 89 89 89 90 90 90 91 91 91 92 92 94 96 97 99
Chapter 5. Data Protection . . . . . . 101

Data Protection for Domino . . . . . . . . Tracing the Data Protection client (Domino) . . Locating solutions for the Data Protection client (Domino) . . . . . . . . . . . . . . Gathering information before calling IBM (Domino) . . . . . . . . . . . . . . Gathering files before calling IBM (Domino) . . Data Protection for Snapshot Devices for mySAP Troubleshooting Data Protection for Snapshot Devices for mySAP . . . . . . . . . . Tracing Data Protection for Snapshot Devices for mySAP . . . . . . . . . . . . . . Locating solutions for Data Protection for Snapshot Devices for mySAP . . . . . . . Gathering information before calling IBM (DP for Snapshot Devices for mySAP) . . . . . . Files to gather before calling IBM (DP for Snapshot Devices for mySAP) . . . . . . . Data Protection for Oracle . . . . . . . . . Tracing Data Protection for Oracle . . . . . Locating solutions for Data Protection for Oracle Gathering information before calling IBM (Data Protection for Oracle) . . . . . . . . . . Gathering files before calling IBM (Data Protection for Oracle) . . . . . . . . . . 101 101 101 101 102 103 103 116 118 118 119 120 120 120 121 121
Chapter 4. Administration Center . . . 65

Establishing a connection between the Administration Center and a Tivoli Storage Manager server . . . . . . . . . . . . . . . . Investigating failed installations . . . . . . . Resolving ISC user authority problems . . . . . Resolving an ISC server crash . . . . . . . . Using the collector tool . . . . . . . . . Using the log analyzer tool (showlog) . . . . ISC server has excessive memory consumption . . Installation log files . . . . . . . . . . . . Configuring the IP address . . . . . . . . . Problems with tutorials . . . . . . . . . . Administration Center performance tuning suggestions . . . . . . . . . . . . . . Workaround for blank pages in the Administration Center . . . . . . . . . . . . . . . . Unable to access the server from a Web browser . . Health Monitor . . . . . . . . . . . . . How the Health Monitor works . . . . . . 65 67 69 71 71 72 73 73 75 75 77 78 80 81 81
iv
IBM Tivoli Tivoli Storage Manager: Problem Determination Guide
| | | | | | | | | | | |
Data Protection for Exchange . . . . . . . . Determining if the problem is a Data Protection for Exchange issue or an Exchange issue . . . Tracing Data Protection for Exchange . . . . Locating solutions for Data Protection for Exchange . . . . . . . . . . . . . . Gathering information before contacting IBM (Data Protection for Exchange) . . . . . . Gathering files before calling IBM (Data Protection for Exchange). . . . . . . . . What to gather if the silent install fails (Data Protection for Exchange). . . . . . . . . Data Protection for Exchange with VSS backup/restore support . . . . . . . . . . Determining if the problem is a Data Protection for Exchange issue or a general VSS issue . . . Tracing the Data Protection client when using VSS technology . . . . . . . . . . . . Gathering information before calling IBM (Exchange with VSS) . . . . . . . . . . Gathering files before calling IBM (Exchange with VSS) . . . . . . . . . . . . . Troubleshooting Data Protection for Exchange VSS and SAN Volume Controller . . . . . . Data Protection for SAP . . . . . . . . . . Troubleshooting Data Protection for SAP . . . Tracing Data Protection for SAP . . . . . . Locating solutions for Data Protection for SAP Gathering information before calling IBM (Data Protection for SAP) . . . . . . . . . . Gathering files before calling IBM (Data Protection for SAP) . . . . . . . . . . Data Protection for SQL . . . . . . . . . . Tracing Data Protection for SQL . . . . . . Locating solutions for Data Protection for SQL Gathering information before calling IBM (Data Protection for SQL) . . . . . . . . . . Gathering files before calling IBM (Data Protection for SQL) . . . . . . . . . . Gathering items if the silent install fails (Data Protection for SQL) . . . . . . . . . . Data Protection for SQL with VSS backup-restore support . . . . . . . . . . . . . . . Determining if the problem is a Data Protection for SQL issue or a general VSS issue . . . . Tracing the Data Protection client when using VSS technology (SQL) . . . . . . . . . Gathering information before calling IBM (SQL with VSS) . . . . . . . . . . . . . Gathering files before calling IBM (SQL with VSS) . . . . . . . . . . . . . . . Troubleshooting Data Protection for SQL VSS and SAN Volume Controller . . . . . . .
122 122 123 123 124 124 125 125 126 128 128 129 131 132 132 134 135 135 136 136 136 137 137 138 138 139 139 141 142 143 144
Storage agent diagnostic tips . . . . . . . Storage agent LAN-free setup . . . . . . . Show commands for the server or storage agent Enabling trace for the Tivoli Storage Manager device driver . . . . . . . . . . . . . Tracing from the server console/admin client Tracing from a command shell (only valid for 5.2.2 and above) - all platforms . . . . . . Tracing for the client . . . . . . . . . . . Client and Journal Daemon traceflags . . . . Enabling trace for the client . . . . . . . Enabling backup-archive client trace . . . . . Determining if data is encrypted or compressed during backup/archive through trace . . . . Tracing for the API . . . . . . . . . . . ODBC tracing . . . . . . . . . . . . .
174 175 178 191 191 192 193 194 194 204 213 214 215
Chapter 8. Hints and tips . . . . . . 217

Device driver hints and tips . . . . . . . . Adjusting to operating system changes . . . . Adjusting to changes in the HBA or SCSI adapter connecting to the device . . . . . . Adjusting to a loose cable connection . . . . Resolving error messages in the system error log . . . . . . . . . . . . . . . . Supporting 64-bit/32-bit Linux kernel modules for 32-bit/64-bit applications . . . . . . . Running a Tivoli Storage Manager Linux server on x86_64 architecture . . . . . . . . . Adjusting to HBA driver changes on the Linux 2.6.x kernels . . . . . . . . . . . . . Requirements for multiple LUN support on Linux kernels . . . . . . . . . . . . Using Tivoli Storage Manager Version 5.3.2 to perform ddtrace on Linux . . . . . . . . Updating device information of host systems on a dynamic SAN without rebooting . . . . . Requirements for Adaptec SCSI and Qlogic Fibre Channel HBA BIOS settings on Linux . . Hard disk drives and disk subsystems hints and tips . . . . . . . . . . . . . . . . . Tape drives and libraries hints and tips . . . . . Adjusting to operating system changes . . . . Adjusting to device driver changes . . . . . Adjusting to a replaced adapter or other hardware changes . . . . . . . . . . . Adjusting to a loose cable connection . . . . Using error messages . . . . . . . . . . Miscellaneous server information hints and tips Storage area network hints and tips . . . . . . Knowing your SAN configuration . . . . . Devices supported by the SAN . . . . . . Considerations for an HBA . . . . . . . . HBA configuration issues . . . . . . . . FC switch configuration issues . . . . . . Verifying the data gateway port settings . . . Verifying the SAN configuration between devices . . . . . . . . . . . . . . Monitoring the fibre-channel-link error report NDMP filer-to-TSM server operations . . . . . SAN devices . . . . . . . . . . . . .
Contents
217 217 218 218 218 218 219 219 219 220 220 221 221 223 223 223 224 224 224 224 225 225 226 226 226 227 227 228 229 229 230
Chapter 7. Trace . . . . . . . . . . 149

Enabling Administration Center trace . . . . . Trace classes for the Administration Center . . Enabling trace for the server or storage agent . . Enabling trace for specific messages for the server or storage agent . . . . . . . . . Defining trace classes for server or storage agent 149 149 152 153 154
SAN device mapping hints and tips . . . . . . Disabling SAN device mapping . . . . . . Platform-specific information . . . . . . . SAN device mapping problems . . . . . . . SAN devices are missing from the display of QUERY SAN server command . . . . . . Error message ANR1745I: Unable to discover SAN devices. Function is busy. . . . . . . Error messages ANR1786W, ANR1787W or ANR1788W were issued . . . . . . . . . ANR1789W Get HBA target mapping failed . . ANR1790W SAN discovery failed . . . . . ANR1791W HBAAPI wrapper library xxxxx failed to load or is missing . . . . . . . . ANR1792W HBAAPI vendor library failed to load or is missing . . . . . . . . . . . Error message ANR1793W . . . . . . . . ANR1794W Tivoli Storage Manager SAN discovery is disabled by options . . . . . . ANR2034E QUERY SAN: No match found using this criteria . . . . . . . . . . . . . ANR8226E Error detecting version of HBA-API library . . . . . . . . . . . . . . . ANR8227E Fileset devices.common.IBM.fc.hbaapi is not at the required level . . . . . . .
232 233 233 234 238 240 241 241 241 241 242 242 243 243 244 244
Accessing server or storage agent help for commands . . . . . . . . . . . Accessing help for messages . . . . . Command-line interface help for server or storage agent . . . . . . . . . . Help system does not display . . . . . . Verifying that the help system is running . Resolving help system installation problems Starting and stopping the help system . . Adding the publications to the Integrated Solutions Console help server . . . . . Reporting a problem with a help topic . .
. . . . . . . . .
. 261 . 261 . . . . . 261 263 263 264 264
. 265 . 265
Chapter 11. Determining data storage problems . . . . . . . . . . . . . 267

Using data storage diagnostic tips Resolving SCSI device problems . Sequential media volume (tape) . . . . . . . . . . . . . . . . . 267 . 269 . 270
Notices . . . . . . . . . . . . . . 271
Programming interface information . . . . . . 273
Glossary . . . . . . . . . . . . . 275 Trademarks . . . . . . . . . . . . 277
Chapter 9. Tivoli Storage Manager ODBC driver . . . . . . . . . . . . 245

Supported ODBC applications and functions Defining ODBC data sources . . . . Known ODBC problems and limitations ODBC troubleshooting and diagnostics . ODBC configuration details . . . . . . . . . . . . . . . . . . . 245 246 248 250 251
Appendix A. Using gt script . . . . . 279 Appendix B. Installing and running the tsmdiag utility . . . . . . . . . 281
| Appendix C. IBM Global Security Kit | return codes . . . . . . . . . . . . 285
Chapter 10. Help facilities . . . . . . 259

Backup-archive client help . Server or storage agent help . . . . . . . . . . . . . . . 259 . 261
Index . . . . . . . . . . . . . . . 295
vi
Preface
This publication helps you determine the source of problems with the servers and clients in your Tivoli Storage Manager environment. Before using this publication, you should be familiar with the following areas: v The operating systems on which your Tivoli Storage Manager servers and clients reside v The communication protocols installed on your client and server machines
vii
viii
Who should read this guide

This guide was written for anyone administering or managing Tivoli Storage Manager. Similarly, information provided by this guide may be useful to business partners and anyone with the responsibility to support Tivoli Storage Manager. You should be familiar with the Tivoli Storage Manager and the operating systems used for the configured Tivoli Storage Manager environment. This document references error logs, trace facilities, and other diagnostic information for the Tivoli Storage Manager. These trace facilities and diagnostic tools are not a programming interface for the product. The Tivoli Storage Manager product development and support use these tools for diagnosing and debugging problems. For this guide, these are provided only to aid in diagnosing and debugging any problems. Trace facilities are subject to change without notice and may vary depending upon the version and release of the product or the platform on which the product is being run. Information referenced within this guide may not be supported or applicable to all versions or releases of the product. Changes are periodically made to the information herein. IBM may make improvements and changes in the products and the programs described in this publication at any time without notice.
ix
Publications
Tivoli Storage Manager publications and other related publications are available online. You can search all the Tivoli Storage Manager publications in the Information Center: http://publib.boulder.ibm.com/infocenter/tivihelp/v1r1/index.jsp You can download PDF versions of publications from the IBM Publications Center: http://www.elink.ibmlink.ibm.com/public/applications/publications/cgibin/ pbi.cgi You can also order some related publications from this Web site. The Web site provides information for ordering publications from countries other than the United States. In the United States, you can order publications by calling 800-879-2755.
Tivoli Storage Manager publications

Tivoli Storage Manager server publications Publication Title IBM Tivoli Storage Manager Messages IBM Tivoli Storage Manager Performance Tuning Guide IBM Tivoli Storage Manager Problem Determination Guide IBM Tivoli Storage Manager for AIX Installation Guide IBM Tivoli Storage Manager for AIX Administrators Guide IBM Tivoli Storage Manager for AIX Administrators Reference IBM Tivoli Storage Manager for HP-UX Installation Guide IBM Tivoli Storage Manager for HP-UX Administrators Guide IBM Tivoli Storage Manager for HP-UX Administrators Reference IBM Tivoli Storage Manager for Linux Installation Guide IBM Tivoli Storage Manager for Linux Administrators Guide IBM Tivoli Storage Manager for Linux Administrators Reference IBM Tivoli Storage Manager for Sun Solaris Installation Guide IBM Tivoli Storage Manager for Sun Solaris Administrators Guide IBM Tivoli Storage Manager for Sun Solaris Administrators Reference IBM Tivoli Storage Manager for Windows Installation Guide IBM Tivoli Storage Manager for Windows Administrators Guide IBM Tivoli Storage Manager for Windows Administrators Reference IBM Tivoli Storage Manager for z/OS Installation Guide IBM Tivoli Storage Manager for z/OS Administrators Guide IBM Tivoli Storage Manager for z/OS Administrators Reference Order Number SC32-0140 SC32-0141 SC32-0142 GC23-5969 SC32-0117 SC32-0123 GC23-5970 SC32-0118 SC32-0124 GC23-5971 SC32-0119 SC32-0125 GC23-5972 SC32-0120 SC32-0126 GC23-5973 SC32-0121 SC32-0127 GC23-5974 SC32-0122 SC32-0128
IBM Tivoli Storage Manager for System Backup and Recovery Installation SC32-6543 and Users Guide
xi
Tivoli Storage Manager storage agent publications Publication Title IBM Tivoli Storage Manager for SAN for AIX Storage Agent Users Guide IBM Tivoli Storage Manager for SAN for HP-UX Storage Agent Users Guide IBM Tivoli Storage Manager for SAN for Linux Storage Agent Users Guide IBM Tivoli Storage Manager for SAN for Sun Solaris Storage Agent Users Guide IBM Tivoli Storage Manager for SAN for Windows Storage Agent Users Guide Order Number SC32-0129 SC32-0130 SC32-0131 SC32-0132 SC32-0133
Tivoli Storage Manager client publications Publication Title IBM Tivoli Storage Manager for Macintosh: Backup-Archive Clients Installation and Users Guide IBM Tivoli Storage Manager for NetWare: Backup-Archive Clients Installation and Users Guide IBM Tivoli Storage Manager for UNIX and Linux: Backup-Archive Clients Installation and Users Guide IBM Tivoli Storage Manager for Windows: Backup-Archive Clients Installation and Users Guide Order Number SC32-0143 SC32-0144 SC32-0145 SC32-0146
IBM Tivoli Storage Manager for Space Management for UNIX and Linux: SC32-0148 Users Guide IBM Tivoli Storage Manager for HSM for Windows Administration Guide SC32-1773 IBM Tivoli Storage Manager Using the Application Program Interface SC32-0147
Tivoli Storage Manager Data Protection publications Publication Title IBM Tivoli Storage Manager for Advanced Copy Services: Data Protection for Snapshot Devices for SAP Installation and Users Guide for Oracle IBM Tivoli Storage Manager for Advanced Copy Services: Data Protection for Snapshot Devices for DB2 Installation and Users Guide IBM Tivoli Storage Manager for Advanced Copy Services: Data Protection for WebSphere Application Server Installation and Users Guide IBM Tivoli Storage Manager for Databases: Data Protection for Informix Installation and Users Guide IBM Tivoli Storage Manager for Databases: Data Protection for Microsoft SQL Server Installation and Users Guide IBM Tivoli Storage Manager for Databases: Data Protection for Oracle for UNIX and Linux Installation and Users Guide IBM Tivoli Storage Manager for Databases: Data Protection for Oracle for Windows Installation and Users Guide Order Number SC33-8207 SC33-8330 SC32-9075 SH26-4095 SC32-9059 SC32-9064 SC32-9065
IBM Tivoli Storage Manager for Enterprise Resource Planning: Data Protection for SC33-6341 SAP Installation and Users Guide for DB2
xii
Tivoli Storage Manager Data Protection publications Publication Title Order Number
IBM Tivoli Storage Manager for Enterprise Resource Planning: Data Protection for SC33-6340 SAP Installation and Users Guide for Oracle IBM Tivoli Storage Manager for Hardware: Data Protection for Enterprise Storage SC32-9060 Server for DB2 UDB Installation and Users Guide IBM Tivoli Storage Manager for Hardware: Data Protection for Snapshot Devices for Oracle Installation and Users Guide IBM Tivoli Storage Manager for Mail: Data Protection for Lotus Domino for UNIX, Linux, and OS/400 Installation and Users Guide IBM Tivoli Storage Manager for Mail: Data Protection for Lotus Domino for Windows Installation and Users Guide IBM Tivoli Storage Manager for Mail: Data Protection for Microsoft Exchange Server Installation and Users Guide GC32-1772 SC32-9056 SC32-9057 SC32-9058
Publications
xiii
xiv
Support information
You can find support information for IBM products from a number of different sources: v Getting technical training v Searching knowledge bases v Contacting IBM Software Support on page xvi
Getting technical training

Information about Tivoli technical training courses is available online. http://www.ibm.com/software/tivoli/education/
Searching knowledge bases

If you have a problem with Tivoli Storage Manager, there is a variety of knowledge bases you can search. You can begin with the Information Center, from which you can search all the Tivoli Storage Manager publications: http://publib.boulder.ibm.com/infocenter/ tivihelp/v1r1/index.jsp
Searching the Internet

If you cannot find an answer to your question in the information center, search the Internet for the latest, most complete information that might help you resolve your problem. To search multiple Internet resources, go to the support web site for Tivoli Storage Manager: http://www.ibm.com/software/sysmgmt/products/support/ IBMTivoliStorageManager.html. From this section, you can search a variety of resources including: v IBM technotes v IBM downloads v IBM Redbooks If you still cannot find the solution to your problem, you can search forums and newsgroups on the Internet for the latest information that might help you resolve your problem.
Using IBM Support Assistant

The IBM Support Assistant is a free, stand-alone application that you can install on any workstation. You can then enhance the application by installing product-specific plug-in modules for the IBM products you use. The IBM Support Assistant helps you gather support information when you need to open a problem management record (PMR), which you can then use to track the problem. The product-specific plug-in modules provide you with the following resources: v Support links
xv
v Education links v Ability to submit problem management reports For more information, see the IBM Support Assistant Web site at http://www.ibm.com/software/support/isa/
Finding product fixes

A product fix might be available to resolve your problem. You can determine what fixes are available by checking the product support Web site. 1. Go to the IBM Software Support Web site: http://www.ibm.com/software/ tivoli/products/storage-mgr/product-links.html 2. Click the Support Pages link for your Tivoli Storage Manager product. 3. Click Fixes for a list of fixes for your product. 4. Click the name of a fix to read the description and optionally download the fix.
Getting E-mail notification of product fixes

You can sign up to receive weekly E-mail notifications about fixes and other news about IBM products. 1. From the support page for any IBM product, click My support in the upper-right corner of the page. 2. If you have already registered, skip to the next step. If you have not registered, click register in the upper-right corner of the support page to establish your user ID and password. 3. Sign in to My support. 4. On the My support page, click Edit profiles in the left navigation pane, and scroll to Select Mail Preferences. Select a product family and check the appropriate boxes for the type of information you want. 5. Click Submit. 6. For E-mail notification for other products, repeat steps 4 and 5.
Contacting IBM Software Support

Before you contact IBM Software Support, you must have an active IBM software maintenance contract, and you must be authorized to submit problems to IBM. The type of software maintenance contract that you need depends on the type of product you have. v For IBM distributed software products (including, but not limited to, Tivoli, Lotus, and Rational products, as well as DB2 and WebSphere products that run on Windows or UNIX operating systems), enroll in Passport Advantage in one of the following ways: Online Go to the Passport Advantage Web page (http://www.ibm.com/ software/sw-lotus/services/cwepassport.nsf/wdocs/passporthome) and click How to Enroll By phone For the phone number to call in your country, go to the IBM Contacts Web page (http://techsupport.services.ibm.com/guides/contacts.html) and click the name of your geographic region. v For IBM eServer software products (including, but not limited to, DB2 and WebSphere products that run in zSeries, pSeries, and iSeries environments), you can purchase a software maintenance agreement by working directly with
xvi
an IBM sales representative or an IBM Business Partner. For more information about support for eServer software products, go to the IBM Technical Support Advantage Web page: http://www.ibm.com/servers/eserver/techsupport.html. If you are not sure what type of software maintenance contract you need, call 1-800-IBMSERV (1-800-426-7378) in the United States. For a list of telephone numbers of people who provide support for your location, go to the IBM Contacts Web page, http://techsupport.services.ibm.com/guides/contacts.html, and click the name of your geographic region. Perform these actions to contact IBM Software Support: 1. Determine the business impact of your problem. 2. Describe your problem and gather background information. 3. Submit your problem to IBM Software Support.
Determine the business impact

When you report a problem to IBM, you are asked to supply a severity level. Therefore, you need to understand and assess the business impact of the problem you are reporting.
Severity 1 Critical business impact: You are unable to use the program, resulting in a critical impact on operations. This condition requires an immediate solution. Significant business impact: The program is usable but is severely limited. Some business impact: The program is usable with less significant features (not critical to operations) unavailable. Minimal business impact: The problem causes little impact on operations, or a reasonable circumvention to the problem has been implemented.
Severity 2 Severity 3 Severity 4
Describe your problem and gather background information

When explaining a problem to IBM, be as specific as possible. Include all relevant background information so that IBM Software Support specialists can help you solve the problem efficiently. To save time, know the answers to these questions: v What software versions were you running when the problem occurred? v Do you have logs, traces, and messages that are related to the problem symptoms? IBM Software Support is likely to ask for this information. v Can the problem be re-created? If so, what steps led to the failure? v Have any changes been made to the system? For example, hardware, operating system, networking software, and so on. v Are you currently using a workaround for this problem? If so, be prepared to explain it when you report the problem.
Submit your problem to IBM Software Support

You can submit your problem to IBM Software Support online or by phone. Online Go to the Submit and track problems page on the IBM Software Support
Support information
xvii
site http://www.ibm.com/software/support/probsub.html . Enter your information into the appropriate problem submission tool. By phone For the phone number to call in your country, go to the contacts page of the IBM Software Support Handbook on the Web and click the name of your geographic region. If the problem you submit is for a software defect or for missing or inaccurate documentation, IBM Software Support creates an Authorized Program Analysis Report (APAR). The APAR describes the problem in detail. Whenever possible, IBM Software Support provides a workaround for you to implement until the APAR is resolved and a fix is delivered. IBM publishes resolved APARs on the IBM product support Web pages daily, so that other users who experience the same problem can benefit from the same resolutions.
Accessibility features
Accessibility features help a user who has a physical disability, such as restricted mobility or limited vision, to use software products successfully. The major accessibility features of Tivoli Storage Manager are described in this topic. v Server and client command-line interfaces provide comprehensive control of Tivoli Storage Manager using a keyboard. v The Windows client-graphical interface can be navigated and operated using a keyboard. v The Web backup-archive client interface is HTML 4.0 compliant, and accessibility is limited only by the choice of Internet browser. v All user documentation is provided in HTML and PDF format. Descriptive text is provided for all documentation images. v The Tivoli Storage Manager for Windows Console follows Microsoft conventions for all keyboard navigation and access. Drag and Drop support is handled using the Microsoft Windows Accessibility option known as MouseKeys. For more information about MouseKeys and other Windows accessibility options, please refer to the Windows Online Help (keyword: MouseKeys).
xviii
Chapter 1. Resolving client problems

If you are experiencing problems with a Tivoli Storage Manager client, there are several ways to resolve them.
Client diagnostic tips

Resolving problems with the client application can involve reproducing the error, connecting to the server, or changing policy settings. Some possible errors may be resolved through the following suggestions: v Changes to the Tivoli Storage Manager client options are not recognized by the client scheduler until the scheduler is restarted. Stop and restart the client scheduler. v The QUIET processing option for the Tivoli Storage Manager client suppresses all messages. Restart the client with the QUIET option turned off. This allows all messages to be issued and will provide a more complete understanding of the problem. v The include/exclude processing option impacts which files are sent to the server for a backup or archive operation. Refer to the include/exclude section of this document for more information on client include/exclude syntax and ordering. v If you are working in an environment with multiple Tivoli Storage Manager servers, a client may use different servers for different operations. Ensure that the client is connected to the appropriate server for the operation that was attempted.
Examining error messages

One way to resolve problems is to examine the error messages that are generated during operation. Check for any ANSnnnnx messages issued to the console, dsmsched.log, or dsmerror.log. Additional information for ANSnnnnx messages is available in either the Tivoli Storage Manager Messages or from the client HELP facility.
Examining the server activity log messages

Check for server activity log using QUERY ACTLOG for messages issued for this Tivoli Storage Manager client session. The messages from the server activity log may provide additional information about the symptoms for the problem or may provide information about the actual cause of the problem that the client encountered.
Identifying when and where the problem can occur

Tivoli Storage Manager client processing problems often only occur when performing specific operations, at certain times, or only on certain client machines. To further isolate when and where a problem occurs, determine the following: v Does this problem occur for a single client, some clients, or all clients for a given server? v Does this problem occur for all clients running on a specific operating system?
v Does this problem occur for specific files, files in a specific directory, files on a specific drive, or all files? v Does this problem occur for clients on a specific network, subnet, or all parts of the network? v Does this problem occur only for the command-line client, the GUI client, or the web client? v Does the Tivoli Storage Manager always fail when processing the same file or directory, or is it different from run to run?
Reproducing the problem

If the problem can be reproduced, try to minimize the circumstances under which it can occur. You can help Tivoli Storage Manager support by minimizing the complexity of the environment in which you want to recreate the problem. The following are some options you can use to minimize the complexity of the environment: v Use a minimal options file consisting of only TCPSERVERADDRESS, TCPPORT, and NODENAME. v If the problem occurs for a file during incremental backup, try to reproduce the problem with a selective backup of just that file. v If the problem occurs during a scheduled event, try to reproduce the problem by manually running the command.
Collecting documentation to resolve problems with the client application

The Tivoli Storage Manager client creates valuable information in a number of different sources. If any of the information in the documentation relates to your problem, provide it to support. Tivoli Storage Manager client problems and configuration information may be found in one or more of the following documents: v Error log. The client error log file is dsmerror.log. v Scheduler log. The error log for the client scheduler is dsmsched.log. v Web client log. The error log for the web client is dsmwebcl.log. v Options files. The client may use a combination of files for its configuration. These files are dsm.opt, dsm.sys for UNIX systems, and the include/exclude file. v Trace data. If tracing was active, the file containing the trace data could be provided to support. v Application dump. If the Tivoli Storage Manager client crashed, many platforms will generate an application dump. The operating system provides the application dump. v Memory dump. If the Tivoli Storage Manager client hung, a memory dump can be generated that can then be used to help with diagnosis. The type of system determines how the memory dump occurs, and the operating system provides the memory dump. Starting with Tivoli Storage Manager Version 5.2, the DSMC QUERY SYSTEMINFO command is available and will collect most of this information in the dsminfo.txt file. The following are items that can be helpful in determining Tivoli Storage Manager problems:
v A list of all the software installed on the client system. The client may experience problems due to interactions with other software on the machine or because of the maintenance levels of software that the client uses. v Client option sets defined on the server that apply to this client node. Issue the QUERY CLOPTSET command to search for the client option sets. v Server options. There are a number of server options that are used to manage the interaction between the client and server. An example of one such server option is TXNGROUPMAX. v Information about this node as it is defined to the server. To collect this information, issue the QUERY NODE nodeName F=D command using an administrative client connected to the server. v Schedule definitions for the schedules that apply to this node. These can be queried from the server using the QUERY SCHEDULE command. v The policy information configured for this node on the IBM Tivoli Storage Manager server. This information can be queried from the server using the QUERY DOMAIN, QUERY POLICYSET, QUERY MANAGEMENTCLASS, and QUERY COPYGROUP commands.
dsmc/dsmadmc/dsmj/... will not start

The dsmc/dsmadmc/dsmj/... occasionally will not start. Processing stops and the following message is displayed if the dsmc/dsmadmc/dsmj/... will not start:
ANS1398E Initialization functions cannot open one of the Tivoli Storage Manager logs or a related file: /dsmerror.log. errno = 13, The file access permissions do not allow the specified action.
Note: The dsmerror.log file is used only as an example file in the message. With Tivoli Storage Manager Version 5.3 and later, client applications will not run without a log to which you can write and the system will deny write access to the log file named in the message. If the log does not exist, it will be created with default permissions. The following rules apply: v The name and the directory specified by the ERRORLOGNAME option are used. v If the option is absent, the name dsmerror.log in the directory specified in the DSM_LOG environment variable, if present, is used. Otherwise, the name dsmerror.log in the current working directory is used. | | | | Because default permissions are used, a log created by the root user may not be written to by any other user. If this is the case, the root user must set the proper permissions or access control lists (ACLs) to allow free use of the client application by all users who need to use it. If the log is successfully created, an error-free session will leave a zero-length (empty) log file. The Tivoli Storage Manager client does not try to create logs in the root directory. Message ANS1398E is displayed when the method in the first rule, above, directs the log file to be created in the root directory.
If a log file exists and can be located, Tivoli Storage Manager uses the method from the first rule. It can also be in the root directory if you so choose. Furthermore, whatever permissions you give that log file will be preserved by Tivoli Storage Manager code.
Recommendations
Create your log file in advance of first use, ensuring that all eligible users have write access to it. Define the ERRORLOGNAME option or the DSM_DIR environment variable to designate your predefined log file. Attention: A system log error indicates that you cannot write to the dsmerror.log file. Certain background Tivoli Storage Manager applications might not start due to errors writing to dsmerror.log. When these errors occur, a number of errors are recorded in the Windows system event log.
C:\Program Files\Tivoli\Tsm\baclient>net start "TSM Sched" The server scheduling service is starting. The server scheduling service could not be started. A service specific error occurred: 12.
Additional setup steps are required for non-root users in order for them to be able to run Tivoli Storage Manager applications or Tivoli Storage Manager for Data Protection applications. You will receive the ANS1398E error if you attempt to run Tivoli Storage Manager applications using an error log which has already been generated by root, that is left with default permissions. For data protection clients, you may only receive a Tivoli Storage Manager API error. Here is one method for setting up dsmerror.log for use by non-root users: 1. Set ERRORLOGNAME in dsm.sys. For example, errorLogName /var/msgs/tsm/dsmerror.log 2. Generate dsmerror.log. dsmc q sess 3. Modify the permissions on dsmerror.log to allow writing by all users. chmod 666 /var/msgs/tsm/dsmerror.log
Client option sets

Client option sets allow administrators to specify additional options that may not be included in the clients option file. The client can use these defined options during a backup, archive, restore, or retrieve process.
What are the client option sets?

An administrator for the Tivoli Storage Manager can create a set of client options to be used by a client node on Tivoli Storage Manager Version 3.0 or later. The client options are defined on the Tivoli Storage Manager server. The client options specified in the client option set are used in conjunction with the client options file.
Where to find detailed information

Detailed information can be found at the following locations: v General overview: Tivoli Storage Manager Administrators Guide: Managing Client Nodes Managing Client Option Files. v Tivoli Storage Manager server commands: Tivoli Storage Manager Administrators Reference: Administrative Commands DEFINE CLOPTSET
DELETE CLOPTSET COPY CLOPTSET QUERY CLOPTSET UPDATE CLOPTSET REGISTER NODE QUERY NODE UPDATE NODE
DEFINE CLIENTOPT DELETE CLIENTOPT UPDATE CLIENTOPT v Tivoli Storage Manager client options: Tivoli Storage Manager Backup-Archive Client Installation and Users Guide. Note: A list of supported Tivoli Storage Manager client options can be found in the Tivoli Storage Manager Administrators Reference in the DEFINE CLIENTOPT command.
INCLEXCL
You can use the INCLEXCL option instead of INCLUDE or EXCLUDE. Options that are passed to the Tivoli Storage Manager client from the Tivoli Storage Manager server are provided in groups. This means that if the INCLUDE and EXCLUDE options are supported on the Tivoli Storage Manager server, that all INCLUDE options would be sent in a group and all EXCLUDE options would be sent in a group. You could not intermix these options to get desired results of including some files from excluded directories. Using the INCLEXCL option allows you to intermix and order the include and exclude options.
The order of option processing

The order in which the options are processed can be controlled. Multiple options can be defined and assigned a sequence number, with these options then processed from low to high sequence. The following example displays the INCLEXCL options:
Option Sequence Override Option Value number ------------------------- -------- -------- ---------------------------------INCLEXCL 0 No exclude sys:\...\* INCLEXCL 1 No include sys:\system\* INCLEXCL 2 No include sys:\tmp\*
This sequence results in the exclusion of all files on the Novell NetWare SYS while the files in the SYS:\SYSTEM\* and SYS:\TMP\* paths are backed up.
Client and client option traceflag settings

Trace settings for the client option sets are specified in the Tivoli Storage Manager option file for all clients. For releases prior to 5.4, the exception is MAC, which uses the Tivoli Storage Manager User Preference file. v TRACEFILE <trace file name> v TRACEFLAGS config For the 5.4.0 release, the MAC client uses the same option file names as the rest of the clients.
Examples of when to use client options sets

The following are customer scenarios which can take advantage of the client option set. Included are examples of required commands. The customer has multiple NetWare servers and wants to control the specific options below for backup: v Wants SYS: volume to be backed up. v Wants the Novell NetWare server which has the root level NDS to be backed up. v Wants all Tivoli Storage Manager NetWare clients to use standard display on client commands with scrollines and scrollprompt. This will control the display so things do not scroll off the screen. v Wants to exclude all license objects from NDS backup. License objects can not be backed up or restored by Tivoli Storage Manager. The license must be reinstalled through Novell NetWare. Note: Domain and inclexcl options can not use the force parameter. As a result, the client can override the client option set through the Tivoli Storage Manager option file. Tivoli Storage Manager server command:
Define Define Define Define Define Define Update cloptset netware description="NetWare Server Option Sets" clientopt netware domain sys: clientopt netware domain nds: clientopt netware scrollprompt yes clientopt netware scrollline 20 clientopt netware inclexcl " exclude NDS:*License ID* " node mock cloptset=netware
The customer has a critical environment and restoring is a high priority. v The user wishes to use collocatebyfilespec so that all filespec data is stored on as few tapes as possible. This will enhance restore processing because fewer tape mounts are needed. v The user does not want the client to be able to override this option. Tivoli Storage Manager server command:
Define cloptset crit_rest description="Critical Restore Option Sets" Define clientopt crit_rest collocatebyfilespec yes force=yes Update node dale cloptset=crit_rest
The customer has machines on a slow network and limited space on the Tivoli Storage Manager server for data. v Use the compression option to limit the amount of data sent and stored. Tivoli Storage Manager server command:
Define Define Define Update cloptset space_rest description="Space Restriction Option Sets" clientopt space_rest compressalways no force=yes clientopt space_rest compression yes force=yes node mark cloptset=space_rest
The customer has a 24/7 database which cannot be stopped. Since the files are open, the server cannot back them up. The customer wants to exclude all files and subdirectories from Tivoli Storage Manager backups and add this to the existing space_rest client option set. v exclude.dir, specify the directory path to be excluded. Tivoli Storage Manager server command:
Define clientopt space_rest inclexcl "exclude.dir c:\lnotes\data"
The customer has a fast network and wants to make the best possible use of Tivoli Storage Manager client resources to complete backups. v Wants resourceutilization set to the maximum amount. Tivoli Storage Manager server command:
Define cloptset unix_srv description="Unix Server Option Sets" Define clientopt unix_srv resourceutilization 10 force=yes
Authenticating the Client

There are three common errors that may occur in relation to your password or authentication method: authentication failure, login failure due to the NetWare target service agent, or login failure due to the NetWare file server.
ANS1025E Session rejected: authentication failure

This error generally occurs when the password expires. It can also occur, however, if either the server or the client is renamed, or if the Tivoli Storage Manager administrative user identification password expires. If you receive the ANS1025E error message during an interactive session, the probable cause is an incorrect password. v The Tivoli Storage Manager administrator might reset the nodes password by issuing the UPDATE NODE command. v Issue a DSMC QUERY SESSION command and, when prompted, enter the new password. If you are receiving error message ANS1025E during a non-interactive session, such as central scheduling, ensure that the client option is PASSWORDACCESS GENERATE. This option causes the client to store the password locally. The password is encrypted and stored either in the registry for Windows clients or in a file named TSM.PWD for Macintosh, UNIX , and NetWare clients. You should not edit the registry or the TSM.PWD file. Instead, see the following actions: v Make sure that PASSWORDACCESS GENERATE is set in the option file. v Issue a DSMC QUERY SESSION command. This command will force-set the locally stored password. v If this does not resolve the problem, update the nodes password by issuing the UPDATE NODE administrative command. v Reissue the DSMC QUERY SESSION command, providing the new password. To see the password expiration setting for a particular node, issue the QUERY NODE F=D administrative command. Look for the Password Expiration Period field. Note: If this field is blank, the default password expiration period of 90 days is in effect. To change the password expiration period for a particular node, issue the administrative UPDATE NODE command with the option PASSEXP=n , where n is the number of days. A value of 0 will disable the password expiration. | | | If a Windows client node is unable to connect to the server after having been renamed, verify that the node name was changed in both the client options file and Windows registry. When the client scheduler runs as a foreground process
| | | |
using the DSMC SCHED command, Tivoli Storage Manager uses the nodename in the client options file to contact the server. However, when the scheduler runs as a Windows service, Tivoli Storage Manager uses the node name in the Windows registry instead. Issue the DSMCUTIL UPDATE SCHEDULE command to achieve the following results: v With the node parameter, address how to change the nodename used with the Tivoli Storage Manager scheduler service on Windows v With the validate:yes parameter, contact the Tivoli Storage Manager server to authenticate (and store the updated password) For more information, see the Changing the processing options used by the scheduler service section of the Tivoli Storage Manager for Windows: Backup-Archive Clients Installation and Users Guide.
ANS1874E login denied to NetWare target service agent server-name or ANS2025E login failed to NetWare file server server-name
The ANS1874E and ANS2025E messages indicate an authentication problem between the Tivoli Storage Manager NetWare client and the NetWare server. The most probable cause of this error is either not ensuring that the proper Target Service Agent (TSA) is loaded, or not using the Novell distinguished name. The password for the NetWare TSA is encrypted and stored in the same file that stores the Tivoli Storage Manager password. The process of storing the password in this file is not related to the client option PASSWORDACCESS GENERATE. If you do not want the password stored locally, enter the NWPWFILE NO option in the dsm.opt file. Some things to verify: v LOAD TSAFS (depending on NetWare OS version). v For NDS backups, LOAD TSANDS. v Use the Novell typeful name. For example, instead of Admin, use .CN=Admin.O=IBM. v Ensure that NWPWFILE YES (the default setting) is in the dsm.opt options file. v Issue the DSMC QUERY TSA command to check to see if the Tivoli Storage Manager can connect to the file system TSA. v Issue the DSMC QUERY TSA NDS command to check to see if the server can connect to the NDS TSA. v Note: The DSMC QUERY TSA command can be used to store the password in the local password file and also to test to see if the stored password is valid. The following is a list of other things to check for when experiencing NetWare login failures: v The NetWare user-id has been disabled. v The NetWare user-id/password is invalid or expired. v The NetWare user-id has inadequate security access. v The NetWare user-id has insufficient rights to files and directories. v The NetWare user-id that is specified has a login restriction based on time-of-day. v The NetWare user-id that is specified has a Network address restriction.
v The NetWare user-id that is specified has a login restriction based on the number of concurrent connections. v NetWare is not allowing logins (DISABLE LOGIN was issued at the console). v The temporary files in SYS:\SYSTEM\TSA are corrupt, which can prevent logging in. Shut down all Tivoli Storage Manager processes and SMS modules and then either move or delete these temporary files. The following Novell TID reviews this issue: error: FFFDFFD7 when the tape software tries to login in order to backup nds v The SMDR configuration is corrupt. Reset the SMDR by issuing the following command: LOAD SMDR /NEW at the NetWare console. v If the message is displayed intermittently during a multiple session backup or restore, the probable cause is that there are not enough available Novell licenses. Each Tivoli Storage Manager session requires at least one licensed connection to the file server. Either add more NetWare licenses or reduce the RESOURCEUTILIZATION setting. The NetWare utility, Nwadmn32, can be used to determine the current number of licenses. v Check the SYS volume for free space. A lack of free space can cause the authentication failure. v If the above items have checked out, contact Novell for additional support to resolve this issue.
Client scheduling
The Tivoli Storage Manager administrator can schedule the Tivoli Storage Manager to perform tasks automatically. If you are experiencing problems with your client scheduler, there are several diagnostic steps to help determine the cause of the problem. The following are methods to help find the cause of the problem: v Issue the QUERY EVENT command to determine the status of a scheduled event. The server maintains a record of all scheduled events, which is useful when managing Tivoli Storage Manager schedules on numerous client machines. The QUERY EVENT command allows an administrator to view the event records on the server. A useful query that shows all of the event results for the previous day is:
query event * * begind=today-1 begint=00:00:00 endd=today-1 endt=23:59:59
Or, the query results can be limited to exception cases:

query event * * begind=today-1 begint=00:00:00 endd=today-1 endt=23:59:59 exceptionsonly=yes
The query results include a status field that gives a summary of the result for a specific event. By using the format=detailed option you can also see the result of an event that is the overall return code passed back by the Tivoli Storage Manager client. The following define the status and meaning of the event status codes that might exist for a scheduled event that has already taken place: Completed The scheduled client event ran to completion without a critical failure. There is a possibility that the event completed with some errors or warnings. Query the event with detailed format to inspect the event result for more information. The result can either be 0, 4, or 8.
Missed The schedule start window has elapsed without action from the Tivoli Storage Manager client. Common explanations for this result include the schedule service not running on the client, or a previously scheduled event not completing for the same or a different schedule. Started Typically, this indicates that a scheduled event has begun processing. However, if an event showing a status of Started is followed by one or more Missed events, it is possible that the client scheduler encountered a hang while processing that event. One common cause for a hanging client schedule is the occurrence of a user interaction prompt such as a prompt for an encryption key that is not responded to. Failed The client event ran to completion; however, a critical failure occurred. v If a scheduled event is missed but other consecutive scheduled events for that node show a result of Completed, check for errors in the server activity log and the client schedule log for more information. When checking the server activity log, narrow the query results down to the time frame surrounding the scheduled event. Begin the event log query at a time shortly before the start window of the scheduled event in question. For example, if investigating the following suspect event:
Scheduled Start Actual Start Schedule Name Node Name Status -------------------- -------------------- ------------- ------------- ------08/21/2003 08:27:33 HOURLY NODEA Missed
You could use one of the following queries:

query actlog begind=08/21/2003 begint=08:25:00 query actlog begind=08/21/2003 begint=08:25:00 originator=client node=nodea
The Tivoli Storage Manager client keeps a detailed log of all scheduled activities. If queries of the servers activity log cannot explain a failed scheduled event, check the clients local schedule log. You must have access to the client machine in order to inspect the schedule log. The schedule log file is typically stored in the same directory as the dsmerror.log in a file named dsmsched.log. The location of the log file can be specified using client options, so you may need to refer to the options file to see if the SCHEDLOGNAME option was used to relocate the log file. On Windows, the schedule log can also be relocated by an option setting which is part of the schedule service definition. You can issue the DSMCUTIL QUERY command to check if this option has been set. When you locate the schedule log, search through the file to find the time period corresponding with the start date and time of the scheduled event in question. The following are some tips on what you might search: - If you are investigating a missed event, check the details of the previous event, including the time at which the previous event completed. - If you are investigating a failed event, look for error messages that explain the failure (such as the server session limit being exceeded). - When an explanation is still not clear, the last place to check is the clients error log file (usually named dsmerror.log). v Use the SHOW PENDING diagnostic tool to display schedules, nodes, and when they should next run. v Other helpful information you can use when diagnosing the reasons why a node missed a scheduled event include viewing the dsm.sys stanza for the node and the MANAGEDSERVICES, PRESCHEDCMD, and POSTSCHEDCMD option values from the client options file.
10
v Start and stop the client service. When you manage a large number of clients running scheduler processes, you might want to be able to start and stop the client service from a remote machine. The Tivoli Storage Manager client for Windows provides a utility to assist with remote management of the scheduler service. For other platforms, standard operating system utilities are required. Windows To remotely manage the client scheduler service using dsmcutil with the /machine: option, you must have administrative rights in the domain of the target machine. To determine whether the scheduler service is running on a remote machine, check the Current Status field from a query similar to the following:
dsmcutil query /name:"TSM Client Scheduler" /machine:ntserv1.ibm.com
Issue the following queries to restart a scheduler service that is missing schedules:
dsmcutil stop /name:"TSM Client Scheduler" /machine:ntserv1.ibm.com dsmcutil start /name:"TSM Client Scheduler" /machine:ntserv1.ibm.com
Consequently, if you use the CAD to manage the scheduler, you may have to restart the CAD service or stop the scheduler service and restart the CAD service with the following queries:
dsmcutil dsmcutil dsmcutil dsmcutil dsmcutil query /name:"TSM Client Scheduler" /machine:ntserv1.ibm.com query /name:"TSM Client Acceptor" /machine:ntserv1.ibm.com stop /name:"TSM Client Scheduler" /machine:ntserv1.ibm.com stop /name:"TSM Client Acceptor" /machine:ntserv1.ibm.com start /name:"TSM Client Acceptor" /machine:ntserv1.ibm.com
UNIX A shell script can be written to search for and stop running Tivoli Storage Manager schedulers or client acceptor processes, and then restart the processes. Software products such as Symarks Power Broker allow Tivoli Storage Manager administrators limited access to UNIX servers for the purpose of managing the scheduler processes, and copying off the schedule log file. The following shell script is an example of how to recycle the Tivoli Storage Manager scheduler process:
#!/bin/ksh # Use the following script to kill the currently running instance # of the TSM scheduler, and restart the scheduler in nohup mode. # # This script will not work properly if more than one scheduler # process is running. # If necessary, the following variables can be customized to allow an # alternate options file to be used. # export DSM_DIR= # export DSM_CONFIG= # export PATH=$PATH:$DSM_DIR # Extract the PID for the running TSM Scheduler PID=$(ps -ef | grep "dsmc sched" | grep -v "grep" | awk {print $2}); print "Original TSM scheduler process using PID=$PID" # Kill the scheduler kill -9 $PID # Restart the scheduler with nohup, redirecting all output to NULL # Output will still be logged in the dsmsched.log nohup dsmc sched 2>&1 > /dev/null & # Extract the PID for the running TSM Scheduler PID=$(ps -ef | grep "dsmc sched" | grep -v "grep" | awk {print $2}); print "New TSM scheduler process using PID=$PID"
11
Including or excluding client files during backup processing

If you implicitly or explicitly indicate that a file should be included or excluded during backup processing and it was not processed correctly, there are several possible causes.
Excluding files automatically from backup processing

The backup application should not back up some files. Two reasons for this could be that there are files that were identified by the operating system as not necessary for backup, or there are files that the Tivoli Storage Manager uses for internal processing. If these files must be included in the backup processing, Tivoli Storage Manager can include them by putting include statements in the client options set on the server. Attention: Because these files were explicitly identified as files not being backed up, including them in the server client options set is not recommended. Issue the Backup-Archive clients DSMC QUERY INCLEXCL command to identify these files. The output from this command (below) shows Operating System as the source file for files that were automatically excluded from backup processing.
tsm> q inclexcl *** FILE INCLUDE/EXCLUDE *** Mode Function Pattern (match from top down) Source File ---- --------- ------------------------------ ----------------Excl All C:\WINDOWS\Registration\*.clb Operating System Excl All C:\WINDOWS\netlogon.chg Operating System
The following files are automatically excluded: Windows v Files enumerated in the HKLM\SYSTEM\CurrentControlSet\Control\ BackupRestore\FilesNotToBackup registry key v The client staging directory C:\ADSM.SYS v RSM database files (these files are processed in the system object or system state backup) v IIS metafiles (these files are processed in the system object or system state backup) v Registry files (these files are processed in the system object or system state backup) v Client trace file UNIX Client trace file NetWare Client trace file Macintosh v Volatile, temporary, and device files used by the operating system v Client trace file
Excluding Windows system files

Windows system files are silently excluded from the system drive backup processing and cannot be included.
12
| | |
To process these Windows system files, you must issue a DSMC BACKUP SYSTEMOBJECT command (Windows 2000 and Windows XP) or a DSMC BACKUP SYSTEMSTATE command (Windows 2003 and Windows Vista). The Windows system files are excluded from the system drive backup processing because they are usually sent during the system object or system state backups. System files are boot files, catalog files, performance counters, and files protected by the Windows system file protection (sfp). These files will not be processed during backup of the system drive but are excluded from the system drive processing internally instead of relying on explicit exclude statements. This is due to the sheer number of exclude statements that would be needed to represent all of these files. Backup performance can be adversely affected. You can issue the Backup-Archive client DSMC QUERY SYSTEMINFO command to identify the Windows system files. The output of this command is written to the dsminfo.txt file.
(partial contents of the dsmfino.txt file) ===================================================================== SFP c:\windows\system32\ahui.exe (protected) c:\windows\system32\apphelp.dll (protected) c:\windows\apppatch\apphelp.sdb (protected) c:\windows\system32\asycfilt.dll (protected)
Excluding files with the EXCLUDE.DIR statement

EXCLUDE.DIR statements exclude all directories and files under the parent directory. If you want to include all files based on a file pattern, regardless of their location within a directory structure, do not use the EXCLUDE.DIR statements. For example, consider this set of UNIX include/exclude statements:
exclude.dir /usr include /.../*.o
The include statement in this example indicates that all files with a .o extension should be included, but the preceding exclude.dir statement will exclude all files in the /usr directory, even if they have a .o extension. This would be true regardless of the order of the two statements. If you want to back up all the files ending with .o use the following syntax:
exclude /usr/.../* include /.../*.o
When using wildcards in INCLUDE/EXCLUDE, use * if you want all the files rather than *.*.*.*, which means to include/exclude all files containing at least one dot (.) character, while * means to include/exclude all files. If you use *.*, files containing no dot characters (such as C:\MYDIR\MYFILE) will not be filtered.
Backing up a single file does not honor EXCLUDE.DIR

If you want to perform a selective backup of a single file from the command-line client, it will not be affected by the EXCLUDE.DIR option.
13
If you issue a selective backup from the command-line client of a single file, the file is processed, even if there is an EXCLUDE.DIR statement which excludes one of the parent directories. For example, consider the following UNIX include/exclude statement which is used in subsequent command-line actions:
exclude.dir /home/spike
The selective backup shown below will always result in the file being processed:
dsmc selective /home/spike/my.file
If you issue a selective backup using a wildcard, no files are processed because the directory is excluded:
dsmc selective "/home/spike/my.*"
Note: A subsequent incremental backup of the /home file system will inactivate the file /home/spike/my.file. EXCLUDE.DIR statements should not be terminated with a directory delimiter. The following are examples of invalid EXCLUDE.DIR statements, due to a terminating directory delimiter:
exclude.dir exclude.dir exclude.dir exclude.dir /usr/ (UNIX) c:\directory\ (Windows) SYS:\PUBLIC\ (NetWare) Panther:User: (Macintosh)
The following examples show the correct coding of EXCLUDE.DIR:

exclude.dir exclude.dir exclude.dir exclude.dir /usr (UNIX) c:\directory (Windows) SYS:\PUBLIC (NetWare) Panther:User (Macintosh)
Using include/exclude statements in the client option set

The Tivoli Storage Manager administrator can include or exclude files on behalf of the client. Include or exclude statements that come from the server will override include and exclude statements entered in the local client option file. Contact the Tivoli Storage Manager server administrator to correct the problem. You can issue the Backup-Archive client DSMC QUERY INCLEXCL command to identify files that are included or excluded by the server client options set. The output from this command shows Operating System as the source file for files that have been automatically excluded from backup processing. See the output below:
tsm> q inclexcl *** FILE INCLUDE/EXCLUDE *** Mode Function Pattern (match from top down) Source File ---- --------- ------------------------------ ----------------Excl All /.../*.o Server Incl All /.../*.o dsm.sys
In the example above, the users indicated that they wanted all files that end with a .o extension to be included in the local options file, but the server sent the client an option to exclude all files that end with a .o extension. The server-provided option prevails.
14
Dictating incremental copy frequency through a server policy

Tivoli Storage Manager server policy dictates an incremental copy frequency that is non-zero. The copy frequency attribute of the current copygroup management class for the file you specified dictates the minimum number of days that must elapse between successive incremental backups. If you are trying to perform an incremental backup on a file and this number is set higher than 0 days, then the file will not be sent to the Tivoli Storage Manager server, even if it has changed. A number of steps can be taken to correct this problem: v Contact the Tivoli Storage Manager server administrator to change the copy frequency attribute v Issue a selective backup of the file. For example, DSMC SELECTIVE C:\FILE.TXT You can issue the following administrative client QUERY COPYGROUP command to determine the setting of the copy frequency parameter:
tsm: WINBETA>q copygroup standard active f=d Policy Domain Name: STANDARD ... Copy Frequency: 1 ...
Compression, encryption, and subfile backup statements

Include and exclude statements for compression, encryption, and subfile backup do not imply that the files will be included for backup processing Include and exclude statements for compression (INCLUDE.COMPRESS), encryption (INCLUDE.ENCRYPT), and subfile backup (INCLUDE.SUBFILE) do not imply that the file will be included for backup processing. You can use the include and exclude statements in combination with the COMPRESS, ENCRYPT, and SUBFILE statements to produce your desired results. Consider the following UNIX example:
exclude /usr/file.o include.compress /usr/*.o
This statement indicates that the /usr/file.o file is excluded from backup processing. The include.compress statement indicates that if a file is a candidate for backup processing and matches the pattern /usr/*.o; then compress the file. The include.compress statement should not be interpreted as backup all files that match the pattern /usr/*.o and compress them. If you want to back up the /usr/file.o file in this example, you must remove the exclude statement.
Using platform-specific include/exclude statements

A platform-specific include/exclude statement contains syntax for everything and all files under a specific directory. If the volume delimiters or directory delimiters are not correct, this might cause include and exclude statements not to function properly.
15
If you want to use an include statement for all files under a specific directory, ensure that the slashes and volume delimiters are correct. If you want to exclude all of the files under a directory called home, or simply all files, see the following examples: Using the backwards slash \ and the volume delimiter : (Windows)
*include everything in the c:\home directory include c:\home\...\* *include everything include *:\...\*
Using the forward slash / (UNIX and Macintosh OS X)

*include everything in the /home directory include /home/.../* *include everything include /.../*
Using either slash (NetWare) but the volume name is required with the volume delimiter :
*include everything in the SYS:\HOME directory include sys:\home\...\* or include sys:/home/.../* *include everything include *:\...\* or include *:/.../*
Resolving errors due to the incorrectly coded include/exclude list

Due to the complexity or number of include/exclude statements, you might experience the unintentional inclusion or exclusion of a file. A new trace statement was added to the Tivoli Storage Manager backup-archive client Version 5.2.2 that can help determine why a file was included or excluded. This is done by configuring the client with the INCLEXCL traceflag. For example, the user believes that the c:\home\file.txt file should be included in the backup processing. The trace shows that there is an exclude statement that excludes this file:
polbind.cpp (1026): File C:\home\file.txt explicitly excluded by pattern Excl All c:\home\*.txt
Further investigation into using the backup-archive client DSMC QUERY INCLEXCL command shows that this statement is coming from the Tivoli Storage Manager server client options set:
tsm> q inclexcl *** FILE INCLUDE/EXCLUDE *** Mode Function Pattern (match from top down) Source File ---- --------- ------------------------------ ----------------Excl All c:\home\*.txt Server
Resolving image backup errors

The two errors that might occur with the image backup process are a Linux image backup failure or a Linux Snapshot image backup failure.
Resolving Linux image backup errors

You can resolve Linux image backup errors by following the specific steps provided.
16
The following error was generated during image backup:

paris:#dsmc b image /dev/system/lv01 Backup Image Function Invoked. ANS1228E Sending of object /dev/system/lv01 failed ANS1584E Error loading system library libdevmapper.so required for image operations for LVM2 volumes. ANS1813E Image Backup processing of /dev/system/lv01 finished with failures. Total number of objects inspected: 1 Total number of objects backed up: 0 Total number of objects updated: 0 Total number of objects rebound: 0 Total number of objects deleted: 0 Total number of objects expired: 0 Total number of objects failed: 1 Total number of bytes transferred: 0 B Data transfer time: 0.00 sec Network data transfer rate: 0.00 KB/sec Aggregate data transfer rate: 0.00 KB/sec Objects compressed by: 0% Elapsed processing time: 00:00:29 paris# cat dsmerror.log 11/15/2006 13:07:53 ANS1228E Sending of object /dev/system/lv01 failed 11/15/2006 13:07:56 ANS1584E Error loading system library libdevmapper.so required for image operations for LVM2 volumes. 11/15/2006 13:07:56 ANS1813E Image Backup processing of /dev/system/lv01 finished with failures.
Ensure that the system has installed the correct version of the libdevmapper. Perform the following steps to determine the installed version: 1. Issue the # DMSETUP VERSION command. The output will display similar to the following output:
Library version: 1.00.09-ioctl (2004-03-31) Driver version: 4.4.0
or Issue the following to determine the version using the rpm:

dumas:/develop/rem/peterman # rpm -q -a |grep device-mapper
The output will display similar to the following output:

device-mapper-1.00.09-17.5
The library version must be version 1.01 or above. If you have a lower version, please upgrade the devmapper rpm at the following Web site: http://www.novell.com. 2. Verify the installation after the upgrade.
dumas:/develop/rem/peterman # rpm -Uvh device-mapper-1.01.01-1.6.i586.rpm Preparing... ########################################### [100%] 1:device-mapper ########################################### [100%] dumas:/develop/rem/peterman # rpm -q -a |grep device-mapper device-mapper-1.01.01-1.6
You can also check the /lib directory to see the correct versions installed. A system with the right levels will have the following:
17
# ls -l /lib/libdev* lrwxrwxrwx 1 root root 20 Jul 5 11:42 /lib/libdevmapper.so ->libdevmapper.so.1.01 -rwxr-xr-x 1 root root 24490 May 23 2005 /lib/libdevmapper.so.1.00 -rwxr-xr-x 1 root root 28216 May 23 2005 /lib/libdevmapper.so.1.01
Linux Snapshot image backup fails with error message ANS1258E

To resolve a failed Linux Snapshot image backup, validate that the system is set up to create a Snapshot. Try to create a Snapshot from a shell command prompt by issuing the following command:
/sbin/lvcreate -L 16384K -n <snapname eg. tsmsnap>-s <volume devname eg /dev/system/lv01>
If this command fails with error Snapshot: Required device-mapper target(s) not detected in your kernel, this means that the :dm_snapshot kernel module is not loaded. This command could also fail for other reasons as well, which may result in similar Tivoli Storage Manager behavior. The following is an example of the output generated when an image backup fails with error message ANS1258E, The image snapshot operation failed.
dsmerror.log : 05/31/2006 15:14:36 ANS1259E The image snapshot operation failed. Diagnostic text: tsmStartSnapshot. 05/31/2006 15:14:38 ANS1259E The image snapshot operation failed. Diagnostic text: tsmTerminateSnapshot. 05/31/2006 15:14:38 ANS1228E Sending of object /fs1 failed 05/31/2006 15:14:38 ANS1258E The image snapshot operation failed.
Perform the following steps to load the modules: 1. Verify that the module is not loaded. See the following example:
suzie:/home2/rat # lsmod |grep dm_ dm_mod 112104 6
2. Load the module. See the following example:

suzie:/home2/rat # modprobe dm_snapshot
3. Verify that the previous step is successful. See the following example:
suzie:/home2/rat # lsmod |grep dm_ dm_snapshot 44024 0 dm_mod 112104 6 dm_snapshot suzie:/home2/rat #
4. Create a Snapshot from the shell prompt. See the following example:
suzie:/etc # /sbin/lvcreate -L 16384K -n tsmsnap -s /dev/system/lv01 Logical volume "tsmsnap" created
5. Remove the Snapshot that was created in the previous step. See the following example:
suzie:/etc # lvremove /dev/system/tsmsnap Do you really want to remove active logical volume "tsmsnap"? [y/n]: y Logical volume "tsmsnap" successfully removed suzie:/etc #
You can now run Snapshot image backups.
18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Resolving errors during AIX JFS2 snapshot-based backup-archive and image backup
During Tivoli Storage Manager termination, the Tivoli Storage Manager client deletes the AIX enhanced journaled file system (JFS2) snapshot that is created during the backup process. However, there are situations in which AIX might fail the snapshot delete request made by Tivoli Storage Manager. The following are situations where a snapshot delete request might fail: v The Control-c keystroke is issued during a Tivoli Storage Manager snapshot backup process. The JFS2 snapshot unmount request might fail with a Device Busy error, due to the Tivoli Storage Manager process being in the middle of accessing the snapshot. v Two Tivoli Storage Manager snapshot backup requests are invoked concurrently for the same file system. For example, if the dsmc backup image /fs1 backup request is submitted from one console, and at the same time a dsmc backup image /fs1 backup request is issued from another console. If the process from the first console creates the first snapshot for /fs1 and the second process from the second console creates the second snapshot for /fs1, and if the second process finishes first and tries to delete the snapshot, AIX fails the delete request. v Two Tivoli Storage Manager snapshot backup requests are invoked concurrently for two virtual mount points whose source file system is the same. For example, issuing dsmc incr /fs1/level1/dir1 from one console and dsmc incr /fs1/level2/level3/dir3 from a second console, concurrently. AIX expects snapshot delete requests to be issued in a certain order with the oldest snapshot deletion requested first, and the next oldest snapshot deletion requested next, and so on. If Tivoli Storage Manager cannot honor the sequence due to concurrent processes creating snapshots for the same file system, AIX fails the delete requests. In the previous examples, Tivoli Storage Manager logs a warning message asking the user to delete the snapshots manually. Issue the following commands, in order, to perform a manual snapshot deletion: 1. SNAPSHOT -Q -C <SRCFS> 2. DF -K 3. UNMOUNT -F /TSM* 4. RMDIR /TSM* 5. SNAPSHOT -D /DEV/TSM* If the snapshot delete process fails with Device Busy or some other error message, issue the - UNMOUNT -F <SRCFS> command to unmount the source file system. Retry snapshot delete. 6. LS -L /DEV/TSM* If any /DEV/TSM* logical volumes remain, issue the - RMLV -F TSM* command. 7. If you have an unmounted source file system, issue the - MOUNT <SRCFS> command to mount it. If any snapshots are not deleted during a previous Tivoli Storage Manager process, Tivoli Storage Manager tries to delete the snapshots during its next invocation. The reason for this is that while older snapshots remain, AIX fails deletion requests for newer snapshots for a given file system. The following are some cases where Tivoli Storage Manager does not try to delete older snapshots: v If the snapshot was not created by Tivoli Storage Manager. Tivoli Storage Manager names its snapshots with a tsm prefix to distinguish them from other snapshots created for the same file system. If the snapshot was not created by
19
| | | | | | | | | | | | | | | | | | | | |
Tivoli Storage Manager, Tivoli Storage Manager produces an error message that asks the user to delete the older snapshot and retry the operation. v If the snapshot is created by Tivoli Storage Manager but is still mounted, the snapshot is being used by some other Tivoli Storage Manager process. v If the snapshot is created by Tivoli Storage Manager, is not mounted, but is newly created, the snapshot may have just been created by some other Tivoli Storage Manager process. In all such cases, you might have to perform a manual deletion. If any unused older snapshots are existing, this causes subsequent Tivoli Storage Manager backups to fail to delete snapshots. Note: There are AIX defect fixes related to JFS2 snapshots in AIX 5.3.0.70 or later and AIX 6.1 or later. If the fixes are not applied, an AIX system crash can occur or Tivoli Storage Manager might hang during snapshot deletion and snapshot query processes. It may also cause data corruption during used block image backup. Therefore, Tivoli Storage Manager will not perform snapshot monitoring, will not perform the snapshot deletion feature described above, and will not perform used block image backup unless AIX is at the AIX 5.3.0.70 or later and AIX 6.1 or later levels. In order to exploit these features, ensure that your operating system level is at AIX 5.3.0.70 or later and AIX 6.1 or later.
API problem determination

There are several methods you can use to determine the source of problems with the application programming interface (API).
Support solutions for the Tivoli Storage Manager API

A number of resources are available to learn about or to diagnose the Tivoli Storage Manager application programming interface (API). The IBM Support Solutions database contains a knowledge base of articles and information on issues related to backup-restore problems. API instrumentation will only be activated if the testflag INSTRUMENT: API is set in the configuration file and the dsmSetUp / dsmCleanUp calls are used in the application. See Support for IBM Tivoli Storage Manager to review API information.
Gathering information before calling IBM support (API)

By collecting all the necessary information about your environment, you can significantly help to determine the problem. Gather as much of the following information as possible before contacting IBM Support: v On what operating system is the problem being experienced? v What is the exact level of the operating system, including all service packs and hot fixes that were applied? v What is the exact level of the Tivoli Storage Manager API? v What is the exact level of the Tivoli Storage Manager server? v What is the Tivoli Storage Manager Server platform and operating system level?
20
v What is the exact level of the Tivoli Storage Manager storage agent (if LAN-free environment)? v What is the Tivoli Storage Manager Storage Agent platform and operating system level (if LAN-free environment)? v List all applications running on the system. v List the steps required to recreate the problem (if the problem can be recreated). v If you cannot recreate the problem, list the steps that caused the problem.
Gathering files before calling IBM support (API)

A number of log files and other data may be created by the Tivoli Storage Manager API. Gather as many of the following files as possible before contacting IBM Support: v Tivoli Storage Manager API options file. The following are the two options files used on UNIX and OS/400 operating systems: dsm.opt The client options file dsm.sys The system options file For Windows, find the dsm.opt default options file or the file referenced by the DSMI_CONFIG environment variable. For UNIX, the default options file is dsm.sys and is found in the directory referenced by the DSMI_DIR environment variable. On other operating systems, the client options file (dsm.opt) contains all of the options. The environment variables that describe the location of the option files and other API components are below: DSMI_CONFIG The fully-qualified name for the client options file. DSMI_DIR This points to the API installation directory and also is used to find the dsm.sys file (on Unix). Wherever the DSMI_DIR is set, ensure that a dsm.sys file exists in this same directory. DSMI_LOG Points to the path for the dsierror.log file. Note: If this variable points to a directory for which the user does not have write permission, dsmSetup and dsmInitEx will fail with return code DSM_RC_ACCESS_DENIED (106). If the ERRORLOGNAME option is set in the options file (dsm.sys/dsm.opt), its value will be used as the error log name instead of the default value dsierror.log. Perform the following steps to verify that the API uses the correct option file and/or server stanza in dsm.sys: 1. Insert an erroneous option or value in the client option file or server stanza in dsm.sys. For example, if it is uncertain whether the API uses the srvr1.cmpron server, insert ERRORNEOUS_OPTION 12345 line into the srvr1.cmpron server stanza of the dsm.sys file. See the following example: SERVERNAME srvr1.cmproff COMPRESSION NO TCPSERVERADDRESS machine.company.com
21
SERVERNAME srvr1.cmpron COMPRESSION YES ERRORNEOUS_OPTION 12345 TCPSERVERADDRESS machine.company.com SERVERNAME srvr1.pwdfl PASSWORDACCESS GENERATE PASSWORDDIR .
TCPSERVERADDRESS machine.company.com 2. Verify that the API detects this error. You can use the sample API program dapismp for this purpose.
# dapismp ... Enter selection ==>0 Node name:node1 Owner name: Password: API Config file: Session options: User Name: User pswd: Are the above responses correct (y/n/q)? Doing signon for node node1, owner, with password *** Init failed: ANS0220E (RC400) An invalid option was found during option parsing.
The wrong options file was updated if no error is reported. 3. Check the environment variable values that were previously mentioned or repeat Steps 1 on page 21 and step 2 with a different options file/server stanza. 4. Remove the option inserted in Step 1 on page 21. Tivoli Storage Manager API error log file. The default API error log is dsierror.log. Any trace files created for the API (the recommended trace flags are api, api_detail, or verbdetail). Output from any failed command or operation. This may be either the console output redirected to a file or an actual screen image of the failure. The output from the Tivoli Storage Manager server QUERY SYSTEM command. Tivoli Storage Manager server activity log. The IBM Tivoli Storage Manager administrator can view this log for you if you do not have a Tivoli Storage Manager administrator user ID and password. If the API client is configured for LAN-free data movement, collect the options file for the Tivoli Storage Manager storage agent. The default name for this options file is dsmsta.opt. A short program or sections of the application source code invoking the Tivoli Storage Manager API function calls that are suspected to cause the problem.
v v v v v
Determining if data is sent to the IBM Tivoli Storage Manager storage agent rather than the server
You need to know if your data is being sent to the Tivoli Storage Manager storage agent, rather than a server. Perform the following steps to verify that the data is being sent to the Tivoli Storage Manager storage agent, rather than the server:
22
1. Add the following trace options to the client options file prior to backing up or archiving objects: v TRACEFILE <trace file name> v TRACEFLAGS api api_detail verbdetail 2. Examine the trace file after the operation and locate a statement that looks similar to the following statement:
dsmSendObj ENTRY:... objNameP: <the file name>
The statement above is followed by the trace statement below:

tsmEndSendObjEx: Total bytes sent * *, encryptType is *** encryptAlg is *** compress is *, totalCompress is * * totalLFBytesSent * *
The trace statement indicates whether the object totalLFBytesSent was sent to the Tivoli Storage Manager storage agent. If totalLFBytesSent is 0 0, the data was sent directly to the Tivoli Storage Manager server. Alternatively, your application itself can determine whether the data was sent through a LAN-free path by using the dsmEndSendObjEx function call and the dsmEndSendObjExOut_t data structure.
/*-------------------------------------------------------------------------+ | Type definition for dsmEndSendObjExOut_t +-------------------------------------------------------------------------*/ typedef struct dsmEndSendObjExOut_t { dsUint16_t stVersion; /* structure version */ dsStruct64_t totalBytesSent; /* total bytes read from app */ dsmBool_t objCompressed; /* was object compressed */ dsStruct64_t totalCompressSize; /* total size after compress */ dsStruct64_t totalLFBytesSent; /* total bytes sent LAN Free */ dsUint8_t encryptionType; /* type of encryption used */ }dsmEndSendObjExOut_t; totalLFBytesSent - The total LAN-free bytes that were sent. For example: ... rc = dsmEndSendObjEx(&endSendObjExIn, &endSendObjExOut); if (rc) { printf("*** dsmEndSendObjEx failed: "); rcApiOut(dsmHandle, rc); } else { dI64toCh(&endSendObjExOut.totalLFBytesSent,t,10); format_number(t,t2); printf("LAN-free bytes sent: %s\n", t2); } ...
See API Function Calls in Using the Application Programming Interface for more details.
Running applications that use the API as a non-root user

You must perform specific steps if you are a non-root user trying to run an application on UNIX that uses the application programming interface (API). Perform the following steps if you are a non-root user trying to run an application on UNIX that uses the API: 1. Set the DSMI_CONFIG environment variable. Verify that the non-root user has read permission for the client options file specified by DSMI_CONFIG. Otherwise, dsmInit/dsmInitEx will fail with return code DSM_RC_NO_OPT_FILE
23
(406). For example, the following options file is not readable by a non-root user, therefore the files permissions need to be updated:
$ ls -l $DSMI_CONFIG -rwx-----1 root sys 86 Oct 7 13:07 /testfsapi/callmt_nr/dsm.opt $ su root Password: # chmod a+r /testfsapi/callmt_nr/dsm.opt # exit $ ls -l $DSMI_CONFIG -rwxr--r-1 root sys 86 Oct 7 13:07 /testfsapi/callmt_nr/dsm.opt
2. Set the DSMI_DIR environment variable to the API installation directory. Verify that the non-root user has read permission for the system options file specified by $DSMI_DIR/dsm.sys.
$ export DSMI_DIR=/opt/tivoli/tsm/client/api/bin64 $ ls -l $DSMI_DIR/dsm.sys -rw-r--r-1 root sys 4712 Oct 19 18:07 /opt/tivoli/tsm/client/api/bin64/dsm.sys
3. Set the DSMI_LOG environment variable. Verify that the non-root user has write permission for this directory. For example, the following DSMI_LOG directory is owned by a non-root user:
$ ls -ld $DSMI_LOG drwxr-xr-x 2 apitest users 96 Oct 19 17:56 /testfsapi/callmt_nr/logs
If PASSWORDACCESS GENERATE is set in system options file dsm.sys, perform steps 4 and 5, otherwise go to step 6. 4. Check the ownership and permissions of the Trusted Communication Agent (TCA). This information is in the directory indicated by the DSMI_DIR environment variable. For example, the following TCA has the correct ownership and permissions:
$ ls -l $DSMI_DIR/dsmtca -rwsr-xr-x 1 root bin 5021160 Oct 14 09:48 /opt/tivoli/tsm/client/api/bin64/dsmtca
Wrong permissions or ownership will result in a DSM_RC_AUTH_FAILURE (137) returned from dsmInit. Additionally, it is imperative that you use the same version of the API library and dsmtca. Mixed versions will result in errors.
Error : calling program and dsmtca are not compatible calling program build date : Mon Oct 18 21:15:59 2004 Mon Oct 18 21:15:59 2004 TCA build date : Wed Oct 13 16:48:03 2004 Wed Oct 13 16:48:03 2004 *** Init failed: ANS0282E (RC168) Password file is not available.
5. The root user must generate the TSM.PWD password file using either the IBM Tivoli Storage Manager backup-archive client or the dapismp sample API application. Location of the password file is determined by the PASSWORDDIR option in the dsm.sys system options file. In the following example, the sample API application generates the TSM.PWD password file for a node whose password is oddesy.
24
# dapismp ************************************************************************* * Welcome to the sample application for the Tivoli Storage Manager API. * * API Library Version = 5.3.0.0 * ************************************************************************* Choose one of the following actions to test: 0. Signon 1. Backup 2. Restore 3. Archive 4. Retrieve 5. Queries 6. Change Password 7. Utilities : Deletes, Updates, Logevent, SetAccess, RetentionEvent 8. Set preferences, envSetUp 9. Exit to system 10. Restore/Retrieve Without Offset Prompt 11. Extended Signon Enter selection ==>0 Node name: Owner name: Password:oddesy API Config file: Session options: User Name: User pswd: Are the above responses correct (y/n/q)? Doing signon for node, owner, with password oddesy Handle on return = 1 Choose one of the following actions to test: 0. Signon 1. Backup 2. Restore 3. Archive 4. Retrieve 5. Queries 6. Change Password 7. Utilities : Deletes, Updates, Logevent, SetAccess, RetentionEvent 8. Set preferences, envSetUp 9. Exit to system 10. Restore/Retrieve Without Offset Prompt 11. Extended Signon Enter selection ==>9 # ls -l TSM.PWD -rw------- 1 root sys 121 Oct 19 18:28 TSM.PWD Function call dsmInit returns DSM_RC_NO_PASS_FILE (168), if the password file is not present in the directory specified by the PASSWORDDIR option.
6. If tracing is enabled, verify that the non-root user has write permission for the file indicated by issuing the TRACEFILE option.
JBB problem determination

Journal Based Backup (JBB) is appropriate for backing up files systems with small or moderate amounts of change activity between backup cycles. | See the Journal Based Backup FAQ site for JBB problem determination help.
Determining if a backup will be journal-based

Before implementing a backup, you need to determine if your backup is going to be journal-based. The following conditions must be met in order for a backup to be journal-based: v The journal daemon must be configured to journal the file system being backed up.
25
v The journal for the file system being backed up must be in the valid state. v The Tivoli Storage Manager node and server that the backup is using must match the node and server for which the journal is valid. The journal daemon journalizes a file system after you list the file system in the tsmjbbd.ini configuration file. See the following configuration information:
[JournaledFileSystemSettings] ; ; List of journalized file systems JournaledFileSystems=c:
If a journal is to be valid, you must perform a full incremental backup on the corresponding file system while the file system is actively being journalized. This full incremental backup must set the Last Backup Completed date on the Tivoli Storage Manager server file space in order for the journal to be set to valid. You can view the Last Backup Completed date by issuing the QUERY FILESPACE server command. After the journal is set to the valid state, subsequent backups by the same node to the same Tivoli Storage Manager server will be journal-based. If a backup uses a different node and/or server, the backup will be non-journal-based but the journal will remain valid for the original node and server, and backups to the original node and server will be journal-based. The following message is an example of what is written to the Windows Application Event Log when a journal is initially set to valid:
Journal set to valid for fs H: and will be used for backup by node GSHLAGER3 to server GSHLAGER2_SERVER1.
The Journal Database Viewing Utility may also be used to determine the current state of a journal. If a valid journal is restarted, backups will be non-journal based until the journal is re-validated. The following message is written to the Windows Application Eventlog when a journal is restarted:
Journal database c:\tsmjournal\tsmH__.jdb for fs H: has been deleted and reset to the invalid state.
Specific messages detailing the specific cause of the journal restart are also written to both the eventlog and the journal errorlog (jbberror.log). The following are reasons for restarting a valid journal: v Error conditions in the journal daemon buffer overflow errors caused by excessive change activity on the journal file system being monitored for changes journal database access errors (disk full errors, etc.) v Request by a backup client v Clients will issue a journal restart request when it is determined that a journal file system lacks integrity for one of the following reasons: The server filespace no longer exists The server filespace was deleted after the last backup The node policy set was updated after the last backup The Last Backup Completed or Last Backup Started dates are not valid (not set)
Running the Journal Daemon in the foreground

You can improve the diagnostic capabilities and your ability to test by running the Journal Daemon in the foreground, rather than as a Windows service.
26
The Journal Daemon may be started from a Windows command prompt as follows: tsmjbbd.exe i
Using the journal database viewing utility

The journal database viewing utility can provide you with valuable information. The Journal Database Viewing Utility provides the following information: v The current state of the journal v The file system tracked by the journal v The journal activation timestamp v The journal validation timestamp v The maximum supported journal size v The node and server for which the journal is valid v The number of entries currently in the journal | | Note: You must have the 5.4 version of the utility (dbviewb.exe) in order to view the 5.4 journal databases. This can be found at the IBM software ftp site. This utility also allows searching, inserting, or deleting specific entries in a journal database. The syntax of this utility is:
dbviewb <fully qualified journal database basefile name> dbviewb <fully qualified journal database basefile name> <i> D:\tsm540c\debug\bin\winnt_unicode>dbviewb c:\tsmjournal\tsmh__.jdb IBM Tivoli Storage Manager Journal Database Viewing Utility Version 5, Release 4, Level 0.0 Last Update: Nov 28 2006 Querying Journal DB ... Journal Database Information: Database File c:\tsmjournal\tsmh__.jdb Database File Disk Size 81 KB (83754 Bytes) Journal File System H: Journal Activation Date Tue Nov 28 11:49:05 2006 Journal Validation Date Wed Nov 29 16:41:11 2006 Maximum Journal Size 8191 PB (9223372036854775807 Bytes) Journal Type Change Journal Journal State Valid Valid for Server GSHLAGER2_SERVER1 Valid for Node GSHLAGER3 Number of DB Entries 22 D:\tsm540c\debug\bin\winnt_unicode> D:\tsm540c\debug\bin\winnt_unicode>dbviewb c:\tsmjournal\tsmh__.jdb i IBM Tivoli Storage Manager Journal Database Viewing Utility Version 5, Release 4, Level 0.0 Last Update: Nov 28 2006 Querying Journal DB ... Journal Database Information: Database File c:\tsmjournal\tsmh__.jdb Database File Disk Size 81 KB (83754 Bytes) Journal File Syst em H: Journal Activation Date Tue Nov 28 11:49:05 2006 Journal Validation Date Wed Nov 29 16:41:11 2006 Maximum Journal Size 8191 PB (9223372036854775807 Bytes) Journal Type Change Journal Journal State Valid Valid for Server GSHLAGER2_SERVER1
27
Valid for Node GSHLAGER3 Number of DB Entries 22 Enter request on a single line, in the following format: Req-Type [Entry-key] Req-type may be one of the following: Del Delete a row from the database. The fully-qualified case sensitive file name is required. Find Find the entry whose key is the argument. List Print all the entries to stdout. No arguments are required. Quit Please enter your request: find H:\dbview.example\Dir3Depth1\F2.txt Located Journal Database Record: ----------------------------------------Object Name : H:\dbview.example\Dir3Depth1\F2.txt Action : Modify Object Type : File Inserted : Fri Dec 01 10:15:28 2006 Object Time : Fri Dec 01 14:15:28 2006 Hit Count : -2110169276 ----------------------------------------Please enter your request: quit
Using open file support and the logical volume snapshot agent
There are several methods in determining problems with the open file support (OFS) and logical volume snapshot agent (LVSA).
Examining the Windows system event log

In Tivoli Storage Manager 5.3.0, support was added to the tsmlvsa.sys driver to write critical information for problem determination to the Windows system event log. Examining the event log is the first step in isolating potential problems with the logical volume snapshot agent (LVSA) in the context of online image or open file backup.
Obtaining debug print output

Many problems require that you obtain trace data from the tsmlvsa.sys driver to supplement what is obtainable from a client service trace. You can obtain the logical volume snapshot agent (LVSA) debug print output by using the DebugView tool. Perform the following steps to install the DebugView tool: 1. Download the latest version of the DebugView tool. Make sure you use the version that is labeled you plan on using DebugView on WinNT/2K/XP. 2. Install the DebugView tool. This is a simple extraction of the files from the dbgvnt.zip file. 3. Place the following text line in the dsm.opt file:
TRACEFLAG SERVICE TRACEFILE trace.txt
Ensure that you direct the trace file to a location with many GB of free space. 4. Run the dbgview.exe executable prior to running the Tivoli Storage Manager client. 5. Configure dbgview.exe to log the output to a file through the File Log to File option.
28
6. Perform the failing operation.
Configuring the system for a full dump

When a system bug check occurs, you must obtain a full memory dump to assist in the diagnosis of possible logical volume snapshot agent (LVSA) problems. The failing system must be configured to take a full memory dump. The following are the proper configuration steps: Open the control panel. Open the system icon. Select the advanced tab. Select the startup and recovery button. Ensure that the following checkboxes are selected in the System Failure section: v Write an event to the system log v Send an administrative alert 6. Ensure that the Automatically reboot checkbox is not selected. 7. In the write debugging information section select Complete Memory Dump. 1. 2. 3. 4. 5. Note: Make a note of where the file will be written (%SystemRoot%\ MEMORY.DMP). Ensure that the Overwrite any existing file checkbox is selected. Windows will ask you to restart so that the new setting can take effect. 8. Restart. If you did not select the Overwrite any existing file checkbox, after the restart rename the previous dump file (if any) to a new name. When a bug check occurs, note the contents of the bug check screen (commonly referred to as the Blue Screen Of Death or BSOD). Collect the memory.dmp file upon reboot for examination.
Forcing a dump for a system hang when a logical volume snapshot agent problem is suspected
If you have ensured proper configuration and a dump is not taken, a dump may need to be forced when the system hangs. There are two methods you might employ: 1. If you have the opportunity to restart and recreate the hang, see Microsoft Knowledge Base article 244138 Windows feature allows a Memory.dmp file to be generated with the keyboard. This method requires a registry change and reboot to enable an on-demand dump when the right CTRL key is held and SCROLL LOCK key is pressed twice. This method may also be required if the BANG! Tool mentioned below is unable to cause a bug check and memory dump. 2. If the system is hung and you cannot afford a reboot and recreate: a. Download and install BANG! from the Windows Driver Developers website. Follow instructions provided in the BANG! package/website. b. Run BANG! and press the Crash Now button. The system should get a blue screen and generate a full memory dump.
29
Note: IBM does not support the BANG! utility. Any questions or problems regarding the BANG! utility should be reported to the Windows Driver Developers website.
Best practices for open file support

The Tivoli Storage Manager Client open file support (OFS) technotes outline current limitations and known problems and to list steps that may help to diagnose problems in the setup and use of OFS See the following technote for more information: v IBM Tivoli Storage Manager v5.3
Windows Volume Shadow Copy Services

The Tivoli Storage Manager Windows client uses the Volume Shadow Copy Services (VSS) of Windows 2003 and Windows Vista to perform system state and system services backup. VSS can also be used as a snapshot provider for Open File Support (OFS) and online image operations.
Determining VSS transient errors

The Tivoli Storage Manager client considers several Volume Shadow Copy Services (VSS) errors to be transient. The Tivoli Storage Manager client considers the following VSS errors to be transient. When one of these errors occurs, the client will, by default, retry the VSS backup process three times at 30-second intervals. The number of retries and retry interval are configurable via two test flags. VSS_E_MAXIMUM_NUMBER_OF_VOLUMES_REACHED VSS_E_SNAPSHOT_SET_IN_PROGRES VSS_E_MAXIMUM_NUMBER_OF_SNAPSHOTS_REACHED VSS_E_PROVIDER_VETO VSS_E_UNEXPECTED VSS_E_FLUSH_WRITES_TIMEOUT VSS_E_HOLD_WRITES_TIMEOUT VSS_E_WRITERERROR_TIMEOUT VSS_E_WRITERERROR_RETRYABLE VSS_E_WRITERERROR_OUTOFRESOURCES VSS_E_WRITER_NOT_RESPONDING VSS_E_VOLUME_IN_USE VSS_E_PROVIDER_IN_USE VSS_E_UNEXPECTED_PROVIDER_ERROR VSS_E_UNEXPECTED_WRITER_ERROR
Defining Windows VSS test flags

The Tivoli Storage Manager client uses test flags to specify the number of retries and how long between retries. SETVSSMAXRETRY Specifies the number of times the Volume Shadow Copy Services (VSS) backup process is retried if a transient error occurs. The default value is to retry three times.
30
SETVSSDELAY Specifies the number of seconds to wait between retries of the VSS backup process, should a transient error occur. The default value is 60 seconds. Option file example:
retry 10 times at 300 second intervals TESTFLAG SETVSSMAXRETRY:10 TESTFLAG SETVSSDELAY:300
Windows 2003 VSS hot fixes

Microsoft issued several hot fixes for Volume Shadow Copy Services (VSS). IBM technote 1242128 lists the known hotfixes, but is not guaranteed to be current. You need to contact Microsoft for the most current VSS hotfixes.
Microsofts VSS tuning recommendations

Microsoft issued several tuning recommendations for Volume Shadow Copy Services (VSS).
Controlling the VSS diff Area Size

After you apply these fixes, one of the following events occurs: v The shadow copy of volume C: took too long to install v The shadow copy of volume C: was aborted because the diff area file could not grow in time. Consider reducing the I/O load on this system to avoid this problem in the future. If these events still occur, you can use the following registry key to control the size of the diff area used by VSS: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\VolSnap\ MinDiffAreaFileSize : REG_DWORD: <size in MB> (the default size is 300, but can be increased to 3000).
Recommended Event Log Maximum Size

Microsoft indicates that if the event logs are sufficiently large, the copy operation can take longer than the timeout for systems with high I/O load or high memory load. Microsoft recommends that logs are kept below 64 MB.
VSS diagnostic information for Microsoft assistance

Microsoft support can assist you with diagnostic information for Volume Shadow Copy Services (VSS) failures. If the VSS failure is outside of the scope of Tivoli Storage Manager, gather the following information for Microsoft support: v Windows application event log v Windows system event log v VSS trace (see instructions below)
31
Examine the application and system event logs focusing on the error events created by the VolSnap and VSS sources at the time of failure. You might want to extract the germane events from the log to isolate the problem and have a more productive interaction with MS support. Perform the following steps to perform a VSS trace: 1. Create a tracing.reg file using the contents shown below and change the TraceFile entry to point to a volume that is not going to have a shadow copy created. Note the double-backslash delimiter usage; you must enter \\ as the delimiter for each backslash in the path you wish to specify. 2. Double-click the file from within Windows Explorer to install tracing.reg. 3. Reproduce the problem. 4. Turn off tracing by deleting the HKEY_LOCAL_MACHINE\SYSTEM\ CurrentControlSet\Services\VSS\Debug\Tracing key. The following are the contents of the tracefile.reg registry file:
Windows Registry Editor Version 5.00 [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\VSS\Debug\Tracing] "TraceFile"="c:\\trace.txt" "TraceLevel"=dword:ffffffff "TraceEnterExit"=dword:00000001 "TraceToFile"=dword:00000001 "TraceToDebugger"=dword:00000000 "TraceFileLineInfo"=dword:00000001 "TraceForceFlush"=dword:00000000
Using the vsreq.exe sample program

The Volume Shadow Copy Services (VSS) Software Developers Kit (SDK) contains the vsreq (VSS requestor) sample program which performs a sequence of VSS API calls similar to the VSS API calls that are performed by the Tivoli Storage Manager backup-archive client. You might want to compile and run vsreq.exe on the failing system to determine if vsreq and Tivoli Storage Manager encounter the same problem. If vsreq can reproduce the same problem as Tivoli Storage Manager, then the output of vsreq can be supplied to MS support to help in diagnosing the VSS problem. In some cases, Microsoft provides an I/O subsystem analysis tool (yapt) to gather I/O performance data for analysis. vshadow is a tool that is also available as an alternative to vsreq.exe.
Comparing Tivoli Storage Manager and Ntbackup.exe interaction with VSS

Ntbackup.exe does not fully exploit Volume Shadow Copy Services (VSS) and cannot always be considered as a benchmark for Tivoli Storage Manager interaction with VSS. The following is the known difference between Ntbackup.exe and Tivoli Storage Manager in the context of VSS: Ntbackup.exe does not use VSS to back up the Active Directory (NTDS). Although Ntbackup.exe uses VSS to take a snapshot, it still uses the legacy NTDS backup API to read data from the disk. Tivoli Storage Manager uses the VSS interface to read NTDS data from the disk. If there is a problem with the VSS writer responsible for NTDS, it does not evidence itself with Ntbackup.exe.
32
| |
Issue the VSSADMIN LIST command to query the VSS writer state to ensure that VSS is in a stable/ready state.
Show commands for the backup-archive client

SHOW commands are undocumented and unsupported diagnostic commands that are used to display information about in-memory control structures and other run-time attributes. The SHOW commands are used by development and service only as diagnostic tools. Several SHOW commands exist for the backup-archive client. Depending upon the information that a SHOW command displays, there may be instances where the information is changing or cases where it may cause the application (client, server, or storage agent) to crash. The SHOW commands should only be used at the recommendation of development or service. The SHOW commands that are documented are not all of the available SHOW commands.
CACHE
Description
Displays information about the subfile cache.
Recommendation
For Microsoft Windows clients configured to use subfile backup, this is useful for displaying information about the configured subfile cache.
CLUSTER
Description
Displays information about the disk mappings in a Microsoft Cluster.
Recommendation
Useful for displaying information about the disk mapping (configuration) in a Microsoft Cluster environment.
DOMAIN
Description
Displays information about the configured domains to use for incremental backup processing.
Recommendation
Useful for displaying and summarizing the DOMAIN, DOMAIN.IMAGE, and DOMAIN.NAS client options.
OPTIONS
Description
Displays the client options.
33
Recommendation
Useful to determine the settings of client options.
OPTTABLE
Description
Displays information about options that are administered by the server versus those that are managed by the client option file.
Recommendation
The client may receive its option settings from either the client option file or from the server. To receive the option from the server, a client option set must be defined using the DEFINECLOPTSET command. This command helps you to determine whether the client is using an option configured from the option file or an option configured from a client option set defined on the server.
PLUGINS
Description
Displays information about installed plug-ins for this client.
Recommendation
The client uses plug-ins to provide additional capabilities, such as image backup. This SHOW command displays the plug-ins that are installed for this client as well as attributes of the various plug-ins, such as their version, type, and location.
SESSION
Description
Displays the capabilities that this client is able to have for this connection to the server.
Recommendation
The client and server report and negotiate the capabilities that each has when a session is started by a client to a server. This SHOW command reports the capabilities available by this server and client.
SYSTEMOBJECT
Description
For Windows 2000 and XP clients, displays the SYSTEM OBJECT data that is available on this client.
Recommendation
It is helpful in determining which SYSTEM OBJECT files are installed on this Windows client and those that could be backed up.
34
SYSTEMSERVICES
Description
For Windows 2003 clients, displays the SYSTEM SERVICES data that is available on this client.
Recommendation
It is helpful in determining which SYSTEM SERVICES files are installed on this Windows client and those that could be backed up. Note: SYSTEMSERVICES is valid for Tivoli Storage Manager version 5.4.
SYSTEMSTATE
Description
For Windows 2003 and Windows Vista clients, displays the SYSTEM STATE data that is available on this client.
Recommendation
It is helpful in determining which SYSTEM STATE files are installed on this Windows machine and those that could be backed up.
TRACEFLAGS
Description
Displays information about trace classes and aggregate trace classes for this client.
Recommendation
It is helpful in determining which trace classes and aggregate trace classes could be used for this client.
VERSION
Description
Displays the version and build date for this client.
Recommendation
It is helpful in determining which client is running and when it was built.
35
36

Problems specific to the server can be attributed to several issues.
Server diagnostic tips

Compiled are some tips to help you resolve server problems.
Checking the server activity log

Check the server activity log 30 minutes before and 30 minutes after the time of the error for other messages. To review the messages in the server activity log, issue the QUERY ACTLOG command (see the Tivoli Storage Manager Administrators Reference for more details). Often, other messages can offer additional information about the cause of the problem and how to resolve it.
Checking HELP for Tivoli Storage Manager messages issued for the problem
Tivoli Storage Manager messages provide additional information beyond just the message itself. The Explanation, System Action, or User Response sections of the message may provide additional information about the problem. Often, this supplemental information about the message may provide the necessary steps to resolve the problem.
Recreating the problem

If a problem can be easily or consistently recreated, it may be possible to isolate the cause of the problem to a specific sequence of events as causing it. Many problems occur as a result of a combination of events. For example, expiration running along with nightly scheduled backups for twenty clients. In some cases, by changing the timing or order of implementation of events, this may prevent the problem from reoccurring. In this example, one way to change the timing would be to run expiration at a time when the nightly scheduled backups for the twenty clients was not running.
Resolving errors from reading or writing to a device

If the problem is an error created by reading or writing data from a device, many systems and devices record information in a system error log. Some examples of system error logs are errpt for AIX, Event Log for Windows, and the System Log for z/OS. If a device or volume that is being used by the Tivoli Storage Manager is reporting some sort of error to the system error log, it is likely a device issue. The error messages recorded in the system error log may provide enough information to resolve the problem.
37
Changing server options or settings creates errors

Changes to options in the server options file, or configuration changes to the server using SET or UPDATE commands, may cause failures for operations that previously succeeded. Changes on the server to device classes, storage pools, and policies may also cause failures to operations that previously succeeded. If there have been one or more configuration changes to the server, try reverting the settings back to their original values and retry the failing operation. If the operation succeeds, try to make one change at a time and retry the operation until the attribute change that caused the failure is identified.
Failing a scheduled client operation

Scheduled client operations are influenced by the schedule definitions on the server as well as the scheduling service (dsmsched) that runs on the client machine itself. In many cases, if a schedule changes on the server, it may require the scheduling service on the client to be stopped and restarted. | | Note: If the scheduling service is managed by the Client Acceptor daemon (dsmcad), stop and restart only the dsmcad.
Resolving server space issues

The Tivoli Storage Manager servers primary function is to store data. If it runs out of space in the database or storage pools, operations may fail. To determine if the database is out of space, issue the QUERY DATABASE command. If the %Util is at or near 100% then more space should be defined. Typically, if the database is running out of space, other server messages are issued to indicate this. To add more space to the database, allocate one or more new database volumes, define them to the server using the DEFINE DBVOL command, and then extend the available database space by issuing the EXTEND DB command. To determine if a storage pool is out of space, issue the QUERY STGPOOL command. If the %Util is at or near 100%, then more storage space should be made available. Typically, if operations fail because a storage pool is out of space, other server messages are issued to indicate this. To add more space to a DISK storage pool, allocate one or more new storage pools and define them to the server using DEFINE VOLUME. To add more space to a sequential media storage pool, evaluate the tape library and determine if more scratch tapes can be added. If so, add the additional scratch volumes to the library and update the MAXSCR parameter for the storage pool by issuing the UPDATE STGPOOL command.
Server error messages indicate a code page conversion failure

The Tivoli Storage Manager server uses operating system functions to convert between Unicode and the server code page. If the system is not set up correctly, this will fail. This problem is most likely to occur when you are using the Administrative Center to access the server.
38
To get more information on the failure, begin tracing the UNICODE trace class and repeat the action that caused the error message to occur. Check the server README file for any platform-specific requirements for language installation. Ensure that the locales indicated by the problem code pages are installed and any requirements listed in the README file are installed.
Resolving a server crash

You can resolve several problems if you know the source of your server crash. The server might crash for one of the following reasons: v A processing error causes memory to be overwritten or some other event triggers the system trap handler to terminate the server process. v The server processing has validation algorithms throughout the application that check various conditions prior to continuing running. As part of this validation checking, there are cases where if the validation check fails, the server will actually terminate itself instead of allowing processing to continue. These catastrophic validation failures are referred to as an assert. If the server terminates due to an assert, the following message is issued:
ANR7837S Internal error XXXNNN detected.
where XXXNNN is an identifier assigned to the assertion failure. Examples of this identifier are DBREC106, BUF011, or TBUNDO096. Other server messages that are indicative of a crash are ANR7836S and ANR7838S. Whether the server terminated as a result of an assert or the system trap handler, the following information should be collected and reported to IBM service for the situation to be diagnosed: v Server error file (dsmserv.err) When the server crashes, it appends information to the dsmserv.err file which is located in the same directory as the server. For the Windows platform, if the server is running as a service, then the file will be named dsmsvc.err. For the z/OS platform, there is no separate error file. The information will be in the JES joblog and in the system log. For z/OS they are generally written by WTO with routecode=11. For HP-UX, it is possible to use the following script to get basic thread information from the core on the customer system without sending it to IBM. The procedure is as follows: 1. Make sure gdb is installed on the customer system. 2. Copy the gt shell script to the server bin directory (where the server executable and core file are located). 3. Make sure the script is an executable (chmod a+x gt). 4. Invoke the script with the paths/names of the executable (default is ./dsmserv) and the core (default is ./dsmcore). Output will be in file dsm_gdb.info (which should be sent to IBM). For Linux and AIX, version 5.3 and later, the trap handler is disabled. This prevents the function traceback from printing on the console and in dsmserv.err. This change is required in order to ensure that we will get a more complete core file. As part of disabling the trap handler, a new script, getcoreinfo, is in the Linux packages. The getcoreinfo script gets the function
39
traceback for the failing thread and registers values and function traceback for all other threads. The amount of information available in the core for other threads is still incomplete on some Linux platforms/distributions. See the getcoreinfo script (in the server bin directory) for more details. v System image (core file) Typically, a core file or other system image of the memory is in use by the Tivoli Storage Manager at the time of the failure. In each case, this file should be renamed to prevent it from being overwritten by a later crash. For example a file should be renamed to core.Aug29 instead of just core. The type and name of the core file varies depending upon the platform: For UNIX systems, such as AIX, HP-UX, Sun Solaris, PASE and Linux, there is typically a file created called core. Be sure there is enough space in the server directory to accommodate a dump operation. It is common to have a dump file as large as 2 GB for the 32-bit Tivoli Storage Manager server. Additionally, make sure the ulimit for core files is set to unlimited to prevent the dump file from being truncated. For Windows systems before 5.2, a drwtsn.dmp file is created if the system is configured to do so. From a DOS prompt, issue drwtsn32 -i to enable dumps. Beginning with 5.2, the dump is performed automatically through a system API call. If the server is running as a service, the dump file is called dsmsvc.dmp. Otherwise, the dump is called dsmserv.dmp. For a z/OS server, a dump should have been created based upon what was specified in the SYSMDUMP DD statement for the JCL used to start the server. A SYSMDUMP should be used instead of a SYSABEND or SYSUDUMP dump, because a SYSMDUMP contains more useful debugging information. Ensure that the dump options for SYSMDUMP include RGN, LSQA, TRT, CSA, SUM, SWA, and NUC on systems where a Tivoli Storage Manager address space will run. To display the current options, issue the z/OS command DISPLAY DUMP,OPTIONS. You can issue the CHNGDUMP command to alter the SYSMDUMP options. Note that this will only change the parameters until the next IPL is performed. If the system was not configured to capture a core file or the system did not have sufficient space to create a complete core file, it may be of limited use in determining the cause of the problem. v Libraries and other files For the UNIX platforms, core files are specific to the application, libraries, and other system resources in use by the application on the system where it was running. To accurately read the core file on our system, we may need all of the below files (located in the directory where the server is installed): dsmserv dsmlicense ndmpspi dsmcored dsmaio centera The needed library files vary between all the platforms: For AIX/PASE systems collect the following files: - /usr/ccs/lib/libpthreads.a - /usr/ccs/lib/libc.a - Collect any other loaded libraries such as message exits. To see what libraries are loaded, invoke dbx by issuing the DBX DSMSERV CORE_FILE
40
command. Then, from the dbx prompt, issue the MAP command. This will show all of the libraries that are loaded and needed for core analysis. For HP-UX, issue the CHATR DSMSERV command. Send in all the dynamic shared libraries. For example: - /usr/lib/libpthread.1 - /usr/lib/libm.2 - /usr/lib/libstd.2 - /usr/lib/libstream.2 - /usr/lib/libCsup.2 - /usr/lib/libcl.2 - /usr/lib/libc.2 - /usr/lib/libdld.2 For Linux, issue the LDD DSMSERV command. Send in all the dynamic shared libraries. For example: - libm.so.6 =>/lib64/libm.so.6 - libnsl.so.1 =>/lib64/libnsl.so.1 - libpthread.so.0 =>/lib64/libpthread.so.0 - libdl.so.2 =>/lib64/libdl.so.2 - libc.so.6 =>/lib64/libc.so.6 - /lib64/ld64.so.1 =>/lib64/ld64.so.1 For Solaris, issue the following commands to collect the needed libraries: - sh - cd /usr - (find . -name ld.so -print ; \ - find . -name ld.so.? -print ; \ - find . -name libm.so.? -print ; \ - find . -name libsocket.so.? -print ; \ - find . -name libnsl.so.? -print ; \ - find . -name libthread.so.? -print ; \ - find . -name libthread_db.so.? -print ; \ - find . -name libdl.so.? -print ; \ - find . -name libw.so.? -print ; \ - find . -name libgen.so.? -print ; \ - find . -name libCrun.so.? -print ; \ - find . -name libc.so.? -print ; \ - find . -name libmp.so.? -print ; \ - find . -name libc_psr.so.? -print ; \ - find . -name librtld_db.so.? -print) > runliblist - tar cfh runliblist.tar -I runliblist v System logs For the AIX platform redirect the output from the command errpt -a into a file: errpt -a >errpt.txt. For the HP-UX platform, copy the /var/adm/syslog/syslog.log file. For the Linux platform, copy the /var/log/messages file. For the Solaris platform, copy the /var/adm/messages file.
41
For the Windows platform, save a copy of the Event Logs, as seen from the Event Viewer. For the z/OS platform, gather a copy of the System Log. v Activity log: Collect the activity log at least two hours prior to the crash and 30 minutes after the crash using the QUERY ACTLOG command. See the Tivoli Storage Manager Administrators Reference for more details. v Package all the data (files) collected and contact IBM service to report this problem. For all files from z/OS, make sure to use xmit to create a fixed-blocked file because it will be moved by ftp to an open systems machine, which will cause the loss of DCB attributes for the file. For the open systems platforms, package the files in a zip or tar file.
Resolving database errors

Server errors may be caused by database irregularities. Some more common issues are running out of space and errors caused by an insert, update, or delete operation.
ANR0102E, ANR0103E, or ANR0104E message issued

The ANR0102E, ANR0103E, or ANR0104E messages are used to report errors during database operations. These messages are used to report errors during an insert (ANR0102E), update (ANR0103E), and delete (ANR0104E) operations. Database operations typically fail for the two reasons listed below: v There is a timing error between server threads. In order for this error to occur, two or more operations have to be working on the same or similar resources within the server. v The database was damaged. A typical symptom for the ANR0102 message is Error 1, which indicates a duplicate row. A typical error for the ANR0103E message may be either an Error 1, indicating a duplicate row, or an Error 2, which indicates a missing row. A typical error for the ANR0104E message is Error 2, which indicates a missing row. For the table reported, issue the SHOW TBLSCAN tableName command, where tableName is the name of the table reported in the message. Note that the table name used by this command is case sensitive and it should be entered exactly as it was displayed in the message. This command will scan the entire table and evaluate the structure of that table. If it detects an error in the structure of the table, it will report that an error was found. In cases where the database table is severely damage, this command may cause the server to crash. Also, if this is a large table within the database, this evaluation may require a significant amount of I/O and may degrade the server performance. If SHOW TBLSCAN reports a problem, contact support. Support will first work with you to try to determine the cause of the database damage. In most cases, the cause is external to the Tivoli Storage Manager and is usually an error in the storage device where the one or more server database volumes are allocated. After helping to determine the cause of the database damage, support will instruct you on the necessary steps to repair this damage in the database.
Maximum reduction value does not increase as the percentage utilized decreases
Issue the QUERY DB command over a period of time when the Pct. Util decreases.
42
The server allocates database space from the beginning of the database, which is the lowest allocated page, to the end of the database, which is the highest allocated page. Because allocation is ascending, it is possible for database space to be freed and become available for use between the lowest and highest allocated pages. The Maximum Reduction reported by QUERY DATABASE cannot reduce the database below the highest allocated page, even though free space may exist. The server uses the free space between the lowest and highest allocated page as necessary. It is possible to reorganize the server database. Reorganizing the database will typically reduce the pages used, which results in reduced percentage utilization. Similarly, reorganizing the server database may increase the maximum reduction value after it completes. Database reorganization is done by issuing DSMSERV UNLOADDB, DSMSERV LOADFORMAT, and then a DSMSERV LOADDB. The database reorganization requires the server to be offline or not available for client use. Review these commands and this procedure in the Tivoli Storage Manager Administrators Reference and Tivoli Storage ManagerAdministrators Guide in the Tivoli Storage Manager publications for your platform.
Out of database space

Issue the QUERY DB command. If the Pct. Util is very high, such as close to 100%, the server will report out of database space during operations that need to add or change database information. You can take several steps to provide more database space. v If the Maximum Extension value reported by the QUERY DB command is greater than zero, issue EXTEND DB nn, where nn is the available space that the database can be extended as reported by QUERY DB. v Define additional database volumes using the DEFINE DBVOLUME command. After defining one or more additional database volumes, issue QUERY DB and extend the database by all or some of the available extension amount using EXTEND DB nn, where nn is the extension amount. v Issue QUERY STATUS and evaluate the Activity Log Retention Period and the Activity Summary Retention Period. By reducing either or both of these values, the database space consumed by this information is reduced. These can be set to lower values by issuing the SET ACTLOGRETENTION and SET SUMMARYRETENTION commands.
Unable to read the database during server restart

If the server fails to read a required page during startup processing, the server fails to start. You can take the following steps to resolve this: v Were the database volumes mirrored as DBCOPY volumes? If so, change or add the MIRRORREAD DB option to MIRRORREAD DB VERIFY and try to restart the server. This forces the server to try to read the database page from all possible database volumes that would contain that page and to compare them to each other. If it can read a valid copy of the database page from one of the database volumes, it will use that copy and may allow the server to restart in this case. v Check the disk drives, adapters, and connections between the server computer and the disk storage where the Tivoli Storage Manager database volumes are defined. If any of the hardware between the server computer and Tivoli Storage Manager database volumes is reporting a failure or other error, investigate and resolve the error and then try to restart the Tivoli Storage Manager server.
43
Resolving a hang or loop

A hang is a situation where the server does not start or complete a function and is not using any central processing unit (CPU) power. A hang could be just one session or process that is not processing, or it could be the entire Tivoli Storage Manager server not responding. A loop is a situation where no progress is being made, but the server is using a high amount of CPU power. A loop can affect just one session or process, or it could affect the entire Tivoli Storage Manager server. You might collect documentation to resolve this type of problem, depending on whether the server is able to respond to commands. It is helpful to schedule the SHOW command list to run intermittently so that you can then see the behavior that precedes the hang situation. v For a hang or a loop where the server can respond to commands, issue the following commands to help determine the cause of the hang: QUERY SESSION f=d QUERY PROCESS SHOW RESQ SHOW THREADS SHOW DEADLOCK SHOW TXNT SHOW DBTXNT SHOW LOCKS SHOW LIBR SHOW MP SHOW SESS SHOW ASQ SHOW ASVOL SHOW DBV SHOW SSS SHOW CSV (do this only if the problem appears to be related to scheduling) v In addition to the output from the listed commands, or in the case of a server that cannot respond to commands, collect a dump. The way that you collect a dump depends on the operating system. For UNIX -type operating systems, issue the KILL -11 command on the dsmserv process to create a core file. The process ID to perform the kill is obtained by issuing the PS command. See Resolving a server crash on page 39 for information on additional libraries or executables that need to be packaged along with the core file. For the Windows operating system, refer to the Microsoft Knowledge Base item 241245 at microsoft.com for instructions on installing and using the userdump.exe program to obtain a dump. For the z/OS platform, issue the operating system DUMP command. The SDUMP options should specify RGN, LSQA, TRT, CSA, SWA, and NUC.
Using recovery log information

The recovery log is a log of updates that are about to be written to the database. The recovery log can be used to recover from system and media failures.
44
ANR0314W Recovery log usage... message received with LOGMODE=ROLLFORWARD

The server reports that the recovery log usage is getting high and issues the ANR0314W message when the Pct. Util exceeds 80%. If the server is running with LOGMODE=ROLLFORWARD, this may put the server in jeopardy of running out of recovery log space. The server retains all the recovery log information between database backups when the LOGMODE=ROLLFORWARD. If a database backup is not run manually, scheduled, or triggered, the server will run out of recovery log space. Consider the following steps to reduce the size of your recovery log: v Manually run a database backup by issuing the BACKUP DB command. The database backup may be run as TYPE=FULL or TYPE=INCREMENTAL. Note that the recovery log utilization continues to increase until the database backup completes, so adequate time must be given for the operation to complete before the server runs out of recovery log space. v Issue the QUERY DBBACKUPTRIGGER command. If a DBBACKUPTRIGGER is not defined, define one by issuing the DEFINE DBBACKUPTRIGGER command. This allows the server to automatically start a database backup based upon the trigger setting for the LOGFULLPCT. If a DBBACKUPTRIGGER is defined: - Evaluate and consider lowering the LOGFULLPCT. - If the LOGFULLPCT is higher than the current Pct. Util for the recovery log, wait until the LOGFULLPCT is reached and a database backup is triggered. If the LOGFULLPCT is greater than 80%, the ANR0314W message will always be issued until the database backup is triggered and completed. - Check previous database backups to make sure they were successful. If there are no volumes available for the triggered database backup to use or there is an error processing the database backup, the server will not be able to relieve the filling recovery log situation. v Evaluate whether or not LOGMODE=ROLLFORWARD is needed in your environment. If it is not needed, issue SET LOGMODE=NORMAL. This allows space in the recovery log to be recovered without requiring a database backup to be run.
Out of recovery log space

Issue the QUERY LOG command. If the Pct. Util is very high (close to 100%), the server reports out of recovery log space during operations that need to add or change database information. You can take several steps to provide more recovery log space: v If the Maximum Extension value reported by the QUERY LOG command is greater than zero, issue the EXTEND LOG nn command where nn is the available space that the recovery log can be extended as reported by QUERY LOG. v Define additional recovery log volumes by issuing the DEFINE LOGVOLUME command. After defining one or more additional recovery log volumes, issue the QUERY LOG command and extend the recovery log by all or some of the available extension amounts by issuing the EXTEND LOG nn command, where nn is the extension amount.
45
v If the server crashes because it is out of recovery log space and fails to start, you might do an emergency extend of the recovery log. Do this by issuing the DSMSERV EXTEND LOG volumeName amount command, where volumeName is a formatted recovery log volume and amount is the amount (in megabytes) to extend the recovery log. The recovery log has a maximum size of 13 GB. The recovery log might only be defined or extended to 12 GB so that there is 1GB in reserve if an emergency extend needs to be performed. In the event that an emergency extend is performed, reduce the recovery log after restarting the server to preserve the 1 GB reserve amount. The recovery log might appear to be out of space, when in fact it is being pinned by an operation or combination of operations on the server. A pinned recovery log is where space in the recovery log cannot be reclaimed and used by current transactions because an existing transaction is processing slowly or is hung. To determine if the recovery log is pinned, repeatedly issue the SHOW LOGPINNED command over many minutes. If this reports the same client session or server processes as pinning the recovery log, you might take action to cancel or terminate that operation in order to keep the recovery log from running out of space. To cancel or terminate a session or process that is pinning the recovery log, issue the SHOW LOGPINNED CANCEL command. Server version 5.1.7.0 and later have additional support for the recovery log to automatically recognize that the recovery log is running out of space and, where possible, to detect and resolve a pinned recovery log using the SHOW LOGPINNED processing.
Unable to read the recovery log during server restart

If the server cannot read from the recovery log during startup processing, the server does not start. Consider the following steps: v Check to see if the recovery volumes are mirrored. If so, change or add the MIRRORREAD LOG option to MIRRORREAD LOG VERIFY and try to restart the server. This forces the server to try to read the recovery log information from all possible recovery log volumes that contain the startup information and compare them to one another. If it can read the requested information from one of the recovery log volumes, it will be used and may allow the server to restart. v Check the disk drives, adapters, and connections between the server computer and the disk storage where the Tivoli Storage Manager recovery log volumes are defined. If any of the hardware between the server computer and Tivoli Storage Manager recovery log volumes is reporting a failure or other error, investigate and resolve the error and then try to restart the Tivoli Storage Manager server.
Defining processes
A server process is a task performed on the server. This task is assigned to perform a specific operation, such as migrating data from a storage pool to the NEXT storage pool in the hierarchy. Or it may be the process for backing up the server database to a sequential media device for the purpose of safeguarding it. A process has the following general characteristics: Process name The name of the process as it appears in the process related messages. Means of initiation Server processes can be initiated in several ways. Typically, they are either
46
an automated process on the server that may or may not be influenced by a server options or other setting, or they are started by a command. Can run in foreground? Many server processes can be run in the FOREGROUND or synchronously. Processes that run in the FOREGROUND can be initiated by a command using the WAIT=YES parameter. Commands that start server processes that do not allow the WAIT=YES parameter or those specified with WAIT=NO are run in the BACKGROUND or asynchronously. Can run as multiple processes Some server processes can initiate multiple processes simultaneously to accomplish the task. Description A description of what this task does.
AUDIT VOLUME
Description
Audit the contents of a volume to validate that the data can still be read and that the server database definitions describing the data are correct.
Recommendation
The audit is initiated by issuing the AUDIT VOLUME command.
BACKUP DB
Description
Back up the server database (FULL or INCREMENTAL).
Recommendation
The backup is initiated by issuing the BACKUP DB command. The backup may also be automatically triggered if the server is in ROLLFORWARD mode for recovery log processing and a DBBACKUPTRIGGER is defined.
Other
The BACKUP DB can run as a synchronous process by specifying WAIT=YES.
BACKUP STGPOOL
Description
Backup a primary server storage pool to a copy storage pool for the purpose of making duplicate copies of the data and potentially taking duplicate copies to an off-site location.
Recommendation
The backup is initiated by issuing the BACKUP STGPOOL command.
47
Other
The BACKUP STGPOOL can run as a synchronous process by specifying WAIT=YES. BACKUP STGPOOL may be run using multiple concurrent processes. This is controlled by the MAXPROCESS parameter specified on the BACKUP STGPOOL command.
CHECKIN LIBVOLUME
Description
Check a tape volume into a tape library.
Recommendation
The check-in is initiated by issuing the CHECKIN LIBVOLUME command.
CHECKOUT LIBVOLUME
Description
Check a tape volume out from a tape library.
Recommendation
The check-out is initiated by issuing the CHECKOUT LIBVOLUME command.
EXPIRATION
Description
Expire (delete) client backup and archive files from the server, based on the policies defined to manage those files.
Recommendation
You can run expiration automatically by specifying EXPINTERVAL=n in the server options file, where n is any number other than zero. Expiration can also be initiated by issuing the EXPIRE INVENTORY command. Please note that it is not possible to have more than one expiration process running at a time.
Other
The EXPIRATION command can run as a synchronous process by specifying WAIT=YES.
IMPORT
Description
Import data from sequential media volumes or directly from another server using TCP/IP communication connections between the servers.
Recommendation
Import processing can be started by any of the following commands: IMPORT ADMIN, IMPORT NODE, IMPORT POLICY, and IMPORT SERVER.
48
LABEL LIBVOLUME
Description
Label one or more library volumes in a library.
Recommendation
Labeling is initiated by issuing the LABEL LIBVOLUME command.
MIGRATION
Description
Migrate data from one storage pool to the next in the storage hierarchy.
Recommendation
Migration starts and stops, based on the HighMig and LowMig thresholds defined for the storage pool. Whenever UPDATE STGPOOL is issued, these values are reexamined and, if appropriate, MIGRATION is started. Otherwise, the server monitors the percentage utilization for non-migrated data in a storage pool, and as needed, it starts migration processing for that storage pool when the HighMig threshold is exceeded. You can also issue the MIGRATE STGPOOL command to manually start migration processing.
Other
Migration may be configured to run multiple concurrent processes. This is controlled by the MIGPROCESS attribute of the storage pool and may be updated by issuing the UPDATE STGPOOL command.
MOVE DATA
Description
Move data from one volume to other volumes in the same storage pool or to a different storage pool.
Recommendation
The move is initiated by issuing the MOVE DATA command.
Other
The MOVE DATA command can run as a synchronous process by specifying WAIT=YES.
MOVE DRMEDIA
Description
Manage the disaster recovery media by moving onsite volumes offsite, or by bringing offsite volumes back onsite. Disaster recovery media is the database backup and storage pool backup volumes necessary to protect and recover the server.
49
Recommendation
The move is initiated by issuing the MOVE DRMEDIA command.
Other
The MOVE DRMEDIA command can run as a synchronous process by specifying WAIT=YES.
MOVE MEDIA
Description
Move volumes from a tape library to the overflow location to prevent a library from becoming full.
Recommendation
The move is initiated by issuing the MOVE MEDIA command.
MOVE NODEDATA
Description
Move all the data for the node or nodes specified to other volumes in the same storage pool or to a different storage pool.
Recommendation
The move is initiated by issuing the MOVE NODEDATA command.
Other
The MOVE NODEDATA command can run as a synchronous process by specifying WAIT=YES.
PREPARE
Description
Create a recovery plan file.
Recommendation
The recovery plan file is initiated by issuing the PREPARE command.
Other
The PREPARE command can run as a synchronous process by specifying WAIT=YES.
RECLAMATION
Description
Reclaim space from tape volumes by moving active data to other volumes and returning the volume back to empty and private, or else back to scratch.
50
Recommendation
The server monitors the RECLAMATION THRESHOLD defined for a storage pool. It starts a reclamation process for that storage pool to reclaim any eligible volumes if it determines that one or more eligible volumes exist.
RESTORE STGPOOL
Description
Restore all files for a given storage pool from a copy storage pool.
Recommendation
The restore is initiated by issuing the RESTORE STGPOOL command.
Other
The RESTORE STGPOOL can run as a synchronous process by specifying WAIT=YES. RESTORE STGPOOL may be run using multiple concurrent processes. This is controlled by the MAXPROCESS parameter specified on the RESTORE STGPOOL command.
RESTORE VOLUME
Description
Restore all files for a given volume from a copy storage pool.
Recommendation
The restore is initiated by issuing the RESTORE VOLUME command.
Other
The RESTORE VOLUME command can run as a synchronous process by specifying WAIT=YES. RESTORE VOLUME may be run using multiple concurrent processes. This is controlled by the MAXPROCESS parameter specified on the RESTORE VOLUME command.
Using the process messages

Server processes, whether run in the FOREGROUND or BACKGROUND, will always issue a process started message and a process ended message. The process ended message that is issued varies, depending upon whether or not the process needs to report information about no items or bytes processed, items and bytes processed, items processed, or just bytes processed. The following are the server process messages: In addition to the general process messages, each process will typically report its own messages relating to when it started and ended.
Process started
The server might run tasks as processes. Processes are assigned an identification message and report that they have started by issuing the following message:
51
ANR0984I Process process_id for process_name started in the process_state at time The following are definitions for the variables in the previous message: process_id Numeric process identifier. process_name The name of the process. process_state FOREGROUND or BACKGROUND. If the process is running in the FOREGROUND, the command was issued with the WAIT=YES parameter. FOREGROUND processing causes the administrative session that issued the command to wait until the processing completes. A process running in the BACKGROUND returns immediately to the administrative session that issued the command, indicating that a process was started while the process still runs. Processes running in the BACKGROUND may be monitored with the QUERY PROCESS command. time The time the process was started.
Process ended
When a process completes and it does not have bytes or number of files to report, the following message is issued: ANR0985I Process process_id for process_name running in the process_state completed with the completion_state at time The following are definitions for the variables in the previous message: process_id Numeric process identifier. process_name The name of the process. process_state FOREGROUND or BACKGROUND. If the process is running in the FOREGROUND, the command was issued with the WAIT=YES parameter. FOREGROUND processing causes the administrative session that issued the command to wait until the processing completes. A process running in the BACKGROUND returns immediately to the administrative session that issued the command, indicating that a process was started while the process still runs. Processes running in the BACKGROUND may be monitored with the QUERY PROCESS command. completion_state SUCCESS or FAILURE time The time the process was started.
Process ended with items and bytes

When a process completes and has bytes and items processed to report, the following message is issued:
52
ANR0986I Process process_id for process_name running in the process_state processed number_of_items items for a total of bytes_processed bytes with a completion state completion_state at time The following are definitions for the variables in the previous message: process_id Numeric process identifier. process_name The name of the process. process_state FOREGROUND or BACKGROUND. If the process is running in the FOREGROUND, the command was issued with the WAIT=YES parameter. FOREGROUND processing causes the administrative session that issued the command to wait until the processing completes. A process running in the BACKGROUND returns immediately to the administrative session that issued the command, indicating that a process was started while the process still runs. Processes running in the BACKGROUND may be monitored with the QUERY PROCESS command. number_of_items The number of items processed. bytes_processed The number of bytes processed. completion_state SUCCESS or FAILURE time The time the process was started.
Process ended with items

When a process completes and has items processed to report, the following message is issued: ANR0987I Process process_id for process_name running in the process_state processed number_of_items items with a completion state completion_state at time The following are definitions for the variables in the previous message: process_id Numeric process identifier. process_name The name of the process. process_state FOREGROUND or BACKGROUND. If the process is running in the FOREGROUND, the command was issued with the WAIT=YES parameter. FOREGROUND processing causes the administrative session that issued the command to wait until the processing completes. A process running in the BACKGROUND returns immediately to the administrative session that issued the command, indicating that a process was started while the process still runs. Processes running in the BACKGROUND may be monitored with the QUERY PROCESS command.
53
completion_state SUCCESS or FAILURE time The time the process was started.
Process ended with bytes

When a process completes and has bytes processed to report, the following message is issued: ANR0988I Process process_id for process_name running in the process_state processed bytes_processed bytes with a completion state completion_state at time The following are definitions for the variables in the previous message: process_id Numeric process identifier. process_name The name of the process. process_state FOREGROUND or BACKGROUND. If the process is running in the FOREGROUND, the command was issued with the WAIT=YES parameter. FOREGROUND processing causes the administrative session that issued the command to wait until the processing completes. A process running in the BACKGROUND returns immediately to the administrative session that issued the command, indicating that a process was started while the process still runs. Processes running in the BACKGROUND may be monitored with the QUERY PROCESS command. bytes_processed The number of bytes processed. completion_state SUCCESS or FAILURE time The time the process was started.
Using the process symptoms to resolve problems

The process symptoms are sometimes helpful in determining where your errors occur.
ANR1221E COMMAND: Process processID terminated insufficient space in target copy storage pool
Issue the QUERY STGPOOL stgpoolName F=D command. Issue the following SQL select statement from an administrative client to this server: select stgpool_name,devclass_name,count(*) as VOLUMES from volumes group by stgpool_name,devclass_name. Compare the number of volumes reported by the select statement to the maximum scratch volumes allowed (as reported by the QUERY STGPOOL command). If the number of volumes reported by the select is equal to or exceeds the Maximum Scratch Volumes Allowed, update the storage pool and allow more scratch volumes by issuing the UPDATE STGPOOL stgpoolName MAXSCR=nn command,
54
where stgpoolName is the name of the storage pool to update and nn is the increased number of scratch volumes to make available to this copy storage pool. Note: The tape library should have this additional number of scratch volumes available, or you need to add scratch volumes to the library prior to issuing this command and retrying the BACKUP STGPOOL operation.
ANR2317W Audit Volume found damaged file on volume volumeName: Node nodeName, Type fileType, File space fileSpaceName, fsId fileSpaceID, File name fileName is number version of totalVersions versions
Issue the QUERY VOLUME volumeName F=D command. Issue the following SQL select statement from an administrative client to this server: select* from VOLHISTORY where VOLUME_NAME=volume_name AND TYPE=STGNEW. The results of the QUERY VOLUME command indicate when this volume was last written. The information from the select operation reports when this volume was added to the storage pool. Often, AUDIT VOLUME may report files as damaged because, at the time that the data was written, the hardware was malfunctioning and did not write the data correctly, even though it reported to the Tivoli Storage Manager server that the operation was successful. As a result of this device malfunction, many files on many different volumes may be affected. You can take the following steps to resolve this: v Evaluate the system error logs or other information about this drive to determine if it still reports an error. If errors are still reported, they must first be resolved. To resolve a hardware issue, work with the hardware vendor to correct the problem. v If this is a copy of a storage pool volume, simply delete this volume using the DELETE VOLUME volumeName DISCARDDATA=YES command. The next time a storage pool backup is run for the primary storage pool or storage pools where this damaged data resides, it will be backed up again to this copy storage pool and no further action is necessary. v If this is a primary storage pool volume and the data was written directly to this volume when the client stored the data, then it is likely that there are no undamaged copies of the data on the server. If possible, back up the files again from the client. v If this is a primary storage pool volume but the data was put on this volume by MIGRATION, MOVE DATA, or MOVE NODEDATA, there might be an undamaged copy of the file on the server. If the primary storage pool that contained this file was backed up to a copy storage pool prior to the MIGRATION, MOVE DATA, or MOVE NODEDATA then an undamaged file may exist. If this is the case, issue the UPDATE VOLUME volumeName ACCESS=DESTROYED command and then issue RESTORE VOLUME volumeName command to recover the damaged files for this volume from the copy storage pool.
Files are not expired after reducing the number of versions that need to be kept
The server policies were updated to reduce the number of versions of a file to retain. Issue the QUERY COPYGROUP domainName policySetName copyGroupName F=D command. If either the Versions Data Exists or Versions Data Deleted parameters were changed for a TYPE=BACKUP copy group, it might affect expiration.
55
If the Versions Data Exists or Versions Data Deleted values for a TYPE=BACKUP copy group were reduced, the server expiration process may not immediately recognize this and expires these files. The server only applies the Versions Data Exists and Versions Data Deleted values to files at the time they are backed up to the server. When a file is backed up, the server will count the number of versions of that file and if that exceeds the number of versions that should be kept, the server will mark the oldest versions that exceed this value to be expired.
Migration does not run for sequential media storage pool

Issue the QUERY STGPOOL stgpoolName F=D command. Migration from sequential media storage pools calculates the Pct. Util as the number of volumes in use for the storage pool, relative to the total number of volumes that can be used for that storage pool. Similarly, it calculates the Pct. Migr as the number of volumes with data, that can be migrated, in use for the storage pool, relative to the total number of volumes that can be used for that storage pool. Because it might be considering unused scratch volumes in this calculation, there may not appear to be sufficient data, that can be migrated, in the storage pool to require migration processing.
Migration only uses one process

Issue the QUERY STGPOOL stgpoolName F=D and QUERY OCCUPANCY * * STGPOOL= stgpoolName command. The following are reasons why only one migration process is being run: v The Migration Processes setting for the storage pool is set to one or is not defined (blank). If this is the case, issue the UPDATE STGPOOL stgpoolName MIGPROCESS=n command, where n is the number of processes to use for migrating from this pool. Note that this value must be less than or equal to the number of drives (mount limit) for the NEXT storage pool where migration is storing data. v If the QUERY OCCUPANCY command only reports a single client node and file space in this storage pool, migration can only run a single process even if the Migration Processes setting for the storage pool is greater than one. Migration processing is partitioning data, based on client node and file space. In order for migration to run with multiple processes, data for more than one client node needs to be available in that storage pool.
Process running slow

Issue the QUERY DB F=D command. If the Cache Hit Ratio that is reported is less then 98%, the process may be running slow because it is waiting for database I/O processing, due to the small size of the buffer pool. You can address this by changing or adding either of the following server options to the options file: v Set the BUFPOOLSIZE option to half the available physical memory on the machine. v Set SELFTUNEBUFPOOLSIZE to YES. After setting either of these options, halt and restart the server.
56
Resolving storage pool issues

Storage pools are critical to the server operation. Tivoli Storage Manager database contains information about registered client nodes, policies, schedules, and the client data in storage pools.
ANR0522W Transaction failed... message received

The ANR0522W message is displayed when the server is unable to allocate space in the storage pool that is identified to store data for the specified client. There are a number of possible causes for running out of space in a storage pool. Likely causes and solutions include the following: v Issue QUERY VOLUME volname F=D for the volumes in the referenced storage pool. For any volumes reported with access different from Read/Write, check that volume. A volume may be marked Read/Only or Unavailable because of a device error. If the device error was resolved, issue the UPDATE VOLUME volname ACCESS=READWRITE command to allow the server to select and try to write data to that volume. v Issue QUERY VOLUME volname for the volumes in the referenced storage pool. Volumes that report pending for the volume status are volumes that are empty but waiting to be reused again by the server. This is controlled by the REUSEDELAY setting for the storage pool and displayed as Delay Period for Volume Reuse on the QUERY STGPOOL command. Evaluate the REUSEDELAY setting for this storage pool and, if appropriate (based upon your data management criteria), lower this value by issuing the UPDATE STGPOOL stgpoolname REUSEDELAY=nn command, where stgpoolname is the name of the storage pool and nn is the new reuse delay setting. The key to getting the data collocated is to have sufficient space in the target storage pool for the collocation processing to select an appropriate volume. This is significantly influenced by the number of scratch volumes in a storage pool.
Storage pool experiences high volume usage after increasing MAXSCRATCH value
For collocated sequential storage pools, increasing the MAXSCRATCH value may cause the server to use more volumes. The server uses more storage pool volumes in this case because of the collocation processing. Collocation groups user data for a client node onto the same tape. During a client backup or archive operation, if no tapes currently have data for this client node, the server selects a scratch volume to store the data. Then, for other client nodes storing data, it makes the same decision when selecting a volume and again selects a scratch volume. The reason this does not occur prior to changing the MAXSCRATCH setting is that if there is no scratch volume available and no preferred volume already assigned for this client node, the volume selection processing on the server ignores the collocation request and stores the data on an available volume.
Storage pool has Collocate?=Yes but volumes still contain data for many nodes
The following are the two possible reasons for this message: 1. Data was stored on volumes in this storage pool prior to setting Collocate=Yes.
57
2. The storage pool ran out of scratch tapes and stored data on the best possible volume, even though it ignored the request to collocate. If data for multiple nodes ends up on the same volume for a storage pool with Collocate=Yes, this can be corrected by one of the following actions: v Issue the MOVE DATA command for the volume or volumes affected. If scratch volumes are available or volumes with sufficient space are assigned to this client node for collocating their data, the process reads the data from the specified volume and moves it to a different volume or volumes in the same storage pool. v Allow migration to move all the data from that storage pool by setting the HIGHMIG and LOWMIG thresholds to accomplish this outcome. By allowing migration to migrate all data to the NEXT storage pool, the collocation requirements are honored, provided that the NEXT storage pool is set to Collocate=Yes, it has sufficient scratch volumes, and it is assigned volumes to satisfy the collocation requirements. v Issue MOVE NODEDATA for the client nodes whose data resides in that storage pool. If scratch volumes are available or volumes with sufficient space are assigned to this client node for collocating their data, the MOVE NODEDATA process reads the data from the volumes that this node has data on and move it to a different volume or volumes in the same storage pool. The key to getting the data collocated is to have sufficient space in the target storage pool for the collocation processing to select an appropriate volume. This is significantly influenced by the number of scratch volumes in a storage pool. Another alternative is to explicitly define more volumes to the storage pool by issuing the DEFINE VOLUME command. Again, the key is to have candidate empty volumes for collocation to use rather than using a volume that already has data on it for a different node.
Unable to store data in an active data pool by using simultaneous write or by issuing the COPY ACTIVEDATA command
You may experience difficulty in storing data in an active data pool, due to policy issues. Before data can be stored in an active data pool, you must establish a policy to allow the data into the pool. The node that owns the data must be assigned to a domain whose active data pool is listed in the domain ACTIVEDESTINATION field. Issue the following command to determine if the node is assigned to a domain that authorizes storing into the active data pool: QUERY NODE node_name F=D The Policy Domain Name field lists the domain to which the node is assigned. Issue the following command to determine if the active data pool is listed in the domain ACTIVEDESTINATION field: QUERY DOMAIN domain_name F=D If the active data pool is not listed, issue the following command to add the active data pool to the list: UPDATE DOMAIN domain_name ACTIVEDESTINATION=activedata_pool_name Note: After you issue the UPDATE DOMAIN domain_name ACTIVEDESTINATION=active-data_pool_name command, all nodes assigned to the domain are authorized to store data in the active data pool. If this is not acceptable, you must create a new domain for those nodes whose data
58
you want stored in the active data pool and assign those nodes to the newly created domain. See the Tivoli Storage Manager Administrators Guide to learn how to establish a new policy domain.
59
60
Chapter 3. Resolving communication problems

Communication errors might be attributed to TCP/IP configuration, client and server connections, and many other causes.
Generating errors when connecting to the server

If you experience problems while connecting to the server, it may be connected to your communication options. Review the following: v Review the changes in the client communication options in the client option file (if they exist) and try to revert back to the previous values. Retry the connection. v If the server communication settings were changed, either update the client communication options to reflect the changed server values or revert the server back to its original values. v If any network settings were changed, such as the TCP/IP address for the client or server (or a firewall), work with the network administrator to understand these changes and update the client, server, or both for these network changes.
Resolving failed connections by clients or administrators

There may be several causes for you having connection problems to the server. The two main cases for connection failure are general failure, where no connections at all are allowed, and an isolated failure where some connections are allowed but others fail. If no connections at all are possible, it may be necessary to run the server in the foreground so that a server console is available, and additional diagnostic steps can be taken. Several settings should be checked to verify the proper configuration for communicating with the server: v Check that the server was able to bind to a port when it is started. If it was unable to bind to a port, then it is likely that some other application is using that port. The server can not bind (use) a given TCP/IP port if another application is already bound to that port. If the server is configured for TCP/IP communications and successfully binds to a port on startup for client sessions, the following message is issued: ANR8200I TCP/IP driver ready for connection with clients on port 1500. If the server is configured for HTTP communications for administrative session and successfully binds to a port on startup, the following message is issued: ANR8280I HTTP driver ready for connection with clients on port 1501. If a given communication method is configured in the server options file, but a successful bind message is not issued during server startup, then there is a problem initializing for that communication method. v Verify that the code TCPPORT setting in the server options file is correct. If this is inadvertently changed, this causes clients to fail to connect. That is because the clients are trying to connect to a different TCP/IP port than the one the server is listening on.
61
v If multiple servers are using the same TCP/IP address, ensure that the TCPPORT and TCPADMINPORT for each server are unique. For example, there are two servers at the same TCP/IP address. The first server has a TCPPORT of 1500 and a TCPADMINPORT of 1500. The second server has a TCPPORT of 1501 and a TCPADMINPORT of 1500. The first server to grab port 1500 locks out the other server from port 1500 and clients can no longer access the first server. Administrative clients always connect to the second server. A better choice of ports for each server would be 1500 and 1501 for TCPPORT; 1510 and 1511 for TCPADMINPORT. v Check that the server is enabled for sessions. Issue the QUERY STATUS command and verify that Availability: Enabled is set. If this reports Availability: Disabled, issue the ENABLE SESSIONS command. v If specific clients are unable to connect to the server, check the communication settings for those clients. For TCP/IP, check the TCPSERVERADDRESS and TCPSERVERPORT options in the client options file. v If only a specific node is rejected by the server, verify that the node is not locked on the server. Issue the QUERY NODE nodeName command, where nodeName is the name of the node to check. If this reports Locked?: Yes, then evaluate why this node is locked. Nodes can only be locked by using the LOCK NODE administrative command. If it is appropriate to unlock this node, issue the UNLOCK NODE nodeName command, where nodeName is the name of the node to unlock. v Work with the local network administrator to determine if there were network changes that account for the failing nodes not connecting to the server. v If the computer that the server is running on is having memory or resource allocation problems, it may not be possible to start new connections to the server. This may be cleared up temporarily by either halting and restarting the server or by halting and rebooting the computer itself. This is a temporary solution, and diagnosis should be continued for either the operating system or the Tivoli Storage Manager server because this may indicate an error in either. | | | | | | | | | | | | | | | | | |
Determining SSL errors

Several errors can occur either during secure sockets layer (SSL) setup or while establishing client and server connections. The following are common SSL errors: Unable to run gsk7capicmd.exe (Windows) In most cases this error is generated by an incorrect environment setup. Please refer to your Users Guide and set up the PATH variable as directed, before running the gsk7capicmd utility. ANS1595E Bad server certificate This error is reported when the server certificate is not known to the client. The bad server certificate error can occur under the following conditions: v The certificate was never imported v The certificate file (cert.arm) was corrupted before being imported v The command to import the certificate was entered incorrectly v The DSM_DIR variable points to wrong directory, which contains an incorrect client key database (dsmcert.kdb) Repeat all of the steps necessary for importing the server certificate and check the DSM_DIR variable. Refer to the client error log for more
62
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
information about the failure. The client error log may also contain information about specific GSKit failure. ANS1592E Failed to initialize SSL protocol This general error occurs on the client and indicates that the SSL connection could not be established. See the client error log for more information on the failure. One of the most common situations that lead to this error message is when the server does not accept SSL sessions on the port to which the client is trying to connect. Determine if the client points to the correct server port (TCPPort), which in most cases is a different port than the default 1500. ANR8583E and GSKit return code 406 This might indicate that a non-SSL enabled client is trying to contact an SSL port. When a client contacts a Tivoli Storage Manager server at a port defined by SSLTCPPORT or SSLTCPADMINPORT, the server establishes a session and initiates an SSL handshake. If the client is not SSL-enabled, it cannot complete the SSL handshake process. The session then appears to hang but will timeout through the server IDLEWAIT option, or end when the server administrator issues the CANCEL SESSION command to manually cancel it. The following is an example of a session in this state, from the server:
TSM:SERVER1>query session ANR2017I Administrator SERVER_CONSOLE issued command: QUERY SESSION Sess Number -----1 Comm. Method -----SSL Sess Wait Bytes Bytes Sess Platform Client Name State Time Sent Recvd Type ------ ------ ------- ------- ----- -------- ------------IdleW 17 S 0 0 Node
Note: Because the computing environment could cause a valid handshake process to take some time, do not assume that the above output always indicates a non-SSL client. ANR8583E and GSKit return code 420, and ANR8581E with GSKit return code 406 occur for the same Tivoli Storage Manager client session When Tivoli Storage Manager server messages ANR8583E with GSKit return code 420 and ANR8581E with GSKit return code 406 occur for the same client session, it is likely that the client has generated an ANS1595E message while attempting to establish a session with the server. If this is the case, follow the guidance in the Tivoli Storage Manager message manual for ANS1595E to eliminate these errors. Attention: For specific GSKit failure information, see Appendix C, IBM Global Security Kit return codes, on page 285.
Chapter 3. Resolving communication problems
63
64
Chapter 4. Administration Center

The Administration Center is a Web-based interface that can be used to centrally configure and manage Tivoli Storage Manager servers. You can achieve problem determination for Administration Center errors through several methods.
Establishing a connection between the Administration Center and a Tivoli Storage Manager server
If you have a problem establishing a connection to a Tivoli Storage Manager server, use the Administration Center to review the steps in this document to try to isolate or resolve the problem. Perform the following steps to establish a connection between the Administration Center and a Tivoli Storage Manager server: 1. Determine if the machine is accessible over the network. Issue the PING command in a command prompt or shell or try connecting using telnet or ftp. If the machine is not accessible over the network, the machine is down or there is a network communication error. Use the same TCP/IP address entered in the Administration Center. 2. Determine if the Tivoli Storage Manager server is running. If the server is not running, the Administration Center will not be able to establish a connection to it. 3. Try connecting to the Tivoli Storage Manager server by using the administrative client. Use the same connection settings as in the administration Center (TCP/IP address, TCP/IP port, admin id, admin password). This is the quickest way to determine if the problem is on the Tivoli Storage Manager server machine or on the machine running the Administration Center. If both the Tivoli Storage Manager server and the Administration Center are running on the same machine, skip this step and go to the directory where the administrative client is installed. Issue DSMADMC with parameters matching those in the connection settings for the Administration Center.
dsmadmc-id=<admin id> -pass=<admin password> -tcpport=<TCP/IP port> -tcps=<TCP/IP address>
4. Determine what port the Tivoli Storage Manager server is using. You can issue the QUERY OPT TCPPORT command on the server to determine the server port. Make sure the port used in the Administration Center is the same port on which the server is running. Go to the directory where the administrative client is installed. Issue DSMADMC with parameters matching those in the connection settings for the Administration Center. This command can also be issued from the server console.
dsmadmc -id=<admin id> -pass=<admin password> -tcpport=<TCP/IP port> -tcps=<TCP/IP address> tsm:SERVER1>QUERY OPT TCPPORT Server Option Option Setting -----------------------------------TCPPort 1500 tsm: SERVER1>
65
5. Determine if the machine you are attempting to access is behind a firewall. If the machine running the Tivoli Storage Manager server is behind a firewall, then the machine running the Administration Center must be authenticated to the firewall. Try using the machine running the Administration Center to connect to the machine running the Tivoli Storage Manager server. Use ping, telnet, or ftp. If this does not work, try connecting to another machine under the same firewall as the server machine. 6. Determine if the TCP/IP address for the Tivoli Storage Manager server is entered correctly in the Administration Center. Select Modify Server Connection from the drop-down list to view the current connection settings for the Tivoli Storage Manager server connection. Verify that the server connection address is correct. a. Issue the PING command in a command prompt or shell. Verify that the machine indicated by the address is accessible. b. Issue the QUERY OPT TCPPORT command in a server console to determine on which port the Tivoli Storage Manager server is running. To determine which port the Tivoli Storage Manager server is using, go to the directory where the administrative client is installed. Issue DSMADMC with parameters matching those in the connection settings for the Administration Center. This command can also be issued from the server console. Verify that the server connection port is correct.
dsmadmc -id=<admin id> -pass=<admin password> -tcpport=<TCP/IP port> -tcps=<TCP/IP address> tsm: SERVER1>QUERY OPT TCPPORT Server Option ----------------TCPPort tsm: SERVER1> Option Setting -------------------1500
7. Determine if the administrator ID for the Tivoli Storage Manager server is entered correctly in the Administration Center. Select Modify Server Connection from the drop-down list to view the current connection settings for the Tivoli Storage Manager server connection. Verify that the administrative ID is correct. 8. Determine if the administrator password for the Tivoli Storage Manager server is entered correctly in the Administration Center. Select Modify Server Connection from the drop-down list to view the current connection settings for the Tivoli Storage Manager server connection. Enter the correct administrative password to connect to the Tivoli Storage Manager server. Retry the operation to see if changing the password resolved the problem. 9. Determine if any messages are issued on the server when you try to connect using the Administration Center. The activity log contains useful information if you happen to experience any trouble connecting to the server. You can see sessions starting and stopping when the Administration Center successfully communicates with the server. If the network is down or the TCP/IP address or port is incorrect, no information shows up in the activity log when you make connection attempts. To check the server activity log, go to the directory where the administrative client is installed. Then issue the DSMADMC command with parameters matching those in the connection settings for the Administration Center. This command can also be issued from the server console.
66
dsmadmc -id=<admin id> -pass=<admin password> -tcpport=<TCP/IP port> -tcps=<TCP/IP address> tsm: SERVER1>QUERY ACTLOG
10. Check the servers database file for corruption. The servers database file is located in the ISC install directory under PortalServer\shared\app\dsm\ tsmservers.xml. If this file is corrupted, problems may occur when you try to connect to the servers. You will, however, typically see other error messages indicating that this file is corrupt. Open this file in an editor (preferably an XML editor), and verify that the end tag has a closing end tag.
Investigating failed installations

You might have problems when installing the Integrated Solutions Console (ISC) or the Tivoli Storage Manager Administration Center.
Resolving a failed Integrated Solutions Console installation

If you have a failed ISC installation, verify the following before reinstalling: v When you extract the package to a temporary directory, the directory must have a short directory path (less than 32 total characters) with no special characters or spaces. v Review the available log files to determine the cause of the installation failure, with special attention to the following: The log.txt file in the <ISC_HOME> directory, where <ISC_HOME> is the base directory for the ISC installation. This is the /opt/IBM/ISC601 directory for UNIX and Linux users and the C:\Program Files\IBM\ISC601 directory for Windows users. All ISC log files (those prefixed with ISC) in the systems temp directory, especially the ISCRuntimeInstall.log file. All log files in the Portal_Logs directory within the systems temp directory. All log files in the <ISC_HOME>/Tivoli/dsm/logs directory. All log files in the <ISC_HOME>/updates directory. See Installation log files on page 73 for more information. v The ISC uses the hosts file to resolve communication with the local host. Ensure the following: There is a reference in the hosts file to localhost directed towards 127.0.0.1 The PING LOCALHOST command succeeds The DNS references resolve to the correct host If these are not true, the following message might be found in the ISCArchiveFixup.log file:
[exec] WASX7023E: Error creating "SOAP" connection to host "localhost"; exception information: com.ibm.websphere.management.exception. ConnectorNotAvailableException: com.ibm.websphere.management.exception. ConnectorNotAvailableException: java.net.UnknownHostException: localhost
v Your operating system is on the supported list. If it is not, a message similar to the following might be found in the ac_install.log file:
[February 22, 2007 7:59:24 AM MST] CWLAA9008: The operating system Linux 2.4.20-8 is not supported.
v You have enough memory allocated to the system. If you do not, a message similar to the following might be found in the ISCArchiveUpdatePortalPorts.log file:
67
BUILD FAILED file:../config/actions/process.xml:162: Execute failed: java.io.IOException: Cannot allocate memory Total time: 45 seconds
v There is enough disk space on the system. To install the ISC requires the following approximate space allocations: New installation with no previous ISC installation Ensure that there is a total of 975 MB free space in the installation directory and a total of 300 MB free space in the temporary directory (75 MB for the Java virtual machine [JVM]). Migrated installation from Tivoli Storage Manager 5.3 and above Ensure that there is a total of 400 MB free space in the installation directory and a total of 225 MB (150 for the installation and another 75 MB for the JVM) free space in the temporary directory. If you do not have sufficient free disk space, the following message might be found in the updatelog.txt file located in a subdirectory of <ISC_ROOT>/ AppServer/logs/update:
There is insufficient free disk space on the system: /opt/IBM/ISC601/AppServer: Required: 400 MB Available: 609 MB /tmp/: Required: 150 MB Available: 102 MB
v Java must be installed on the system and in the path of the process that is running the installation. Note: Do not install the 64-bit Java Runtime Environment (even for 64-bit kernel). Use the 32-bit 1.4.2 or higher version of the IBM Java Runtime Environment or Sun Java Runtime Environment. See the log files located in <lSC_HOME>/updates with the file name of <date>_<time>_IUI_install.log. Issue the JAVA -VERSION command to test your Java installation. Verify that the response indicates that Java Runtime Environment 1.4.2 or later is installed. The following log information might be found for UNIX or Linux users:
Executing install task: isc-test-parms-ptf Setting WAS_HOME=/opt/IBM/ISC601/AppServer Ant Task Executed: java -cp /opt/IBM/ISC601/PortalServer/migration/ant154/lib/ant. jar org.apache.tools.ant.Main -buildfile "/tmp/fixpack/fix/ISC_PTF_6011_Config.xml" PARMS_HIDDEN isc-test-parms-ptf ========================================================= /usr/bin/sh: java: not found. ========================================================= isc-test-parms-ptf failed to execute
The following log information might be found for Windows users:

Executing install task: isc-test-parms-ptf Setting WAS_HOME=C:/Program Files/IBM/ISC601/AppServer Ant Task Executed: java -cp "C:/Program Files/IBM/ISC601/PortalServer\migration\ant154\lib\ant.jar" org.apache.tools.ant.Main -buildfile "C:\DOCUME~1\ADMINI~1\LOCALS~1\Temp\2\fixpack\fix\ISC_PTF_6011_Config.xml" PARMS_HIDDEN isc-test-parms-ptf
68
========================================================= java is not recognized as an internal or external command, operable program or batch file. ========================================================= isc-test-parms-ptf failed to execute
Note: For Linux Enterprise systems, problems were reported regarding the GNU Java that is natively installed which causes the installation to stop at 90% completion. Instead, install IBM Java Runtime Environment or Sun Java Runtime Environment version 1.4.2 (or later). Ensure that the correct version is installed by issuing the JAVA -VERSION command.
Resolving a failed Tivoli Storage Manager Administration Center installation

If you have a failed Tivoli Storage Manager installation, verify the following: v The ISC is installed. v The configuration information listed in the Review Integrated Solutions Console Configuration Information section of the Tivoli Storage Manager Administration Center Installation Wizard is correct. v ISC server and Console Help server are running.
Resolving ISC user authority problems

You can use the Settings tab of the Integrated Solutions Console (ISC) to create, delete, and view information for users and groups, as well as assign groups or users permissions to access resources. You can use the pages on the Settings tab to view settings for access permissions, users, and user groups. You can also define new users and assign the users to groups that are authorized to access console resources. ISC provides help for the Settings tab. This help assists you when navigating the Settings tab and its pages. ISC access permissions are role-based. Each role represents a set of actions that an authorized user can perform on a specific resource. A role is always defined in relationship to a resource. In ISC, a resource can be a page, a portlet on a page, or an organizational node in the navigation tree. To simplify the assignment of access permissions, ISC supports propagating and inheriting roles. ISC users can access only the pages for which they are authorized. In addition, the user must have permission to access one or more portlets on the page. Otherwise, the page displays as a blank page to the user. If a page contains multiple portlets, the user can be given access to some or all of the portlets. Access permissions can be set for each individual portlet or for the concrete portlet application to which the portlet belongs. Installing the Administration Center creates a new ISC user group for the Tivoli Storage Manager Administration Center. The new user group is called TSM_AdminCenter and new ISC users should be created in the group.They will have access to the Administration Center. The Administration Center resource is composed of several concrete portlet applications with all of its portlets being associated with one of these applications. The TSM_AdminCenter group is assigned the Administrator role for these concrete
69
portlet applications. Administrator is the highest level of authority for a resource. Users and groups that have Administrator authority can perform all of the permitted actions. Perform the following steps to create an ISC user that has access to the Tivoli Storage Manager Administration Center: 1. Log into the ISC console. 2. Select the Settings Tab. 3. Select User and Group Management. a. Search for the User Groups phrase. b. Search for the All Available phrase. c. Click the Search button. The table should now include the TSM_AdminCenter group. d. Click on TSM_AdminCenter. e. Click on the New User button. f. Fill in the form and click the OK button. If an ISC user does not have access to the Administration Center, the following is a list of items that should be verified: v Ensure that the group to which the user belongs has the necessary permissions to the resource. Ideally, the group is TSM_AdminCenter, but that is not a requirement. Select the Settings Tab. Select User and Group Permissions. Select User Groups. - Search for the phrase: All Available. - Click the Search button. - The table should now include the User Group of the given user. - Click on the Select Resource Type button of that group. - Select Portlet Applications and verify the permissions of the user. - Select Portlets and verify the permissions of the user. - Select Pages and verify the permissions of the user. v Ensure that the TSM_AdminCenter group does exist. v Ensure that TSM_AdminCenter group still has the necessary permissions to all the concrete portlet applications of the Administration Center. v Ensure that the user has access to all the portlets and pages of the Administration Center. Select the Settings Tab. Select User and Group Permissions. Select Users. - Search for keyword: uid. - Search for keyword: user_id. - Click the Search button. The table should now include the specified user. - Click on Select Resource Type. - Select Portlet Applications and verify the permissions of the user. - Select Portlets and verify the permissions of the user. - Select Pages and verify the permissions of the user.
70
Resolving an ISC server crash

The Integrated Solutions Console (ISC) is built on top of a WebSphere Application Server and WebSphere Portal. If the Websphere Portal server were to crash, ISC would no longer be accessible from a browser. Before restarting the ISC server, gather information about the crash so that the problem can be reported. See Unable to access the server from a Web browser on page 80 for additional information. WebSphere includes several troubleshooting tools that are designed to help you isolate the source of problems. Many of these tools are designed to generate information that IBM Support can use, and their output might not be understandable by the customer. This section only discusses tools that are bundled with ISC. A wide range of tools which address a variety of problems are available from the WebSphere Technical Support Web site.
Using the collector tool

The collector tool gathers information about your WebSphere server installation and packages it in a Java archive (JAR) file that you can send to Websphere Customer Support to assist in determining and analyzing your problem. Information in the JAR file includes logs, property files, configuration files, operating system and Java data, and the presence and level of each software prerequisite. There are two ways to run the collector tool. The collector tool can be run to collect summary data or to traverse the system to gather relevant files and command results. The collector tool produces a Java archive (JAR) file of information needed to determine and solve a problem. The collector summary option produces a lightweight collection of version and other information that is useful when first reporting the problem to Websphere Support. There are two phases of using the collector tool. The first phase involves running the collector tool on your WebSphere server and producing a JAR file . The Websphere Support team performs the second phase, which involves analyzing the JAR file. The collector program runs to completion as it creates the JAR file, despite any errors that it might find. Errors might include missing files or commands. The collector tool collects as much data in the JAR file as possible. Perform the following steps to run the collector tool: 1. Log on to the system as root (or Administrator on Windows). 2. Verify that Java 1.2.2 or higher is available in the path. 3. If there are multiple JDKs on the system, verify that the JDK that the WebSphere server uses is the one in the path for the collector program. The collector program requires Java to run. It also collects data about the Java Development Kit (JDK) in which it runs. 4. Verify that all necessary information is in the path being used by the collector program and that you are not running the program from within the ISC installation root directory.
71
a. If this system is a Linux or UNIX-based platform, verify that the path contains the following: v /bin v /sbin v /usr/bin v /usr/sbin b. If this system is a Windows platform, include regedit in the path. The collector tool is located in the isc_root/AppServer/bin directory. 5. Create a work directory where you can start the collector program. 6. Make the work directory, the current directory. The collector program writes its output JAR file to the current directory and also creates and deletes a number of temporary files in the current directory. By creating a work directory to run the collector program, the collector program avoids naming collisions and makes cleanup easier. You cannot run the collector tool in a directory under the ISC installation directory. 7. Run the collector program by issuing the COLLECTOR command from the command line. For example: c:\work>collector Issuing the COLLECTOR command with no additional parameters gathers one copy of the node data and data from each server in the node, and stores them in a single JAR output file. To gather data from a specific server in the node, issue the COLLECTOR servername command, where servername is the name of the problem server. The name of the ISC server is ISC_Portal. The collector program creates a log file, Collector.log, and an output JAR file in the current directory. The name of the JAR file is based on the hostname and package of the Websphere server in the following format: hostname-ND-WASenv.jar or hostname-BaseWASenv.jar. For example, if you run the collector tool on the server ws-laceweb within a Network Deployment cell, the filename is ws-laceweb-ND-WASenv.jar. The Collector.log file is one of the files collected in the hostname-ND-WASenv.jar or hostname-Base-WASenv.jar file. Note: Contact Websphere Support for assistance in deciphering the output of the collector tool.
Using the log analyzer tool (showlog)

The Log Analyzer takes one or more service or activity logs, merges all of the data, and displays the entries. Based on its symptom database, the tool analyzes and interprets the event or error conditions in the log entries to help you diagnose problems. The Log Analyzer has a special feature, enabling it to download the latest symptom database from the IBM Web site. About the Service or Activity log
72
The Websphere server creates the service or activity log file from the activity of the various WebSphere server components. The Log Analyzer is used to view the service or activity log file and can merge service or activity log files into one log file. The service or activity log file (activity.log), is a binary file in the logs directory of the install_root. You cannot view the service or activity log with a text editor. The Log Analyzer tool lets you view the file. Using the Log Analyzer The Websphere server on which ISC runs does not include the Java administrative console, so there is no graphical interface available for viewing a service or activity log file. The alternate viewing tool, showlog, must be used to view the service or activity log file: 1. Change to the directory your_isc_root/AppServer/bin where your_isc_root is the root directory for your ISC installation. 2. Run the showlog tool with no parameters to display usage instructions: v On Windows systems, run showlog.bat v On UNIX systems, run showlog.sh To direct the service or activity log (activity.log) contents to stdout, issue the showlog activity.log command. To dump the service or activity log to a text file for viewing with a text editor, issue the SHOWLOG ACTIVITY.LOG TEXTFILENAME command. Note: Contact Websphere Support for assistance in deciphering the output of the collector tool.
ISC server has excessive memory consumption

The Integrated Solutions Console (ISC) is built on top of a WebSphere application server and WebSphere Portal. If the Websphere server is using a large amount of memory, there are a few actions that can be taken. To determine if a larger amount of memory is in use, use the operating system tools that provide such information. For example, on Windows, the Task Manager shows memory use. Log off from ISC on a regular basis or if you determine that the memory usage of the Websphere Server is high. Logging out at the end of each day should help minimize the memory use. Logging out instead of only closing the browser is recommended. The default value of Session Timeout is 30 minutes. Reduce this value to a lower number to help reduce memory consumption requirements. The value can be changed with the Administration Center support utility on page 99.
Installation log files

The log files that are created when installing or uninstalling are located in several locations. Depending on the operating system, files stored in the OS temporary directory might not be saved after the system is shut down. For the Integrated Solutions Console Version 6.0.1, the installation logs are collected in the ISCLogs__.jar file.
73
This file can be found in the /Tivoli/dsm/logs directory. The table below describes the files that are created when installing and uninstalling, and recommends when and where to check the file for information that might assist you in troubleshooting problems:
Table 1. File Name ISCRuntimeInstall.log Description Contains messages for the Integrated Solutions Console Runtime Installation. Symptoms Check this log if the Integrated Solutions Console Runtime installation failed. Location This log is located in the OS temporary directory. For Windows, this is the value of the %TEMP% system variable. For all other systems, this is the /tmp directory. This log is located in the OS temporary directory. For Windows, this is the value of the %TEMP% system variable. For all other systems, this is the /tmp directory. This log is located in the OS temporary directory. For Windows, this is the value of the %TEMP% system variable. For all other systems, this is the /tmp directory. This log is located in the isc_runtime_root/ Tivoli/dsm/logs directory.
ISCRuntimeUninstall.log
Contains trace output for the runtime uninstall program.
Check this log if the runtime uninstall failed. Look for messages that indicate that a command failed.
EWASE.rsp
Contains the settings specified for the Integrated Solutions Console Runtime installation.
Check this log and verify that the settings are correct if the installation was not successful.
ac_install.log
Contains output from the Integrated Solutions Console Runtime installation and Administration Center installation. Contains error messages for the ISMP installation wizards for both the Integrated Solutions Console and the Administration Center.
Check this log if the Administration Center installation failed or when the ISCRuntimeInstall.log does not exist.
log.txt
Check this file to determine the value of the exception thrown. The exception error thrown helps determine where the install failed.
This log is located in the installation root directory (isc_runtime_root).
74
Configuring the IP address

The Administration Center assumes that the host system uses a static IP address instead of a dynamically-assigned IP address. A static IP address is necessary because the Administration Center server must be listed on the domain name servers, which map the host name to the physical address of the system. On a system that is not connected to the network, you must configure your system so that the IP loopback port is mapped to the fully-qualified host name. To enable that mapping, perform the following steps: 1. Go to the system where the Administration Center will be installed. 2. Locate the TCP/IP hosts file on your system. For Windows systems, look in WINNT\system32\drivers\etc. For AIX, Linux, and Solaris systems, look in path/etc. 3. Use a text editor to open the file named hosts. 4. At the end of the hosts file, add lines similar to the following:
127.0.0.1 localhost 127.0.0.1 your.server.name
where your.server.name is the fully-qualified host name for the Administration Center system. 5. Save the hosts file.
Problems with tutorials

You must verify certain configuration settings related to tutorials.
Is the help system functioning?

The tutorials are part of the help system, therefore when the help system is not running, you cannot view the tutorials. Determine if the help server process is running. Windows Open the Task Manager and select the process tab. UNIX Run a command like the following: ps -ef | grep eclipse Look for either eclipse.exe or EclipseSrv.exe in the list of processes. When you do not see either process name in the list, the help system is not running. You must start the help system. You must investigate further when there are multiple processes with the same names in the list or continued attempts to start the help server fail. This situation can arise if you are using the publications CD on the same system or if some other application is using the port. If, by chance, the random port number the publications CD chooses is 8423, then the Integrated Solutions Console help server does not start. Try pointing your Web browser to http://<machine name>:8423. For example, http://wscheid.tucson.ibm.com:8423 would display the help system on wscheid.tucson.ibm.com. You should see the title, Integrated Solutions Console, on the top left hand corner of the screen. If you do not see that title, some other help system or program is running on port 8423. Stop that application and restart the Integrated Solutions Console help server. See Starting and Stopping the help server for more information.
75
Are pop-ups allowed for the site?

After you verify that the help server is running, verify that pop-ups are allowed for the site. Every modern web browser now contains some pop-up blocker. Third-party additions can also perform this function. The Administration Center uses JavaScript to open the tutorial windows. Many pop-up blockers consider this action to be something that should be blocked. Consult the help for the pop-up blocker software to determine how to allow a Web site to display pop-ups. For example, Mozilla will place a little icon (looks like an exclamation point running from an explosion) in the status bar indicating that the pop-up is blocked. To allow the site to open pop-ups, click on the icon and press the add button in the dialog. Internet Explorer has a similar feature, although the browser version and the operating system version can change how you configure the setting or whether there is a setting. Another way to test whether a pop-up blocker is causing a problem is to run the tutorials directly in the help system. Open a browser window to http://<machine name>:8423. Expand the Tivoli Storage Manager branch of the contents. Expand the Tutorials branch of the contents. Select a tutorial to view. The tutorial runs in the work area to the right of the Contents tree. If the tutorial starts using this method, the pop-up blocker is probably causing the problem. If the tutorial does not start, you most likely have a problem with the MacroMedia Flash plug-in for the browser.
Is MacroMedia Flash player working?

To view the tutorials, the browser must have the MacroMedia Flash plug-in at version 6.0 or later. To obtain a copy of the plug-in, go to MacroMedia. Follow the instructions on the Web site to install the MacroMedia Flash player. After installation completes, the site provides a way to test the installation. Consult the support section of the MacroMedia Web site if you have problems during installation or are unable to get the plug-in to work.
Are the tutorials installed correctly?

The Integrated Solutions Console uses Eclipse to provide a help system. The help system contents are stored in a number of plug-ins, which are directories with a defined structure. The Administration Center uses the following two plug-ins for its help topics: v com.tivoli.dsm.admincenter.help.doc.5.3.0.0 v com.tivoli.dsm.admincenter.demo The second of these plug-ins contains the tutorials. Verify that the tutorials plug-in exists and contains a doc.zip file. Plug-ins are located in <isc root>/PortalServer/ISCEclipse/plugins where <isc root> is the Integrated Solutions Console installation directory. For example, on Windows look in c:\program files\IBM\ISC\PortalServer\ISCEclipse\plugins for the com.tivoli.dsm.admincenter.demo directory. On UNIX, look in /opt/IBM/ISC/PortalServer/ISCEclipse/plugins for that directory. If you see that the directory contains the doc.zip file, the plug-in is correctly installed. Otherwise, run the Administration Center installation again to install the help topics.
76
Conclusion
If you were unable to identify the problem through any of the previous actions, collect the following information before calling support: v View the source of the help pop-up window. On most browsers, a right-mouse click will provide a menu with a View Source option. A window appears with the HTML source code for that window. Write down the title of that window, which will be the URL or the name of the file the help system is attempting to display. v Turn on tracing for the SNAPIN trace class and try running the tutorial again. Stop tracing and run the collection option of the service tool.
Administration Center performance tuning suggestions

The Administration Centers performance can be improved in several ways through the suggestions in this topic.
How do I size the number of Tivoli Storage Manager Administration Center installations?
While there are a number of variables that can influence this sizing, most variables can be safely ignored if the Administration Center installation meets the recommended installation minimums and the following guidelines: v The Administration Center is being installed on a system dedicated to this function and no other applications will be installed on this system that require significant processing or memory resources. v The Administration Center is being installed on a system that is based on recent hardware (not more than 18 months has elapsed since general availability of the specific hardware models). If these conditions are met, it is likely that the system has sufficient processing capacity. Uni-processor systems or dual-processor systems are expected to be sufficient to meet the processing needs for a single Administration Center installation. More processors are not expected to be beneficial, as the major resource demand will be for memory. The Administration Center disk I/O and network I/O requirements are not particularly demanding and there is no need for sustained high I/O throughput. However, application response time suffers if network delays or disk I/O delays occur. A low-latency network provides the best administrator response time. The Integrated Solutions Console and Administration Center should be installed close to the administrators, rather than close to the Tivoli Storage Manager servers, because more of the network traffic during administration activities will occur between the ISC system and the administrators browser than between the ISC system and a Tivoli Storage Manager server.
How can I best optimize performance of the Administration Center?

The main resource requirement for supporting a significant number of the Tivoli Storage Manager administrators in the Administration Center is memory. Performance testing shows that the Administration Center process working set memory is determined by using the following equation: MWorkingSet = 100 + MJavaHeapSize
77
where: MWorkingSet is the Administration Center process working set memory in MB. MJavaHeapSize is the Administration Center Java heap size memory in MB. This working set memory must remain less than 2 GB, which is the maximum supported process virtual memory for many 32-bit operating systems. This is not likely to be a problem, since the maximum Java heap size allowed is 1536 MB. The required working set memory must be available in real memory (RAM) or significant response time degradation may occur as the result of system memory paging. Thus, the process working set memory requirement is effectively determined by the amount of Java heap memory required. The system memory sizing must include memory for the operating system and any other applications running on the system. Most operating systems require 256 MB, so this much memory should be added to the process working set memory requirement to obtain the system memory requirement. See the Administration Center Support Utility section in this guide for more tuning suggestions.
Where do I go for more information on performance tuning?

For the Tivoli Storage Manager Administration Center performance information, search the knowledge base for the Tivoli Storage Manager from the support Web site. Use 1193443 as your search term.
Workaround for blank pages in the Administration Center

One problem that has been found after installing the Integrated Solutions Console (ISC) and Administration Center, is that after logging in to the ISC, the Administration Center work items on the left display correctly, but when you click on them, the resultant pages are blank. This may be due to using equipment at or close to the minimum hardware requirements for this product or on a system under heavy load. The following error is seen in <isc_home>\PortalServer\log\SystemOut.log:
Exception: com.ibm.websphere.management.exception.DocumentNotFoundException for admincenter<some_number>.ear
The following errors are seen in <isc_home>/Tivoli/dsm/logs/ac_install.log:

[31 August 2005 08:29:20 GMT+00:00] **** Precompile Startup **** [31 August 2005 08:29:20 GMT+00:00] WPS script file [/opt/IBM/ISC/AppServer/bin/wpsSetupCmd.sh] exists - UNIXzzz [31 August 2005 08:29:20 GMT+00:00] ERROR: AdminCenter failed to deploy, unable to locate enterprise name. [31 August 2005 08:29:20 GMT+00:00] **** Precompile Failed **** [31 August 2005 08:29:20 GMT+00:00] Completed step number: 10zzz [31 August 2005 08:29:20 GMT+00:00] WARNING: Failed to precompile the JSP pages. There will be a performance hit while running.
<isc_home> The root directory in which the ISC was installed (default for Windows: C:\Program Files\IBM\ISC Default for UNIX: /opt/IBM/ISC) <userID> The user ID used to install the ISC (default is iscadmin). <password> The password used to install the ISC.
78
localhost:8421 The host name or IP address of where the ISC was installed and the port that was used (8421 is default). <fully qualified path> This is the fully-qualified path to the AdminCenter.war file that should be located where the Admin Center package was un-zipped or un-tared.
Workaround for Windows

1. Make sure the ISC server is running. You can verify this by looking in Windows Services for IBM WebSphere Application Server V5 - ISC Runtime Service. 2. Open a command prompt window and go to the <isc_home>\PortalServer\bin directory. 3. Issue the following: iscremove com.tivoli.dsm.admincenter <userID><password>http://localhost:8421/deploy/deploy This should return with a 0 when successfully removed. 4. Change your directory (cd) to <isc_home>\PortalServer\shared\app\config\ services 5. Open the DeploymentService.properties file. 6. Change was.notification.timeout = 60 to was.notification.timeout = 300 7. Stop and restart the ISC Service by completing the following steps: a. cd to <isc_home>\PortalServer\bin b. Issue stopISC.bat ISC_Portal c. Issue startISC.bat ISC_Portal 8. Issue iscdeploy <fully qualified path>\AdminCenter.war <userID><password>http://localhost:8421/deploy/deploy This should return with a 0 when successfully deployed.
Workaround for UNIX (AIX, Solaris, and Linux)

1. Ensure that the ISC server is running. You can verify this by going to <isc_home>/AppServer/bin and issuing serverStatus.sh ISC_Portal 2. Open a command prompt and go to the <isc_home>/PortalServer/bin directory. 3. Issue ./iscremove.sh com.tivoli.dsm.admincenter <userID><password>http:// localhost:8421/deploy/deploy This should return with a 0 when successfully removed. 4. cd to <isc_home>/PortalServer/shared/app/config/services 5. Open the DeploymentService.properties file. 6. Change was.notification.timeout = 60 to was.notification.timeout = 300 7. Stop and restart the ISC Service by completing the following steps: a. Cd to <isc_home>/PortalServer/bin b. Issue ./stopISC.sh ISC_Portal c. Issue ./startISC.sh ISC_Portal 8. Issue /iscdeploy.sh <fully qualified path>/AdminCenter.war <userID><password>http://localhost:8421/deploy/deploy This should return with a 0 when successfully deployed.
79
Unable to access the server from a Web browser

You might encounter a problem when you try to access the Integrated Solutions Console (ISC) and the Tivoli Storage Manager Administration Center from a Web browser. The following are examples of the messages that you could receive: v The page cannot be displayed. v The page you are looking for is currently unavailable. v Error: Cannot find server or DNS Error. v The connection was refused when attempting to contact host:port. These types of errors can occur for a number of reasons. The following are ways in which you can correct these errors: v Ensure that the system on which the browser is running is connected to the network and the target system on which ISC is installed. v Ensure that the browser settings are correct. v If you are behind a firewall and Internet security or proxy software is active, try disabling it temporarily. If ISC is now accessible, the firewall settings should be investigated. v Ensure that ISC is installed on the target system. v Ensure that ISC is running on the target system. ISC may have been terminated either. v Ensure that the Tivoli Storage Manager Administration Center was deployed into ISC. ISC is actually composed of two servers: the console server and the console help server. If the system on which ISC is installed is shut down, any ISC servers that were started will stop. If the system is configured to automatically start the console server and console help server when the system restarts, you should not have to take any action. The following table lists the commands for starting and stopping the servers on AIX, Linux, and Solaris systems.
Servers Start command (on a single line) Stop command (on a single line) your_isc_root/PortalServer/bin/ stopISC.sh ISC_Portal administrator password
Console server and your_isc_root/PortalServer/ console help server bin/startISC.sh ISC_Portal Console server only
your_isc_root/AppServer/bin/ your_isc_root/AppServer/bin/ startServer.sh ISC_Portal stopServer.sh ISC_Portal -username -username administrator-password password administrator-password password your_isc_root/PortalServer/ ISCEclipse/StartEclipse.sh your_isc_root/PortalServer/ISCEclipse/ StopEclipse.sh
Console help server only
Note: In the commands, the variables have the following meanings: your_isc_root The root directory for your Integrated Solutions Console installation.
80
administrator The administrator user ID for Integrated Solutions Console if the installation uses the WebSphere Application Server embedded in the console runtime. If the installation uses a separate WebSphere Application Server installation, administrator is the administrator user ID for the WebSphere Application Server installation. password The password for the administrator.
The table below lists the commands for starting and stopping the servers on a Windows system.
Servers Console server and console help server Console server only Start command (on a single line) Stop command (on a single line) your_isc_root\PortalServer\bin\ startISC.bat ISC_Portal your_isc_root\AppServer\bin\ startServer.bat ISC_Portal -username administrator-password password your_isc_root\PortalServer\ ISCEclipse\StartEclipse.bat your_isc_root\PortalServer\bin\ stopISC.bat ISC_Portal administrator password your_isc_root\AppServer\bin\ stopServer.bat ISC_Portal -username administrator-password password your_isc_root\PortalServer\ ISCEclipse\StopEclipse.bat
Console help server only
Note: In the commands, the variables have the following meanings: your_isc_root The root directory for your Integrated Solutions Console installation. administrator The administrator user ID for Integrated Solutions Console if the installation uses the WebSphere Application Server embedded in the console runtime. If the installation uses a separate WebSphere Application Server installation, administrator is the administrator user ID for the WebSphere Application Server installation. password The password for the administrator.
Health Monitor
A health monitor presents a view of the overall status of multiple servers and their storage devices. From the health monitor, you can link to details for a server, including a summary of the results of client schedules, and a summary of the availability of storage devices.
How the Health Monitor works

The Administration Center offers the functions of most administrative commands, as well as unique functions such as the health monitor and wizards to help you perform complex tasks. When you install the Tivoli Storage Manager Server 5.3.0 or upgrade the server to Tivoli Storage Manager 5.3.0 or a later version, the administrator ID ADMIN_CENTER is created and set to a lock status. When you configure the health monitor or add a server connection and check the unlock ADMIN_CENTER
81
administrator check box, the ID becomes unlocked to allow the health monitor to contact the server for status. If for some reason the ADMIN_CENTER does not exist, the health monitor registers the id on the server. The initial password is the same as the ID. Change the password through the Configure Health Monitor action from the Health Monitor Server table. The password and the time interval set are saved in the health monitor configuration file (dsmhealthmon.properties). This file is located in the {ISCROOT}/PortalServer/shared/app/config/services directory. The health monitor uses this ID and password to contact each server that was added to the Administration Center. When it completes contacting each server, the health monitor sleeps for the duration specified in the refresh interval. Only unique instances of servers are contacted. For instance, if two administrators each have WIN64SRV at address 9.11.232.90 on port 1500 configured, that server is only contacted once. The length of time it takes for the health monitor to complete contacting each server in the list may vary widely if servers are down. This is because the health monitor must wait for each connection to time-out before proceeding to the next server in the list. The refresh interval countdown displayed on the health monitor work page is not intended to match the actual cycle that the health monitor is using to contact servers. Instead, this is when the work page is refreshed. For example, there are 177 servers connections added to the Administration Center. Of these 177 servers, over half are test servers and are not active. It takes the health monitor approximately 10 minutes to work through the list of 177 servers to complete the check. If the refresh interval for the health monitor is 15 minutes and it takes 10 minutes to cycle through the servers, the servers are contacted about once every 25 minutes (10 minutes for the completion of the contact cycle and 15 minutes of sleep). In the meantime, the administrator left the screen on the health monitor work page and the screen refreshes every 15 minutes, whether a new status is available or not. The refresh rate can be changed in the Administration Center. Click Health Monitor and then select the Configure Health Monitor action from the drop-down list. From there you can change the refresh rate. When server connections are added, the health monitor cycles through its list of servers rather than waiting for the refresh interval. Avoid unnecessarily adding or removing server connections. An update of the configuration settings of the health monitor causes the health monitor to wake up and contact the list of servers. Avoid unnecessarily updating configuration settings. Note: The health monitor must be configured only one time, regardless of how many administrators and servers you defined. That configuration applies to all servers.
Conditions that can cause an unknown server status

v The server is down: if the server is down, the health monitor is unable to obtain status. v The ADMIN_CENTER ID not defined: If the ADMIN_CENTER administrative ID was deleted from the server, the health monitor is unable to obtain status. Redefine the ID.
82
v The ADMIN_CENTER password for that server does not match the password for the health monitor. All servers must use the ADMIN_CENTER ID and the same password. If the password that the health monitor uses to contact the server is invalid, the health monitor cannot obtain status. v The ADMIN_CENTER ID is locked. The ID must be unlocked or the health monitor is unable to obtain status. Use the Administrator tab of the server properties notebook to update the ID. When adding additional server connections, if you select the Unlock ADMIN_CENTER ID check box, this unlocks the ID. v The health monitor worker is no longer running. Tracing is the only way to determine if the worker thread is running. Follow the Administration Center tracing instruction to activate tracing for the HEALTH trace class. Search the PortalServer/logs/trace.log file for strings like Time for a nap or Back to work. If you find those messages in the trace file, it indicates that the health monitor is running. You might need to wait some time before seeing those messages, depending on if the health monitor worker is sleeping for the given interval. When you do not see any activity for a period of time longer than the refresh interval, try restarting the Integrated Solutions Console. This restarts the health monitor worker.
Conditions that can cause a warning or critical database status

The health monitor determines the server database status by evaluating a set of rules. These rules are fixed (cannot be changed by you). No single condition can cause the change in status, but rather, the health monitor looks at a variety of items to determine the overall status. Conditions are evaluated on a point system, and if the score is between zero and four, the status is Normal. Points of between five and nine cause a Warning, and points ten and above indicate a Critical status. The rules may change over time, as based on customer feedback. The following items are evaluated: v Dont have a space trigger - 2 points v Extension of database is 0 and maximum utilization is at 90% - 5 points v No space trigger, no extendable space for the DB, and current utilization of DB over 90% - 10 points v v v v v v v v Extension space exists and utilization is over 85% but less than 95% - 5 points Extension space exists and utilization is over 94% - 10 points No database backup in the last 24 hours - 3 points Cache hit ratio is less than 98% - 3 points Log utilization is currently over 80% but less than 90% - 3 points Log utilization is currently over 89% but less than 95% - 5 points Log utilization is currently over 94% - 5 points Log maximum utilization is over 90% - 3 points
Conditions that can cause a warning or critical storage status

Storage status works on the same principles as the database status. Perform the following steps to determine the overall health of the storage hardware: 1. Determine the percentage of drives that are offline in a given library.
83
2. Determine the percentage of paths that are offline for each drive in a given library. 3. Find the average of the numbers in the above step, which represents the score for the drive path status. 4. Determine the percentage of library paths that are offline for the given library. The score for the library is the maximum of the three scores. The number should be somewhere in the range from zero and ten. 5. Repeat the above steps for every library on the server. The overall health of the storage hardware is the average of the sum of all library scores. This is the overall health of the storage hardware on a scale of from zero to ten. A score of four or less represents Normal status. A score greater than four but less than eight represents Warning status. A score greater than seven represents a Critical status. If you take offline all of the drives in the library, this makes the status Critical for the library. If you take offline all of the paths for all of the drives in a single library, this causes the status to be Critical. If you take offline the only path to the library, this makes the library status Critical. Depending on the number of libraries on the server, the Critical status for a single library is diluted by other healthy libraries. For example, if the status of one library is Critical and two libraries are Normal, then the storage status is Warning. To view the individual status of the library, use the Storage Devices work item.
When to resynchronize the ADMIN_CENTER administrator ID password

When you configure the health monitor, the password for the ADMIN_CENTER administrator ID is saved in the health monitor configuration. The same password is used to contact all the servers on the list. As long as the password that is being updated is updated through the Configure Health Monitor action from the Health Monitor Server table, the password is synchronized for all the servers. When the ADMIN_CENTER password gets changed from the command line or the Server Properties notebook, however, the health monitor knows the new password in order to update the configuration file. The following occurs: 1. The health monitor contacts the server and the server returns an invalid credential message. 2. After the third try, the ADMIN_CENTER ID is locked. You cannot unlock the ADMIN_CENTER ID through the command line or the Server Properties notebook because the health monitor does not have a valid password. This action simply unlocks the ID; then after three tries, the ID is locked again. If you re-configure the health monitor to update the password, this also does not work because the health monitor attempts to log on using the password it knows about - which is no longer valid on that server. In either situation, resynchronize the ADMIN_CENTER password on all the servers to solve the problem. The health monitor will update the ADMIN_CENTER password to the current health monitor password for all the servers.
84
Administration Center task fails with a message

The Administration Center provides both informational and error messages when a failure occurs. Most failure messages will typically provide information about what went wrong. The Administration Center failure messages often explicitly diagnose the problem and include suggestions about how to manage it. The messages can contain information about Administration Center internal errors or server errors. In some cases, errors can be resolved by retrying a failed task. Before retrying the task, close any error messages. This ensures that only relevant error messages are displayed if the task fails again. Some Administration Center failure messages include one or more error messages returned from the server. The server messages are displayed in the language that was enabled for the server, which might not match the language enabled for the Web browser. In such cases, consult the Message Manual for the message number identified for the specific server platform. Further distinction between Administration Center messages and Tivoli Storage Manager server messages is outlined in Administration Center messages versus Tivoli Storage Manager server messages on page 92. To better understand the problem, check the Activity Log for the server chosen when the failure message occurred. To view the activity log, select a server and select the Server Properties table action. In the properties notebook, click the Activity Log tab. This is very useful for cases where the failure message does not seem related to the behavior of the task, it does not include server messages at all, or only the last server message is shown when multiple tasks were performed. In these cases, multiple commands are typically issued to the server. Look carefully in the Activity Log for the activity at the time frame of the failure. Keep in mind that the server machine and the machine used to access the Administration Center might be running in different time zones. Activity log information is displayed using the servers time. If errors are encountered during command routing, check the activity log of the source server first. If it does not provide sufficient information to diagnose the problem, check the activity log of the target server. Make sure that the two servers are running, that they are set up correctly to communicate with each other, and that the source server has appropriate authority level for the commands performed on the target server at the time of failure. If you need to contact customer support to resolve an Administration Center issue, you can use the tracing features provided by the Administration Center Support Utility to obtain additional information. You can find further information about the activity log and Administration Center internal errors in Administration Center task fails or returns unexpected results on page 91. Note: Some messages in the Administration Center tell you to see the Problem Determination Guide for more information. If the problem still cannot be resolved, then see Enabling Administration Center trace on page 149 for instructions on how to gather a trace before contacting customer support.
85
ANRW0089E
The following command generates a confirmation request from the server: <command>. The program cannot process this request. See Chapter 4, Administration Center, on page 65 for more information.
Description
This error message indicates that a command was issued that was expecting a reply. Some commands require a confirmation. This can be seen on the administrative client as commands that prompt for additional responses such as Are you sure you want to delete this? The API that the Administration Center uses does not have a mechanism to process confirmation requests.
Resolve
Try issuing the same command using the command line.
ANRW0090E
The following command is unknown to the server: <command>. See the Administration Center section of the Problem Determination Guide for more information.
Description
This error message indicates that an unknown command was issued to the server. This could be caused by a variety of reasons. The command may not exist on the version of server the command is being issued on. It may be an unsupported command on the platform the server is running on. The command may not have been constructed correctly.
Resolve
Visually inspect the command that was attempted. If the command is run together such as, QUERYNODE instead of QUERY NODE, this indicates the command was not constructed correctly. Attempt to reconstruct the command and issue it using the command line.
ANRW0126E
The administrative API encountered an internal error while processing the request. See the Administration Center section of the Problem Determination Guide for more information.
Description
This error message indicates that the administrative API that the Administration Center uses had an unrecoverable error while attempting to issue a command to the server.
Resolve
Try issuing the same command using the command line.
86
ANRW0127E
The administrative API is unable to process the command document sent from the server. See the Administration Center section of the Problem Determination Guide for more information.
Description
The Administration Center transfers a command document file from the server. This XML file indicates all the valid commands and their syntax for each version and platform of the server. This file is only transferred once during the first connection with a server of a particular version and platform. The error reports that the file could not be processed and may be corrupt. The file is located in the ISC install directory under PortalServer\shared\app\dsm\CmdFileCache. The name of the file depends on the version and platform. For example a 5.3.0 server for AIX would have the name aix_5_3_0.xml. Since the Administration Center is having trouble reading this file, it may be corrupt. Navigate to the CmdFileCache directory.
Resolve
Since all the files in this directory are only cached files, it is safe to delete them. Delete the file in question, or delete all of the files in the cache directory. This will force the Administration Center to pull the file from each server. Also check the permissions for this directory and make sure that the Administration Center can read and write files in this directory.
ANRW0128E
The administrative API is unable to process the command because it cannot be found in the command document. See the Administration Center section of the Problem Determination Guide for more information.
Description
The Administration Center transfers a command document file from the server. This XML file indicates all the valid commands and their syntax for each version and platform of the server. This file is only transferred once during the first connection with a server of a particular version and platform. The error reports that the command in question could not be found in the command file. This indicates the command file may be corrupt or out of date (wrong version). The file is located in the ISC install directory under PortalServer\shared\app\dsm\ CmdFileCache. The name of the file depends on the version and platform. For example a 5.3.0 server for AIX would have the name aix_5_3_0.xml.
Resolve
Since the Administration Center is are having trouble reading this file, it may be corrupt. Navigate to the CmdFileCache directory. Since all the files in this directory are only cached files, it is safe to delete them. Delete the file in question, or delete all of the files in the cache directory. This will force the Administration Center to pull the file from each server. Also check the permissions for this directory and make sure that the Administration Center can read and write files in this directory.
87
ANRW0129E
The following command contains one or more invalid parameters: <command>. See the Administration Center section of the Problem Determination Guide for more information.
Description
This error message indicates that the Administration Center was attempting to run a command on a server but the call to the API contains one or more invalid parameters. This type of error should typically be caught during validation before the command is submitted to the API to run.
Resolve
Check the parameters in the command. If you are filling out text entry fields on a page, you may be able to find the error in the parameters and correct it. Viewing the activity log in the servers properties notebook may also help determine the problem.
ANRW0130E
The administrative API encountered invalid parameters while processing the request. See the Administration Center section of the Problem Determination Guide for more information.
Description
This error message indicates that a command was being run through the administrative API but one of the parameters to an API method was invalid.
Resolve
Although this is typically an internal error, it may be caused by unusual parameters while constructing the command. Check the parameters in the command. If you are filling out text entry fields on a page, you may be able to find the error in the parameters and correct it. If you are entering a value with unusual characters such as <.> &, this may be the source of the problem. Viewing the activity log in the servers properties notebook may also help determine the problem.
ANRW0131E
The administrators authority level on this server cannot be determined. See the Administration Center section of the Problem Determination Guide for more information.
Description
This error message indicates that the authority level for the administrator used for connecting to this server cannot be determined.
Resolve
Try using a different administrator ID. Viewing the activity log in the servers properties notebook may also help determine the problem. To use a different
88
administrator ID, use the Modify Server Connection table action. Then specify a different administrator ID.
ANRW0132E
The administrative API cannot locate the command document. See the Administration Center section of the Problem Determination Guide for more information.
Description
The Administration Center transfers a command document file from the server. This XML file indicates all the valid commands and their syntax for each version and platform of the server. This file is only transferred once during the first connection with a server of a particular version and platform. The error reports that this file could not be located. The file is located in the ISC install directory under PortalServer\shared\app\dsm\CmdFileCache. The name of the file depends on the version and platform. For example a 5.3.0 server for AIX would have the name aix_5_3_0.xml. Check the directory permissions for the CmdFileCache directory.
Resolve
Administration Center must have permission to read and write to this directory.
ANRW0209E
An unknown failure occurred while the health monitor was evaluating the server. See the Administration Center section of the Problem Determination Guide for more information.
Description
The health monitor encountered an error that it could not recover from.
Resolve
Not much can be done in this case except to verify the health monitor is set up correctly. You also may want to try to stop and restart ISC. The only way the health monitor thread can come back is by restarting ISC in some cases.
ANRW0212E
The health monitor cannot evaluate the health of the server. Correcting problems that are indicated in the Detailed Server Health information might solve the problem. See the Administration Center section of the Problem Determination Guide for more information. The health monitor encountered an error while running reports on the servers.
Description
This error can occur if file specifying how to run the health monitor reports is modified or corrupted.
89
Resolve
Verify the health monitor reports file is not modified. This file is located in the ISC install directory under PortalServer\shared\app\dsm\reports\dsmhealthmon.xml. If this file has been modified or corrupted, replace this file with a valid version of the file. This can be done by installing the Administration Center again.
ANRW0213E
The health monitor is unable to locate information about the health of the server. See the Administration Center section of the Problem Determination Guide for more information.
Description
This error can occur when a server has recently been removed from the Administration Center, but the health monitor still thinks the server is there. This can happen if the health monitor and the rest of the Administration Center become out of sync on which servers are defined. It can also occur if the report files generated by the health monitor cannot be found.
Resolve
If the health monitor and the rest of the Administration Center become out of sync then restarting ISC will help. Verify the health monitor report directory exists and has read and write privileges. This directory is located in the ISC install directory under PortalServer\shared\app\dsm\reports. You may also want to try to run the new resync health monitor password (action from the Health Monitor table of servers entry screen). This password tries to resolve any credential issues with the system.
ANRW0221E
The health monitor reports an internal error. See the Administration Center section of the Problem Determination Guide for more information.
Description
The health monitor encountered an internal error that it could not recover from.
Resolve
Not much can be done in this case except to verify the health monitor is set up correctly. You may also want to try to restart ISC when the health monitor setup is correct.
ANRW0257E
The status information that is displayed is not valid because the health monitor service is unavailable. See the Administration Center section of the Problem Determination Guide for more information.
Description
This error occurs when the health monitor service cannot be found. This can happen when the health monitor service is not set up correctly.
90
Resolve
Verify the health monitor service is set up correctly.
Administration Center task fails or returns unexpected results

There are several methods to determine a problem with a task that fails on the Administration Center with unexpected results.
Check the server activity log

The Administration Center processes its work by issuing the Tivoli Storage Manager query, select, and other commands to the server. Check the activity log of the server that was selected for the failing task. To view the activity log, select a server and select the Server Properties table action. In the properties notebook, click the Activity Log tab. Ensure that commands related to the task appear in the activity log and are being issued to the correct server. If there is no evidence of activity related to the Administration Center task, there may be a connection problem or the incorrect server might have been selected. Look for failures related to commands that were issued in the time frame of the failing Administration Center task. These failures could be due to the incorrect formatting of the command or invalid parameter values. In some cases, a parameter value comes directly from the Administration Center, but in many others it is specified by the user through a text box, radio button, or other input method. Correct the invalid value if possible, but if an invalid parameter or incorrectly constructed command cannot be changed by the user, an internal error might be the cause.
Errors caused by starting or stopping a wizard or portlet

Some problems occur because of the manner in which wizards or portlets are started or restarted. Some of the wizards commit objects to the server database as you proceed through the wizard, instead of only when you finish the wizard. When the wizard defines objects before you click Finish, a summary panel listing the committed undo the committed changes. If the wizard is then canceled, the objects that were created are not deleted. You cannot restart the wizard from the point at which it was canceled. If you cancel and then restart a wizard, you cannot duplicate the work done earlier. For example, if you use the Add a Storage Device wizard to add a tape device, you will encounter several such summary panels that show the objects that are created before the wizard is finished. The first summary panel is for the creation of a device class, a library, and paths from the server to the library. If you cancel this wizard following the first summary panel and restart it, you cannot create another library with the same name because it now already exists. After an object is created with a wizard, any changes to that object must be made through a properties notebook. In the Add a Storage Device example, changes to the device class, library, or storage pools must be made through the device class, library, or storage pool properties notebooks. Typically, you can click an objects name to open its properties notebook. The properties notebook contains tabbed pages for modifying all of the attributes created by the wizard. Attributes that are set to the default by the wizard can be changed only in these properties notebooks.
91
Working with multiple portlets

The Administration Center allows you to open multiple work pages and multiple portlets within work pages. Updates to a server made in one portlet do not necessarily propagate to all other open portlets. In particular, the settings in a properties notebook are gathered when the properties notebook is first launched. If server objects in the properties notebook are created in a different portlet while the properties notebook is active, the newer objects will not be automatically added to the properties notebook. Likewise, a wizard that offers a selection of objects will only list those objects that existed at the start of the wizard. However, in some cases, object lists can be manually refreshed. Some portlets contain many tabs and list many server objects and their attributes. It can be difficult to ascertain whether changes made in another table, notebook, or wizard affected objects in a particular portlet. Furthermore, commands issued directly to the server through the server console or command line client can also add objects that will not appear in a notebook or wizard started before the commands were issued. In general, it is best to limit the number of portlets running at once, and to not use the command line client or server console while using the Administration Center. If a missing or changed object becomes apparent, cancel and restart the current task.
Problems caused by internal errors

An internal error is an error in the Administration Center itself, which can be more difficult to diagnose. In most cases, an error message is displayed. Another symptom is when an OK or Cancel button does not close the current panel. Finally, the server activity log might contain commands issued by the Administration Center that are improperly formatted or contain invalid parameter values. These errors can only be accurately diagnosed by support personnel. In addition to the server activity log, the Integrated Solutions Console trace log can often provide information relating to the error. If the problem is repeatable, the tracing level of the trace log can be increased. The information in the trace log will likely be useful to support personnel. See Enabling Administration Center trace on page 149 for more information about the Integrated Solutions Console trace log and setting the tracing level.
Administration Center messages versus Tivoli Storage Manager server messages

Messages shown in the Administration Center are typically either Tivoli Storage Manager server messages or messages specific to the Administration Center. Tivoli Storage Manager server messages are displayed in some cases when the Administration Center performed a command on a Tivoli Storage Manager server. In some cases, both types of messages are displayed. To distinguish between Administration Center messages and Tivoli Storage Manager server messages, observe the prefix of the message number. Messages that do not show any message number are Administration Center informational messages specific to the Integrated Solutions Console. Administration Center messages have the prefix ANRW. Informational and error messages are distinguished by the suffix of the message, I E.
92
For example: ANRW0022I The operation completed successfully. ANRW0023E An internal error occurred during validation. Informational messages, for example ANRW0022I, provide some general information or identify limitations or requirements for the task being performed. Error messages, for example ANRW0023E, identify that there is a problem completing the requested task. The message can prompt the user to perform additional setup (such as server-to-server-communication), identify that the requested action is invalid in the current work item or task, or report an internal error. Resolving an internal error, such as ANRW0023E, generally requires assistance from customer support. You will typically be asked to provide tracing information, as well as information about the sequence of actions performed before the failure occurred. Messages issued by the Administration Center are displayed in the language selected for the Web browser, or its closest default. Administration Center message numbers are not server dependent, unlike Tivoli Storage Manager server messages. Administration Center messages are generally based on the return code specified when a Tivoli Storage Manager server command was performed. In some cases the server return code that is the basis for the Administration Center message will be shown, along with the server message. In rare cases, the return code and message displayed will be inconsistent. If you use tracing to capture information for a server command implementation problem, the return codes identified with the server messages can help customer support diagnose the issue. Some Tivoli Storage Manager server messages are dependent on the server operating system. When the Administration Center displays a Tivoli Storage Manager server message, the message number can be associated with different messages for servers running on different operating systems. A single message number might have different variations for different platforms. Tivoli Storage Manager messages use different message prefixes, typically ANR or ANE. The following is a sample message that occurs if a DEFINE LIBRARY command fails on a z/OS server: ANR2022E DEFINE LIBRARY: One or more parameters are missing. Use the command line in the Administration Center or on the server machine where the failure occurred to obtain further information about the server message. To request help for the specific Tivoli Storage Manager server message number, issue the HELP ANR2022E command: As an alternative, use the Messages Manual and identify the message by its message number. If the message is platform specific, locate the message for the server platform of interest. For z/OS, for example, you should consider the version of the message showing (S/390) right after the message number. For the sample server message above, locate message number ANR2022E in the Message Manual.
93
Understanding Tivoli Storage Manager messages

The following examples illustrate the format used to describe the Tivoli Storage Manager messages: v Messages that begin with an ANE prefix and are in the 4000-4999 range originate from the backup-archive client. These messages (or events) are sent to the server for distribution to various event-logging receivers. v The client may send statistics to the server that is providing information about a backup or restore. These statistics are informational messages that may be enabled or disabled to the various event-logging receivers. These messages are not published here. v Messages that begin with an ANR prefix originate from the server or storage agent. v Messages that begin with an ANS prefix are from one of the following clients: Administrative clients Application program interface clients Backup-archive clients Space Manager (HSM) clients v Messages that begin with an ACD prefix are from Data Protection for Lotus Domino. v Messages that begin with an ACN prefix are from Data Protection for Microsoft Exchange Server. v Messages that begin with an ACO prefix are from Data Protection for Microsoft SQL Server. v Messages that begin with an ANU prefix are from Data Protection for Oracle. v Messages that begin with a BKI prefix are from Data Protection for SAP for DB2 UDB and Data Protection for SAP for Oracle. v Messages that begin with a DKP prefix and are in the 0001-9999 range are from Data Protection for WebSphere Application Server. v Messages that begin with an EEO prefix and are in the 0000-9999 range are from Data Protection for IBM Enterprise Storage Server (ESS) for Oracle. v Messages that begin with an EEP prefix and are in the 0000-9999 range are from Data Protection for Enterprise Storage Server (ESS) for DB2 UDB. v Messages that begin with an IDS prefix and are in the 1000-1999 range are from Data Protection for Enterprise Storage Server (ESS) for SAP.
Message format
The following example displays the Tivoli Storage Manager message format:
94
Message Prefix Message Number Message Type* ANR 0992 I Message text
Server installation complete. Explanation: The server installation procedure has completed successfully. The server is now available for normal operation. System Action: Server completes installation processing. User response: None. * I = Information E = Error S = Severe Warning W = Warning K = Kernel message that originates from the hierarchical storage management (HSM) client
Explanation
System Action
User Response
Message variables in the message text appear in italics. The server and client messages fall into the following categories: v Common messages that pertain to all Tivoli Storage Manager server platforms v Platform-specific messages that pertain to each operating environment for the server and the client v Messages that pertain to application clients
How to read a return code message

Many different commands can generate the same return code. The following examples are illustrations of two different commands that are issued that result in the same return code; therefore, you must read the descriptive message for the command. The QUERY EVENT command:
95
Message Prefix Message Number Message Type ANS NNNN T TSM:SERVER 1> q event standard dddd ANR2034I: QUERY EVENT: No match found for this query. ANS5I02I: Return code 11 Command issued Descriptive message (This value is explained in the descriptive message.)
The DEFINE VOLUME command:
Message Prefix Message Number Message Type ANS NNNN T TSM:SERVER 1> def vol cstg05 primary ANRxxxxx: DEFINE VOLUME: Storage pool CSTG05 is not defined. ANS5I02I: Return code 11 Command issued Descriptive message (This value is explained in the descriptive message.)
Problems with the Tivoli Storage Manager server command definition file
The Administration Center uses command definition files which specify the commands supported for each server operating system. There is a different command definition file for each server operating system. If this file was corrupted on the Integrated Solutions Console system or on the server system, parts of the interface might not be able to validate creation of new objects and might be unable to correctly construct server commands. Administration Center messages that identify server command syntax errors or validation errors (for text entry or the selection of interface controls) can sometimes be caused by problems with the command definition files. The Administration Center acquires a command definition file the first time it establishes a connection with a server. The next time the Administration Center contacts a server running on the same operating system, it uses the command file that was previously obtained for that operating system. If the command file is not available in the appropriate location (due to renaming or deletion) then any request following the disappearance of the file triggers another transfer of the command file for that specific operating system.
96
Do not attempt to update the command definition file. If problems occur that could be related to the command definition file, verify that the appropriate command definition file is available to the Administration Center for the servers operating system. If this file does not exist, verify that the Tivoli Storage Manager server can communicate with the Administration Center API and that it has the platform-specific command file and dsmcmd.xml available in the server directory hierarchy. Though it is not recommended, there is nothing to prevent the command definition file from being modified or deleted on the Integrated Solutions Console system, or the Tivoli Storage Manager server system. Verify that the appropriate platform-specific command file is available to the Administration Center for the platform of the server of interest. If not, verify that the Tivoli Storage Manager server can communicate with the Administration Center API and it has the platform-specific command file and dsmcmd.xml available and valid in the server directory hierarchy. To verify that a connection with the server can be established, see Using data storage diagnostic tips on page 267. One scenario is, for example, when a failure occurs as a task is being performed on an AIX 5.3.0 server. When a connection is established to that server or a command needs to be generated or verified, the Administration Center looks for the command definition file. The platform-specific command file is copied from the server to the following location in the Integrated Solutions Console Portal Server hierarchy: PortalServer\shared\app\dsm\CmdFileCache\aix_5_3_0_0.xml. Other platforms will have similar file name construction and are located in the same Portal Server directory. Any changes made to the command definition file can corrupt the XML defined objects in the file, which can cause Administration Center tasks to fail. If you identify a corrupted command definition file and cannot recover the original command definition file, you can use a connection to another server running the same operating system to obtain a valid command definition file, after renaming the original corrupted file. If other Administration Center .xml or .xsd files in the WEB-INF directory are corrupted, similar issues can occur. For example, the Administration Center might not be able to construct tables or launch portlets.
Diagnosing red error messages

The Administration Center produces red error messages, which are messages that appear in red print in order to draw your attention.
This portlet is unavailable. If the problem persists, please contact the portal administrator.
This error message is typically displayed when the Java Server Page (JSP) has code initializing a table model or a list model. If there is a connection problem with the Tivoli Storage Manager server, it is possible that the bean that holds all the initial values for the fields on this page was not properly initialized. For example, the reference to a table model in the bean for the Client Nodes Notebook is not initialized and has a null value. This causes an exception when the page is rendered. The message is displayed by the portal server and not the Tivoli Storage Manager Administration Center. Refer to tracing instructions for the
97
Administration Center to turn on the trace and look for any exceptions in the trace. The following message is an example of what you might see:
[12/7/04 13:43:37:859 PST] 1a33b798 snapin e com.tivoli.dsm.snapin.DsmLaunchedPortlet doView ERROR: view class call failed [12/7/04 13:43:37:859 PST] 1a33b798 snapin e com.tivoli.dsm.snapin.DsmLaunchedPortlet doView The following exception was logged com.ibm.wps.pe.pc.legacy.impl.TransferPortletException: Cannot set a null Table Model at com.ibm.wps.pe.pc.legacy.impl.PortletContextImpl.include (PortletContextImpl.java:270) at com.tivoli.dsm.snapin.views.PolicyDomainsView.performView (PolicyDomainsView.java:184) at com.tivoli.dsm.snapin.DsmLaunchedPortlet.doView (DsmLaunchedPortlet.java:727) at org.apache.jetspeed.portlet.PortletAdapter.service (PortletAdapter.java:154) ... ... ... ... Nested Exception is java.lang.IllegalArgumentException: Cannot set a null Table Model at com.ibm.psw.wcl.components.table.WTable.setModel (WTable.java:2376) at com.ibm.psw.wcl.components.table.WTable.setModel (WTable.java:2362) at org.apache.jsp._node_5F_prop_5F_notebook._jspService (_node_5F_prop_5F_notebook.java:2098) at com.ibm.ws.webcontainer.jsp.runtime.HttpJspBase.service (HttpJspBase.java:89)
In the case above, check that you can connect to the Tivoli Storage Manager server. Close the portlet and retry the same action. Another cause might be a compile error of the JSP, which should not have happened unless the page was modified. Look for any exceptions in the trace. You may see messages similar to the following:
[10/31/04 23:11:32:484 PST] 4c5b22a4 snapin e com.tivoli.dsm.snapin.DsmMainPortlet doView The following exception was logged com.ibm.wps.pe.pc.legacy.impl.TransferPortletException: /jsp/console/tpvw_tbl_libraries.jsp(155,4) Unable to load class com.tivoli.dsm.taglib.DsmRefreshTag at com.ibm.wps.pe.pc.legacy.impl.PortletContextImpl.include (PortletContextImpl.java(Compiled Code)) at com.tivoli.dsm.snapin.views.TpvwView.performView(TpvwView.java:348) ... ... ... Nested Exception is org.apache.jasper.compiler.CompileException: /jsp/console/tpvw_tbl_libraries.jsp(155,4) Unable to load class com.tivoli.dsm.taglib.DsmRefreshTag at com.ibm.ws.webcontainer.jsp.compiler.BasicTagBeginGenerator.init (BasicTagBeginGenerator.java:81) at org.apache.jasper.compiler.JspParseEventListener$GeneratorWrapper.init (JspParseEventListener.java(Compiled Code)) at org.apache.jasper.compiler.JspParseEventListener.addGenerator (JspParseEventListener.java(Compiled Code))
The error above cannot be recovered unless the JSP is fixed. Make sure the JSP was not modified. If the problem persists, consider reinstalling the IBM Tivoli Storage Manager Administration Center.
98
There has been an application error! Please contact to your system administrator to report this error.
Typically, this error message is caused by the user clicking a button or link more than once before the action is completed. For example, pressing the Cancel button on a notebook. If the request is not completed, and the Cancel button is pressed again, the application might fail and issue this error message. Close the application and log on to ISC again.
Administration Center support utility

The Administration Center Support Utility is a command-line driven utility designed to assist in performing basic support tasks with the Administration Center. This utility is located in the Integrated Solutions Console installation directory under Tivoli\dsm\bin. On Windows, start the tool by issuing supportUtil.bat. On UNIX, start the tool by issuing supportUtil.sh.
C:\Program Files\IBM\ISC\Tivoli\dsm\bin>supportUtil.bat Administration Center Support Utility - Main Menu ================================================== 1. Turn all tracing on. 2. Turn all tracing off. 3. Turn a single trace class on. 4. Update the maximum memory size Administration Center can use. 5. Update the Administration Center session timeout setting. 6. Collect trace files, logs and system information to send to support. 7. View the log file for this utility. 8. Exit.
99
100
Chapter 5. Data Protection

The Data Protection client uses the Tivoli Storage Manager API to communicate with the Tivoli Storage Manager server to provide data management functions.
Data Protection for Domino

Data Protection for Domino provides support for protecting Lotus Domino databases and transaction logs.
Tracing the Data Protection client (Domino)

The Data Protection client uses the Tivoli Storage Manager API to communicate with the Tivoli Storage Manager server and provide data management functions. To enable tracing, specify both of the following options on the command line when launching Data Protection for Domino or GUI executable:
/TRACEFILE= trace file name /TRACEFLAG= trace flags
trace file name The name of the file to which the trace data is written. Note that on UNIX environments, the domdsmc executable is launched by Domino startup script. This script changes the current working directory to the Domino Data directory; therefore, if a full path name is not specified, the trace-file-name is created in the Domino Data directory. trace flags The list of trace flags to enable. Trace flags are separated by a comma. For example, issue the following for the command line client:
DOMDSMC SELECTIVE database.nsf /TRACEFILE=trace.log /TRACEFLAG=SERVICE,API
Issue the following for the GUI client:

DOMDSM /TRACEFILE=trace.log /TRACEFLAG=ALL
Locating solutions for the Data Protection client (Domino)

Several resources are available to help you learn about or to diagnose the Data Protection client. The IBM Support Solutions database contains a knowledge base of articles and information on issues related to backup and restore problems. See Support for Data Protection for Domino. The product plug-in for the Tivoli Storage Manager for Mail: Data Protection for Domino, provides additional product-specific problem determination tools. The IBM Support Assistant information is found at the Software Support Web site.
Gathering information before calling IBM (Domino)

The Data Protection client is dependent upon the operating system as well as the Domino application.
101
You can significantly assist in determining the problem by collecting all the necessary information about the environment. Gather as much of the following information as possible before contacting IBM Support. v The exact level of the operating system, including all patches that have been applied. v The exact level of the Domino server. v v v v v v v v v | | | | | v The exact level of Data Protection for Domino. The exact level of the Tivoli Storage Manager API. The exact level of the Tivoli Storage Manager server. The exact level of the Tivoli Storage Manager backup-archive client. The exact level of the Tivoli Storage Manager storage agent (if LAN-free environment). The Tivoli Storage Manager server platform and operating system level. The output from the Tivoli Storage Manager server QUERY SYSTEM command. The output from the Data Protection for Domino DOMDSMC QUERY ADSM command. The output from the Data Protection for Domino DOMDSMC QUERY DOMINO command. The output from the DB2 BACKUP DB <dominodatabase> USE TSM command where dominodatabase is the name of the Domino DB2 database when encountering problems backing up DB2-enabled Lotus Notes databases. The value of the DSMI_CONFIG, DSMI_DIR, and DSMI_LOG environment variables (Domino DB2). Permissions and name of the user ID being used to run backup and restore operations.
v v
v A list of the steps needed to recreate the problem (if the problem can be recreated). v If the problem cannot be recreated, list the steps that caused the problem. v Is the problem occurring on other Domino servers?
Gathering files before calling IBM (Domino)

Several log files and other data can be collected by the Data Protection client. Gather as many of the following files as possible before contacting IBM Support: v Data Protection for Domino configuration file (default: domdsm.cfg). v Data Protection for Domino log file (default: domdsm.log). This file and should be monitored daily and indicates the date and time of a backup, data backed up, and any error messages or completion codes. This log file is commonly renamed in scheduled backup scripts by the /LOGFILE parameter specified with the DOMDSMC BACKUP command. v Tivoli Storage Manager API options file (default: dsm.opt). v Tivoli Storage Manager API error log file (default: dsierror.log). v Output from failed command or operation (redirected console or screen image). v The Domino NSD file contains a crash report that can help determine if a Domino crash occurred (nsd.exe for Windows, nsd.sh for Unix).
102
v Tivoli Storage Manager Server activity log. The Data Protection client sends information to the server activity log. A server administrator can view this log for you if you do not have a Tivoli Storage Manager administrator user ID and password. v If the Data Protection client is configured for LAN-free data movement, also collect the options file for the Tivoli Storage Manager storage agent. The default name for this file is dsmsta.opt. v Any command files and scripts used to run Data Protection for Domino. v If the Tivoli Storage Manager client scheduler is being used, also collect the client schedule log.
Data Protection for Snapshot Devices for mySAP

Data Protection for Snapshot Devices for mySAP (formerly Data Protection for FlashCopy Devices for mySAP or Data Protection for IBM ESS for mySAP) provides support for protecting Oracle or DB2 databases in a mySAP environment using the storage subsystems DS8000 and DS6000 series, IBM Enterprise Storage Server 800 (ESS800), the IBM System Storage SAN Volume Controller, and IBM System Storage N Series devices. | | The information gathered for Data Protection for Snapshot Devices for mySAP pertains to Tivoli Storage Manager version 5.4.
Troubleshooting Data Protection for Snapshot Devices for mySAP

In general, problems related to Data Protection for Snapshot Devices for mySAP can be categorized as setup and configuration problems or defects in the product or in interfacing products. When encountering an error, first check the description of the respective error message in the applicable Data Protection for Snapshot Devices for mySAP Installation and Users Guide and perform the recommended action. Error messages can appear both on the screen and in log files. If the system has performed successfully on numerous previous occasions but a problem seems to occur repeatedly without apparent reason and can be reproduced readily, ask yourself the following questions: v What changed in the setup of Data Protection for Snapshot Devices? v What changed in the setup of Data Protection for SAP? v What changed in the setup of the related components, such as Oracle, DB2, SAP, IBM Tivoli Storage Manager, the ESS/DS CIM Agent, operating system, network, or hardware? v Was the production database expanded by implementing new data files, file systems, logical volumes, or volume groups, for example? v Were patches or updates applied to any components in the system? If you are sure that nothing changed, ask the same questions of your coworkers and system administrators (DBAs, storage administrator, network administrator, IBM Tivoli Storage Manager administrator, and so on). If this does not resolve the problem, check when the configuration files (init<SID>.fcs, init<SID>.fct, vendor.env, init<SID>.utl, dsm.sys, dsm.opt, /etc/services, /etc/inittab, for example) were last changed.
103
The following UNIX command lists all files in the directory /etc that were modified within the last 5 days:
find /etc -type f -ctime 5 -print
Also, the system might have caused changes. The following are examples of such changes: v Disks are getting full (check with UNIX command df) v Networks are slower (check for causes such as additional hosts, additional applications, defects in software or hardware) v The IBM Tivoli Storage Manager server is slower (check to see if clients and/or operations were added and look at the IBM Tivoli Storage Manager servers activity log) If changes were made to the system, back them out one-at-a-time and try to reproduce the problem. In most cases you will find the one change or set of changes that caused the problem to occur. You can then decide if you need those changes or can fix the symptoms. Otherwise, if you need to implement those changes and cannot prevent their symptoms, contact support for the affected components. If a problem seems to occur randomly for an operation that was normally successful on previous occasions, try to find out what is different when the problem occurs. Compare the logs of the application in question to determine the differences between successful and unsuccessful runs. The following logs are available for comparison: v tdpdb2.<SID>.<nodename>.log (DB2) v v v v db2diag.log (DB2) tdphdwdb2_<p/b>_<function>_<timestamp>/.log/.trace (DB2) splitint_<p/b>_<function>_<timestamp>.log/.trace IBM Tivoli Storage Manager activity log
(where <p/b> indicates either _p_ for the production system or _b_ for the backup system). The following questions can be asked to try to find a pattern for the occurrence of the problem: v Does it always occur at the same time? v Does it always occur after you run some other operation or the same operation? v Does it always occur if some other application or process is running in parallel? If the problem always occurs at the same time, check if there are any scheduled processes (such as virus checker, automatic updates, batch jobs). The logs mentioned above may help you with this. Troubleshooting a FlashCopy backup (DB2) The following table has descriptions of troubleshooting tips for a FlashCopy backup in a DB2 environment, with ESS or DS storage systems as an example. Problem descriptions citing messages denoted as EExnnnn apply to both DB2 (EEPnnnn) and Oracle (EEOnnnn) environments. A more detailed description, with actual output examples and also troubleshooting information specific to an N Series environment, is shown in the Users Guide.
104
Table 2. Phases of a FlashCopy backup Phase Functions
Initialization In the Initialization phase, tdphdwdb2 is called on the backup system with the backup or flashcopy function. The Initialization phase includes the following steps: v Check the profile parameter. v Check the DB2 instance on backup system. v Check the Data Protection for SAP version. v Check the DB2 client connection to the production database. v Get a list of database files from the production database. v Check the connection to the production system through the FlashCopy backup Library (formerly the rexec mechanism). v Check connection to the ESS/DS CIM Agent.
Get Resources
In the Get Resources phase, splitint is started on the production system with the getresources function. The Get Resources phase includes the following steps: v Check the connection to the ESS Copy Services server. v Check the connection to the ESS/DS CIM Agent. v Get the status and other properties of the source and target volumes through the ESS/DS CIM Agent. v Determine the FlashCopy source volumes (from a list of database files). v Search for a matching target volume for each of the source volumes. v Check LVM mirroring configurations for VGs and LVs.
Flashcopy
In the Flashcopy phase, splitint/tdphdwdb2 runs on the production system with the flashcopy function. TRhe FlashCopy phase includes the following steps: v Suspend the production base v Flush the filesystems on the production system v Initiate the Flashcopy with the ESS/DS CIM Agent v Resume the production database write activities
Set Resources
In the Set Resources phase, tdphdwdb2/splitint is running on the backup system after the Get Resources and FlashCopy phases. The Set Resources phase has the following steps: v Clean up by removing the target volumes from the AIX ODM. v Configure all disk devices in the AIX ODM. v Clean up by removing all volumes with duplicate PVIDs from the AIX ODM. v Import/recreate the volume groups. v Mount the filesystems.
IBM Tivoli Storage Manager backup
In the IBM Tivoli Storage Manager backup phase, tdphdwdb2 calls DB2 on the backup system to start a backup of the FlashCopy-processed database to the IBM Tivoli Storage Manager.
INITIALIZATION PHASE
105
The following are the most common problems that can occur in the Initialization phase of a FlashCopy backup: v AUTHORIZATION PROBLEMS A typical problem that can occur on a regular basis is one involving the authentication of the db2<sid> user on the backup and/or production system. Passwords must be changed on a regular basis, and if this password change is not made known to Data Protection for Snapshot Devices, the next FlashCopy backup will fail.
Problem User name and/or password invalid Description This error occurs if an incorrect db2<sid> user name is specified in the profile parameter LOGON_HOST_PROD or if the password specified with the tdphdwdb2 configure function is incorrect. Solution Perform the following steps to verify: Check if the db2<sid> user specified in the profile parameter LOGON_HOST_PROD is the same as the database user on the production system. If the users are identical, the password of the db2<sid> user is no longer correct. To resolve this problem, correct the profile parameter LOGON_HOST_PROD. Change the password stored in the IDS_CONTROL_FILE (.fcp) by issuing the following command: tdphdwdb2 -f configure -p <profile> Password expired This error can occur on a To resolve this problem, change the regular basis if the password of expired password on the production the db2<sid> user on the and/or backup system with the passwd production system or on the AIX command. backup system is changed. Change the password stored in the IDS_CONTROL_FILE (.fcp) by issuing the following command: tdphdwdb2 -f configure -p <profile>
106
Problem
Description
Solution Perform the following steps to verify: On the production system: db2 get dbm cfg | grep AUTHENTICATION Database manager authentication (AUTHENTICATION) = SERVER_ENCRYPT On the backup system: db2 list db directory Database 2 entry: Database alias = R_E01 Database name = E01 Node name = REME01 Database release level = a.00 Comment = the production database Directory entry type = Remote Authentication = SERVER Catalog database partition number = -1 To resolve this problem, proceed as follows: 1. On the backup system, remove the database directory entries R_<SID> and R_<SID>_<NNNN> (where <NNNN> must be replaced by all DB2 partition numbers). 2. Set the Data Protection for Snapshot Devices profile (.fcs) parameter DB2_AUTHENTICATION to SERVER_ENCRYPT in the DB2 section of the profile and start a new FlashCopy backup run. This will create new catalog entries R_<SID> and R_<SID>_<NNNN> with the correct authentication method used on the production system.
Unsupported This error is the result of a function mismatch of the DB2 authentication method between the production and backup database instances. For example, on the production database instance the authentication is set to SERVER_ENCRYPT (DB2 database manager parameter AUTHENTICATION), but on the backup server the database directory entry R_<sid> is configured with authentication SERVER.
PROBLEMS INVOLVING A COMPONENT THAT IS NOT ACTIVE OR NOT READY

Problem DB2 was not started on the production system Data Protection for Snapshot Devices socket server not started on PS Description If a FlashCopy backup is started before the production database instance, Data Protection for Snapshot Devices fails with message IDS2502E. If a FlashCopy backup is started when the Data Protection for Snapshot Devices socket server on the production system is not active, Data Protection for Snapshot Devices will fail with message IDS2555E. Solution Start the production database and instance with the db2start command on the production system and restart the FlashCopy backup command. Start the Data Protection for Snapshot Devices for mySAP socket server on the production system and restart the FlashCopy backup command.
107
Problem
Description
Solution Eliminate the cause for the state of the tablespaces, or wait until the backup is finished and restart the FlashCopy backup.
Tablespace not If a FlashCopy backup is in normal started when one or more state tablespaces are not in a normal state on the production system, Data Protection for Snapshot Devices for mySAP will fail with message IDS2517E. The most common states are backup in progress and Offline and not accessible.
CIM AGENT CONNECTION PROBLEMS

Table 3. The most common errors that occur when the connection to the ESS/DS CIM Agent fails Problem Bad ESS/DS CIM Agent server address specified ESS/DS CIM agent down, Incorrect TCP/IP port specified, Description Invalid locator error Solution Correct the value of the Data Protection for Snapshot Devices profile parameter PRIMARY_COPYSERVICES_SERVERNAME.
These 3 configuration failures result in the same error message: Cannot connect to...
If the CIM Agent is not running, start it by issuing the following command: /opt/IBM/cimom/startcimom If the CIM Agent is running, check the TCP/IP port on which it is running. This can be verified in the log file of the CIM Agent (cimom.log), which is normally located in /opt/IBM/cimom. Correct the Data Protection for Snapshot Devices profile parameter COPYSERVICES_SERVERPORT as necessary. If the CIM Agent is running and the TCP/IP port of the CIM Agent is correct, modify the CIM Agent address in the parameter PRIMARY_COPYSERVICES_SERVERNAME if necessary.
To verify the current Incorrect CIM problem, first check that Agent server the ESS/DS CIM Agent (CIMOM) daemon is address installed and started by specified entering the following command: # ps -ef | grep CIMOM
GET RESOURCES PHASE

Table 4. Typical problems that can occur in the Get Resources phase of a FlashCopy backup Problem ESS Copy Services server not accessible Description Message EEx2730E is displayed if the ESS Copy Services servers (primary and secondary) are both not available. The most common reasons can be that either the ESS Copy Services servers are not started or there are network problems with the connection to the ESS Copy Services Server. Solution Make sure that the ESS Copy Services servers are started and that the TCP/IP connection to the ESS Copy Services servers is possible.
108
Table 4. Typical problems that can occur in the Get Resources phase of a FlashCopy backup (continued) Problem No target volume found for source volume Description Error message EEx2060W is typically displayed if a source volume does not have a matching target volume in the target volumes file (init<SID>.fct). Solution Add a new target volume to the target volumes file with the same size as the source volume. The new target volume(s) must be added in the correct target set (in the case that multiple target sets are used, e.g. for AIX LVM mirroring environments). To identify which target set needs to be adapted, check the target data container number in the error message. This number must match the volumes_set_<number>. Add the parameter HARDWARE_ID_LVM_MIRROR in the target volumes file in the correct volumes_set_<number> section. Specify the serial number of the mirrored hardware unit as the value for this parameter.
LVM mirrored Message IDS1133E indicates a setup but not specified problem with AIX LVM mirroring. in .fct It reports that the production database is placed on source volumes that are LVM-mirrored to another storage hardware unit but that the target volumes file does not reflect this with the HARDWARE_ID_LVM_MIRROR profile parameter. This parameter configures Data Protection for Snapshot Devices for environments where the production source volumes are LVM-mirrored to another hardware unit. Not LVM mirrored but specified in .fct Error message IDS1138E indicates a setup problem with AIX LVM mirroring. It indicates that the production database is implemented on source volumes that are not LVM-mirrored to another hardware unit but that in the target volumes file the HARDWARE_ID_LVM_MIRROR profile parameter is set. This parameter configures Data Protection for Snapshot Devices for environments where the production source volumes are LVM-mirrored to another hardware unit. In this case, however, the production system is not LVM mirrored.
Remove the parameter HARDWARE_ID_LVM_MIRROR from the target volumes file in the correct volumes_set_<number> section.
LVs must have 2 Error message EEx0166E indicates a LVM copies setup problem with AIX LVM mirroring. For example, the production database is placed on source volumes that are LVM-mirrored across two hardware units, but some logical volumes have one LVM mirror copy only.
Create a second copy for each of the affected logical volumes. Check the output to determine on which unit the second mirror copy must be created.
109
Table 4. Typical problems that can occur in the Get Resources phase of a FlashCopy backup (continued) Problem LVs must not have stale PPs Description Error message EEx0170W indicates a setup problem with AIX LVM mirroring. For example, the production database resides on source volumes that are LVM-mirrored across two hardware units, but the logical volumes have stale physical partitions on one of the LVM mirror copies. Depending on the target volumes set specified in the command line with -n <TargetSet> or automatically selected by Data Protection for Snapshot Devices, one of the two volumes causes the FlashCopy backup run to fail with error message EEx0176E. Error message EEx0174E indicates a setup problem with AIX LVM mirroring. For example, the production database resides on source volumes that are LVM-mirrored across two units, but at least one logical volume is not mirrored correctly. Solution Synchronize both logical volumes. To allow for high availability failovers, stale PPs should be avoided in any case.
Copy set must completely reside on one hardware unit
Remove one copy of the first logical volume and create a new second copy of this logical volume on the other unit.
LVM mirroring scheduling policy must be parallel
Error message EEx0168E indicates a Change the LVM mirroring setup problem with AIX LVM scheduling policy to one of the mirroring. For example, the valid values. production database resides on source volumes that are LVM-mirrored across two units, but the LVM mirroring scheduling policy (sequential) selected for the first volume group is not supported by Data Protection for Snapshot Devices. Valid values for scheduling policy are: parallel, parallel/parallel or striped. Error message EEx0172E indicates a Turn the LVM mirroring write setup problem with AIX LVM consistency on for the specified mirroring. For example, the volume group. production database resides on source volumes that are LVM-mirrored across two units, but the LVM mirroring write consistency (off) selected for one volume group is not supported by Data Protection for Snapshot Devices.
LVM mirror write consistency must be on
110
Table 4. Typical problems that can occur in the Get Resources phase of a FlashCopy backup (continued) Problem Quorum of VGs must be off Description Solution
Error message EEx0164E indicates a Turn the quorum off for the setup problem with AIX LVM specified volume group. mirroring. For example, the production database resides on source volumes that are LVM-mirrored across two units, but the quorum for one volume group is not supported by Data Protection for Snapshot Devices. Error message EEx0648E is displayed if the user and/or password specified for the ESS/DS CIM Agent is incorrect. The ESS/DS CIM Agent user is specified with the Data Protection for Snapshot Devices profile parameter COPYSERVICES_USERNAME. Check this parameter first. To check if the password is correctly known, enter the following command: # <dest-path>/setuser -u <CIM-user> -p <password> The password can be changed with the chuser command, but for that an admin user password of the CIM Agent is needed. After changing the password or verifying that the password is correct, Data Protection for Snapshot Devices must be updated with the correct password information. To do this, the following command must be started as the DB2 instance user: cd /db2/SIM/dbs ./splitint f password p <profile>
ESS/DS CIM Agent authentication problem
where <dest-path> is the destination This command calls the function directory where the CIM agent is password of Data Protection for installed. Snapshot Devices, which asks for If the password is correct, the the passwords of the DB2 instance following prompt is displayed: owner and the CIM Agent. After the passwords are updated Application setuser started successfully with Data Protection in interactive mode To terminate the application for Snapshot Devices, the enter: exit authentication problem should be To get a help message enter: resolved. help >>> If the password is incorrect, an output similar to the following is shown: E CMMOM0002E CIM_ERR_ACCESS_DENIED followed by traceback information.
111
Table 4. Typical problems that can occur in the Get Resources phase of a FlashCopy backup (continued) Problem CIM Agent query failed -default timeout 6 minutes Description Error message EEx0648E is displayed if the ESS/DS does not respond in a defined period of time (by default, 6 minutes). The error message can be found in the trace file of the failed FlashCopy backup (splitint_p_getresources_ <timestamp>.trace) Several lines above this error message, the time difference of 6 minutes can be noted between the following messages: DsHWInterface::HwGetFCS(): Entering..... DsHWInterface::HwGetFCS(): Exiting with error: ... It may happen that a query to the ESS/DS takes more than 6 minutes to respond if a very large number of volumes are used in a FlashCopy backup or restore operation. Solution Increase the timeout (specified with the Data Protection for Snapshot Devices profile parameter COPYSERVICES_TIMEOUT) to a value that is sufficient.
112
Table 4. Typical problems that can occur in the Get Resources phase of a FlashCopy backup (continued) Problem CIM Agent query failed -error in ESS/DS volume list Description Error messages EEx0648E and IDS1401E are displayed if incorrect ESS/DS volumes are specified in the FlashCopy target volumes file (.fct). There is something wrong with one of the target sets in the target volumes file. For example, when looking at the trace file of the FlashCopy backup run (splitint_p_getresources_ <timestamp>.trace) and searching for the error message EEx0648E, the reason for the error message can be found several lines above: Cant find System for LUN <1C824376> The list of known ESS/DS storage subsystems is shown in previous output lines. When comparing the list of ESS/DS with the LUN 1C824376, there is a clear mismatch between the LUN ID and the ESS/DS serial number. There is no ESS/DS with serial number 24376. For this reason, a typographical error in the .fct file must be assumed as the reason for the failed FlashCopy backup. Solution Edit the target volumes file (.fct) and correct the ESS/DS serial number.
SET RESOURCES PHASE The following describes the most typical problems that can occur in the Set Resources phase of a FlashCopy backup.
113
Problem No target volume found (ESS/DS CIM Agent)
Description The error message hdwmap.sh not found is issued for one of the following reasons: v The symbolic link hdwmap.sh does not exist in /db2/<SID>/dbs v The symbolic link hdwmap.sh does not point to /usr/tivoli/tsm/tdpessr3/db2/ <release_level>/hdwmap.sh v Data Protection for Snapshot Devices script hdwmap.sh does not exist in /usr/tivoli/tsm/ tdpessr3/db2/ <release_level> v the $PATH variable of the DB2 instance owner does not include . (the local directory)
Solution For the first 3 reasons: Check the symbolic link pointing to /usr/tivoli/tsm/tdpessr3/db2/< release_level>/hdwmap.sh and check for the existence of hdwmap.sh in /usr/tivoli/tsm/tdpessr3/db2/ <release_level> For reason 4: Check the $PATH variable of the DB2 instance owner and add . if necessary. Verify that all target volumes used as FlashCopy targets are attached to the backup system. In addition, as user root, run the following command: /db2/<SID>/dbs/hdwmap.sh vpath0 11E23376 hdisk2 hdisk56 hdisk110 vpath1 11F23376 hdisk3 hdisk57 hdisk111 vpath2 12023376 hdisk4 hdisk58 hdisk112 vpath35 31C23376 hdisk37 hdisk91 hdisk145 ... If not all target volumes can be found in the output of the command listed above, you need to attach the missing volumes with the ESS Specialist or the ESS CLI.
Importation of volume group failed -- missing volumes
Error message EEx0402I appears when not all target volumes are attached to the backup system. The result may be that some volume groups cannot be imported.
114
Problem Importation of volume group failed -- volumes not accessible
Description Error message EEx0402I appears when target volumes attached to the backup system cannot be accessed. The result may be that some volume groups cannot be imported. This error message is potentially the result of corruption in the AIX ODM. For example, the physical volume vpath79 is one of the target volumes specified in the target volumes file (.fct). This volume is attached to the backup system and is visible for AIX. Data Protection for Snapshot Devices for mySAP detects this target volume and tries to import a volume group from this volume. However, the low-level AIX command within the importvg command fails to access the volume.
Solution The reason for this cannot be determined exactly. Normally a manual cleanup of the AIX ODM would fix this problem: 1. Unmount all filesystems belonging to volume groups residing on ESS/DS volumes (take care that rootvg is not touched). 2. Vary off all volume groups residing on ESS/DS volumes (take care that rootvg is not touched). 3. Remove all multi path devices (vpath<NN> in case of SDD) (take care that rootvg is not touched). 4. Remove all hdisks of type IBM FC 2105 and IBM FC 2107. Be careful not to remove the disks of rootvg or local volume groups. 5. Afterward, you should have a clean set of hdisks in the AIX ODM (a check with lspv should show only disks of rootvg and local volume groups). 6. Run the cfgmgr command. 7. All hdisks and vpath devices should now be visible again. 8. Vary on all volume groups that were varied off previously. 9. If this step fails, the manual cleanup did not fix the AIX ODM problem, or the problem resides elsewhere in AIX, possibly the AIX kernel. In this case, a reboot of the backup system is needed to clean up the system. After the reboot, everything should work properly. 10. If the above procedure does not resolve the problem, restart the backup system (if this was not already done).
115
Problem Filesystem check/mount fails - FS is unknown
Description Error message EEx0402I can be shown in different error scenarios. The following two scenarios show typical configuration problems that will result in an error indicating that a file system cannot be found. v The first configuration problem results in a volume group in which a logical volume does not have a mount point in the output but the logical volume itself exists in the volume group. The reason for this can be a configuration problem on the production system or a problem with some FlashCopy target volumes.
Solution On the production system add the missing LABEL to the logical volume with the following command: chlv L <mount point> <LV name> In the above example the command would look like: chlv L /db2/E01/sapdata1 lvE01sapdata1 A new FlashCopy backup will then run correctly.
The second configuration problem results in a volume group that was imported (with warnings but no v Looking at the volume group on errors), but the detailed view of the physical volumes in the volume the production system and comparing two logical volumes groups shows that one of the target of this volume group, the output volumes is missing on the backup shows a difference. One of them system. does not have a LABEL defined. Verify that all target volumes used When copying a volume group as FlashCopy targets are attached with such an unlabeled logical volume, the importvg command to the backup system. In addition, as user root, run the following on the backup system cannot completely import the filesystem command: from this logical volume. The label is needed in order to import this filesystem correctly. /db2/<SID>/dbs/hdwmap.sh vpath0 11E23376 hdisk2 hdisk56 hdisk110 vpath1 11F23376 hdisk3 hdisk57 hdisk111 vpath2 12023376 hdisk4 hdisk58 hdisk112 vpath35 31C23376 hdisk37 hdisk91 hdisk145 If not all target volumes can be found in the output of the command listed above, you need to attach the missing volumes with the ESS Specialist or the ESS CLI.
IBM Tivoli Storage Manager BACKUP PHASE Refer to the troubleshooting information for Data Protection for SAP on page 132 as well as other relevant sections of the Problem Determination Guide.
Tracing Data Protection for Snapshot Devices for mySAP

Data Protection for Snapshot Devices for mySAP uses the IBM Tivoli Storage Manager API to communicate with the IBM Tivoli Storage Manager server and to provide data management functions. Refer to the backup-archive client section for a list of available trace classes.
116
To enable tracing, specify the following options in the user setup file (init<SID>.fcs) when you invoke the Data Protection for Snapshot Devices for mySAP command-line executables: v LOG_TRACE_DIR complete path where complete path is the directory where all of the run logs and trace files are placed. v TRACE YES Activate the trace for production and backup. Both parameters are also described with more details in the Data Protection for Snapshot Devices for mySAP Installation and Users Guide. You can find the logs and traces in the directories specified in the LOG_TRACE_DIR parameter of the Data Protection for Snapshot Devices profile. If no parameter is specified, the logs and traces are placed in the directory as specified in the WORK_DIR parameter of the Data Protection for Snapshot Devices profile. The file-naming convention for logs and traces is as follows: v splitint_b_<splitint function>_<date time stamp>.log v splitint_p_<splitint function>_<date time stamp>.log v splitint_b_<splitint function>_<date time stamp>.trace v splitint_p_<splitint function>_<date time stamp>.trace where v _b_ indicates that the function is running on the backup system v _p_ indicates that the function is running on the production system For more information, see also Summary of Log and Trace Files in the Data Protection for Snapshot Devices for mySAP Installation and Users Guide. COLLECTING DATA FROM THE DS OPEN API COM AGENT WHEN AN ERROR IS ENCOUNTERED Starting with IBM Tivoli Storage Manager Version 5.3, the communication between Data Protection for Snapshot Devices and the storage subsystem (except N Series) is handled via the CIM agent. The installation and configuration information for the CIM agent is contained in Data Storage Open Application Programming Interface Reference (GC35-0493-03). Perform the following steps to collect the log data: 1. Go to the CIM agent directory. The default directory is /opt/IBM/cimagent 2. Issue the collectLogs target directory command to compress all *.log files that are in the CIM agent directory and all sub-directories. These will be collected in the collectedLogs.zip file in the target directory. If a target directory is not specified, the default is the cimagent directory. If the specified target directory does not exist, it will be created. The logging settings are controlled by the logger.properties file in the cimagent directory. Any changes to this file require a restart of cimom to become effective. Default values: message.file.maxFiles = 20 This setting indicates the maximum number of 1MB cimom.log files to be kept. Adjust according to the level of activity and your available disk space.
117
message.logger.level = DEBUG_MAX This allows all debugging messages to be logged and could slow performance. This can be reduced to DEBUG_MIN when the system stabilizes. trace.file.maxFiles = 40 This setting indicates the maximum number of 1MB providerTrace.log files to be kept. Adjust this according to the level of activity and your available disk space. trace.logger.level = DEBUG_MAX This setting allows all debugging traces to be logged and could slow performance. This can be reduced to DEBUG_MID when the system stabilizes. indication.file.maxFiles=3 This setting indicates the maximum number of 1MB indication.log files to be kept. Adjust according to the level of activity and your available disk space. indication.logger.level=DEBUG_MAX This setting allows all debugging traces to be logged and could slow performance. This can be reduced to DEBUG_MID when the system stabilizes.
Locating solutions for Data Protection for Snapshot Devices for mySAP
Several resources are available for you to learn about or to diagnose Data Protection for Snapshot Devices for mySAP. The first reference to consider is the Data Protection for Snapshot Devices for mySAP Installation and Users Guide for the applicable DBMS. This is available from the Tivoli Information Center under Storage Manager for Advanced Copy Services. For the latest news, the IBM Support Solutions database contains a knowledge base of articles and information on issues related to backup and restore problems. Refer to the IBM support site for Tivoli Storage Manager for Advanced Copy Services (new product designation) for more information. Additional information can be found at the following Web sites: v For SAP, more information can be found at the SAP Service Marketplace. v For Oracle, search the Oracle Technology Network or visit the Oracle Web site. v For DB2, search the DB2 Web site.
Gathering information before calling IBM (DP for Snapshot Devices for mySAP)
Data Protection for Snapshot Devices for mySAP depends on the operating system as well as the Oracle or DB2 application. If you collect all the necessary information about the environment, this can significantly help to determine the problem. Gather as much of the following information as possible before contacting IBM Support:
118
v The exact level of the operating system, including the bit width (32/64) and all the maintenance levels and patches that were applied. v Are you running in an HACMP environment? If so, determine the exact level of the HACMP software, including all the maintenance levels and patches that were applied. v The exact level of the Oracle or DB2 server, including the bit width and all of the maintenance levels and patches that were applied. v Are you running in a single server or an OPS or RAC environment? If so, how many hosts are participating in this clustered environment? v The exact level of the ESS Server and ESS Copy Services CLI, if using an ESS storage device. Is your ESS FlashCopy feature enabled? v Are you using SDD? If so, the exact level of the SDD software. v v v v v v Is your database server attached to ESS using SCSI or Fibre Channel? The exact level of Data Protection for SAP. The exact level of the IBM Tivoli Storage Manager API. The exact level of the IBM Tivoli Storage Manager server. The exact level of the IBM Tivoli Storage Manager backup-archive client. The exact level of the IBM Tivoli Storage Manager storage agent (if LAN-free environment).
v The IBM Tivoli Storage Manager server platform and operating system level. v Output from the IBM Tivoli Storage Manager server query system command. v Output from the failing Data Protection for Snapshot Devices commands on the production and backup systems. v A list of the steps needed to recreate the problem. Note that if the problem cannot be easily reproduced, then list the steps that caused the instance of the problem that was encountered. v Does the problem occur on other database servers? v The type of file system (JFS, JFS2). v The layout of the file system of the SAP server in relation to the layout of the storage device (file systems, logical volumes, volume groups, LUNs, LSSs). v Does the problem occur on other SAP database servers?
Files to gather before calling IBM (DP for Snapshot Devices for mySAP)
Several log files and other data may be collected by Data Protection for Snapshot Devices for mySAP. Gather as many of the following files as possible before contacting IBM Support: v The error log files that DSM_LOG and DSMI_LOG point to on the production and backup systems. The Data Protection for Snapshot Devices error log file (default: tdpess.log) indicates the date and time of the commands and any error messages or completion codes. This file should be monitored daily. v The log files for SAP BR*Tools. v The trace files on the production system and the backup system. v The Data Protection for Snapshot Devices profile. This file ends in .fcs. v The Data Protection for Snapshot Devices target volumes file as specified by the VOLUMES_FILE parameter. This file ends in .fct. v The options files that are identified by DSM_DIR, DSMI_DIR, DSM_CONFIG, DSMI_CONFIG on both the production and backup systems.
119
v The IBM Tivoli Storage Manager Server activity log. Data Protection for Snapshot Devices for m<SAP logs information to the server activity log. An IBM Tivoli Storage Manager administrator can view this log for you if you do not have a IBM Tivoli Storage Manager administrator ID and password. v If Data Protection for SAP is configured for LAN-free data movement, also collect the options file for the IBM Tivoli Storage Manager storage agent. The default name for this file is dsmsta.opt. v Any command files and scripts used to run Data Protection for Snapshot Devices commands. v Any operating system error information (run the command: errpt -a). v If theIBM Tivoli Storage Manager client scheduler is used, collect the client schedule.
Data Protection for Oracle

Data Protection for Oracle provides support for protecting Oracle databases.
Tracing Data Protection for Oracle

The data protection client uses the Tivoli Storage Manager API to communicate with the Tivoli Storage Manager server to provide data management functions. Add the following lines to the tdpo.opt file to enable tracing for Data Protection for Oracle:
tdpo_trace_file trace file name tdpo_trace_flags trace flags
tdpo_trace_file= The name of the file to which the trace data is written. | | | tdpo_trace_flags= The list of trace flags to enable. Trace flags are separated by a space. The trace flags specific to Data Protection for Oracle are orclevel0, orclevel1, orclevel2, all, and orcbuff. For example:
tdpo_trace_file /home/dpOracle/log/trace.out tdpo_trace_flags orclevel0 orclevel1 orclevel2
Locating solutions for Data Protection for Oracle

Several resources are available to help you learn about or to diagnose the data protection client. The IBM Support Solutions database contains a knowledge base of articles and information on issues related to backup-restore problems. See Support for Data Protection for Oracle for information on Oracle backup/restore problems. Enter a term in order to search through solutions to previously-encountered issues. The product plug-in for the Tivoli Storage Manager for Mail: Data Protection for Domino, provides additional product-specific problem determination tools. The IBM Support Assistant information is found at the Software Support Web site. You can also search Oracles Technology Network, or visit the Oracle Web site.
120
Gathering information before calling IBM (Data Protection for Oracle)

The data protection client depends upon the operating system as well as the Oracle database. If you collect all of the necessary information about the environment, this can significantly help to determine the problem. Gather as much of the following information as possible before contacting IBM Support: v On what operating system is the problem being experienced? v The exact level of the operating system, including all service packs and hotfixes that were applied. v The exact level of the Oracle database, including all service packs and hotfixes that were applied. v The exact level of Data Protection for Oracle. v The exact level of the Tivoli Storage Manager API. v The exact level of the Tivoli Storage Manager server. v The exact level of the Tivoli Storage Manager backup-archive client. v The exact level of the Tivoli Storage Manager storage agent (if LAN-free environment). v Rman output. v Tivoli Storage Manager server platform and operating system level. v The output from the Tivoli Storage Manager server QUERY SYSTEM command. v The output from the Data Protection for Oracle TDPOCONF SHOWENV command. v A list of the other applications running on the system. v A list of the steps needed to recreate the problem (if the problem can be recreated). v If the problem cannot be recreated, list the steps that caused the problem. v Is Data Protection for Oracle running in a Oracle OPS or RAC environment? v Is the problem occurring on other Oracle databases? | | | | v v v v The location of all libobk links (libobk.a, libobk.so...etc.) and where they point. Whether or not the TDP for Oracle implementation is using an rman catalog. The bit version you are using (32-bit, 64-bit). The rman script.
Gathering files before calling IBM (Data Protection for Oracle)

Several log files and other data may be collected by the data protection client. Gather as many of the following files as possible before contacting IBM Support: v The Data Protection for Oracle configuration file. The default configuration file is tdpo.opt. v The Data Protection for Oracle log file. The default log file is tdpoerror.log. v The Data Protection for Oracle Tivoli Storage Manager API options file. The default options file is dsm.opt. v The Tivoli Storage Manager API error log file. The default API error log is tdpoerror.log.
121
v The output from failed command or operation. This may be either the console output redirected to a file or an actual screen image of the failure. v Any .trc files in Oracles user_dump_destination directory. v Sbtio.log, which should be in the $ORACLE_HOME/rdbms/log or user_dump_destination. v The Tivoli Storage Manager Server activity log. The data protection client logs information to the server activity log. A Tivoli Storage Manager administrator can view this log for you if you do not have a Tivoli Storage Manager administrator user ID and password. v If the data protection client is configured for LAN-free data movement, collect the options file for the Tivoli Storage Manager storage agent. The default name for this file is: dsmsta.opt. v Any command files and scripts used to run Data Protection for Oracle. v If the Tivoli Storage Manager client scheduler is being used, collect the client schedule log.
Data Protection for Exchange

Data Protection for Exchange provides legacy support for protecting Microsoft Exchange databases. If you encounter a problem during Data Protection for Exchange processing, perform the following steps as your first attempt to resolve the problem: 1. Retry the operation that failed. 2. If the problem occurred during an incremental, differential, or database copy backup, run a full backup. If the full backup completes successfully, retry the operation that failed. 3. If the problem still exists, close other applications, especially those applications that interact with Exchange (antivirus applications, for example). Retry the operation that failed. 4. If the problem still exists: a. Shut down the Exchange server. b. Restart the Exchange server. c. Run the operation that failed. 5. If the problem still exists: a. Shut down the entire machine. b. Restart the machine. c. Run the operation that failed. 6. If the problem still exists, determine if it is occurring on other Exchange servers.
Determining if the problem is a Data Protection for Exchange issue or an Exchange issue
The Data Protection client interacts closely with Microsoft Exchange. Because of this close interaction, it is necessary to first determine if the problem is with Microsoft Exchange or with the Data Protection client. Perform the following steps to try to isolate the source of the error: 1. Try recreating the problem with the Microsoft NTBACKUP utility. This utility uses a call sequence similar to Data Protection for Exchange to run an online
122
| | | | | | |
backup. If the problem can be recreated with NTBACKUP, then the problem most likely exists within the Exchange server. 2. Try recreating the problem with the Microsoft BACKREST (Exchange 2000 Server, Exchange Server 2003, Exchange Server 2007) application. This application can run backups using the Microsoft Exchange APIs. If the problem can be recreated with BACKREST, then the problem most likely exists within the Exchange server. Microsoft ships BACKREST with the Exchange Software Developers Kit (SDK). IBM Service can provide a copy of BACKREST if you encounter problems obtaining or building this application. 3. If the error message "ACN5350E An unknown Exchange API error has occurred" is displayed, the Exchange server encountered an unexpected situation. Microsoft assistance may be needed if the problem continues. 4. Data Protection for Exchange error messages occasionally contain an HRESULT code. Use this code to search Microsoft documentation and the Microsoft Knowledge Base for resolution information. Some Exchange SDK files contain these messages ESEBKMSG.H (Exchange 2000 Server, Exchange Server 2003, Exchange Server 2007).
Tracing Data Protection for Exchange

The Data Protection client uses the Tivoli Storage Manager API to communicate with the server and provide data management functions. Refer to the backup-archive client section for a list of available trace classes. To enable tracing, specify both of the following options on the command line when launching the Data Protection for Exchange command-line or GUI executable:
where trace file name= The name of the file to which the trace data is written. trace flags= The list of trace flags to enable. Trace flags are separated by a comma. Use the SERVICE and API trace classes to turn on a subset of trace flags that the service group normally requires. The following examples show you how to enable tracing, depending on your environment: Command line client:
TDPEXCC BACKUP SG1 FULL /TRACEFILE=trace.log /TRACEFLAG=SERVICE,API
GUI client:
TDPEXC /TRACEFILE=trace.log /TRACEFLAG=SERVICE,API
Locating solutions for Data Protection for Exchange

Several resources are available to learn about or to diagnose issues with the Data Protection client. The IBM Support Solutions database contains a knowledge base of articles and information on issues related to backup-restore problems. Refer to Support for Data Protection for Exchange to view this information.
123
The Microsoft Knowledge Base contains articles related to backup-restore problems. To review the Microsoft Knowledge Base, visit Microsoft Support.
Gathering information before contacting IBM (Data Protection for Exchange)

The Data Protection client is dependent upon the operating system and the Exchange application. Collecting all the necessary information about the environment can significantly assist in determining the problem. Gather as much of the following information as possible before contacting IBM Support: v The exact level of the Windows operating system, including all service packs and hotfixes that were applied. v The exact level of the Exchange Server, including all service packs and hotfixes that were applied. v The exact level of Data Protection for Exchange. v The exact level of the Tivoli Storage Manager API. v The exact level of the Tivoli Storage Manager server. v The exact level of the Tivoli Storage Manager backup-archive client. v The exact level of the Tivoli Storage Manager storage agent (if LAN-free environment). v The Tivoli Storage Manager server platform and operating system level. v The output from the Tivoli Storage Manager server QUERY SYSTEM command. v The output from the Data Protection for Exchange TDPEXCC QUERY EXCHANGE command. v The device type (and connectivity path) of the Exchange databases and logs. v Permissions and the name of the user ID being used to run backup and restore operations. v The name and version of antivirus software. v A list of third-party Exchange applications running on the system. v A list of other applications running on the system. v A list of the steps needed to recreate the problem (if the problem can be recreated). v If the problem can not be recreated, list the steps that caused the problem. v Is Data Protection for Exchange running in a Microsoft Cluster Server (MSCS) environment? v Is the problem occurring on other Exchange servers?
Gathering files before calling IBM (Data Protection for Exchange)

A number of log files and other data may be collected by the Data Protection client. Gather as many of the following files as possible before contacting IBM Support: v The Data Protection for Exchange configuration file. The default configuration file is tdpexc.cfg. v The Data Protection for Exchange log file. The default log file is tdpexc.log. This file indicates the date and time of a backup, the data that was backed up, and any error messages or completion codes. This file is important and should be
124
v v v
v v v
monitored daily. This log file is commonly renamed in scheduled backup scripts by the /LOGFILE parameter specified with the TDPEXCC BACKUP command. The Data Protection for Exchange Tivoli Storage Manager API options file. The default options file is dsm.opt. The Tivoli Storage Manager API error log file. The default error log file is dsierror.log. The Windows Event Log for Application and System. The Exchange Server logs information to the Windows Event Log. Exchange server error information can be obtained by viewing the Windows Event Log. The Tivoli Storage Manager registry hive export. The Exchange Server registry hive export. The Tivoli Storage Manager Server activity log. The Data Protection client logs information to the server activity log. A Tivoli Storage Manager administrator can view this log for you if you do not have a Tivoli Storage Manager administrator user ID and password. If the Data Protection client is configured for LAN-free data movement, also collect the options file for the Tivoli Storage Manager storage agent. The default name for this file is dsmsta.opt. If the Tivoli Storage Manager client scheduler is being used, also collect the client schedule log.
What to gather if the silent install fails (Data Protection for Exchange)
The silent install may not report information about the cause of the failure. To isolate or diagnose a failed silent installation, additional steps must be taken. Gather the following information to assist Customer Support when evaluating your situation: v Operating system level. v List of service packs or other fixes applied for the operating system. v A description of the hardware configuration. v Installation package (CD-ROM or electronic download) and level. v Any Windows event log relevant to the failed installation. v Windows services active during the failed installation (for example, antivirus software). v Whether or not you are logged on to the local machine console (not through terminal server). v Whether or not you are logged on as a local administrator, not a domain administrator. The Tivoli Storage Manager does not support cross-domain installations. v You can create a detailed log file (setup.log) of the failed installation. Run the setup program (setup.exe) in the following manner:
setup /v"l*v setup.log"
Data Protection for Exchange with VSS backup/restore support

Data Protection for Exchange provides support for protecting Microsoft Exchange databases through two different methods. The most common method is through Microsofts Legacy API. Data Protection for Exchange also can use Microsofts Virtual Shadow Copy Service (VSS).
125
If you encounter a problem during Data Protection for Exchange processing using VSS for Backup and Restore, follow the steps below as your first attempt to resolve the problem: 1. Retry the operation that failed. 2. If the problem still exists, close other applications, especially those applications that interact with Exchange (antivirus applications, for example). Retry the operation that failed. 3. If the problem still exists: a. Shut down the Exchange server. b. Restart the Exchange server. c. Run the operation that failed. 4. If the problem still exists: a. Shut down the entire machine. b. Restart the machine. c. Run the operation that failed. 5. If the problem still exists, determine if it is occurring on other Exchange servers.
Determining if the problem is a Data Protection for Exchange issue or a general VSS issue
The Data Protection client interacts closely with the backup-archive client (DSMAGENT), which performs all of the Virtual Shadow Copy Service (VSS) operations. Determine first if the problem is with Microsofts VSS service or with the Tivoli Storage Manager. Perform the following steps to try to isolate the source of the error: 1. Test the connectivity between the Data Protection client and the Tivoli Storage Manager dsmagent. Issue the TDPEXCC QUERY EXCHANGE command on the machine where the Exchange server is installed to verify that your installation and configuration is correct. This command returns information on the following: v Exchange Server status v Storage groups v Circular logging v VSS components The following is an example of the output generated by the TDPEXCC QUERY EXCHANGE command:
Volume Shadow Copy Service (VSS) Information -------------------------------------------Writer Name : Microsoft Exchange Writer Local DSMAgent Node : SERVERA Remote DSMAgent Node : SERVERX Writer Status : Online Selectable Components : 4
If the TDPEXCC QUERY EXCHANGE command does not return all of this information, you might have a proxy configuration problem. Refer to your Users manual for instructions on setting up proxy configurations. If all of the information returned to you seems correct, proceed to the next step. 2. Use the vssadmin or vshadow utility to recreate the VSS operation without the Tivoli Storage Manager intervening. When VSS operations are failing, use these
126
programs to recreate the error to determine if this is a general VSS problem or a problem within the Tivoli Storage Manager code. vssadmin A utility that is pre-installed with your operating system. It can display current volume shadow copy backups and all installed shadow copy writers and providers in the command window. The following are example VSSADMIN commands:
VSSADMIN LIST WRITERS VSSADMIN LIST PROVIDERS VSSADMIN LIST SHADOWS
vshadow A utility included with Microsofts Volume Shadow Copy Services SDK that can be used to exercise most of the VSS infrastructure, such as creating/querying/deleting shadow copies. You can also use vshadow to create both persistent and non-persistent shadow copies, transportable snapshots, as well as assign a drive letter or mount point to a shadow copy. v The following can be determined by using the vssadmin or vshadow utility: Verify VSS Provider configurations and setup Rule out any possible VSS problems before running the Tivoli Storage Manager VSS functions That you might have a VSS configuration problem or a real hardware problem if an operation does not work with vshadow/vssadmin That you might have a Tivoli Storage Manager problem if an operation works with vshadow/vssadmin but not with the Tivoli Storage Manager v Perform the following tests to ensure that VSS is working correctly: Test non-persistent shadow copy creation and deletion a. Run VSHADOW k: l: where k: and l: are the Exchange Server database and log volumes. b. Repeat the step above four times. c. Inspect the Windows Event Log to ensure that things look appropriate. Test persistent shadow copy creation and deletion a. Run VSHADOW -p k: l: (where k: and l: are the Exchange Server database and log volumes. You may have to run VSHADOW -da to remove this if you do not have enough space. b. Repeat the previous step four times. c. Inspect the Windows Event Log to ensure that things look appropriate. Test non-persistent transportable shadow copy creation and deletion (VSS Hardware Provider environments only) a. Run VSHADOW -p -t=export.xml k: l: where k: and l: are the Exchange Server database and log volumes. b. Copy the resultant export.xml file from machine 1 to machine 2 before performing the next step. c. On the machine you have set aside for offload, run VSHADOW -i=export.xml d. Inspect the Windows Event Log to ensure that things look appropriate. If any of these tests fail repeatedly, there is hardware configuration problem or a real VSS Problem. Consult your hardware documentation for known problems or search Microsoft Knowledge Database for any information. If all tests pass, continue to Step 3 on page 128.
127
3. Recreate your specific problem by using vshadow. If you can only recreate your problem through a series of steps (for example: a backup fails only when you perform two consecutive local backups), try to perform those same tests by using vshadow. v Exchange VSS backups to Local are simulated by running a vshadow persistent snapshot. v Exchange VSS backups to the Tivoli Storage Manager are simulated by running a vshadow non-persistent snapshot. v Exchange VSS backups to Local and to the Tivoli Storage Manager are simulated by running a vshadow persistent snapshot. v Offloaded Exchange VSS backups to the Tivoli Storage Manager are simulated by running a vshadow non-persistent, transportable snapshot. Refer to the VSHADOW documentation for the specific commands for performing backups. If you can recreate the problem, it most likely is a general VSS issue. Refer to Microsoft Knowledge Database for information. If your operation passes successfully with vshadow, it most likely is a Tivoli Storage Manager/Data Protection for Exchange client problem.
Tracing the Data Protection client when using VSS technology

You must gather traces for Data Protection for Exchange, the TSM API, and the DSMAGENT processes to ensure proper diagnosis of the Virtual Shadow Copy Service (VSS) operation. The following are the different traces to gather when you diagnose Data Protection for Exchange VSS operational problems: v Data Protection for Exchange trace To create the trace flag, issue the /TRACEFILE and /TRACEFLAGS command-line options with the following example command:
TDPEXCC BACKUP SG1 FULL /TRACEFILE=DPTRACE.TXT /TRACEFLAG=SERVICE TDPEXC /TRACEFILE=DPTRACE.TXT /TRACEFLAG=SERVICE
v Tivoli Storage Manager API trace Enable tracing with the DP/Exchange DSM.OPT file and the TRACEFILE and TRACEFLAGS keywords. The following is an example of the entry in the DP/Exchange DSM.OPT file:
TRACEFILE APITRACE.TXT TRACEFLAG SERVICE
v DSMAGENT trace Enable tracing with the DSMAGENT DSM.OPT file and the TRACEFILE and TRACEFLAGS keywords. The following is an example of the entry in the DSMAGENT DSM.OPT file:
TRACEFILE AGTTRACE.TXT TRACEFLAG ALL_VSS
The trace flag, in this instance, is ALL_VSS (you might need different traceflags, depending on the circumstance).
Gathering information before calling IBM (Exchange with VSS)

The Data Protection client is dependent upon the operating system and the Exchange application. Collecting all the necessary information about the environment can significantly assist in determining the problem.
128
Gather as much of the following information as possible before contacting IBM Support: v The exact level of the Windows operating system, including all service packs and hotfixes that were applied. v The exact level of the Exchange Server, including all service packs and hotfixes that were applied. v The exact level of Data Protection for Exchange with Virtual Shadow Copy Service (VSS) Backup/Restore support. v The exact level of the Tivoli Storage Manager API. v The exact level of the Tivoli Storage Manager server. v The exact level of the Tivoli Storage Manager backup-archive client. v The exact level of the Tivoli Storage Manager storage agent (if LAN-free environment). v The Tivoli Storage Manager server platform and operating system level. v The output from the Tivoli Storage Manager server QUERY SYSTEM command. v The output from the Data Protection for Exchange TDPEXCC QUERY EXCHANGE command. v The device type (and connectivity path) of the Exchange databases and logs. v The specific hardware that is being used (for example: HBA, driver levels, microcode levels, SAN Volume Controller levels, DS6000, DS8000 hardware details). v Permissions and the name of the user ID being used to run backup and restore operations. v The name and version of antivirus software. v The IBM VSS hardware provider level. The IBM CIM agent level for DS6000/DS8000 or the SAN Volume Controller. A list of third-party Exchange applications running on the system. A list of other applications running on the system. A list of the steps needed to recreate the problem (if the problem can be recreated). v If the problem can not be recreated, list the steps that caused the problem. v Is Data Protection for Exchange running in a Microsoft Cluster Server (MSCS) environment? v Is the problem occurring on other Exchange servers? v v v v Some operating system information that is listed above can be gathered by using the MPS Reports Tool. This is a Microsoft Reporting tool which lets you gather levels of operating system, software installed, driver versions, and other data. You can download the MPS Reports Tool at Microsoft.com.
Gathering files before calling IBM (Exchange with VSS)

Several log files and other data can be collected for Data Protection for Exchange server diagnosis. Gather as many of the following files as possible before contacting IBM Support: v The Data Protection for Exchange configuration file. The default configuration file is tdpexc.cfg. v The Data Protection for Exchange Tivoli Storage Manager API options file. The default options file is dsm.opt. v The Tivoli Storage Manager registry hive export.
129
v The Exchange Server registry hive export. v The Tivoli Storage Manager Server activity log. The Data Protection client logs information to the server activity log. A Tivoli Storage Manager administrator can view this log for you if you do not have a Tivoli Storage Manager administrator user ID and password. v If the Data Protection client is configured for LAN-free data movement, also collect the options file for the Tivoli Storage Manager storage agent. The default name for this file is dsmsta.opt. v Any screen shots or command-line output of failures or problems. Log files can indicate the date and time of a backup, the data that is backed up, and any error messages or completion codes that could help to determine your problem. The following are the Tivoli Storage Manager log files to gather: v The Data Protection for Exchange log file. The default location of this file is C:\Program Files\Tivoli\TSM\TDPExchange\tdpexc.log v The Tivoli Storage Manager API Error log file. The default location of this file is C:\Program Files\Tivoli\TSM\TDPExchange\dsierror.log v The DSMAGENT error log file. The default location of this file is C:\Program Files\Tivoli\TSM\baclient\dsmerror.log The Windows event log receives information from the Exchange Server and many different components involved during a Virtual Shadow Copy Service (VSS) operation. Microsoft provides a tool in the system32 directory on Windows XP and Windows 2003, called eventquery.vbs (it is a VBS script, not a binary executable) to retrieve event log data. You might use the tool as an alternative method of getting Windows event log data. To invoke the utility, issue the following CSCRIPT command:
cscript c:\windows\system32\eventquery.vbs parameters> outfilename
The utility, by default, produces a tabular listing of all event log records in three sections (one section per event log type). To get all log data in default format, issue the following command:
cscript c:\windows\system32\eventquery.vbs >eq.out
Specify the type of event log you require by using one of the following /L parameters: /L Application /L Security /L System The following example generates output only for the application and system event logs:
cscript c:\windows\system32\eventquery.vbs /L Application >eq_app.out cscript c:\windows\system32\eventquery.vbs /L System >eq_sys.out
You can utilize the /V parameter to receive detailed (verbose) output:

cscript c:\windows\system32\eventquery.vbs /V >eq.out cscript c:\windows\system32\eventquery.vbs /L System /V >eq_sys.out
You can utilize the /FO parameter to specify tabular, list, or comma-separated (CSV) output. The following are the different methods of specifying the output: /FO TABLE /FO LIST /FO CSV
130
The default format is TABLE. The LIST output puts each column of the record on a separate line, similar to how the Tivoli Storage Manager adminitrators command-line interface (CLI) displays output when it is too wide for tabular display. The CSV output can be loaded into a spreadsheet or database tool for easier viewing. The following example generates a detailed CSV file of the application log:
cscript c:\windows\system32\eventquery.vbs /L Application /FO CSV /V >eq_app.out
You can get additional help information for the tool by using the following example:
cscript c:\windows\system32\eventquery.vbs /?
The following VSS provider log files can also be helpful, if applicable: v System Provider - (Windows Event Log) v IBM System Storage SAN Volume Controller, DS6000, DS8000 - D:\Program Files\IBM\Hardware Provider for VSS\IBMVss.log v NetApp - D:Program Files\SnapDrive\*.log
Troubleshooting Data Protection for Exchange VSS and SAN Volume Controller
The troubleshooting tips included here are designed to help you accelerate your problem determination task. Note: If you plan to perform Data Protection for Exchange Virtual Shadow Copy Service (VSS) Instant Restore operations using SAN Volume Controller or DS6000/DS8000 disk subsystems, you must use non-SSL communications with the CIMOM. By default, the CIMOM is configured to use SSL communications. The Tivoli Storage Manager only supports non-SSL communications with the CIMOM. See the Data Protection for Exchange Users Guide for information on enabling non-SSL communications. The following are some areas which you can troubleshoot when you are having VSS and SAN Volume Controller or DS6000/DS8000 problems: v CIMOM (Common Information Model Object Manager) Connectivity issues To verify connectivity to the CIMOM, perform the following steps: 1. Refer to your SAN Volume Controller documentation. 2. Run the IBMVCFG LIST command. The default location is D:\Program Files\IBM\Hardware Provider for VSS-VDS. 3. Issue the IBMVCFG SHOWCFG command to view the provider configuration information. 4. Check that the CIMOM is properly configured. Run verifyconfig.bat -u username -p password on the SAN Volume Controller Master Console. 5. Check the username and password. If there is a problem with the truststore, follow the procedure in the documentation to generate a new truststore. 6. Set up the CIMOM properties file in non-SSL mode if you are using the SAN Volume Controller or DS6000/DS8000 and you plan to use Instant Restore. v CIMOM operational issues If your backup or restore fails, check the IBMVSS.log file. If the failure is due to a CIMOM failure, the log displays output similar to the following:
Wed Wed Wed Wed Jan Jan Jan Jan 11 11 11 11 17:34:34.793 17:34:35.702 17:34:35.702 17:34:35.718 Calling AttachReplicas AttachReplicas: 909ms returnValue: 34561 AttachReplicas returned: 34561
131
java.util.MissingResourceException: Cant find resource for bundle java.util.PropertyResourceBundle, key 1793 at java.util.ResourceBundle.getObject(ResourceBundle.java:329) at java.util.ResourceBundle.getString(ResourceBundle.java:289) at com.ibm.cim.CIMException.<init>(CIMException.java:472) at ESSService.executeFlashCopy(ESSService.java:3168) Wed Jan 11 17:34:35.779 - IBMVSS: AbortSnapshots
A return value of 0 means that it was successful. To determine why it failed, look at the log files generated by the CLI or graphical user interface (GUI), depending on how you run your operation. These may provide more information on the failure. v Host configuration issues If the failure seems to be for a different reason than a CIMOM failure, verify your configuration. Run the latest support levels of the software for the SAN Volume Controller or DS6000/DS8000. Check the IBM Storage web site for details. v Collecting logs in this environment If you are unable to resolve these problems, provide the following information to IBM Support: Information Listed in Collection TSM diagnostic information section HBA type, firmware and driver levels SDD version SAN Volume Controller microcode version (if applicable) DS6000/DS8000 microcode version (if applicable) SAN Volume Controller Master Console version (if applicable) For DS6000/DS8000, the CIM Agent version (if applicable) IBMVSS.log Application Event Log System Event Log If the problem appears related to CIMOM, you also need the CIMOM logs. Run CollectLogs.bat and send the file that is created (CollectedLogs.zip) to IBM Support. The default location for the SAN Volume Controller is C:\Program Files\IBM\svcconsole\support, and the default location for the DS6000/DS8000 is C:\Program Files\IBM\cimagent.
Data Protection for SAP

Data Protection for SAP provides support for protecting Oracle or DB2 databases in an SAP environment.
Troubleshooting Data Protection for SAP

Data Protection for SAP provides support for protecting Oracle or DB2 databases. The troubleshooting tips included here are designed to accelerate your problem determination. Note: For detailed troubleshooting information, refer to the Troubleshooting section of the respective Data Protection for SAP Installation and Users Guide (Oracle or DB2). When Data Protection for SAP is used in conjunction with Data Protection for Snapshot Devices, see also Data Protection for Snapshot Devices for mySAP.
132
In general, problems related to Data Protection for SAP can be categorized as one of the following: v Setup and configuration problems v Defects in the product or in interfacing products v Performance degradation The first step when encountering an error is to check the description of the respective error message in the applicable Data Protection for SAP Installation and Users Guide (Oracle or DB2) and perform the recommended action. Error messages can appear both on the screen and in log files. If a problem seems to occur repeatedly without apparent reason and can be reproduced readily when the system performed successfully on numerous previous occasions, ask yourself the following questions: v What changed in the setup of Data Protection for SAP? v What changed in the setup of related IBM Tivoli Storage Manager components such as Data Protection for Snapshot Devices? v What changed in the setup of the other components such as DB2 or Oracle, SAP, IBM Tivoli Storage Manager, operating system, network, or hardware? v Was the production database expanded by implementing new data files, filesystems, logical volumes, or volume groups, for example? v Were patches or updates applied to any components in the system? If you are sure that nothing was changed, ask the same questions of your coworkers and system administrators (DBAs, storage administrator, network administrator, IBM Tivoli Storage Manager administrator, etc.). If this does not resolve the problem, it might prove helpful to check when the configuration files (for example vendor.env, init<SID>.utl, init<SID>.sap, dsm.sys, dsm.opt, /etc/services, /etc/inittab) were last changed. The following UNIX command lists all files in the directory /etc that were modified within the last 5 days:
find /etc -type f -ctime 5 -print
Also, there might have been changes that were caused by the system. Examples of such changes are: v Disks are getting full (check with UNIX command df) v Networks are slower (check for causes such as additional hosts, additional applications, defects in software or hardware). Compare runs in the Performance Monitor history view in the Administration Assistant or check the Data Protection for SAP log files. v The IBM Tivoli Storage Manager server is slower (o see if clients and/or operations were added, and look at the IBM Tivoli Storage Manager servers activity log) If you discover that changes were made to the system, back them out, one-at-a-time, and try to reproduce the problem. In most cases you will find the one change or set of changes that caused the problem to occur. You can then decide if you need those changes or can fix the symptoms. Otherwise, if you need to implement those changes and cannot prevent their symptoms, you can contact support for the affected components. On the other hand, if a problem seems to occur randomly for an operation that was normally successful on previous occasions, you should try to find out what is different when the problem occurs.
133
Compare the logs of the application component in question to determine the differences between successful and unsuccessful runs, such as v back<SID>.log, rest<SID>.log, arch<SID>.log (Oracle) v tdpdb2.<SID>.<nodename>.log (DB2) v v v v v backom.log (DB2) db2diag.log (DB2) alert_<SID>.log (Oracle) sbtio.log (Oracle RMAN) IBM Tivoli Storage Manager activity log
Try to find a pattern for the occurrence of the problem: v Does it always occur at the same time? v Does it always occur after you run some other operation or the same operation? v Does it always occur if some other application or process is running in parallel? If it always occurs at the same time, check if there are any scheduled processes (such as virus checker, automatic updates, batch jobs). The logs mentioned above may help you with this.
Tracing Data Protection for SAP

Data Protection for SAP uses the IBM Tivoli Storage Manager API to communicate with the IBM Tivoli Storage Manager server and to provide data management functions. To enable tracing, specify both of the following options in the user setup file when invoking the Data Protection for SAP (Oracle and DB2) command-line executables:
TRACE=trace flags TRACEFILE=backint_%BID.trace
where trace flags= The list of trace flags to enable. Trace flags are separated by a space. The Data Protection for SAP trace flags are: ALL FILEIO FILEIO_MAX COMPR COMPR_MAX SYSCALL SYSCALL_MAX MUX MUX_MAX TSM TSM_MAX ASYNC ASYNC_MAX APPLICATION APPLICATION_MAX COMM COMM_MAX DEADLOCK DEADLOCK_MAX PROFILE PROFILE_MAX BLAPI BLAPI_MAX. If an error occurs early within an operation, it is suggested to set the ALL flag. Note: Setting this flag to trace a complete backup can result in a large trace file. The trace file in this case may be hundreds of megabytes. backint_%BID.trace= The value of this parameter specifies a directory and file where all the run logs and trace file are placed. This should be a fully qualified directory. The characters %BID are replaced during execution by the current backup ID. Both parameters are also described with more details in the Data Protection for SAP Installation and Users Guide. An example:
134
tracefile /home/tdpess/log/trace.out traceflags tdph tdph_detail api api_detail appl verbinfo timestamp
Be sure to remove the corresponding tracefile and traceflags options from the files that the DSM_CONFIG and DSMI_CONFIG environment variables identify.
Locating solutions for Data Protection for SAP

Several resources are available to help you learn about or to diagnose Data Protection for SAP. The first reference to consider is the Data Protection for SAP Installation and Users Guide for the underlying database (DB2 or Oracle). This is available from the Tivoli Information Center under Storage Manager for Enterprise Resource Planning. Refer to the Troubleshooting section. For the latest news, the IBM Support Solutions database contains a knowledge base of articles and information on issues related to backup and restore problems. Refer to IBM Tivoli Storage Manager for Enterprise Resource Planning. Additional information can be found at the following Web sites: v For SAP, you can find more information at the SAP Service Marketplace. v For DB2 UDB, you can visit the DB2 Product Family Web site. v For Oracle, you can search Oracle Technology Network, or visit the Oracle web site.
Gathering information before calling IBM (Data Protection for SAP)

Data Protection for SAP depends upon the operating system, as well as the Oracle or DB2 application. If you collect all the necessary information about the environment, this can significantly help to determine the problem. Gather as much of the following information as possible before contacting IBM Support: v On what operating system is the problem being experienced? v The exact level of the operating system, including all service packs and hotfixes that were applied. v The exact level of the Oracle or DB2, including all fix packs that were applied. v The exact level of Data Protection for SAP. v The exact level of the IBM Tivoli Storage Manager API. v The exact level of the IBM Tivoli Storage Manager server. v The exact level of the IBM Tivoli Storage Manager backup-archive client. v The exact level of the IBM Tivoli Storage Manager storage agent (if LAN-free environment). v The IBM Tivoli Storage Manager server platform and operating system level. v The output from the IBM Tivoli Storage Manager server query system command. v A list of other applications running on the system. v A list of the steps needed to recreate the problem (if the problem can be recreated). v If the problem is not re-creatable, list the steps that caused the problem. v Is the problem occurring on other Oracle or DB2 databases?
135
When using DB2: v Is Data Protection for SAP running in a DB2 (EEE or DPF) or ESE environment?
Gathering files before calling IBM (Data Protection for SAP)

Several log files and other data may be collected by Data Protection for SAP. Gather as many of the following files as possible before contacting IBM Support: v The Data Protection for SAP profile. The default profile is init<SID>.utl. v The log files of the SAP BR*Tools. v If using DB2, the Data Protection for SAP log file (tdpdb2.<SID>.<nodename>.log). v If using Oracle RMAN, the Data Protection for SAP default log file (sbtio.log). v If using DB2, Data Protection for SAP vendor environment file (the default vendor environment file is vendor.env). v The Data Protection for SAP IBM Tivoli Storage Manager API options file. The default options file is dsm.opt. v The IBM Tivoli Storage Manager API error log file. The default API error log is dsierror.log. v Output from failed commands or operations. This may be either the console output that is redirected to a file, or an actual screen image of the failure. v With DB2, use the db2support utility to collect environment data about either a client or server machine. This program generates information about a DB2 server, including information about its configuration and system environment. Collect any files in the DB2 dump directory (start with the DB2 diagnostic log; db2diag.log). v IBM Tivoli Storage Manager Server activity log. Data Protection for SAP records information to the server activity log. A IBM Tivoli Storage Manager administrator can view this log for you if you do not have a IBM Tivoli Storage Manager administrator user ID and password. v If Data Protection for SAP is configured for LAN-free data movement, collect the options file for the IBM Tivoli Storage Manager storage agent. The default name for this file is dsmsta.opt. v Any command files and scripts used to run Data Protection for SAP. v If the IBM Tivoli Storage Manager client scheduler was used, collect the client schedule log.
Data Protection for SQL

Data Protection for SQL provides support for protecting SQL databases.
Tracing Data Protection for SQL

The Data Protection client uses the Tivoli Storage Manager API to communicate with the Tivoli Storage Manager server and provide data management functions. To enable tracing, specify both of the following options on the command line when launching the Data Protection for SQL command-line or GUI executable:
where
136
trace file name= The name of the file to which the trace data is written. trace flags= The list of trace flags to enable. Trace flags are separated by a comma. Use the SERVICE and API trace classes to turn on a subset of trace flags that the service group normally requires. See the following examples: Command line client:
TDPSQLC BACKUP pubs FULL /TRACEFILE=trace.log /TRACEFLAG=SERVICE,API
GUI client:
TDPSQL /TRACEFILE=trace.log /TRACEFLAG=SERVICE,API
Locating solutions for Data Protection for SQL

Several resources are available to help you learn about or to diagnose the Data Protection client. The IBM Support Solutions database contains a knowledge base of articles and information on issues related to backup-restore problems. See Support for Data Protection for SQL for information on SQL backup-restore problems. The Microsoft Knowledge Base contains articles related to backup-restore problems. To review the Microsoft Knowledge Base, visit Microsoft Support.
Gathering information before calling IBM (Data Protection for SQL)

The Data Protection client depends upon the operating system as well as the SQL application. By collecting all of the necessary information about the environment, you can significantly assist in determining the problem. Gather as much of the following information as possible before contacting IBM Support: v The exact level of the Windows operating system, including all service packs and hotfixes that were applied. v The exact level of the SQL Server, including all service packs and hotfixes that were applied. v The exact level of Data Protection for SQL. v The exact level of the Tivoli Storage Manager API. v The exact level of the Tivoli Storage Manager server. v The exact level of the Tivoli Storage Manager backup-archive client. v The exact level of the Tivoli Storage Manager storage agent (if LAN-free environment). v The Tivoli Storage Manager server platform and operating system level. v The output from the Tivoli Storage Manager server QUERY SYSTEM command. v The output from the Data Protection for SQL TDPSQLC QUERY SQL command. v The device type (and connectivity path) of the SQL databases and logs. v Permissions and name of the user ID used to run backup/restore operations.
137
v A list of third-party SQL applications running on the system. v A list of other applications running on the system. v A list of the steps needed to recreate the problem (if the problem can be recreated). v If the problem cannot be recreated, list the steps that caused the problem. v Is Data Protection for SQL running in a Microsoft Cluster Server (MSCS) environment? v Is the problem occurring on other SQL servers?
Gathering files before calling IBM (Data Protection for SQL)

Several log files and other data may be collected by the Data Protection client. Before contacting IBM Support, gather as many of the following files as possible to resolve your problem: v The Data Protection for SQL configuration file (tdpsql.cfg is the default). v The Data Protection for SQL log file (tdpsql.log is the default), which indicates the date and time of a backup, data backed up, and any error messages or completion codes. This file should be monitored daily and is commonly renamed with the TDPSQLC BACKUP command in scheduled backup scripts by specifying the /LOGFILE parameter. v The Data Protection for SQL Tivoli Storage Manager API options file (dsm.opt is the default). v The Tivoli Storage Manager API error log file (dsierror.log is the default). v The SQL Server log file. The SQL Server records information to the SQL Server log file. You can view SQL Server log information by using the Enterprise Manager. Select Server Management SQL Server Logs Current or Archive #n. v The Windows Event Log for the Application and System event logs. v The Tivoli Storage Manager registry hive export. v The SQL Server registry hive export. v The Tivoli Storage Manager Server activity log. The Data Protection client records information to the server activity log. A Tivoli Storage Manager administrator can view this log for you if you do not have a Tivoli Storage Manager administrator user ID and password. v If the Data Protection client is configured for LAN-free data movement, also collect the options file for the Tivoli Storage Manager storage agent. The default name for this file is: dsmsta.opt. v If the Tivoli Storage Manager client scheduler is used, collect the client schedule log.
Gathering items if the silent install fails (Data Protection for SQL)
The silent install may not report information about the cause of the failure. To isolate or diagnose a failed silent installation, gather the following information to assist Customer Support when they evaluate your situation: v The operating system level. v A list of service packs or other fixes applied to the operating system. v A description of the hardware configuration. v The installation package (CD-ROM or electronic download) and level.
138
v Any Windows event log relevant to the failed installation. v The Windows services that were active during the failed installation (for example, antivirus software). v Whether or not you are logged on to the local machine console (not through terminal server). v Whether or not you are logged on as a local administrator, not a domain administrator. Tivoli Storage Manager does not support cross-domain installations. v You can create a detailed log file (setup.log) of the failed installation. Run the setup program (setup.exe) by issuing the SETUP /VL*V SETUP.LOG command. | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Data Protection for SQL with VSS backup-restore support

Data Protection for SQL provides support for protecting Microsoft SQL databases through two different methods. The most common method is through Microsofts Server Managed Objects (SMO) application programming interface (API). Data Protection for SQL also can use Microsofts Virtual Shadow Copy Service (VSS). If you encounter a problem during Data Protection for SQL processing using VSS for Backup and Restore, follow the steps below as your first attempt to resolve the problem: 1. Retry the operation that failed. 2. If the problem still exists, close other applications, especially those applications that interact with SQL (antivirus applications, for example). Retry the operation that failed. 3. If the problem still exists: a. Shut down the SQL server. b. Restart the SQL server. c. Run the operation that failed. 4. If the problem still exists: a. Shut down the entire machine. b. Restart the machine. c. Run the operation that failed. 5. If the problem still exists, determine if it is occurring on other SQL servers.
Determining if the problem is a Data Protection for SQL issue or a general VSS issue
The Data Protection client interacts closely with the backup-archive client (DSMAGENT), which performs all of the Virtual Shadow Copy Service (VSS) operations. Determine first if the problem is with Microsofts VSS service or with the Tivoli Storage Manager. Perform the following steps to try to isolate the source of the error: 1. Test the connectivity between the Data Protection client and the Tivoli Storage Manager dsmagent. Issue the TDPSQLC QUERY SQL command on the machine where the SQL server is installed to verify that your installation and configuration is correct. This command returns information on the following: v SQL Server status v VSS components
139
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
The following is an example of the output generated by the TDPSQLC QUERY SQL command:
Volume Shadow Copy Service (VSS) Information -------------------------------------------Writer Name : SqlServerWriter Local DSMAgent Node : SERVERA Remote DSMAgent Node : SERVERX Writer Status : Online Selectable Components : 10
If the TDPSQLC QUERY SQL command does not return all of this information, you might have a proxy configuration problem. Refer to your Users manual for instructions on setting up proxy configurations. If all of the information returned to you seems correct, proceed to the next step. 2. Use the vssadmin or vshadow utility to recreate the VSS operation without the Tivoli Storage Manager intervening. When VSS operations are failing, use these programs to recreate the error to determine if this is a general VSS problem or a problem within the Tivoli Storage Manager code. vssadmin A utility that is pre-installed with your operating system. It can display current volume shadow copy backups and all installed shadow copy writers and providers in the command window. The following are example VSSADMIN commands:
VSSADMIN LIST WRITERS VSSADMIN LIST PROVIDERS VSSADMIN LIST SHADOWS
vshadow A utility included with Microsofts Volume Shadow Copy Services SDK that can be used to exercise most of the VSS infrastructure, such as creating/querying/deleting shadow copies. You can also use vshadow to create both persistent and non-persistent shadow copies, transportable snapshots, as well as assign a drive letter or mount point to a shadow copy. v The following can be determined by using the vssadmin or vshadow utility: Verify VSS Provider configurations and setup Rule out any possible VSS problems before running the Tivoli Storage Manager VSS functions That you might have a VSS configuration problem or a real hardware problem if an operation does not work with vshadow/vssadmin That you might have a Tivoli Storage Manager problem if an operation works with vshadow/vssadmin but not with the Tivoli Storage Manager v Perform the following tests to ensure that VSS is working correctly: Test non-persistent shadow copy creation and deletion a. Run VSHADOW k: l: where k: and l: are the SQL Server database and log volumes. b. Repeat the step above four times. c. Inspect the Windows Event Log to ensure that things look appropriate. Test persistent shadow copy creation and deletion a. Run VSHADOW -p k: l: (where k: and l: are the SQL Server database and log volumes. You may have to run VSHADOW -da to remove this if you do not have enough space. b. Repeat the previous step four times.
140
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
c. Inspect the Windows Event Log to ensure that things look appropriate. Test non-persistent transportable shadow copy creation and deletion (VSS Hardware Provider environments only) a. Run VSHADOW -p -t=export.xml k: l: where k: and l: are the SQL Server database and log volumes. b. Copy the resultant export.xml file from machine 1 to machine 2 before performing the next step. c. On the machine you have set aside for offload, run VSHADOW -i=export.xml d. Inspect the Windows Event Log to ensure that things look appropriate. If any of these tests fail repeatedly, there is hardware configuration problem or a real VSS Problem. Consult your hardware documentation for known problems or search Microsoft Knowledge Database for any information. If all tests pass, continue to Step 3. 3. Recreate your specific problem by using vshadow. If you can only recreate your problem through a series of steps (for example: a backup fails only when you perform two consecutive local backups), try to perform those same tests by using vshadow. v SQL VSS backups to Local are simulated by running a vshadow persistent snapshot. v SQL VSS backups to the Tivoli Storage Manager are simulated by running a vshadow non-persistent snapshot. v SQL VSS backups to Local and to the Tivoli Storage Manager are simulated by running a vshadow persistent snapshot. v Offloaded SQL VSS backups to the Tivoli Storage Manager are simulated by running a vshadow non-persistent, transportable snapshot. Refer to the VSHADOW documentation for the specific commands for performing backups. If you can recreate the problem, it most likely is a general VSS issue. Refer to Microsoft Knowledge Database for information. If your operation passes successfully with vshadow, it most likely is a Tivoli Storage Manager/Data Protection for SQL client problem.
Tracing the Data Protection client when using VSS technology (SQL)
You must gather traces for Data Protection for SQL, the TSM API, and the DSMAGENT processes to ensure proper diagnosis of the Virtual Shadow Copy Service (VSS) operation. The following are the different traces to gather when you diagnose Data Protection for SQL VSS operational problems: v Data Protection for SQL trace To create the trace flag, issue the /TRACEFILE and /TRACEFLAGS command-line options with the following example command:
TDPSQLC BACKUP DATABASE1 FULL /TRACEFILE=DPTRACE.TXT /TRACEFLAG=SERVICE TDPSQL /TRACEFILE=DPTRACE.TXT /TRACEFLAG=SERVICE
v Tivoli Storage Manager API trace Enable tracing with the DP/SQL DSM.OPT file and the TRACEFILE and TRACEFLAGS keywords. The following is an example of the entry in the DP/SQL DSM.OPT file:
141
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
TRACEFILE APITRACE.TXT TRACEFLAG SERVICE
v DSMAGENT trace Enable tracing with the DSMAGENT DSM.OPT file and the TRACEFILE and TRACEFLAGS keywords. The following is an example of the entry in the DSMAGENT DSM.OPT file:
TRACEFILE AGTTRACE.TXT TRACEFLAG ALL_VSS
The trace flag, in this instance, is ALL_VSS (you might need different traceflags, depending on the circumstance).
Gathering information before calling IBM (SQL with VSS)

The Data Protection client is dependent upon the operating system and the SQL application. Collecting all the necessary information about the environment can significantly assist in determining the problem. Gather as much of the following information as possible before contacting IBM Support: v The exact level of the Windows operating system, including all service packs and hotfixes that were applied. v The exact level of the SQL Server, including all service packs and hotfixes that were applied. v The exact level of Data Protection for SQL with Virtual Shadow Copy Service (VSS) Backup/Restore support. v The exact level of the Tivoli Storage Manager API. v The exact level of the Tivoli Storage Manager server. v The exact level of the Tivoli Storage Manager backup-archive client. v The exact level of the Tivoli Storage Manager storage agent (if LAN-free environment). v The Tivoli Storage Manager server platform and operating system level. v The output from the Tivoli Storage Manager server QUERY SYSTEM command. v The output from the Data Protection for SQL TDPSQLC QUERY SQL command. v The device type (and connectivity path) of the SQL databases and logs. v The specific hardware that is being used (for example: HBA, driver levels, microcode levels, SAN Volume Controller levels, DS6000, DS8000 hardware details). v Permissions and the name of the user ID being used to run backup and restore operations. v The name and version of antivirus software. v The IBM VSS hardware provider level. v The IBM CIM agent level for DS6000/DS8000 or the SAN Volume Controller. v A list of third-party SQL applications running on the system. v A list of other applications running on the system. v A list of the steps needed to recreate the problem (if the problem can be recreated). v If the problem can not be recreated, list the steps that caused the problem. v Is Data Protection for SQL running in a Microsoft Cluster Server (MSCS) environment? v Is the problem occurring on other SQL servers?
142
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Some operating system information that is listed above can be gathered by using the MPS Reports Tool. This is a Microsoft Reporting tool which lets you gather levels of operating system, software installed, driver versions, and other data. You can download the MPS Reports Tool at Microsoft.com.
Gathering files before calling IBM (SQL with VSS)

Several log files and other data can be collected for Data Protection for SQL server diagnosis. Gather as many of the following files as possible before contacting IBM Support: v The Data Protection for SQL configuration file. The default configuration file is tdpsql.cfg. v The Data Protection for SQL Tivoli Storage Manager application programming interface (API) options file. The default options file is dsm.opt. v The Tivoli Storage Manager registry hive export. v The SQL Server registry hive export. v The Tivoli Storage Manager Server activity log. The Data Protection client logs information to the server activity log. A Tivoli Storage Manager administrator can view this log for you if you do not have a Tivoli Storage Manager administrator user ID and password. v If the Data Protection client is configured for LAN-free data movement, also collect the options file for the Tivoli Storage Manager storage agent. The default name for this file is dsmsta.opt. v Any screen shots or command-line output of failures or problems. Log files can indicate the date and time of a backup, the data that is backed up, and any error messages or completion codes that could help to determine your problem. The following are the Tivoli Storage Manager log files to gather: v The Data Protection for SQL log file. The default location of this file is C:\Program Files\Tivoli\TSM\TDPSql\tdpsql.log v The Tivoli Storage Manager API Error log file. The default location of this file is C:\Program Files\Tivoli\TSM\TDPSql\dsierror.log v The DSMAGENT error log file. The default location of this file is C:\Program Files\Tivoli\TSM\baclient\dsmerror.log The Windows event log receives information from the SQL Server and many different components involved during a Virtual Shadow Copy Service (VSS) operation. Microsoft provides a tool in the system32 directory on Windows XP and Windows 2003, called eventquery.vbs (it is a VBS script, not a binary executable) to retrieve event log data. You might use the tool as an alternative method of getting Windows event log data. To invoke the utility, issue the following cscript command:
cscript c:\windows\system32\eventquery.vbs parameters> outfilename
The utility, by default, produces a tabular listing of all event log records in three sections (one section per event log type). To get all log data in default format, issue the following command:
cscript c:\windows\system32\eventquery.vbs >eq.out
Specify the type of event log you require by using one of the following /L parameters: /L Application /L Security
143
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
/L System The following example generates output only for the application and system event logs:
cscript c:\windows\system32\eventquery.vbs /L Application >eq_app.out cscript c:\windows\system32\eventquery.vbs /L System >eq_sys.out
You can utilize the /V parameter to receive detailed (verbose) output:

cscript c:\windows\system32\eventquery.vbs /V >eq.out cscript c:\windows\system32\eventquery.vbs /L System /V >eq_sys.out
You can utilize the /FO parameter to specify tabular, list, or comma-separated (CSV) output. The following are the different methods of specifying the output: /FO TABLE /FO LIST /FO CSV The default format is TABLE. The LIST output puts each column of the record on a separate line, similar to how the Tivoli Storage Manager administrators command-line interface (CLI) displays output when it is too wide for tabular display. The CSV output can be loaded into a spreadsheet or database tool for easier viewing. The following example generates a detailed CSV file of the application log:
cscript c:\windows\system32\eventquery.vbs /L Application /FO CSV /V >eq_app.out
You can get additional help information for the tool by using the following example:
cscript c:\windows\system32\eventquery.vbs /?
The following VSS provider log files can also be helpful, if applicable: v System Provider - (Windows Event Log) v IBM System Storage SAN Volume Controller, DS6000, DS8000 - D:\Program Files\IBM\Hardware Provider for VSS\IBMVss.log v NetApp - D:Program Files\SnapDrive\*.log
Troubleshooting Data Protection for SQL VSS and SAN Volume Controller
The troubleshooting tips included here are designed to help you accelerate your problem determination task. Note: If you plan to perform Data Protection for SQL Virtual Shadow Copy Service (VSS) Instant Restore operations using SAN Volume Controller or DS6000/DS8000 disk subsystems, you must use non-SSL communications with the CIMOM. By default, the CIMOM is configured to use SSL communications. The Tivoli Storage Manager only supports non-SSL communications with the CIMOM. See the Data Protection for SQL Users Guide for information on enabling non-SSL communications. The following are some areas which you can troubleshoot when you are having VSS and SAN Volume Controller or DS6000/DS8000 problems: v CIMOM (Common Information Model Object Manager) Connectivity issues To verify connectivity to the CIMOM, perform the following steps: 1. Refer to your SAN Volume Controller documentation.
144
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
2. Run the IBMVCFG LIST command. The default location is D:\Program Files\IBM\Hardware Provider for VSS-VDS. 3. Issue the IBMVCFG SHOWCFG command to view the provider configuration information. 4. Check that the CIMOM is properly configured. Run verifyconfig.bat -u username -p password on the SAN Volume Controller Master Console. 5. Check the username and password. If there is a problem with the truststore, follow the procedure in the documentation to generate a new truststore. 6. Set up the CIMOM properties file in non-SSL mode if you are using the SAN Volume Controller or DS6000/DS8000 and you plan to use Instant Restore. v CIMOM operational issues If your backup or restore fails, check the IBMVSS.log file. If the failure is due to a CIMOM failure, the log displays output similar to the following:
Wed Jan 11 17:34:34.793 - Calling AttachReplicas Wed Jan 11 17:34:35.702 - AttachReplicas: 909ms Wed Jan 11 17:34:35.702 - returnValue: 34561 Wed Jan 11 17:34:35.718 - AttachReplicas returned: 34561 java.util.MissingResourceException: Cant find resource for bundle java.util.PropertyResourceBundle, key 1793 at java.util.ResourceBundle.getObject(ResourceBundle.java:329) at java.util.ResourceBundle.getString(ResourceBundle.java:289) at com.ibm.cim.CIMException.<init>(CIMException.java:472) at ESSService.executeFlashCopy(ESSService.java:3168) Wed Jan 11 17:34:35.779 - IBMVSS: AbortSnapshots
A return value of 0 means that it was successful. To determine why it failed, look at the log files generated by the command line interface (CLI) or graphical user interface (GUI), depending on how you run your operation. These may provide more information on the failure. v Host configuration issues If the failure seems to be for a different reason than a CIMOM failure, verify your configuration. Run the latest support levels of the software for the SAN Volume Controller or DS6000/DS8000. Check the IBM Storage web site for details. v Collecting logs in this environment If you are unable to resolve these problems, provide the following information to IBM Support: Information Listed in Collection TSM diagnostic information section HBA type, firmware and driver levels SDD version SAN Volume Controller microcode version (if applicable) DS6000/DS8000 microcode version (if applicable) SAN Volume Controller Master Console version (if applicable) For DS6000/DS8000, the CIM Agent version (if applicable) IBMVSS.log Application Event Log System Event Log If the problem appears related to CIMOM, you also need the CIMOM logs. Run CollectLogs.bat and send the file that is created (CollectedLogs.zip) to IBM Support. The default location for the SAN Volume Controller is C:\Program Files\IBM\svcconsole\support, and the default location for the DS6000/DS8000 is C:\Program Files\IBM\cimagent.
145
146
Chapter 6. Creating a DB2UDB transaction after backup

Important: You MUST run at least a minor transaction after taking a backup and before performing a restore. Otherwise, your roll-forward recovery will fail with the following error:
SQL4970N Roll recovery on database "dbname" cannot reach the specified stop point (end of log or point-in-time) because of missing log file(s) on node(s) "0."
This problem was reported to DB2 but DB2 recommended that we ALWAYS run a minor transaction on the database after backup. DB2 did not consider the problem serious enough to fix because most production databases will have plenty of transactions before a possible restore. Perform the following steps to run a DB2 transaction: 1. Issue the following command to connect to the database:
db2 connect to <database name>
2. Issue the following command to list the tables:

db2 list tables
3. Issue the following command to pick a table and find its columns:
db2 describe table <tablename>
4. Issue the following command to view the contents of the table:

db2 "select * from <table name>"
5. Issue the following command to update one or more columns:

db2 "update <table name> set <column name> = <value>"
where <row id> = <row id value>
147
148
Chapter 7. Trace
You can determine the problem you are having with the Tivoli Storage Manager through trace.
Enabling Administration Center trace

Tracing is available for the Tivoli Storage Manager Administration Center. Perform the following steps to enable tracing for the Administration Center: 1. Start the Administration Center support utility on page 99. This utility is located in the Integrated Solutions Console directory under Tivoli\dsm\bin. Start the utility by typing supportUtil.bat on Windows or supportUtil.sh on Unix. 2. Choose which trace classes to enable or simply turn all tracing on. See Trace classes for the Administration Center for more information about the different trace classes. Follow the on-screen instructions to enable tracing.
C:\Program Files\IBM\ISC\Tivoli\dsm\bin>supportUtil.bat Administration Center Support Utility - Main Menu ================================================== 1. Turn all tracing on 2. Turn all tracing off 3. Turn a single trace class on 4. Update the maximum memory size Administration Center can use 5. Update the Administration Center session timeout setting 6. Collect trace files, logs and system information to send to support 7. View the log file for this utility 9. Exit Enter Selection: 3 Administration Center Support Utility - Turn Single Class On ============================================================= 1. ALL - all trace classes 2. SNAPIN - this includes CONTROL, VIEWS, ACTIONS, UTIL, and PS 3. CONTROL - traces control objects 4. VIEWS - traces view objects 5. ACTIONS - traces action objects 6. UTIL - traces utility objects 7. PS - traces Presentation Services extended objects 8. API - traces communications with the IBM Tivoli Storage Manager server 9. HEALTH - traces the health monitor 10. TAGLIB - traces JSP tag library extensions 11. XMLVAL - traces the reading and writing of XML files Enter Selection: 2
3. Perform the operation that is causing the problem. This causes a trace of the problem to be written to the trace file. The trace file is located in the Integrated Solutions Console installation directory under PortalServer\log\trace.log. 4. Disable tracing. Leaving tracing enabled may cause performance problems. 5. Run Collect trace files, logs and system information to send to support. This packages up the trace information collected and log files into a single file called AdminCenterSupport.zip. This file can then be sent to support.
Trace classes for the Administration Center

The Administration Center provides individual and aggregate trace classes. Aggregate classes are a shortcut for enabling many related trace classes by simply specifying the aggregate trace class name.
149
The trace classes documented here are those that are most typically requested or used for diagnosing problems. Those trace classes that are included does not entail all possible trace classes that are available. The trace class name should be used when activating tracing using the Administration Center Support Utility.
ACTIONS
Description
Displays processing information for the action classes.
Recommendation
Useful for problems related to action objects. Use this trace class to debug problems that occur as an action, before displaying a page.
ALL
Description
Displays all processing information for the Administration Center. An aggregate that includes all of the trace classes listed in the bulleted list above.
Recommendation
Useful in most cases. Generally recommended when the nature of the problem is unknown. If the ALL trace flag is used, it is not necessary to specify any other trace flags because it already includes all of the trace classes.
API
Description
Displays information in the processing of the API interface. The API interface is used for communications between the Administration Center and the Tivoli Storage Manager server.
Recommendation
Use this trace class for problems related to command implementation on the Tivoli Storage Manager server.
CONTROL
Description
Displays processing information for the connection manager, the portlet factory and other key control components.
Recommendation
Useful for problems related to control objects. Use this trace class to debug problems that occur when you perform actions that use internal control objects, such as the connection manager.
150
HEALTH
Description
Displays all information about the health monitor.
Recommendation
Use this trace class for problems related to the Health Monitor.
PS
Description
Displays general information in the processing of Presentation Services object extensions. This traces all listener, validator, and JavaServer Page (JSP) bean classes.
Recommendation
Use this trace class for problems related to graphical user interface objects, such as tables and buttons.
SNAPIN
Description
Displays processing information for the snapin portlet classes. An aggregate that includes most of the Administration Center trace classes documented here.
Recommendation
Useful in many cases. Generally recommended for diagnosing problems with the Administration Center when the Health Monitor and XML files are working fine.
Other
Aggregate of all trace classes except HEALTH, XMLVAL, and TAGLIB.
TAGLIB
Description
Displays all information related to the JSP tag library extensions.
Recommendation
Use this trace class for problems related to JSP pages.
UTIL
Description
Displays processing information for the utility classes.
Recommendation
Useful for problems related to utility objects. Use this trace class to debug problems that occur in various internal utility routines.
Chapter 7. Trace
151
VIEWS
Description
Displays processing information for the view classes.
Recommendation
Useful for problems related to view objects. Use this trace class to debug problems that occur while you are displaying a page.
XMLVAL
Description
Displays all information in the processing of XML files.
Recommendation
Use this trace class for problems related to the file controlling the servers that the Administration Center knows about (tsmservers.xml) and the command file cache.
Enabling trace for the server or storage agent

Trace commands can be issued from the server console, storage agent console, administrative client connected to either the server or storage agent, server options file (dsmserv.opt), or the storage agent options file (dsmsta.opt). Trace commands apply to the server or storage agent to which the command was submitted. Trace commands in the options files are used to trace during startup and initialization of the applications or to provide a default set of trace classes. Starting with Tivoli Storage Manager Version 5.3, there are two trace classes that are always enabled by default, whether they appear on the options file or not. These are ADDMSG and SYSTIME. Perform the following steps to enable trace classes for the server or storage agent: 1. Determine the trace classes to enable. See Defining trace classes for server or storage agent on page 154 for possible trace classes. To have trace messages issued for a given trace class, that trace class needs to be enabled either prior to beginning the trace or after the tracing has begun. 2. Issue the following command to enable one or more trace classes: TRACE ENABLE <traceClassName> Note that <traceClassName> may be a space-delimited list of trace classes. For example, this could be entered as TRACE ENABLE TM SESSION. The TRACE ENABLE command is cumulative, such that extra trace classes can be enabled by issuing TRACE ENABLE numerous times. For example, if you wanted to add the PVR trace class in addition to those that were already enabled, issue: TRACE ENABLE PVR. To stop having trace messages issued for a given trace class, that trace class needs to be disabled either prior to beginning the trace or after the tracing has begun. 3. Issue the following command to disable one or more trace classes: TRACE DISABLE <traceClassName> Note: <trace class name> may be a space delimited list of trace classes. For example, this could be entered as TRACE DISABLE TM SESSION.
152
Additional trace classes can also be disabled by issuing TRACE DISABLE. For example, if you wanted to remove the PVR trace class in addition to those that were already disabled, issue: TRACE DISABLE PVR. By issuing TRACE DISABLE without specifying any trace classes, all currently enabled trace classes will be disabled. 4. Tracing can occur to the console or to a file. Perform the following to begin tracing: v For tracing to the console, issue: TRACE BEGIN v For tracing to a file with no size limitation, issue: TRACE BEGIN <fileName> v For tracing to a file with a size limitation, issue: TRACE BEGIN <fileName> MAXSIZE=<maximum size in megabytes> Note: The fileName can be a fully-qualified path such as /opt/tmp or c:\temp. If a full path is not given, the trace file will be located in the same directory as the running executable. 5. Perform the operation that is causing the problem. 6. Issue the TRACE END command to stop trace messages from being issued. If tracing was being done to a file, ending the trace will write any remaining trace messages to the file and then close the file. It is best to trace to a file. Typically, the tracing for the server or storage agent will generate a large amount of output. It is possible to enable tracing and begin it using the server or storage agent options file. The commands and syntax discussed above are the exact same for the server or storage agent options file, and they are generally used to trace startup and initialization of the server. For example, if the following lines were added to the servers option file, tracing would be started for the DB, TM, and LOG trace classes, and the trace messages written to the file MYTRACE.OUT.
TRACE ENABLE DB TM LOGTRACE BEGIN MYTRACE.OUT
If the tracing is no longer needed after startup, the tracing is ended by issuing the TRACE END command from the console or from an administrative client.
Enabling trace for specific messages for the server or storage agent
A stack trace reveals information about an application that can help IBM Support help you to diagnose your problems faster. Note: Stack trace can be extremely robust and may flood the activity log, depending on the frequency of the failure. You may not be able to view the activity log, therefore you may want to disable stack trace. IBM Support may find it helpful to enable stack trace on specific messages issued by the server or storage agent. The following are types of messages on which a stack trace can be enabled: v Server console v Storage agent console v Administrative client connected to either the server or storage agent
Chapter 7. Trace
153
To get a stack trace when a specific message is issued by the server or storage agent, enable the message for stack trace. Issue the MSGSTACKTRACE ENABLE <messageNumber> command to enable one or more messages for stack trace. Note: <messageNumber> may be a space-delimited list of message numbers. This could be entered as MS ENABLE 2017. The MSGSTACKTRACE ENABLE command is cumulative, such that extra messages are enabled by issuing the MSGSTACKTRACE ENABLE command additional times. If you want to add message 985, in addition to those that are already enabled, issue MS ENABLE 985. Notice that only the number part of the message is allowed in the MSGSTACKTRACE command. To stop getting stack trace for messages issued by the server or storage agent, the stack trace for those messages needs to be disabled. Issue the MSGSTACKTRACE DISABLE <messageNumber> command to disable one or more messages. Note that <messageNumber> may be a space-delimited list of message numbers. For example, this could be entered as MSGSTACKTRACE DISABLE 2017 985. Additional messages can also be disabled by issuing MS DISABLE. For example, if you wanted to remove message number 7837 in addition to those that are already disabled, issue MSGSTACKTRACE DISABLE 7837. The following messages are enabled for stack trace by default:
435 784 884 5021 9602 437 785 1032 5093 9604 486 786 1078 5099 9605 661 790 1092 5100 9606 685 793 1117 5267 9607 727 794 1156 6753 9608 728 860 1227 7823 9999 780 881 5010 7837 781 882 5015 9600 782 883 5019 9601
Defining trace classes for server or storage agent

The server and storage agent provide aggregate trace classes. These are a shortcut for enabling many related trace classes by simply specifying the aggregate trace class name for the TRACE ENABLE command. There are references to trace classes in the table below that are enabled as part of an aggregate trace class, but they are not explicitly documented. In this case, these trace classes are not typically used for problem diagnosis. The trace classes listed are those that are most typically requested or used for diagnosing problems. This list does not discuss all possible trace classes that are available. To find information about a trace class, click the trace class in the table below. The trace class name should be used with the TRACE ENABLE and TRACE DISABLE commands.
154
Table 5. List of trace classes for server or storage agent v ACCNT v ADDMSG on page 157 v ADMCMD on page 158 v ADMNODE v ADMPROC v ADMRES v AS on page 158 v AFCOPY v AFCREATE on page 158 v AFLOCK v AFMOVE on page 158 v AFTXN v ARCH v ARCHD v ARCHUTIL v AS on page 158 v ASALLOC on page 159 v ASDEALLOC on page 159 v ASMOUNT on page 159 v ASRTRV on page 159 v ASTXN on page 160 v ASVOL on page 160 v ASSD on page 160 v AUDIT v BACK v BACKD v BF on page 160 v HT v IC v ICBACK v ICBITS v ICDISK v ICHDRPG v ICLOCK v ICLOG v ICREST v ICSTREAM v ICTRIGGER v ICVOLHST on page 166 v ID v IMDEL v IMEXP v IMFS on page 167 v IMLOCK v IMSORT v INSTR v INTERACTION v LANFREE on page 167 v LANG v LM v LNFDATA v LNFENTRY v LNFMEM v LNFVERB
Chapter 7. Trace
155
Table 5. List of trace classes for server or storage agent (continued) v BFAGGR on page 161 v BFCREATE on page 161 v BFLOCK v BFREMOTE on page 161 v BFRTRV on page 161 v BFSALVAGE v BFTRG v BITVECTOR v BKSET/OBJSET on page 162 v BLKDISK on page 162 v BRNODE on page 162 v CA v CAACOMAX v CAACOMIN v CAENG v CAMEM v CASCSI v CC v CCLOCK v CENTERA v COLLOCATE on page 162 v COLLOCATEDETAIL v CRC on page 163 v CRCDATA on page 163 v CRCPROTO on page 163 v CRCVAL on page 163 v CRYPTO on page 164 v DB v DBALLOC v DBBACKUP v DBBUF v DBBUFPF v DBBUFWR v DBCKPT v DBDISPCOLS v DBDUMP v DBLOG v DBNETDB on page 164 v DBNETENCODE v DBSERVER v LOCK v LOG v LOGCOMP v LOGPARTW v LOGREC v LOGRESIZE v LOGSEG v LVM v MMS on page 167 v MPCOUNT v NA on page 167 v NACONFIG v NADISK v NAHBAAPI v NALOCK v NAMOVER v NAPATH v NQR v NQRDETAIL v NQRSEARCH v BKSET/OBJSET on page 162 v ORM v ORMCMD v ORMDELETE v ORMUPDATE v PERFSTAT v PM v PMCACHE v PMCG v PMLOCK v PMMC v PROXYNODE on page 168 v PVR on page 168 v REDIRECT v RETPROT on page 168 v SCHED on page 169 v SCHEDWAIT v SESSION on page 169 v SESSIOND v SESSREMOTE on page 169
156
Table 5. List of trace classes for server or storage agent (continued) v DBTXN v DEADLOCK v DELTA on page 164 v DF on page 164 v DFCOPY v DFCREATE on page 165 v DFLOCK v DFMOVE on page 165 v DFRTRV on page 165 v DFTXN v DRM v DRMAPPEND v DRMLICCOMP v DRMMACH v DRMMACHCHAR v DRMPR v DRMPRDBBACK v DRMPRDEVCLASS v DRMPRLOGDBVOL v DRMPRMDRM v DRMPRREQPOOL v DRMPRST v DRMPRSTGPOOL v DRMRECOVINST v DRMRECOVMEDIA v DRMSETGA v DS on page 165 v DSALLOC on page 166 v DSDEALLOC on page 166 v DSRTRV on page 166 v DSVOL on page 166 v EXTERNAL v FORMAT v DELTA on page 164 v HSMPFR v SETOBJRANGE v SFACO v SFPERF v SHRED on page 170 v SPI/SPID on page 170 v SS v SSBUFQ v SSCLASS v SSFRAME v SSLDATA on page 170 v SSLINFO on page 171 v SSLOCK v SSPOOL v SSSESS v SSSWINIT v SSSWTHR v SSTRANS v SSVOL v SW v STATXN v TAPE on page 171 v TBPREFETCH v TCP on page 171 v TCPDATA on page 171 v TCPINFO on page 172 v TEC on page 172 v TOC on page 172 v TOCBUILD on page 172 v TOCLOAD on page 173 v TOCREAD on page 173 v TOCUTIL on page 173 v TXN v UNDERSIZE v UNICODE on page 173 v UNICODEDETAIL v XIDETAIL
ADDMSG
Description
Issues console messages (ANR, ANE, and so on) to the trace file.
Recommendation
This trace class is valuable for correlating server messages to trace messages and for preserving the timing for when each was issued.
Chapter 7. Trace
157
Other
Available in V5.2 or V5.1.7.0 and later.
ADMCMD
Description
Traces related to command processing.
Recommendation
Use this trace class to debug the command interpreter, including the PARALLEL and SERIAL command handling.
AF
Description
Displays information about user data stored on sequential media devices.
Recommendation
Use this trace class to diagnose problems reading or writing user files to sequential media volumes.
Other
This is an aggregate trace class that enables AFCREATE, AFMOVE, AFLOCK, AFTXN, and AFCOPY. Typically recommended to issue TRACE DISABLE AFLOCK, unless the locking information is explicitly requested or needed.
AFCREATE
Description
Displays information about storing user data on sequential media volumes.
Recommendation
Use this trace class to diagnose writing user data on sequential media volumes.
AFMOVE
Description
Displays operations that move user data using sequential media volumes. Move operations are performed by MIGRATION, RECLAMATION, MOVE DATA, and MOVE NODEDATA server processes.
Recommendation
Use this trace class to diagnose problems with the data movement server processes.
AS
Description
Displays information volume selection and assignment, coordination of drives (mount points), and management of data placement on volumes.
158
Recommendation
Use this trace class to diagnose many different problems relating to volumes, mount points, or data read and write operations.
Other
This is an aggregate trace class that enables ASALLOC, ASRTRV, ASDEALLOC, ASMOUNT, ASVOL, ASTXN, and ASSD. Typically recommended to issue TRACE DISABLE ASTXN, unless the locking information is explicitly requested or needed.
ASALLOC
Description
Displays information about reserving and allocating space on sequential media volumes for the purpose of storing data. This may be for storing data on behalf of a client session or for server data movement operations such as MIGRATION, RECLAMATION, MOVE DATA, or MOVE NODEDATA
Recommendation
Use this trace class to diagnose problems where the server or storage agent report no space available but there should be space available in the storage hierarchy.
ASDEALLOC
Description
Displays information about releasing and de-allocating space on sequential media volumes for the purpose of storing data. Typical de-allocation operations on the server are EXPIRATION, MIGRATION, RECLAMATION, MOVE DATA, MOVE NODEDATA, AUDIT VOLUME, DELETE VOLUME, and DELETE FILESPACE.
Recommendation
Use this trace class to diagnose during the deletion of data.
ASMOUNT
Description
Displays information about drive (mount point) selection and assignment for sequential media devices.
Recommendation
Use this trace class to diagnose situations where sessions or processes are waiting for mount points or cases where an operation fails because no mount point is available. Also useful in cases where a mount point is preempted.
ASRTRV
Description
Displays information about reading data from sequential media volumes.
Chapter 7. Trace
159
Recommendation
Use this trace class to diagnose problems reading data such as RESTORE or RETRIEVE client by the client, or MIGRATION, RECLAMATION, STORAGE POOL BACKUP, AUDIT VOLUME, GENERATE BACKUPSET, EXPORT, MOVE DATA, or MOVE NODEDATA by the server.
ASTXN
Description
Displays information about transactions used to make database updates relating to information for sequential media volumes, storage pools, device classes, and other attributes.
Recommendation
Use this trace class to diagnose hangs, database operations, failures reported for sequential media operations, or general data storage problems.
ASVOL
Description
Displays information about volume selection and assignment for sequential media volumes.
Recommendation
Use this trace class to diagnose situations where sessions or processes are waiting for volumes, or cases where an operation fails because no volume is available. Also useful in cases where volume access is preempted.
ASSD
Description
Displays information about sequential stream data operations. These are operations that use sequential media device classes, volumes, or mount points but do not store data in the storage hierarchy. Server processes that perform sequential stream data operations are BACKUP DB, EXPORT/IMPORT, and GENERATE BACKUPSET.
Recommendation
Use this trace class to diagnose server processes that perform sequential stream data operations.
BF
Description
Information about user data (files) stored in the storage hierarchy.
Recommendation
Use this trace class to diagnose general data read or write problems for client operations and server processes.
160
Other
This is an aggregate trace class that enables BFCREATE, BFRTRV, BFSALVAGE, BFLOCK, BFAGGR, BFREMOTE, and BFTRG.
BFAGGR
Description
Displays information about server aggregation of user data. The server will aggregate many smaller user files into a larger file in the storage hierarchy to optimize performance for data movement operations such as MIGRATION, MOVE DATA, and MOVE NODEDATA.
Recommendation
Use this trace class to diagnose general data read or write problems for client operations and server processes, or both.
BFCREATE
Description
Displays information about client operations that store data in the storage hierarchy. Typically, these are BACKUP, ARCHIVE, or SPACE MANAGE operations by the client.
Recommendation
Use this trace class to diagnose failures or other problems while a client is storing data.
BFREMOTE
Description
Traces the first stage of NDMP backup and restore processes.
Recommendation
This trace class is used to identify NDMP-related backup or restore operations. These trace classes are specific to the functions which implement the NDMP protocol. The SPID trace class provides more detailed tracing, including tracing all NDMP file history records sent by the NDMP file server.
Other
The BFREMOTE trace class is available in Tivoli Storage Manager Version 5.2 and later.
BFRTRV
Description
Displays information about client operations that read data from the storage hierarchy.
Chapter 7. Trace
161
Recommendation
Use this trace class to diagnose failures or other problems while a client is reading data.
BKSET/OBJSET
Description
Trace class for backupset functions.
Recommendation
Use this trace class to debug problems in the GENERATE BACKUPSET command or during a client restore from a backupset.
Other
The BKSET and OBJSET trace classes are synonymous.
BLKDISK
Description
Trace class for viewing disk I/O activity to storage pool, database, and log volumes.
Recommendation
Use this trace class to view I/O activity to disk to diagnose performance and disk I/O errors.
BRNODE
Description
Trace class for the BACKUP and RESTORE NODE commands, used during NDMP operations.
Recommendation
Use this trace class to debug problems in the BACKUP and RESTORE NODE commands.
Other
Only available in Tivoli Storage Manager 4.2.1.0 and later.
COLLOCATE
Description
Displays information about collocation processing on storage pools.
Recommendation
Use this trace class to diagnose problems with collocation processing.
162
Other
COLLOCATEDETAIL trace class can also be used to get more detailed information about the collocation processing, such as files being processed for a collocation group. This can cause a large amount of output trace statements.
CRC
Description
Displays information about generating and managing CRCs on the server or storage agent.
Recommendation
Use this trace class to diagnose data corruption issues where CRC processing did not report data corruption.
Other
This is an aggregate trace class that enables CRCDATA, CRCPROTO, and CRCVAL.
CRCDATA
Description
Displays information about generating and managing CRCs for data stored in storage pools with CRCDATA=YES set.
Recommendation
CRCPROTO
Description
Displays information about generating and managing CRCs for data exchanged between the client and either the server or storage agent where this node is configured with VALIDATEPROTOCOL=ALL or VALIDATEPROTOCOL=DATAOnly on the server.
Recommendation
CRCVAL
Description
Displays information about generating and comparing CRC values.
Recommendation
Informational for displaying CRC values during processing.
Chapter 7. Trace
163
CRYPTO
Description
Displays information about AES encryption operations and some general encryption settings.
Recommendation
Use this trace class to isolate and identify encryption related problems.
DBNETDB
Description
Displays information about LAN-free operations and the negotiation and management of information between the server and storage agent.
Recommendation
Use this trace class to diagnose problems with LAN-free data movement.
Other
Typically applies to server and storage agent prior to Tivoli Storage Manager Version 5.2. After Tivoli Storage Manager Version 5.2, this still displays information but was superseded by the LANFREE trace class. Typically best to enable this trace class on both the server and storage agent.
DELTA
Description
Trace class for logical group functions.
Recommendation
Use this trace class to debug problems with logical groups, whether delta-base groups (subfile backup) or peer groups (Windows SYSTEM OBJECT or image backups). Group processing is relevant during just about any operation that references backup objects, including client backup and restore, expiration, deletion (DELETE FILESPACE, DELETE VOLUME), export/import, backupset generation and restore, no-query restore, db audit, etc.
Other
The DELTA and GROUP trace classes are synonymous.
DF
Description
Displays information about user data stored on DISK volumes.
Recommendation
Use this trace class to diagnose problems reading or writing user files to DISK volumes.
164
Other
This is an aggregate trace class that enables DFCREATE, DFRTRV, DFMOVE, DFLOCK, DFTXN, and DFCOPY. It is recommended to issue the TRACE DISABLE DFLOCK command unless the locking information is explicitly requested or needed.
DFCREATE
Description
Displays information about storing user data on DISK volumes.
Recommendation
Use this trace class to diagnose writing user data on DISK volumes.
DFMOVE
Description
Displays operations that move user data by using DISK volumes. Move operations are performed by the MIGRATION, MOVE DATA, and MOVE NODEDATA server processes.
Recommendation
Use this trace class to diagnose problems with the data movement server processes.
DFRTRV
Description
Displays information about reading user data from DISK volumes.
Recommendation
Use this trace class to diagnose reading user data from DISK volumes.
DS
Description
Displays information about volume selection, space reservation, assignment, and management of data placement on DISK volumes.
Recommendation
Use this trace class to diagnose many different problems relating to DISK volume data read and write operations.
Other
This is an aggregate trace class that enables DSALLOC DSRTRV DSDEALLOC DSVOL. It is recommended to issue TRACE DISABLE DSTXN unless the locking information is explicitly requested or needed.
Chapter 7. Trace
165
DSALLOC
Description
Displays information about reserving and allocating space on DISK volumes for the purpose of storing data. This may be storing data on behalf of a client session or for server data movement operations such as MIGRATION, MOVE DATA, or MOVE NODEDATA.
Recommendation
Use this trace class to diagnose problems where the server or storage agent report that no space is available, but there should be space available in the storage hierarchy.
DSDEALLOC
Description
Displays information about releasing and de-allocating space on DISK volumes. Typical de-allocation operations on the server are EXPIRATION, MIGRATION, MOVE DATA, MOVE NODEDATA, AUDIT VOLUME, DELETE VOLUME, and DELETE FILESPACE.
Recommendation
Use this trace class to diagnose during the deletion of data.
DSRTRV
Description
Displays information about reading data from DISK volumes.
Recommendation
Use this trace class to diagnose problems reading data such as RESTORE or RETRIEVE client by the client, or MIGRATION, STORAGE POOL BACKUP, AUDIT VOLUME, GENERATE BACKUPSET, EXPORT, MOVE DATA, or MOVE NODEDATA by the server.
DSVOL
Description
Displays information about volume selection and assignment for DISK volumes.
Recommendation
Use this trace class to diagnose situations where sessions or processes are waiting for volumes or cases where an operation fails because no volume is available.
ICVOLHST
Description
Trace class for volume history functions.
166
Recommendation
Use this trace class to debug problems with creating volume history entries (for example: during EXPORT, BACKUP DB, or GENERATE BACKUPSET) or deleting volume history entries (for example: during DELETE VOLHISTORY).
IMFS
Description
Trace class for filespace functions.
Recommendation
Use this trace class to debug problems related to inventory filespaces (e.g. during DELETE FILESPACE).
LANFREE
Description
Displays general information about LAN-free operations on either the server or storage agent. Also displays error information for LAN-free-related operations.
Recommendation
Any LAN-free failure.
Other
This trace class is available only in Tivoli Storage Manager Version 5.2 or later and is best if traced on both the server and storage agent concurrently. This is an aggregate trace class that enables LNFVERB, LNFMEM, LNFENTRY, and LNFDATA.
MMS
Description
Displays information about tape libraries and the server or storage agent use of these.
Recommendation
Used to diagnose problems with tape libraries, library volume inventories, or other general library issues.
Other
This trace class is not Available for z/OS . This is an aggregate trace class that enables MMSBASE, MMSTXN, MMSLIB, MMSDRIVE, MMSOP, MMSMAN, MMSSCSI, MMSFLAG, MMSACSLS, and MMSSHARE. Include NA and PVR trace classes when tracing MMS (suggested).
NA
Description
Displays information about path information for the server or storage agent. This relates to the DEFINE PATH, UPDATE PATH, DELETE PATH, and QUERY PATH
Chapter 7. Trace
167
commands. This trace class is also useful for identifying issues related to operations involving NDMP file servers, for example, DEFINE DATAMOVER, UPDATE DATAMOVER, BACKUP NODE, and RESTORE NODE commands.
Recommendation
Use this trace class to diagnose problems with paths to devices.
Other
This is an aggregate trace class that enables NALOCK, NAPATH, NAMOVER, NADISK, and NACONFIG. It might be best to include MMS and PVR trace classes when tracing NA.
PROXYNODE
Description
Displays information about proxynode sessions and the commands related to proxynode associations (GRANT, REVOKE, QUERY PROXYNODE).
Recommendation
Use this trace class to diagnose problems with proxynode sessions and related commands.
Other
It might be best to include SESSION trace when analyzing proxynode session problems.
PVR
Description
Displays information about sequential media devices and the server or storage agent use of these devices.
Recommendation
Use this trace class to diagnose problems with tape drives, failures reading or writing tape volumes, or other tape-volume-related issues.
Other
This is an aggregate trace class that enables PVRVOL, PVRCLASS, PVRMP, and PVRREMOTE. Additionally, this aggregate enables the following classes for open platforms only: PVR8MMBASE, PVR4MMBASE, PVRQICBASE, PVRDLTBASE, PVRECARTBASE, PVRDTFBASE, PVRTAPIBASE, PVRGTS, PVRFILE, PVRTAPE, PVRODSK, PVRNTP, PVRSHARE, and PVRNAS. It is often useful to include MMS and NA trace classes on open platforms and TAPE on z/OS when tracing PVR.
RETPROT
Description
Trace class for the archive retention protection functions.
168
Recommendation
Use this trace class to debug problems using the RETINIT and RETMIN parameters in the archive copy group. You can also use this trace class for problems caused by using the VB_SignalObject verb (only supported via the client API) to signal an objects event or to hold or release an object, or problems during expiration or deletion of retention protected objects.
Other
This trace class is only available in Tivoli Storage Manager 5.2.2.0 and later.
SCHED
Description
Trace class for the central scheduler functions.
Recommendation
Use this trace class to debug problems related to schedule commands like DEFINE/UPDATE/QUERY SCHEDULE or DEFINE ASSOCIATION. Also use this trace class to debug problems related to the central scheduler background processes, such as the schedule manager and schedule prompter.
Other
This trace class applies to classic and enhanced schedules equally.
SESSION
Description
Displays information about sessions connected to the server, including all verbs sent and received by the server.
Recommendation
This trace class is useful in many cases. It is generally recommended for protocol violations, transaction processing errors, or in cases where the client is hung and not responding.
SESSREMOTE
Description
Traces communication between the Tivoli Storage Manager server and the Tivoli Storage Manager client during NDMP backup and restore operations.
Recommendation
This trace class is used to identify NDMP-related backup or restore operations that are initiated using the Tivoli Storage Manager Web or command line client.
Other
Available in Tivoli Storage Manager Version 5.2 and later.
Chapter 7. Trace
169
SHRED
Description
Displays information related to data-shredding operations on the server.
Recommendation
This trace class is used to diagnose problems with data shredding. Data shredding is only applicable if one or more storage pools on the server has a non-zero value for the SHRED attribute. Activity that is related to data shredding occurs primarily during the EXPIRE INVENTORY, DELETE FILESPACE, DELETE VOLUME, MOVE DATA, MIGRATE, and SHRED DATA commands. Other trace classes that report activity related to data shredding are BFDESTROY, DFDESTROY, DSALLOC, DSDEALLOC, and CRCDATA.
Other
The SHRED command is available in Tivoli Storage Manager, Version 5.4.
SPI/SPID
Description
Traces the Servers NDMP protocol interface.
Recommendation
The SPI and SPID trace classes are used to identify issues related to NDMP backup or restore operations of NAS file servers. These trace classes are specific to the functions that implement the NDMP protocol and communicate with a NAS file server. The SPID trace class provides more detailed tracing, including tracing all NDMP file history records sent by the NAS file server.
Other
Available in Tivoli Storage Manager Version 5.2 and later. | | | | | | | | | | |
SSLDATA
Description
Detailed secure socket layer (SSL) trace used to display byte-level information about data that is sent or received between the backup-archive client and the server.
Recommendation
Use the SSLDATA trace class to debug the session data corruption issues that may be caused by SSL that is running through the SSLTCP or SSLTCPADMIN server options. Because this is a byte-level trace, it can collect a large amount of data.
Other
170
| | | | | | | | | | |
SSLINFO
Description
General secure socket layer (SSL) trace used to display setup and characteristics of SSL sessions between the backup-archive client and the server.
Recommendation
Use the SSLINFO trace class to debug session connection and handshake errors that may be caused by the SSL that is running through the SSLTCP or SSLTCPADMIN server options. This can be used in tandem with the TCPINFO and SESSION trace classes.
Other
TAPE
Description
Displays information about sequential media devices and the server or storage agent use of these devices.
Recommendation
Used to diagnose problems with tape drives, failures reading or writing tape volumes, or other tape-volume-related issues.
Other
This trace class is only available on z/OS. This is an aggregate trace class that enables CART, CARTDATA, CARTINFO, REEL, REELDATA, FLAT, and FLATDATA. It is often useful to include the PVR trace class when tracing TAPE.
TCP
Description
Collects information regarding TCP/IP used between the client and either server or storage agent.
Recommendation
Use this trace class to debug session connection errors or data corruption issues that may be caused by the network.
Other
This is an aggregate trace class. It enables TCPINFO and TCPERROR.
TCPDATA
Description
Detailed TCP/IP trace used to display byte-level information about data that is sent or received.
Chapter 7. Trace
171
Recommendation
Use this trace class to debug session data corruption issues that may be caused by the network.
TCPINFO
Description
General TCP/IP trace used to display setup and characteristics of TCP/IP on the server or storage agent.
Recommendation
Use this trace class to debug session data corruption issues that may be caused by the network.
TEC
Description
Provides information regarding events sent to a TEC server. Corresponds to the tivoli event receiver.
Recommendation
To debug connection issues encountered with TEC event logging.
TOC
Description
General trace class for the Table Of Contents (TOC) component, used during file-level NDMP operations.
Recommendation
Use this trace class to debug problems during file-level NDMP operations, such as an NDMP backup with the TOC=YES parameter, or an NDMP restore with the FILELIST parameter.
Other
This trace class is only available in Tivoli Storage Manager 5.2.0.0 and later. This is an aggregate trace class that enables TOCBUILD, TOCLOAD, TOCREAD, and TOCUTIL.
TOCBUILD
Description
Table Of Contents (TOC) build functions.
Recommendation
Use this trace class to debug problems during an NDMP backup with the TOC=YES parameter.
172
Other
Only available in Tivoli Storage Manager 5.2.0.0 and up.
TOCLOAD
Description
Table Of Contents (TOC) load functions.
Recommendation
Use this trace class to debug problems while displaying files and directories on the client GUI.
Other
TOCREAD
Description
Table Of Contents (TOC) read functions.
Recommendation
Use this trace class to debug problems during a QUERY TOC command or while trying to load a TOC for displaying files and directories on the client GUI.
Other
TOCUTIL
Description
Table Of Contents (TOC) utility functions.
Recommendation
Use this trace class to debug problems related to TOC component initialization or TOC retention.
Other
UNICODE
Description
Displays information about code page conversions and Unicode filespace operations.
Recommendation
Use this trace class to debug problems related to code page conversion problems or unicode filespace problems.
Chapter 7. Trace
173
XI
Description
Displays general information for the IMPORT and EXPORT commands.
Recommendation
Use this trace class to debug problems related to IMPORT and EXPORT commands.
Storage agent diagnostic tips

The storage agent is a program that enables Tivoli Storage Manager to back up and restore client data directly to and from SAN-attached storage.

Check the server activity log for other messages, 30 minutes before and 30 minutes after the time of the error. Storage agents start and manage many sessions to the server. Review the server activity log for messages from the storage agent. To review the activity log messages, issue the QUERY ACTLOG command. See the Tivoli Storage Manager Administrators Reference for more details. If no messages are seen for this storage agent in the server activity log, verify the communication settings: v Issue QUERY SERVER F=D on the server and verify that the high-level address (HLA) and low-level address (LLA) set for the server entry representing this storage agent are correct. v In the device configuration file specified in the dsmsta.opt file, verify that the SERVERNAME as well as the HLA and LLA are set correctly in the DEFINE SERVER line. If messages are seen on the server for this storage agent, do any error messages give an indication about why this storage agent is not working with the server?
Checking HELP for the Tivoli Storage Manager messages

Check HELP for any messages issued by the Tivoli Storage Manager. Tivoli Storage Manager messages provide information beyond just the message itself. The Explanation, System Action, or User Response sections of the message might provide additional information about the problem. Often, this supplemental information provides the necessary steps to resolve the problem.
Resolving an error caused by reading or writing to a device

If the problem is an error involving the reading or writing of data from a device, many systems and devices record information in a system error log. The system error log for AIX is errpt, and for Windows it is the Event Log. If a device or volume that is used by the Tivoli Storage Manager is reporting an error to the system error log, it is likely a device issue. The error messages recorded in the system error log may provide enough information to resolve the problem. Storage agents are particularly vulnerable if path information is changed or not correct. Issue the QUERY PATH F=D command on the server. For each of the
174
storage agents paths, verify that the settings are correct. In particular, verify that the device listed matches the system device name. If the path information is not correct, update the path information with the UPDATE PATH command.
Resolving problems caused by changing storage agent options

Changes to options in the storage agent option file may cause operations to fail, even though they had previously succeeded. Review any changes to the storage agent option file. Try reverting the settings to their original values and retrying the operation. If the storage agent now works correctly, try reintroducing changes to the storage agent option file one-at-a-time and retry storage agent operations until the option file change that caused the failure is identified.
Resolving problems caused by changing server options or settings

Changes to options in the server option file or changes to server settings using the SET commands might affect the storage agent. Review any changes to server option settings. Try reverting the settings to their original values and retrying the operation. If the storage agent now works correctly, try reintroducing changes to the storage agent option file one-at-a-time and retry storage agent operations until the option file change that caused the failure is identified. Review server settings by issuing the QUERY STATUS command. If any settings reported by this query have changed, review the reason for the change and, if possible, revert it to the original value and retry the storage agent operation.
Storage agent LAN-free setup

LAN-free data movement is the direct movement of client data between a client machine and a storage device on a SAN, rather than on a LAN. You might be experiencing problems with the storage agent that are related to your LAN-free setup.
Data is sent directly to the server

The client summary statistics do not report any bytes transferred LAN-free. The client reports the bytes sent LAN-free by issuing the ANE4971I LAN-free Data Bytes: xx KB command. Similarly, the server does not report any instance of ANR0415I Session SESS_NUM proxied by STORAGE_AGENT started for node NODE_NAME for this node and storage agent, indicating that the LAN-free proxy operation was done for this client node. The client will only attempt to send data LAN-free using the storage agent if the primary storage pool destination in the server storage hierarchy is LAN-free. A server storage pool is LAN-free enabled for a given storage agent if one or more paths are defined from that storage agent to a SAN device. To determine if the storage pool destination is configured correctly, perform the following procedures: v Issue the QUERY NODE nodeName command. This reports the policy domain to which this node is assigned. v Issue the QUERY COPYGROUP domainName policySetName mgmtclassName F=D command for the management classes that this node would use from their assigned policy domain. Note that this reports information for backup files. To
Chapter 7. Trace
175
v v v v
v v
query copy-group information for archive files, issue the QUERY COPYGROUP domainName policySetName mgmtclassName TYPE=ARCHIVE F=D command. Issue the QUERY STGPOOL stgpoolName command, where stgpoolName is the destination reported from the previous QUERY COPYGROUP queries. Issue the QUERY DEVCLASS deviceClassName command for the device class used by the destination storage pool. Issue the QUERY LIBRARY libraryName command for the library reported for the device class used by the destination storage pool. Issue the QUERY DRIVE libraryName F=D command for the library specified for the device class used by the destination storage pool. If no drives are defined to this library, review the library and drive configuration for this server and use the DEFINE DRIVE command to define the needed drives. If one or more of the drives report ONLINE=No, evaluate why the drive is offline and, if possible, update it to online by issuing the UPDATE DRIVE libraryName driveName ONLINE=YES command. Issue the QUERY SERVER command to determine the name of the storage agent as defined to this server. Issue the QUERY PATH stgAgentName command, where stgAgentName is the name of the storage agent defined to this server and reported in the QUERY SERVER command. Review this output and verify that one or more paths are defined for drives defined for the device class used by the destination storage pool. If no paths are defined for this storage pool, use DEFINE PATH to define the needed paths. Also, review this output and verify that the path is online. If paths are defined but no paths are online, update the path to online by issuing the UPDATE PATH srcName destName SRCTYPE=SERVER DESTTYPE=DRIVE ONLINE=YES command.
The z/OS LAN-free session failed

If a LAN-free session fails, ensure that the storage pool contains volumes that were formatted for LAN-free operations. To format volumes automatically, set the value of the LFVOLUMEFORMATCOUNT server option from 1 to 255. Volumes are formatted at the time the storage agent makes a request if the storage pool does not contain any formatted volumes. By using the FORMAT LFVOLUME command, you can format volumes before LAN-free sessions begin and have them available in a storage pool.
The storage pool is configured for simultaneous write but will not work LAN-free
The server disqualifies a storage pool as being a LAN-free-enabled storage pool if it was configured for simultaneous write. In this case, data from the client is sent directly to the server which will not be using a LAN-free storage pool. Issue the QUERY STGPOOL stgpoolName F=D command for the destination storage pool for this client. If the storage pool is set for simultaneous write operations, the Copy Storage Pool(s): value references one or more other storage pool names. If this is the case, Tivoli Storage Manager interprets the simultaneous write operation to be a higher priority than the LAN-free data transfer. Because simultaneous write is considered a higher priority operation, this storage pool is not reported as LAN-free enabled and as such, the client will send the data directly to the server. The storage agent does not support simultaneous write operations.
176
SHOW LAN-free and VALIDATE LAN-free - Validating LAN-free definitions on the server
It can be difficult to determine if all the appropriate definitions are defined on the server to allow a given client node to perform LAN-free operations with a storage agent. Similarly, if the configuration changes, it may be difficult to determine if the definitions on the server for a previously configured LAN-free client are still appropriate. Version 5.2.2 of the Tivoli Storage Manager server offers the SHOW LANFREE command. Version 5.3.0 of the Tivoli Storage Manager server offers the equivalent VALIDATE LANFREE command. These commands evaluate the destination storage pools for the domain to which this client node is assigned. The policy destinations are evaluated for BACKUP, ARCHIVE, and SPACEMANAGED operations for this node. For a given destination and operation, SHOW LANFREE and VALIDATE LANFREE report whether or not it is LAN-free capable. If it is not LAN-free capable, these commands give an explanation about why this destination cannot be used. The SHOW LANFREE and VALIDATE LANFREE commands are available from the server. The following shows the syntax for these command:
SHOW LANFREE nodeName stgAgentName VALIDATE LANFREE nodeName stgAgentName
The following are the definitions for the variables in the previous commands: nodeName A client node registered to the server. stgAgentName A storage agent defined to the server. The following is an example of the output from this command:
tsm: SRV1>q lanfree fred sta1 ANR0391I Evaluating node FRED using storage agent STA1 for LAN-free data movement. ANR0387W Node FRED has data path restrictions. Node Storage Operation Name Agent NODE1 STA1 NODE1 STA1 NODE1 STA1 NODE1 STA1 NODE1 STA1 NODE1 STA1 BACKUP BACKUP BACKUP ARCHIVE ARCHIVE ARCHIVE Mgmt Class Name NOLF LF NOLF_SW NOLF LF NOLF_SW Destination LAN-Free Name capable? OUTPOOL LFPOOL PRIMARY OUTPOOL LFPOOL PRIMARY NO YES NO NO YES NO Explanation No available online paths Destination storage pool is configured for simultaneous write No available online paths
Destination storage pool is configured for simultaneous write ANR1706I Ping for server STA1 was able to establish a connection. ANR0392I Node FRED using storage agent STA1 has 2 storage pools capable of LAN-free data movement and 4 storage pools not capable of LAN-free data movement.
Testing LAN-free configuration

The storage agent and client are both able to manage failover directly to the server, depending upon the LAN-free configuration and the type of error encountered.
Chapter 7. Trace
177
Because of this failover capability, it may not be apparent that data is being transferred over the LAN when it was intended to be transferred LAN-free. It is possible to set the LAN-free environment to limit data transfer to only LAN-free. To test a LAN-free configuration, issue the UPDATE NODE nodeName DATAWRITEPATH=LAN-FREE command for the client node whose LAN-free configuration you want to test. Next, try a data store operation such as backup or restore. If the client and storage agent attempt to send the data directly to the server using the LAN, the following error message is received: ANR0416W Session sessionNumber for node nodeName not allowed to operation using path data transfer path The operation reported indicates either READ or WRITE, depending upon the operation attempted. The path is reported as LAN-free. If this message is received when trying a LAN-free operation, evaluate and verify the LAN-free settings. Generally, if data is not sent LAN-free when the client is configured to use LAN-free, the storage pool destination for the policy assigned to this node is not a LAN-free enabled storage pool, or the paths are not defined correctly.
Show commands for the server or storage agent

SHOW commands are undocumented and unsupported diagnostic commands that are used to display information about in-memory control structures and other run-time attributes. The SHOW commands are used by development and service only as diagnostic tools. Several SHOW commands exist for the backup-archive client. Depending upon the information that a SHOW command displays, there may be instances where the information is changing or cases where it may cause the application (client, server, or storage agent) to crash. The SHOW commands should only be used at the recommendation of development or service. The SHOW commands that are documented are not all of the available SHOW commands.
AGGREGATE
Description
Displays information about an aggregate object in the server storage hierarchy.
Recommendation
The AGGREGATE command is used to determine the existence and logical files stored in an aggregate object in the servers storage hierarchy. The offset, length, and active state of backup files is displayed for files within the aggregate. You might use this command if you are having trouble restoring or retrieving files, expiring or moving data, backing up primary storage pools, copying active data to active data pools, or auditing volumes.
Other
The syntax is SHOW AGGRegate aggrID-high aggrID-low. aggrID-high and aggrID-low are the high-order and low-order 32-bit words of the 64-bit aggregate id of the aggregate that is being queried.
178
ASQUEUED
Description
Displays the mount point queue.
Recommendation
In order to use a drive, a client session or server process must first obtain a mount point. The mount point management on the server allows for queuing waiters for mount points if more mount points are needed than are available. The ASQUEUED command is useful for determining the state of a mount point request, especially if a session or process appears to be hung waiting for a mount point.
Other
The syntax is SHOW ASQueued.
ASVOL
Description
Displays assigned volumes.
Recommendation
As sequential media volumes are assigned for use by a session or a process, they are tracked in an in-memory list. You can view this list to determine the state of in-use volumes, as well as hang or deadlock situations where a session or process appears to be stuck waiting for a volume or holding a volume and waiting for something else.
Other
The syntax is SHOW ASVol.
BFOBJECT
Description
Displays information about a bitfile object in the server storage hierarchy.
Recommendation
The BFOBJECT command helps you determine the existence and attributes of a bitfile object in the servers storage hierarchy. You might use this command if you are having trouble restoring, retrieving, expiring, or auditing the object.
Other
The syntax is SHOW BFObject objID-high objID-low. objID-high and objID-low are the high-order and low-order 32-bit words of the 64-bit object id of the object being queried. The high-order word is optional; if not specified, a value of zero is assumed. The object can be a backup object, an archive object, a space-managed object, and so on.
Chapter 7. Trace
179
BUFSTATS
Description
Displays usage statistics for the database buffer pool. The BUFSTATS command shows the cache hit percentage of the buffer pool, which is suggested to be above 98%.
Recommendation
Use the BUFSTATS command to determine if the configured database buffer pool size is large enough. See the Tivoli Storage Manager Performance Tuning Guide for more details on setting the database buffer pool to an appropriate size.
Other
The syntax is SHOW BUFStats.
BUFVARS
Description
Displays database buffer pool global attributes.
Recommendation
Use the BUFVARS command to determine if the configured database buffer pool size is large enough. This may also be useful for diagnosing cases where the server is hung or when the server runs out of recovery log space. The database buffer pool performance and characteristics can influence the servers running out of recovery log space because the ability to write (flush) the changed pages to the database volumes can impact the ability of the recovery log to manage its space.
Other
The syntax is SHOW BUFVars. Note: Ensure that the IMEXP trace class is active.
CONFIGURATION
Description
The CONFIGURATION command is a summary SHOW command that actually issues many different show commands and queries.
Recommendation
Use the CONFIGURATION command to provide a general configuration and other information about the server to IBM service.
Other
The syntax is SHOW CONFIGuration.
180
DBTXNTABLE
Description
Displays information about transactions which are performing database operations.
Recommendation
For transactions performing database operations, the DBTXNTABLE command displays the following information: v Database tables that are open (in-use) v Recovery log usage information such as the number of log records written and the recovery log space used v The first, last, and next recovery log records written v Whether or not a transaction is valid or being aborted (rolled back)
Other
The syntax is SHOW DBTXNTable.
DBVARS
Description
Displays database global attributes.
Recommendation
Use the DBVARS command to view the current state and attributes of the server database.
Other
The syntax is SHOW DBVars.
DEVCLASS
Description
Displays information about device classes.
Recommendation
Use the DEVCLASS command to display the states of allocated drives, device class attributes, and other information. This command is often used to diagnose problems with devices or hangs waiting for a drive, library, or volume. The command SHOW LIBRARY also gives good complementary information about drives and libraries.
Other
The syntax for this command is SHOW DEVCLass.
GROUPLEADERS
Description
Displays all backup group leaders for an object in the servers inventory.
Chapter 7. Trace
181
Recommendation
Use the GROUPLEADERS command to determine the backup group relationships of an object in the servers inventory. You might use this command if you are having trouble restoring, retrieving, expiring, or auditing the object.
Other
The syntax is SHOW GROUPLeaders objID-high objID-low. objID-high and objID-low are the high-order and low-order 32-bit words of the 64-bit object id of the object being queried. The high-order word is optional; if not specified, a value of zero is assumed. The object must be a backup object. Note: Only available in Tivoli Storage Manager 5.2.3.0 and later.
GROUPMEMBERS
Description
Displays all backup group members for an object in the servers inventory.
Recommendation
Use the GROUPMEMBERS command to determine the backup group relationships of an object in the servers inventory. You might use this command if you are having trouble restoring, retrieving, expiring, or auditing the object.
Other
The syntax is SHOW GROUPMembers objID-high objID-low . objID-high and objID-low are the high-order and low-order 32-bit words of the 64-bit object id of the object being queried. The high-order word is optional; if not specified, a value of zero is assumed. The object must be a backup object. Note: Only available in Tivoli Storage Manager 5.2.0.0 and later.
INVOBJECT
Description
Displays information about an inventory object in the server.
Recommendation
Use the INVOBJECT command to determine the existence and attributes of an object in the servers inventory. You might use this command if you are having trouble restoring, retrieving, expiring, or auditing the object. As of Tivoli Storage Manager 5.2.2.0, this command reports the following: v New information for archive retention protected objects. v If the archive object is in deletion hold. v If the object uses event-based retention. v If the: Expiring.Objects row does not exist, then the objects event has not been signaled. Basedate in the Expiring.Objects row is later than the insert date, then this is the date the objects event was signaled.
182
Other
The syntax is SHOW INVObject objID-high objID-low. objID-high and objID-low are the high-order and low-order 32-bit words of the 64-bit object ID of the object being queried. The high-order word is optional; if not specified, a value of zero is assumed. The object can be a backup object, an archive object, a space-managed object, and so on.
LANFREE
Description
Several definitions must be on the server in order for a given client to perform LAN-free data movement operations. In cases where these definitions are not present or are incorrect, it may be difficult to determine if the LAN-free environment is configured correctly.
Recommendation
The LANFREE command evaluates all possible destination storage pools for this client node and reports whether or not the storage pool is capable of LAN-free data movement operations.
Other
The syntax is SHOW LANFREE nodeName storageAgent. Note: The VALIDATE LANFREE command replaces the SHOW LANFREE command for the Tivoli Storage Manager Version 5.3 and later. SHOW LANFREE was introduced in the Tivoli Storage Manager Version 5.2.2, and is only available for Tivoli Storage Manager Version 5.2.0.0.
LIBINVENTORY
Description
Displays the current state of the library inventory for the library specified.
Recommendation
Use the LIBINVENTORY command if there is a problem with the library inventory information. The command will display current in-memory properties of the library inventory.
Other
The syntax is SHOW LIBINVentory libraryName where libraryName is optional, and if left out, the command will return the inventory information for all libraries.
LIBRARY
Description
Use the LIBRARY command to display the current state of the specified library and all of its drives.
Chapter 7. Trace
183
Recommendation
This command is useful to gather a quick view of all in-memory information about a library and its drives. This output can be gathered for any problem related to libraries or drives (e.g. mounting problems).
Other
The syntax is SHOW LIBRary libraryName where libraryName is optional. If left out, the command will return information for all the libraries.
LOCK
Description
Displays lock holders and waiters.
Recommendation
The server and storage agent use locks as a mechanism to serialize access and updates to information and other constructs. This information is used to diagnose hangs or other resource contention issues.
Other
The syntax is SHOW LOCK.
LOGPINNED
Description
Evaluates and determines whether or not the server recovery log is pinned. A pinned recovery log may cause the recovery log to run out of space and possibly cause the server to crash.
Recommendation
The LOGPINNED SHOW command interrogates a number of server control structures and correlates the data to determine if a session, transaction, or process is pinning the recovery log. If it determines that something is pinning the recovery log, it reports this information.
Other
The syntax is SHOW LOGPInned. To recover from a pinned recovery log, issue SHOW LOGPInned Cancel to cause the server to cancel or terminate the session, transaction, or process. Under some conditions, the pinning session or transaction may not terminate after issuing the CANCEL parameter. Note: Only available in Tivoli Storage Manager Version 5.1.7.0 and later.
LOGVARS
Description
Displays recovery log global attributes.
184
Recommendation
Use the LOGVARS command to determine the state of the recovery log.
Other
The syntax is SHOW LOGVars.
MEMTREND
Description
The MEMTREND command reports the memory used by the server, in megabytes, recorded at hourly intervals for the last 50 hours (this is a constant in the server code and is not user-configurable). The command also displays a histogram to help visualize the usage trend.
Recommendation
Use the MEMTREND command to determine if the server has a memory leak. If the memory usage is constantly increasing, this might indicate a leak. Note that for the measurements to be valid, the measurement period (the last 50 hours) should represent normal, steadystate server activity. The reported usage represents the amount of memory that internal server routines request from the pseudo-kernel memory routines. It does NOT represent the total amount of memory that the server is using. Nevertheless, it is still useful in determining the servers memory usage trend.
Other
The syntax is SHOW MEMTREnd. Note: Only available in Tivoli Storage Manager Version 5.3.0.0 and later.
MP
Description
Displays mount points.
Recommendation
Use the MP command to determine which volume is in-use by a given mount point and other attributes for the assigned mount points. SHOW LIBRARY and SHOW DEVCLASS have useful complimentary information with this command to display the current state of drives and current devclass mount point counts.
Other
The syntax is SHOW MP.
NASDEV
Description
Displays the SCSI devices attached to a NAS file server associated with a NAS datamover definition.
Chapter 7. Trace
185
Recommendation
Create an NDMP connection to the specified NAS file server and display the attached SCSI devices on the file server. This command only requires a NAS node and datamover definition.
Other
The syntax is SHOW NASDev.
NASFS
Description
Displays the filesystems on a NAS file server associated with a NAS datamover definition.
Recommendation
Create an NDMP connection to the specified NAS file server and display the file systems defined on the file server. Any filesystems displayed by this command may be backed up by the Tivoli Storage Manager. The NASFS command requires only a NAS node and datamover definition.
Other
The syntax is SHOW NASFs.
NASINFORMATION
Description
Displays configuration information about the NAS file server associated with a NAS datamover definition.
Recommendation
Create an NDMP connection to the specified NAS file server and display general configuration information retrieved from the file server. This command is useful for identifying basic communication problems with NAS file servers such as authentication errors. This command only requires a NAS node and datamover definition.
Other
The syntax is SHOW NASInformation.
NASWORKLOAD
Description
Displays the workload of NAS filers that are used for all Tivoli Storage Manager operations.
Recommendation
Use the NASWORKLOAD command to determine the workload of back-end data movement, as well as backup and restore operations.
186
Other
The syntax is SHOW NASWorkload.
RESQUEUE
Description
Displays the resource queue.
Recommendation
Use the resource queue to monitor common resources on the server. If a resource appears to be hung or holding a resource for an unreasonable amount of time, the resource monitoring algorithms for the server will take action and cancel or terminate the resource user. Typically, this is used to display information about transactions, locks, and other resources used by a storage agent on the database server that it is configured to use.
Other
The syntax is SHOW RESQueue.
SESSIONS
Description
Displays information about sessions connected to the server or storage agent.
Recommendation
Use the SESSIONS command to diagnose hangs or other general session problems while a session is still connected to the server. This is also useful in cases where a session is canceled or terminated and still appears in QUERY SESSION.
Other
The syntax is SHOW SESSions.
SLOTS
Description
Displays the current state of the specified librarys slot information (for example: which volumes are in the library and in which slots).
Recommendation
The information displayed is what is saved directly from the library hardware to in-memory values, and can be used to determine if this information is out-of-sync, incorrect, or to determine if the values returned from the library hardware itself are invalid. Alternatively, use this command to determine the drive element numbers for a SCSI library if QUERY SAN is unavailable for a particular library (for example: 3570 library).
Chapter 7. Trace
187
Other
The syntax is SHOW SLOTS libraryName.
SSPOOL
Description
Displays information storage pools.
Recommendation
Use the SSPOOL command to display the states and attributes of defined storage pools.
Other
The syntax is SHOW SSPool.
THREADS
Description
Displays information about all threads known to the server.
Recommendation
The server displays information about each thread, typically including the Tivoli Storage Manager thread ID, the system thread id, the thread name, mutexes it holds (if any), and mutex or condition it is awaiting (if any). This command is platform-specific, so each platform may have slightly different information. You might want to use this command if the server or a particular server process appears to be hung in order to see if there are threads waiting for resources held by another thread.
Other
The syntax is SHOW THReads. Note: On some platforms (as an example: HP), the information reported is obtained without serialization. On a busy system (one that is not hung), the information may be inconsistent, multiple threads may report holding the same mutex, or a thread may report that it is waiting on a mutex held by another thread that does not claim to hold it. On a Tivoli Storage Manager Server Version 5.3 for AIX, SHOW THReads displays a stack trace of each thread. A stack trace reveals information about the server, which can help developers and service personnel diagnose problems. See the examples below:
Stack trace: 0x09000000000b9a5c select 0x000000010000d06c pkDelayThread 0x00000001007d8290 CsCmdSchedulerThread 0x000000010000a494 StartThread 0x090000000032f2ac _pthread_body Thread 33: SmAdminCommandThread tid=8513, ktid=3686529, ptid=21, det=0, zomb=0, join=0, result=0, sess=0 Holding mutex THRV->mutex (0x11173b9f8), acquired at 10000b5c8 (PkShowThreads)
188
Stack trace: 0x000000010000ba0c PkShowThreads 0x00000001009df84c ShowCommand 0x00000001009df2f0 psCommand 0x0000000100a064a0 AdmShow 0x00000001006f091c AdmCommandLocal 0x00000001006f2758 admCommand 0x000000010087d8d4 SmAdminCommandThread 0x000000010000a494 StartThread 0x090000000032f2ac _pthread_body
TOCSETS
Description
Displays all Table Of Contents (TOC) sets known to the server.
Recommendation
A TOC set is used during file-level NDMP operations. During an NDMP backup with the TOC=YES parameter, a TOC is built in the server db. During a restore, one or more TOCs may be loaded into the server database in order to provide file and directory names to the client GUI. This command displays the status of the TOC set (e.g. building or loading) and how much temporary DB space is in use for each TOC set. You might use this command if you are having trouble doing an NDMP backup with the TOC=YES parameter, or having trouble restoring files from an NDMP backup, or if TOC sets are being retained in the server database too long or not long enough.
Other
The syntax is SHOW TOCSets DELETE=setNum TOUCH=setNum. The DELETE parameter causes the specified TOC set number to be deleted. The TOUCH parameter updates the last used date of the specified TOC set number. A TOC set is retained for the TOC retention period following the last used date (see SET TOCRETENTION command). Only available in the Tivoli Storage Manager Version 5.2.0.0 and later.
TOCVARS
Description
Displays information about the Table Of Contents (TOC) component of the server.
Recommendation
Use the TOCVARS command to determine the status of the TOC component. You might use this command if you are having trouble doing an NDMP backup with the TOC=YES parameter, or having trouble restoring files from an NDMP backup.
Other
The syntax is SHOW TOCVars. Note: Only available in the Tivoli Storage Manager Version 5.2.0.0 and later.
Chapter 7. Trace
189
TXNTABLE
Description
Displays information about transactions that are on the in-use list on the server.
Recommendation
Transactions are used by server processes, sessions, or other operations to read information from the database, make updates to the database (such as insert, update, or delete information), or to manage locks. This information is useful for diagnosing hangs or other transaction related failures while the transaction is still open on the server.
Other
The syntax is SHOW TXNTable.
VALIDATE LANFREE
Description
Several definitions must be on the server in order for a given client to perform LAN-free data movement operations. In cases where these definitions are not present or are incorrect, it may be difficult to determine if the LAN-free environment is configured correctly.
Recommendation
The VALIDATE LANFREE command evaluates all possible destination storage pools for this client node and reports whether or not the storage pool is capable of LAN-free data movement operations.
Other
The syntax is VALIDATE LANFREE nodeName storageAgent. Note: The VALIDATE LANFREE command replaces the unsupported SHOW LANFREE command for the Tivoli Storage Manager Version 5.3 and later.
VERIFYEXPTABLE
Description
Performs verification of the Expiring.Objects table of the server.
Recommendation
Use the VERIFYEXPTABLE command to find and clean up orphaned entries in the Expiring.Objects table (as an example: entries that do not have corresponding entries in the Backup.Objects or Archive.Objects tables). You might use this command if you are having trouble with inventory expiration. Note: The VERIFYEXPTABLE command was changed to CLEANUP EXPTABLE in the following Tivoli Storage Manager releases: v 5.2.4.2 v 5.2.6 v 5.3.0.1
190
v 5.3.1
Other
The syntax is SHOW VERIFYEXPtable.
VOLINUSE
Description
Displays whether or not the volume specified is currently in the servers in-use list. The VOLINUSE command displays additional information that may be helpful including whether the volume is currently pending removal from the in-use list.
Recommendation
Use the VOLINUSE command to determine whether or not a volume is currently on the in-use list and, if necessary, to remove it from that list. Note that operations associated with this volume might fail if the volume is removed from the in-use list.
Other
The syntax is SHOW VOLINUSE volumeName. If the volume needs to be removed from the in-use list, the following additional parameter can be specified to remove the volume from the list: SHOW VOLINUSE volumeName REMOVE=YES.
Enabling trace for the Tivoli Storage Manager device driver

Tracing is available for the Tivoli Storage Manager device driver. The Tivoli Storage Manager device driver can be traced from the server console, an administrative client, or from a shell running on the system where the device driver is installed. The tracing instructions below apply to the Tivoli Storage Manager device driver on all platforms where the device driver is supported. For devices that use device drivers other than the Tivoli Storage Manager device driver, the ability to trace and instructions on how to trace those device drivers is provided by the device vendor.
Tracing from the server console/admin client

To trace the driver from the server, you must first issue the proper commands. Issue the TRACE ENABLE and TRACE BEGIN commands (see Enabling trace for the server or storage agent on page 152) to trace the driver from the server. The Tivoli Storage Manager device driver actually consists of three drivers: one for library-autochanger devices, one for tape devices, and one for optical drives. You may choose which one you want to trace. The following is the syntax for the command: DDTRACE START [ LIBRARYDD | TAPEDD | OPTICALDD] [flags=EE |, FULL |, SYSLOG | BASE ] DDTRACE GET [ LIBRARYDD | TAPEDD | OPTICALDD] DDTRACE END [ LIBRARYDD | TAPEDD | OPTICALDD] The following options are available:
Chapter 7. Trace
191
START Turns on tracing. This writes the trace to a memory buffer based on the default or specified FLAGS option. GET END Writes the memory buffer to the same file that was specified with the server TRACE BEGIN command. Stops writing trace to the memory buffer. This does not wipe out the contents of the buffer, so you may run END before running GET.
LIBRARYDD Traces the device driver that controls library-autochangers. TAPEDD Traces the device driver that controls tape drives. OPTICALDD Traces the device driver that controls optical drives. For the options listed above, you may specify any one device driver or the library device driver, and one of the other two. These are space delimited. For example: DDTRACE START drives. DDTRACE START optical drives. DDTRACE START DDTRACE START drives. DDTRACE START optical drives. TAPEDD - Starts tracing the device driver that controls tape OPTICALDD Starts tracing the device driver that controls LIBRARYDD Starts tracing the library-autochanger. LIBRARYDD TAPEDD Traces both the library and the tape LIBRARYDD OPTICALDD Traces both the library and the
Whichever of these you use, specify the same ones for all commands in the start-get-end series. The FLAGS parameter is optional and usually not required. The following are the values for the FLAGS parameter: EE FULL Traces all device driver routine entries and exits. Turns on more debug tracing. It provides more detail, but because the memory buffer size is fixed, fewer events are traced. It also does not trace routine entry and exit points.
SYSLOG On some platforms, this directs the trace statements to be written to the system log in addition to the memory buffer. This is most useful in debugging kernel crashes or in circumstances where the trace wraps in the memory buffer. BASE This is the default and cannot be specified with any other flags. It is only used to turn off the EE, FULL, and SYSLOG flags without turning off trace.
Tracing from a command shell (only valid for 5.2.2 and above) - all platforms
A stand-alone utility, ddtrace exactly mimics the DDTRACE server commands.
192
A stand-alone utility, ddtrace is installed in the devices directory, which is the same directory as the mttest, lbtest, and optest utilities. Its syntax and options are identical to the DDTRACE server command. For example: $ ddtrace start librarydd tapedd flags=EE - Start tracing both the library and tape drivers, and get additional entry/exit trace. $ ddtrace get librarydd tapedd - Get the trace from memory and write it to the file ddtrace.out. $ ddtrace end librarydd tapedd - Stop tracing to memory. The main use of this stand-alone utility is primarily for cases when the driver needs to be traced during the Tivoli Storage Manager server initialization. It writes the memory buffer to the ddtrace.out file in the current directory. If the file exists, it appends to the file and does not overwrite it.
Tracing for the client

You can enable tracing on the client or client API. Perform the following steps to enable tracing on the client or client API: 1. Determine the trace classes to enable from the table below:
Trace Class Name SERVICE Description Display general processing information for the client. When to use Useful in many cases. Generally recommended for protocol violations, transaction processing errors, or in cases where the client is hung and not responding. To debug protocol violations, transaction processing errors, or in cases where the client is hung and not responding. To debug session data corruption issues that may be caused by the network. This generates a large amount of output. Additional Notes
VERBINFO
Collect information regarding the client-server protocol used by IBM Tivoli Storage Manager. Detailed information regarding the client-server protocol used by IBM Tivoli Storage Manager . This displays internal memory buffers containing the verbs sent and received by the client.
VERBDETAIL
2. Enable the trace by adding the following to the client options file: traceflag <trace class name>. Note: <trace class name> may be a comma-delimited list of trace classes. For example, this could be entered as traceflag service,verbinfo,verbdetail. 3. Configure trace to begin and issue the trace messages to a file by adding the following to the client options file: tracefile <file name>.
Chapter 7. Trace
193
4. Perform the operation that is causing the problem. Note: Tracing may also be configured and started by invoking the client from a command prompt and specifying the flags above. For example, dsm -traceflags=service -tracefile=file.out.
Client and Journal Daemon traceflags

Trace settings for the Journal Daemon are specified in the journal daemon configuration file tsmjbbd.ini. Trace settings for the Journal Daemon are specified in the journal daemon configuration file tsmjbbd.ini as follows:
[JournalSettings] TraceFlags=all_jbb ; ; the following two settings allow tracefile segmentation ; TraceMax=100 TraceSegMax=1 tracefile=tracefiles\trace.out
| | | | |
Journal Daemon specific trace settings: v BTREEDB - low-level BTREE database base class v CACHEDB - disk cache backup and Windows 2003 exclude cache processing v DBPERF - low-level database operation performance v DBSTATS - performance tracking of database query, insert/update, delete, and tree walk operations v FILEOPS - internal database activity v v v v v v v v v JBBCOMM - listening thread JBBDAEMON - process manager JBBFILEMON - file system monitor JBBDBACCESS - database controller thread JBBDBINFO - low-level database access JBBNPCOMM - named pipe communications JBBSERVICE - Windows platform-specific service tracing JBBVERBINFO - detailed verb information ALL_JBB - aggregate traceflag that includes all of the above settings
Trace Settings for the backup-archive client specified in dsm.opt: v JOURNAL - journal based backup tracing
Enabling trace for the client

The client provides individual and aggregate trace classes. Aggregate classes are a shortcut for enabling many related trace classes by simply specifying the aggregate trace class name. In the table below, there may be references to trace classes that are enabled as part of an aggregate trace class but are not explicitly discussed on their own. The trace classes listed are those that are most typically requested or used for diagnosing problems. This list does not discuss all possible trace classes that are available.
194
To find information about a trace class, click the trace class in the table below or view the full information below. The trace class name should be used with the TRACEFLAG options in DSM.OPT.
ALL_BACK
Description
Displays general backup processing information for the client.
Recommendation
Use this trace class for problems related to selective or incremental backups.
Other
Aggregate of TXN, INCR, POLICY, and PFM trace classes and implicitly included in the SERVICEtrace class.
ALL_FILE
Description
Displays general backup processing information for the client.
Recommendation
Use this trace class for problems related to reading and writing of data and obtaining file attribute information.
Other
Aggregate of DIROPS, FILEOPS, and FIOATTRIBS trace classes and implicitly included in the SERVICE trace class.
ALL_IMAGE
Description
Displays image processing information for the client.
Recommendation
Use this trace class for problems related to all aspects of volume image backup-and-restore operations.
Other
Aggregate of several image-related trace classes and implicitly included in the SERVICE trace class.
ALL_JBB
Description
Displays journal-based backup processing information for the client.
Recommendation
Use this trace class for problems related to all aspects of journal-based backups.
Chapter 7. Trace
195
Other
Aggregate of several journal-based backup-related trace classes and implicitly included in the SERVICE trace class.
ALL_NAS
Description
Displays NDMP processing information for the client.
Recommendation
Use this trace class for problems related to all aspects of NDMP backup-and-restore operations.
Other
Aggregate of several NDMP-related trace classes and implicitly included in the SERVICE trace class.
ALL_SESS
Description
Displays all session and verb information sent between the client and the server.
Recommendation
Use this trace class for problems related to the client and server session, such as communication timeouts, protocol violations, and instances where the client appears to be hung waiting for the server, or vice versa.
Other
Aggregate of SESSION, VERBINFO, SESSVERB, VERBADMIN, and VERBDETAIL trace classes. All of the trace classes in this aggregate are implicitly included in the SERVICEtrace class, except for VERBDETAIL
ALL_SNAPSHOT
Description
Displays information relating to volume snapshot operations.
Recommendation
Use this trace class to determine problems related to volume snapshots that are used in online image backup and open file-support operations.
Other
Aggregate of several volume snapshot-related trace classes and implicitly included in the SERVICE trace class.
ALL_WAS
Description
Displays Web Application Server (WAS) processing information for the client.
196
Recommendation
Use this trace class for problems related to all aspects of WAS backup-and-restore operations.
Other
Aggregate of several WAS related trace classes and implicitly included in the SERVICE trace class.
AUDIT
Description
Displays auditing information for backup-and-restore processing.
Recommendation
Use this trace class to keep record of files processed, committed and restored in a file.
Other
Part of the SERVICE trace aggregate.
CLIENTTYPE
Description
Displays client type on each trace output line.
Recommendation
Use this trace class for tracing situations when more then one client component is involved, such as the client acceptor daemon (CAD) and the file system agent.
Other
None.
COMPRESS
Description
Displays compression information.
Recommendation
Use this trace class to determine how much data is compressed on a per-file basis.
Other
DELTA
Description
Displays adaptive subfile backup processing information.
Chapter 7. Trace
197
Recommendation
Use this trace class to determine errors in adaptive subfile backup-and-restore operations.
Other
DIROPS
Description
Displays directory read and write operations.
Recommendation
Use this trace class when problems occur in a directory, read or write.
Other
Part of the SERVICE and ALL_FILE trace aggregates.
DOMAIN
Description
Displays incremental domain processing information.
Recommendation
Use this trace class for determining how DOMAIN statements are resolved during backup processing, such as problems in resolving the ALL-LOCAL domain.
Other
Part of the SERVICE trace aggregates.
ENCRYPT
Description
Displays data encryption information.
Recommendation
Use this trace class to determine if a file is included for encryption processing.
Other
ERROR
Description
Displays operating system specific error information.
198
Recommendation
Use this trace class to determine error codes generated by the operating system.
Other
FILEOPS
Description
Displays file read and write operations.
Recommendation
Use this trace class when problems occur in a file open, read, write, or close operation.
Other
Part of the SERVICE and ALL_FILE trace aggregates.
FIOATTRIBS
Description
Displays comparisons of file attributes between the local client version and the active version on the Tivoli Storage Manager server.
Recommendation
Use this trace class in determining why a file was backed up during an incremental backup.
Other
Part of the SERVICE, ALL_BACK, and ALL_FILE trace aggregates.
INCR
Description
Displays incremental list processing comparisons between the client and server.
Recommendation
Use this trace class for determining if files are candidates for incremental backup, especially in conjunction with the FIOATTRIBS trace class.
Other
Part of the SERVICE and ALL_BACK trace aggregates.
INCLEXCL
Description
Displays include/exclude status for the object being processed. This flag is also used for the Preview function.
Chapter 7. Trace
199
Recommendation
Use this trace class to determine which object (usually file or directory) is included or excluded during backup-archive/preview.
Other
None.
MEMORY
Description
Displays memory allocation and free requests.
Recommendation
Use this trace class to determine memory leaks, memory spikes, and other memory-related problems.
Other
This trace class writes a large amount of information into the trace file and is not included in any aggregate classes.
NETWARE
Description
Displays detailed information relating to the Novell SMS backup API.
Recommendation
Use this trace class for isolating NetWare client problems between the backup application and operating system.
Other
Aggregate of the SMS_DEBUG, SMS_DETAIL, TXN, and TID trace classes.
OPTIONS
Description
Displays current processing options.
Recommendation
Use this trace class to determine which options are in effect for the current session and problems in accepting processing options from server client-options sets.
Other
PASSWORD
Description
Displays password file-access information (does not show passwords).
200
Recommendation
Use this trace class to determine problems with reading the Tivoli Storage Manager server passwords from local storage, for example, PASSWORDACCESS=GENERATE errors.
Other
PID
Description
Displays process ID on each trace statement.
Recommendation
Use this trace class to diagnose problems that may involve multiple processes.
Other
POLICY
Description
Displays policy information available to the backup-archive client.
Recommendation
Use this trace class to see which policies are available during a backup or archive operation.
Other
Part of the SERVICE and ALL_BACK trace aggregates.
SCHEDULER
Description
Displays general processing information for the scheduler. An aggregate that includes most of the client trace classes listed in this table.
Recommendation
Useful in many cases. This trace class is generally recommended for diagnosing scheduler problems when the nature of the problem is unknown. If the SCHEDULER trace flag is used, it will generally not be necessary to specify any other trace flags because it already includes most of the basic trace classes.
Other
Aggregate of all trace classes except MEMORY, THREAD_STATUS, and *DETAIL classes.
Chapter 7. Trace
201
SERVICE
Description
Displays general processing information for the client. An aggregate that includes most of the client trace classes listed in this table.
Recommendation
Useful in many cases. This trace class is generally recommended when the nature of the problem is unknown. If the SERVICE trace flag is used, it will generally not be necessary to specify any other trace flags because it already includes most of the basic trace classes.
Other
Aggregate of all trace classes except MEMORY and *DETAIL classes. The SERVICE trace flag can generate a substantial amount of information. Consider using the TRACEMAX option in conjunction with the SERVICE trace flag.
SESSION
Description
Displays minimal session information between the client and the server.
Recommendation
Use this trace class to give session context to general processing errors, or in conjunction with one of the VERB* trace classes, to determine session problems such as session timeouts and protocol violations.
Other
Part of the SERVICE and ALL_SESS trace aggregates.
SESSVERB
Description
Displays additional session information between the client and the server.
Recommendation
Use this trace class to give session context to general processing errors, or in conjunction with one of the VERB* trace classes, to determine session problems such as session timeouts and protocol violations.
Other
SMS_DETAIL
Description
202
Recommendation
Use this trace class to isolate NetWare client problems between the backup application and operating system.
Other
Part of the SERVICE and NETWARE trace aggregates.
SMS_DEBUG
Description
Recommendation
Use this trace class to isolate NetWare client problems between the backup application and operating system.
Other
Part of the SERVICE and NETWARE trace aggregates.
STATS
Description
Displays final processing statistics in the trace file.
Recommendation
Use this trace class for collecting final processing statistics into a file.
Other
THREAD_STATUS
Description
Displays thread status.
Recommendation
Use this trace class when diagnosing problems related to threading.
Other
TXN
Description
Displays transaction processing information.
Chapter 7. Trace
203
Recommendation
Use this trace class when diagnosing problems related to transaction processing problems on the server such as transaction aborts and retries.
Other
Part of the SERVICE, ALL_BACK and NETWARE trace aggregates.
VERBDETAIL
Description
Displays detailed verb information pertinent to client-server sessions.
Recommendation
Use this trace class to determine contents of verbs sent between the client and server.
Other
Part of the ALL_SESS trace aggregates.
VERBINFO
Description
Displays verb information pertinent to client-server sessions.
Recommendation
Use this trace class in conjunction with the SESSION traceflag to give session context to general processing errors or to determine session problems like session timeouts and protocol violations.
Other
WIN2K
Description
Displays Windows system object or system state processing.
Recommendation
Use this trace class to determine errors with backup or restore of the system state information.
Other
Part of the SERVICE trace aggregates. Only valid on the Windows backup-archive client.
Enabling backup-archive client trace

There are two methods of tracing that are available for the backup-archive (B/A) client. The first method is to configure trace parameters prior to starting the B/A
204
Client. The second, more recent method, is to enable tracing while the client is running. Choose which method of tracing to enable.
Enabling client trace on command-line

You can trace the available backup-archive (B/A) client by enabling client trace on the command line. Perform the following steps to enable client tracing on the command line: 1. Determine the trace classes to enable. See Enabling trace for the client on page 194 for possible trace classes. 2. Choose which trace classes to enable by adding the following to the client options file: traceflags <trace class name> 3. Use a minus sign (-) in front of a trace class to subtract the class from tracing. Make sure that the subtracted trace classes are placed at the end of the trace class list. For example, if you want to collect a SERVICE trace without the SESSION or SESSVERB classes, then specify the following: Correct: traceflags service,-session,-sessverb Incorrect: traceflags -session,-sessverb,service Note: <trace class name> may be a comma-delimited list of trace classes. For example, this could be entered astraceflags service,verbdetail 4. Choose the location of the trace messages output by adding the following to the client options file: tracefile <file name>. The tracefile name should be fully-qualified, for example: tracefile tracefile tracefile tracefile c:\service\trace.out (Windows) /home/spike/trace.out (UNIX) SYS:\PUBLIC\TRACE.OUT (NetWare) trace.txt (Macintosh)
5. You can also set a maximum size for the trace file between 1 and 4,294,967,295 MB by specifying the following in the client options file: tracemax <size in mb> If a maximum value is specified, the client starts writing information from the beginning of the trace file (that is, wrapping) when the trace reaches its maximum size. This can be useful if you are trying to capture an event that happens at the end of a long-running process. For example, to specify a maximum trace file size of 10 MB: tracemax 10 After a tracefile reaches the limit specified with tracemax, Continued at beginning of file is written to the end of the trace file and tracing continues from the top of the file. The end of the tracefile is indicated with END OF DATA. You can locate the end of the trace by searching for this string. If you specify a TRACEMAX size of 1001 or higher and TRACESEGSIZE is not specified, then the trace file is automatically split into multiple segments of 1000 MB per segment (see TRACESEGSIZE discussion immediately below). 6. You can choose to let the client split the trace into smaller segments between 1 and 1,000 MB per segment by specifying the following in the client options file: tracesegsize <trace segment size in MB> This allows you to more easily manage very large amounts of trace data, avoiding the problems associated with zipping very large files and eliminating the need to use a separate file splitter utility. For example, issue the following to specify a trace segment size of 200 MB: tracesegize 200 Trace file segments are named using the name specified with the tracefile option plus an extension indicating the segment number. For example, if you
Chapter 7. Trace
205
specify tracefile tsmtrace.out and tracesegsize 200, then the trace will be segmented into multiple separate files of no more than 200 MB each, with file names tsmtrace.out.1, tsmtrace.out.2, and so on. Note that when specifying the segment size, do not use any comma separators: Correct: tracemax 1000 Incorrect: tracemax 1,000 If you use the tracesegsize option, the trace file segments will be named using the tracefile option with an additional extension using the segment number. For example, trace.out.1 7. Perform the operation that exhibits the problem. Note: Tracing may also be configured and started by invoking the client from a command prompt and specifying the flags above. For example:
dsmc -traceflags=service,verbdetail -tracefile=tsmtrace.out -tracemax=2500 -tracesegsize=200
Enabling tracing while client is running

You can trace the available backup-archive (B/A) client while the client is running. v dsmtrace is included with the Tivoli Storage Manager 5.3 B/A client. The B/A client must be installed to use dynamic tracing. v The DSMTRACELISTEN YES option must be in effect when the client is started. On Windows, this option is specified in the client options file (usually dsm.opt). On Unix, this option is specified in the system options file (dsm.sys) in the stanza that the client uses. v Unix users must be logged in as root in order to use dsmtrace. v Windows users must be logged in as a member of the Administrators group. When the client starts, it launches a separate trace listener thread. This thread listens on a named pipe, waiting to be contacted by the dsmtrace utility. In order to make the named pipe name unique, the client process ID (PID) is a part of the pipe name. When dsmtrace is used to configure tracing, it contacts the client through the named pipe on which the client is listening and passes to it the desired trace configuration operation. The client then passes the results of the operation back to dsmtrace through another similarly-named output pipe. dsmtrace displays the results to the console. The client starts the trace listener thread only when client option DSMTRACELISTEN YES is in effect. If DSMTRACELISTEN NO is in effect, then the listener thread is not started and dynamic tracing is not available to that client. DSMTRACELISTEN NO is currently the default value. Historically, the steps for gathering a Tivoli Storage Manager client trace are as follows: 1. Stop the Tivoli Storage Manager client. 2. Configure the client options file with the desired trace options. 3. Restart the client and reproduce the problem. 4. Stop the client. 5. Remove the trace options from the client options file. 6. Send the resulting trace file to IBM technical support for analysis. Beginning with Tivoli Storage Manager Version 5.3, you can also use the dsmtrace utility to start, stop, and configure client tracing dynamically without having to stop the client or modify the options file. Dynamic tracing is
206
especially useful when you need to trace only the beginning of a long-running client operation, or when you need to start tracing after the client was running for some time. The dsmtrace utility includes the following features: v Identify running processes and their process IDs (PIDs) v Enable client tracing v Disable client tracing v Query client trace status The table below summarizes the availability of this feature:
Client Component Backup-Archive Client (command line) Unix Program Name dsmc Windows Program Name dsmc.exe dsm.exe dsmcad.exe dsmagent.exe dsmcsvc.exe tsmjbbd.exe domdsmc.exe domdsm.exe tdpexcc.exe tdpexc.exe tdpsqlc.exe tdpsql.exe
Backup-Archive Client (GUI) N/A Client Acceptor Daemon Remote Client Agent Scheduler Service Journal Service Data Protection for Domino (command line) Data Protection for Domino (GUI) dsmcad dsmagent N/A N/A domdsmc N/A
Data Protection for Microsoft N/A Exchange (command line) Data Protection for Microsoft N/A Exchange (GUI) Data Protection for Microsoft N/A SQL Server (command line) Data Protection for Microsoft N/A SQL Server (GUI)
Note: v Dynamic tracing is currently unavailable for NetWare clients. v The Unix column above includes Mac OS X. v Tracing for the Data Protection components is for the Tivoli Storage Manager API only. v The Tivoli Storage Manager API tracing is available with any multithreaded application that uses the Tivoli Storage Manager API. The executable name is the name of the application program that loads the API. 1. Identify the process ID (PID) of the client you want to trace (make sure that DSMTRACELISTEN YES is in effect). Issue the following command to show all running instances of the client: dsmtrace query pids Example output:
D:\tsm>dsmtrace query pids IBM Tivoli Storage Manager dsmtrace utility dsmtrace Version 5, Release 3, Level 0.0 dsmtrace date/time: 10/24/2004 21:07:36
Chapter 7. Trace
207
(c) Copyright by IBM Corporation and other(s) 1990, 2004. All Rights Reserved. PROCESS ID PROCESS OWNER DESCRIPTION 4020 andy Backup-Archive Client (CLI) D:\tsm> EXECUTABLE NAME dsmc.exe
Important note for Linux users: the threading model for some versions of Linux is to run each thread as a separate process. This means that when you query process information, you may see several processes for each instance of the client. The process you need to identify is the dsmc parent process. For example:
fvtlinuxppc:/opt/tivoli/tsm/client/ba/bin # dsmtrace q p IBM Tivoli Storage Manager dsmtrace utility dsmtrace Version 5, Release 3, Level 0.0 dsmtrace date/time: 10/24/04 08:07:37 (c) Copyright by IBM Corporation and other(s) 1990, 2004. All Rights Reserved. PROCESS ID PROCESS OWNER DESCRIPTION 28970 root Backup-Archive Client 28969 root Backup-Archive Client 28968 root Backup-Archive Client 28967 root Backup-Archive Client (CLI) (CLI) (CLI) (CLI) EXECUTABLE NAME dsmc dsmc dsmc dsmc
fvtlinuxppc:/opt/tivoli/tsm/client/ba/bin #
In such a situation, use the ps command to identify the parent dsmc process:
linuxppc:~ # ps -ef | grep dsmc root root root root root 28967 28968 28969 28970 24092 1151 28967 28968 28968 24076 0 0 0 0 0 Oct22 pts/16 Oct22 pts/16 Oct22 pts/16 Oct22 pts/16 08:15 pts/93 00:00:00 dsmc 00:00:00 dsmc 00:00:00 dsmc 00:00:00 dsmc 00:00:00 grep dsmc
linuxppc:~ #
Notice that the parent for processes 28969 and 28970 is 28968. The parent for 28968 is 28967. The parent for 28967 is 1151, but the 1151 process does not appear in this display output. Process 1151 is the process that launched dsmc. So, the correct parent process ID is 28967. 2. Issue the following to enable tracing on the client:
dsmtrace enable 4020 -traceflags=service -tracefile=d:\trace.txt
Example output:
C:\program files\tivoli\tsm\baclient>dsmtrace enable 4020 -traceflags=service -tracefile=d:\trace.txt IBM Tivoli Storage Manager dsmtrace utility dsmtrace Version 5, Release 3, Level 0.0 dsmtrace date/time: 10/24/2004 21:45:54 (c) Copyright by IBM Corporation and other(s) 1990, 2004. All Rights Reserved. ANS2805I Tracing has been enabled. C:\program files\tivoli\tsm\baclient> C:\program files\tivoli\tsm\baclient>
Important: When tracing an API application: The -pipenameprefix option must be included.
208
v On Unix, use prefix /tmp/TsmTraceTargetAPI v On Windows systems use prefix \\.\pipe\TsmTraceTargetAPI 3. After sufficient trace data is collected, disable the tracing by issuing the following:
dsmtrace disable 4020
Example output:
C:\program files\tivoli\tsm\baclient>dsmtrace disable 4020 IBM Tivoli Storage Manager dsmtrace utility dsmtrace Version 5, Release 3, Level 0.0 dsmtrace date/time: 10/24/2004 21:47:43 (c) Copyright by IBM Corporation and other(s) 1990, 2004. All Rights Reserved. ANS2802I Tracing has been disabled.
Other Examples Display all running processes whose names are listed in the table in the Background section by issuing the following command:
DSMTRACE QUERY PIDS
Display all running processes by issuing the following command:

dsmtrace query pids -filter=*
Display all running processes whose names begin with dsm by issuing the following command:
dsmtrace query pids -filter=dsm*
Display all running processes whose names are dsm plus one other character by issuing the following command:
dsmtrace query pids -filter=dsm?
Turn on SERVICE tracing for process 2132 by issuing the following command:
dsmtrace enable 2132 -traceflags=service -tracefile=c:\trace.txt
Trace output is written to file c:\trace.txt. Turn off extrc tracing for process 2132 (presumably tracing is already running for this process) by issuing the following command:
dsmtrace enable 2132 -traceflags=-extrc
Turn on FILEOPS tracing for process 4978 by issuing the following command:
dsmtrace enable 4978 -traceflags=fileops -tracefile=/tmp/dsmtrace.out -tracemax=1000 -tracesegsize=200
The trace is written to files /tmp/dsmtrace.out.1, /tmp/dsmtrace.out.2, and so on, with each file being no larger than 200 MB. After 1000 MB are written, tracing wraps back to /tmp/dsmtrace.out.1. Display basic trace information and list trace flags that are turned on for process 4978 by issuing the following command:
dsmtrace query trace 4978 -on
Disable tracing for process 4978 by issuing the following command:

Chapter 7. Trace
209
dsmtrace disable 4978
Disable tracing for API application process 364 by issuing the following command:
dsmtrace disable 364 -pipenameprefix=/tmp/TsmTraceTargetAPI
Known trace problems and limitations

We have gathered the known problems and limitations of trace processes to help you resolve problems you may have encountered when running a trace process. v If tracing is not currently active for a process, and dsmtrace is used only with the -TRACEFLAGS option, e.g. dsmtrace enable 2346 -traceflags=service, then you still see the following message:
ANS2805I Tracing has been enabled.
In this case, the trace flags were enabled (turned on), but tracing is not actually active until a trace file is specified by using the -TRACEFILE option. v Do not use the dsmtrace enable command to start tracing the API for Data Protection applications if the Data Protection application is run in a manner that does not cause it to connect to the Tivoli Storage Manager server. For example, The Data Protection for Lotus Domino command line interface has several such commands: domdsmc help domdsmc set domdsmc query domino domdsmc query pendingdbs domdsmc query preferences If you use dsmtrace to enable tracing for such commands, this can result in a hang of the dsmtrace process and (Unix only) a residual named pipe in the /tmp directory. v On Windows systems, you must be logged in with a local administrative account in order to use dsmtrace. v On Unix systems, you must be logged in as root to use dsmtrace. v On Unix systems, if a client process crashes or is killed, it may leave a named pipe (Unix FIFO) in the /tmp directory. These FIFOs have names beginning with TsmTrace and they include a process ID (PID) number. If this happens, and then a new client process is started whose PID happens to match that of the old residual FIFO, then the trace listener thread will probably not start. A fix for this is planned for a future release. Any old FIFOs with process numbers that do not match those of running the Tivoli Storage Manager processes can be safely deleted. Do NOT delete the FIFO of a running process. v The threading model for some versions of Linux is to run each thread as a separate process. This means that when you query process information, you may see several processes for each instance of the client. The process you need to identify is the dsmc parent process. For an example, refer to the previous example section. v When multiple instances of the same program are running, you must identify the PID of the instance you want to trace. In such a situation, using other clues, such as process information from the operating system, may be available to narrow down the desired PID. For example, if you want to trace dsmc that is being run by user andy and there are two instances of dsmc (one owned by user andy and the other owned by user kevin), you can use the process owner to identify which process to trace.
210
v If an options file contains an invalid option and the client does not start, you might see some named pipe errors in the dsmerror.log file. These error messages may be safely ignored. A fix for this is planned for a future release.
Trace options
Trace has several options that you can employ.
DSMTRACEListen
DSMTRACEListen No | Yes No Yes The client does not start the trace listener thread and dynamic tracing is not available. This is the default. The client starts the trace listener thread and dynamic tracing is available.
On Windows, this option is specified in the client options file (usually dsm.opt). On Unix, this option is specified in the system options file (dsm.sys) in the stanza that the client uses. This option cannot be specified from the command line.
dsmtrace
dsmtrace enable <pid> <options> Use this command to start or modify tracing for a process. pid The process ID (PID) for the client. Use dsmtrace query pids or your operating system facilities to identify the correct PID.
options The client trace options. dsmtrace disable <pid>[<options>] Use this command to stop tracing for a process. The trace file will close and the trace flags, maximum trace size, maximum trace segment size, and trace file name will all be cleared. <pid> This is the process ID (PID) for the client. Use dsmtrace query pids or your operating system facilities to identify the correct PID.
<options> The client trace options. dsmtrace help This command displays basic syntax for dsmtrace. dsmtrace query pids [-Filter=<spec>] <spec> This is the client process name filter specification. This can include the wildcard characters ? (match exactly one character) or * (match zero or more characters). If no filter is specified, then the default behavior is to display process information for any running instances of the program names listed in the table in the Background section above. Note: When using the FILTER option on Unix, put the * symbol before and after the search text. This is necessary because the executable name often includes the path in front of it, and in some cases, the executable name may have additional characters at the end of it. For example: v /opt/tivoli/tsm/client/ba/bin/dsmc v domdsmc_DominoUserID
Chapter 7. Trace
211
Thus, instead of -filter=dsmc or -filter=domdsmc, use -filter=*dsmc* or -filter=*domdsmc*. dsmtrace query trace <pid> [<options>] [<displayType>] [-ALl | -ON | -OFf | -BASic] <pid> This is the process ID (PID) for the client. Use dsmtrace query pids or your operating system facilities to identify the correct PID.
<options> The client trace options. <displayType> The display type can be one of the following: ALl This displays all trace flags and, for each flag, indicates whether it is turned on or off. The information shown with the -BASIC display type is also included. This displays the names of the trace flags that are turned on. The information shown with the -BASIC display type is also included. This displays the names of the trace flags that are turned off. The information shown with the -BASIC display type is also included.
ON
OFf
BASic This displays the name of the trace file and the maximum trace and trace segment sizes. This display type also indicates whether tracing is enabled or disabled.
-PIPENameprefix
-PIPENameprefix=<pipeNamePrefix> This option must be used when tracing API applications: v On Unix, use prefix /tmp/TsmTraceTargetAPI v On Windows, use prefix \\.\pipe\TsmTraceTargetAPI
-TRACEFIle
-TRACEFIle=<traceFileName> This must specify a valid file name to which the trace is written. If tracing is already running, then this option has no effect.
-TRACEFLags
-TRACEFLags=<traceFlags> Specify one or more trace flags. Typically, the trace flag SERVICE is used. Multiple trace flags are separated with a comma. Trace flags can also be turned off by prefixing the flag name with a minus sign. When combining trace flags that you want to turn on with trace flags that you want to turn off, put the flags that you want to turn off at the end of the list. For example, if you want to turn on SERVICE tracing except for VERBDETAIL, specify -TRACEFLAGS=SERVICE,-VERBDETAIL. If tracing is already running, then this option can be used to turn on additional trace flags or turn off trace flags.
-TRACEMax
-TRACEMax=<maximumTraceSize> This option limits the maximum trace file length to the specified value (by
212
default the trace file will grow indefinitely). When the maximum length is reached, then the trace will wrap back to the beginning of the file. Specify a value in MB between 1 and 4095. If tracing is already running, then this option has no effect.
-TRACESegsize
-TRACESegsize=<maximumTraceSegmentSize> This option is used when you anticipate a large trace file and you want the trace file to be written in smaller, more easily-manageable segments. Each segment will be no larger than the specified size. When this option is used, a segment number will be appended to the trace file name for each segment. Specify a value in MB between 1 and 1000. If tracing is already running, then this option has no effect. Note: v To turn tracing on for a process, you must use the -TRACEFLAGS and -TRACEFILE options (and -PIPENAMEPREFIX when tracing an API application). v To modify trace flags for an existing process, use -TRACEFLAGS (and -PIPENAMEPREFIX when tracing an API application). v If you need to modify the trace file name, maximum trace size, or maximum trace segment size, then you need to first disable tracing altogether (see the dsmtrace disable command).
Determining if data is encrypted or compressed during backup/archive through trace

You must perform several steps to determine if the data during backup-archive is compressed or encrypted, or both. 1. Add the trace options listed below to the client options file prior to backing up or archiving objects: v TRACEFILE <trace file name> v TRACEFLAGS api api_detail 2. Examine the trace file after the operation and locate a statement that looks similar to the following statement:
dsmSendObj ENTRY:... objNameP: <the file name>
This output is followed by the following trace message that indicates whether the object is compressed, encrypted, or both compressed and encrypted:
tsmEndSendObjEx: Total bytes send * *, encryptType is *** encryptAlg is *** compress is *, totalCompress is * * totalLFBytesSent * *
+----------------------------------------------------------------------------------+ | encryptType/compress | 0 | 1 | +----------------------------------------------------------------------------------+ | NO | not compressed, not encrypted | compressed, not encrypted | | CLIENTENCRKEY | not compressed, encrypted | compressed, encrypted | | USER | not compressed, encrypted | compressed, encrypted | +----------------------------------------------------------------------------------+
Alternatively, your application itself can determine encryption type/strength and compression of your data by using the dsmEndSendObjEx function call and the dsmEndSendObjExOut_t data structure.
Chapter 7. Trace
213
/*-------------------------------------------------------------------------+ | Type definition for dsmEndSendObjExOut_t +-------------------------------------------------------------------------*/ typedef struct dsmEndSendObjExOut_t { dsUint16_t stVersion; /* structure version */ dsStruct64_t totalBytesSent; /* total bytes read from app */ dsmBool_t objCompressed; /* was object compressed */ dsStruct64_t totalCompressSize; /* total size after compress */ dsStruct64_t totalLFBytesSent; /* total bytes sent LAN Free */ dsUint8_t encryptionType; /* type of encryption used */ }dsmEndSendObjExOut_t; objCompressed - A flag that displays if the object was compressed. encryptionType - A flag that displays the encryption type. For example: ... rc = dsmEndSendObjEx(&endSendObjExIn, &endSendObjExOut); if (rc) { printf("*** dsmEndSendObjEx failed: "); rcApiOut(dsmHandle, rc); } else { printf("Compression: %s\n", endSendObjExOut.objCompressed == bTrue ? "YES" : "NO"); printf("Encryption: %s\n", endSendObjExOut.encryptionType & DSM_ENCRYPT_CLIENTENCRKEY ? "CLIENTENCRKEY" : endSendObjExOut.encryptionType & DSM_ENCRYPT_USER ? "USER" : "NO"); printf("Encryption Strength: %s\n\n", endSendObjExOut.encryptionType & DSM_ENCRYPT_AES_128BIT ? "AES_128BIT" : endSendObjExOut.encryptionType & DSM_ENCRYPT_DES_56BIT ? "DES_56BIT" : "NONE"); } ...
See API Function Calls in Using the Application Programming Interface for more information.
Tracing for the API

You can enable tracing for the application programming interface (API). To enable tracing for the Tivoli Storage Manager API, add the following lines to the dsm.opt file or another file designated as the client options file:
TRACEFILE trace file name TRACEFLAGS trace flags
trace file name The name of the file where you want to write the trace data. trace flags The list of trace flags to enable. Separate each trace flag by a space. The following are the trace flags specific to the Tivoli Storage Manager API: api Information about the API function calls
api_detail Detailed information about the API function calls
214
You can also specify other Tivoli Storage Manager backup-archive client and Tivoli Storage Manager API trace flags. Refer to the backup-archive client documentation for a list of any available trace classes. For example: v TRACEFILE /log/trace.out v TRACEFLAGS api api_detail verbinfo verbdetail timestamp Note: If you do not have write permission for the file pointed by the TRACEFILE option, dsmSetup or dsmInitEx/dsmInit will fail with return code DSM_RC_CANNOT_OPEN_TRACEFILE (426). To enable tracing for the multi-threaded API after an application has been started, use the dsmtrace utility. The dsmtrace utility allows you to turn on tracing while the problem is occurring, without having trace constantly enabled. Refer to the dsmtrace section.
ODBC tracing
There are two components from which you must collect Tivoli Storage Manager ODBC driver trace data. When collecting trace data, you must collect the following: v ODBC Driver Manager trace v Tivoli Storage Manager ODBC driver trace Perform the following steps to collect trace data: 1. Activate the Tivoli Storage Manager ODBC driver tracing. To do this, start the ODBC Data Source Administrator (odbcad32.exe), select the data source, and click Configure. This opens the Configure a Tivoli Storage Manager Data Source dialog. Select the Enable trace check box and enter a fully-qualified trace file name. For example, C:\TSMODBC.TRC. Then click OK. Leave the ODBC Data Source Administrator running. 2. Activate ODBC Driver Manager tracing by clicking the Tracing tab in the ODBC Data Source Administrator and enter a fully-qualified trace file name in the Log file Path text box. For example, C:\ODBCMGR.TRC. Do not use the same name as in step 1. Click Apply. 3. Click Start Tracing Now (the buttons name will change to Stop Tracing Now). If the button already was labeled Stop Tracing Now, click the button to change it to Start Tracing Now, then click it again to restart the trace (the button will once again read Stop Tracing Now). Click OK. 4. Delete (or rename) any existing Tivoli Storage Manager ODBC driver trace. Using the same name as in step 1, delete or rename C:\TSMODBC.TRC (if it exists). 5. Delete (or rename) any existing ODBC Driver Manager trace. Using the same name as in step 2, delete or rename C:\ODBCMGR.TRC (if it exists). 6. Recreate the problem. 7. After the problem is recreated, rename the two trace files as different names. Using the sample names above, you could rename them as follows: C:\TSMODBC.TRC to C:\TSMODBC.KEEP.TRC C:\ODBCMGR.TRC to C:\ODBCMGR.KEEP.TRC 8. Zip up the renamed trace files (they may become large) and be prepared to send them in to IBM/Tivoli support for analysis.
Chapter 7. Trace
215
9. Deactivate ODBC Driver Manager tracing. Using the ODBC Data Source Administrator, select the Tracing tab and click Stop Tracing Now. Then click OK. 10. Deactivate Tivoli Storage Manager ODBC driver tracing. Using the ODBC Data Source Administrator, select the data source and click Configure. Clear the Enable trace check box, and then click OK.
216
Chapter 8. Hints and tips

The hints and tips that are gathered here are from actual problem experiences. You may find that one of the solutions is right for your Tivoli Storage Manager problem.
Device driver hints and tips

There are many possible causes for device driver problems. The problem may be with the operating system, the application using the device, the device firmware, or the device hardware itself. Whenever a device problem is encountered, ask Has anything been changed? If the adapter firmware changed, this may cause a device to exhibit intermittent or persistent failures. Try reverting back to an earlier version of the firmware to see if the problem continues. If cabling between the computer and the device was changed, this often accounts for intermittent or persistent failures. Check any cabling changes to verify that they are correct. A device might exhibit intermittent or persistent failures if the device firmware was changed. Try reverting back to an earlier version of the firmware to see if the problem continues. For SCSI connections, a bent pin in the SCSI cable where it connects to the computer (or device) can cause errors for that device or any device on the same SCSI bus. A cable with a bent pin must be repaired or replaced. Similarly, SCSI buses must be terminated. If a SCSI bus is improperly terminated, devices on the bus may exhibit intermittent problems, or data that is transferred on the bus may be or appear to be corrupted. Check the SCSI bus terminators to ensure that they are correct. Note: If the hints and tips information does not adequately address your device driver issue and/or this is the initial set up of your systems device drivers, please refer to the Tivoli Storage Manager Administrators Guide and Tivoli Storage Manager Installation Guide. Check, also, that your hardware devices aresupported by the Tivoli Storage Manager.
Performing a Tivoli Storage Manager device driver trace from Tivoli Storage Manager 5.2.2
For information on device driver trace from the Tivoli Storage Manager, see Enabling trace for the Tivoli Storage Manager device driver on page 191.
Adjusting to operating system changes

Operating system maintenance can change kernel levels, device drivers, or other system attributes that can affect a device. Similarly, upgrading the version or release of the operating system can cause device compatibility issues. If possible, revert the operating system back to the
217
state prior to the device failure. If this is not possible, check for device driver updates that may be needed based on this fix level, release, or version of the operating system.
Adjusting to changes in the HBA or SCSI adapter connecting to the device

A device driver communicates to a given device through an adapter. If it is a fibre channel-attached device, the device driver will use a host bus adapter (HBA) to communicate. If the device is SCSI attached, the device driver will use a SCSI adapter to communicate. In either case, if the adapter firmware was updated or the adapter itself was replaced, the device driver may have trouble using the device. Work with the vendor of the adapter to verify that it is installed and configured appropriately. The following is a list of other possible steps: v If the adapter was changed, try reverting back to the previous adapter to see if that resolves the issue. v If other hardware in the computer was changed or the computer was opened, reopen the computer and check to make sure that the adapter is properly seated in the bus. By opening and changing other hardware in the computer, the adapter cards and other connections in the computer may have been loosened, which may cause intermittent problems or total failure of devices or other system resources.
Adjusting to a loose cable connection

Problems with the device may occur if a connection is loose from the computer to the cable, or from the cable to the device. Check the connections and verify that the cable connections are correct and secure. For SCSI devices, check that the SCSI terminators are correct and that there are no bent pins in the terminator itself. An improperly terminated SCSI bus may result in difficult problems with one or more devices on that bus.
Resolving error messages in the system error log

A device may try to report an error to a system error log. Examples of various system error logs are: errpt for AIX, Event Log for Windows, and System Log for z/OS . The system error logs can be useful because the messages and information recorded may help to report the problem or the messages may include recommendations on how to resolve the problem. Check the appropriate error log and take any actions based on the messages issued to the error log.
Supporting 64-bit/32-bit Linux kernel modules for 32-bit/64-bit applications

The Linux kernel modules dictate the bit mode of the Linux SCSI generic (sg) device driver, all different Host Bus Adapter (HBA) drivers, and other settings. All of these kernel modules only support applications which have the same bit mode with running kernel modules. In other words, the 64-bit kernel modules only support 64-bit applications on 64-bit Linux systems.
218
If a 32-bit application runs on a 64-bit Linux system and invokes a 64-bit kernel module, the 32-bit application causes a kernel segmentation fault. The same will happen if a 64-bit application invokes a 32-bit kernel module on a 32-bit Linux system. To avoid this kind of problem, ensure that the bit mode of the Linux kernel module and its applications are the same. That means 32-bit applications can only invoke 32-bit kernel modules on 32-bit Linux systems. 64-bit applications can only invoke 64-bit kernel modules on 64-bit Linux systems.
Running a Tivoli Storage Manager Linux server on x86_64 architecture

The 32-bit and 64-bit Linux operating systems can run on the AMD64 and EM64T systems, which are 64-bit systems. A 64-bit Tivoli Storage Manager Linux server and storage agent can only run on a AMD64/EM64T system with a 64-bit Linux operating system. Likewise, a 32-bit Tivoli Storage Manager Linux server and storage agent can only run on an AMD64/EM64T system with 32-bit Linux operating system. A 64-bit Tivoli Storage Manager server issuing the QUERY SAN command requires a 64-bit HBA API on an AMD64/EM64T system. If an AMD64 system is equipped with a Qlogic HBA, it could create a problem since, by default, Qlogic only provides 32-bit HBA API on AMD64 system. You must install the 64-bit HBA API on the system before issuing the 64-bit QUERY SAN command.
Adjusting to HBA driver changes on the Linux 2.6.x kernels

All drivers have ko as a new suffix at 2.6.x kernels. The following are the driver names and locations in 2.6.x kernels: Adaptec The driver (aic7xxx.ko) is located in the /lib/modules/kernel-level/ drivers/scsi/aic7xxx/ directory. Emulex The driver (lpfcdd.ko) is located in the /lib/modules/kernel-level/drivers/ scsi/lpfc/ directory. Qlogic Its driver names are qla2xxx.ko, qla2100.ko, qla2200.ko, qla2300.ko, qla2322.ko, and so on. There is a certain order to load the HBA drivers. The qla2xxx.ko is a base driver and should be loaded first. After loading the qla2xxx.ko driver, the system should then load the qla2300.ko driver if it is equipped with a Qla2300 card. All of drivers are located in the /lib/modules/kernel-level/drivers/scsi/qla2xxx/ directory.
Requirements for multiple LUN support on Linux kernels

In order to configure SCSI devices with multiple LUNs on a Linux system, the Linux kernel must be set to enable multiple LUN support. Multiple LUN support on some Linux distributions, however, is not a default option and requires users to manually add this option to the running kernel. Perform the following steps to set up multiple LUN enabled on IA32 architecture: 1. Add one parameter to a boot loader configuration file.
219
v For LILO boot loader: a. Add append=max_scsi_luns=128 to the /ect/lilo.conf file. b. Run lilo. v For GRUB boot loader: a. Add max_scsi_luns=128 after the kernel image list at /etc/grub.conf file for RedHat distribution. b. Add max_scsi_luns=128 after the kernel image list at /boot/grub/menu.1 file for SuSE distribution. 2. Restart the system.
Using Tivoli Storage Manager Version 5.3.2 to perform ddtrace on Linux

In Tivoli Storage Manager Version 5.3.2 and later, the passthru device driver can be traced by issuing the ddtrace command. To enable trace, issue the following commands from the server console or admin client:
trace enable lpdd <other server trace class names> trace begin <file name>
Select one of the following three options: v ddtrace start librarydd tapedd (to trace both library and drive) v ddtrace start librarydd (library trace only) v ddtrace start tapedd (drive trace only) Note: DDTRACE GET and DDTRACE END are not required. The Tivoli Storage Manager passthru device driver trace cannot be enabled through the ddtrace utility.
Updating device information of host systems on a dynamic SAN without rebooting

When devices in a SAN (Storage Area Network) environment change, the information about this changed environment is not automatically sent to host systems attached to the SAN. If the device information has not been updated to host systems attached to the SAN, previously-defined device paths will no longer exist. If you use the existing device information to define device paths, backup, or restore data, these operations may fail. In order to avoid these kinds of failures, use a different method for different platforms to update the device information on the SAN without rebooting host systems. On AIX, issue the cfgmgr command to force the operating system to re-configure itself. Then run SMIT to re-configure your Tivoli Storage Manager devices. On HP, issue the ioscan command with the -kfn option to force the operating system to re-scan SCSI buses and fibre channels. Then run tsmddcfg to configure your Tivoli Storage Manager devices. On Linux, there is no system command to re-configure the operating system. In order to re-scan SCSI buses and fibre channels, the adapter drivers corresponding
220
to these SCSI adapters and fibre channel adapters must be unloaded and then reloaded into the Linux kernel. After reloading HBA drivers, run autoconf or tsmscsi to re-configure Tivoli Storage Manager devices on Linux. You might issue the lspci command to find out which SCSI adapter and fibre channel adapter is available on the system. The rmmod command unloads a driver from the kernel and the modprobe command loads a driver to the kernel.
Table 6. HBA adapters and corresponding drivers for all architectures of Linux HBA Adapters Adaptec 7892 Qlogic 22xx Qlogic 23xx Qlogic 2362 Emulex HBA Driver Name aix7xxx qla2200 qla2300 qla2362 lpfcdd Available Architectures IA32, AMD64 IA32, AMD64 IA32, AMD64 EM64T IA32, iSeries, pSeries
Requirements for Adaptec SCSI and Qlogic Fibre Channel HBA BIOS settings on Linux
By default, Adaptec SCSI adapters set the multiple LUN (Logical Unit Number) option tooff in their BIOS. This makes the SCSI adapter driver unable to probe a SCSI unit with multi-LUN properly. This option must be turned on. Perform the following steps to turn on the multiple LUN options: 1. Press the Ctrl and A keys at the same time. 2. Select SCSI Device Configuration in theConfigure/View Host Adapter Setting. 3. Change No to Yes for Bios Multiple LUN support. By default, Qlogic Fibre HBAs set the tape enable option as off in their BIOS. This affects the execution of some SCSI commands on several SCSI tape devices, so this option must be turned on. Perform the following steps to turn on the tape enable option: 1. Press the Alt and Q keys at the same time. 2. Select Advanced Settings. 3. Change Disable to Enable for Fibre Channel Tape Support.
Hard disk drives and disk subsystems hints and tips

The Tivoli Storage Manager server requires specific behaviors from hard disk drives, disk subsystems, third-party file systems, and remote file systems. These required behaviors allow Tivoli Storage Manager to appropriately manage and store data by ensuring the integrity of the Tivoli Storage Manager server itself. The following definitions are provided to help you better understand the hard disk drives and disk subsystems: Hard disk drive A hard disk drive storage device is typically installed inside a given computer and used for storage by a Tivoli Storage Manager server on that machine. Disk subsystem An external disk subsystem that connects to a computer through a SAN or
221
some other mechanism. Generally, disk subsystems are outside of the computer to which they are attached and may be located in close proximity or they may be located much farther away. These subsystems may also have some method of caching the I/O requests to the disks. Disk subsystems often have their own configuration and management software. The Tivoli Storage Manager server may define hard disk drives and disk subsystems used by the computer or operating system on the machine where the Tivoli Storage Manager is installed. Typically, a hard disk drive or disk subsystem is defined to the computer where the Tivoli Storage Manager is installed as a drive or file system. After the hard disk drive or disk subsystem is defined to the operating system, Tivoli Storage Manager may use this space by allocating a database, recovery log, or storage pool volume on the device. The Tivoli Storage Manager volume subsequently looks like another file on that drive or file system. Tivoli Storage Manager requires the following conditions for hard disk drives, disk subsystems, third-party file systems, and remote file systems: v When the Tivoli Storage Manager opens database, recovery log, and storage pool volumes, they are opened with the appropriate operating system settings to require data write requests to bypass any cache and be written directly to the device. By bypassing cache during write operations, Tivoli Storage Manager can maintain the integrity of client attributes and data. This is required because if an external event, such as a power failure, causes the Tivoli Storage Manager server or the computer where the server is installed to halt or crash while the server is running, the data in the cache may or may not be written to the disk. If the Tivoli Storage Manager data in the disk cache is not successfully written to the disk, information in the server database or recovery log may not be complete, or data that was supposed to be written to the storage pool volumes may be missing. This is less of an issue for hard disk drives installed in the computer where the Tivoli Storage Manager server is installed and running. In this case, the operating system settings that are used when the Tivoli Storage Manager opens volumes on that hard disk drive generally manage the cache behavior appropriately and honor the request to prevent caching of write operations. Typically, the use and configuration of caching for disk subsystems is a greater issue. The reason for this is that disk subsystems often do not receive information from the operating system about bypassing cache for write operations, or else they ignore this information when a volume opens. Because of this, the caching of data write operations may result in corruption of the Tivoli Storage Manager server database or loss of client data, or both, depending upon which Tivoli Storage Manager volumes are defined on the disk subsystem and the amount of data lost in the cache. Disk subsystems should be configured to not cache write operations when a Tivoli Storage Manager database, recovery log, or storage pool volume is defined on that disk. Another alternative is to use non-volatile cache for the disk subsystem. Non-volatile cache employs a battery backup or some other sort of scheme to allow the contents of the cache to be written to the disk if a failure occurs. v The size and location of the Tivoli Storage Manager database, recovery log, and storage pool volumes (files) can not change after they are defined and used by the server. If the size is changed or the file is moved, internal information that the Tivoli Storage Manager uses to describe the volume may no longer match the actual attributes of the file. If you need to move or change the size of a Tivoli Storage Manager database, recovery log, or storage pool volume, move any existing data to other volumes prior to altering or moving the database.
222
FILE directory mapping between storage agents and servers for shared files
Tivoli Storage Manager servers and storage agents can access the same data in File device classes by defining a set of directories that should be used within a device class definition. For each storage agent that will be sharing FILE access, the PATHs defined to each DRIVE seen by the storage agent must provide access to the same set of directories. When the PATHs are defined, the directories for each storage agent must match in number and ordering for the directories as listed in the device class definition on the server. If these definitions are out of sync, the storage agent may be unable to access the FILE volumes. This results in successful LAN-restores and mount failures for the LANFREE-restores.
Tape drives and libraries hints and tips

There are several possible causes for problems with tape drives and libraries. The problem may be with software on the machine trying to use the device, the connections to the device, or the device itself. Whenever a device problem is encountered, ask, Has anything been changed? Suspect anything from the machine trying to use the device to the device itself. This, especially if the device worked prior to a given change then stopped working after that change. v If the adapter firmware changed, a device might exhibit intermittent or persistent failures. Try reverting back to an earlier version of the firmware to see if the problem continues. v If cabling between the computer and the device was changed, this often accounts for intermittent or persistent failures. Check any cabling changes to verify that they are correct. v If the device firmware has changed, this may cause a device to exhibit intermittent or persistent failures. Try reverting back to an earlier version of the firmware to see if the problem continues.
Adjusting to operating system changes

Operating system maintenance can change kernel levels, device drivers, or other system attributes that can affect a device. Similarly, upgrading the version or release of the operating system can cause device compatibility issues. If possible, revert the operating system to the state prior to the device failure. If this is not possible, check for device driver updates that may be needed based on this fix level, release, or version of the operating system.
Adjusting to device driver changes

A device driver upgrade can result in a tape drive or library device not working. These issues can also occur as a result of the type of driver that you use. When working with IBM libraries or drives, as opposed to using other vendor libraries and drives, the type of device driver that you choose is important. IBM libraries and drives should use the IBM device driver, while other vendor libraries and drives should use the Tivoli Storage Manager device driver.
223
Try reverting to the previous (or earlier) version of the device driver to see if the problem was introduced by the newer version of the driver.
Adjusting to a replaced adapter or other hardware changes

The connecting point for the device to the computer is usually referred to as an adapter. Another term for adapter is card. A SCSI connection to the device uses a SCSI adapter, while a fibre-channel (optical) connection to the device uses a host bus adapter (HBA). In either case, if an actual adapter was changed or the computer was opened and other hardware changed or fixed, this may be the cause of the problem. v If the adapter was changed, try reverting back to the previous adapter to see if that resolves the issue. v If other hardware in the computer was changed or the computer was opened, reopen the computer and check to make sure that the adapter is properly seated in the bus. By opening and changing other hardware in the computer, the adapter cards and other connections in the computer may have been loosened, which may cause intermittent problems or total failure of the devices or other system resources.
Adjusting to a loose cable connection

Problems might occur to the device if a connection is loose from the computer to the cable, or from the cable to the device. Check the connections and verify that the cable connections are correct and secure. For SCSI devices, check that the SCSI terminators are correct and that there are no bent pins in the terminator itself. An improperly-terminated SCSI bus may result in difficult problems with one or more devices on that bus.
Using error messages

A device may try to report an error to a system error log. Examples of various system error logs are: v errpt for AIX v Event Log for Windows v System Log for z/OS The system error logs can be useful because the messages and information recorded may help to report the problem, or the messages may include recommendations on how to resolve the problem. Check the appropriate error log and take any recommended actions based on messages issued to the error log.
Miscellaneous server information hints and tips

Running the AIX Tivoli Storage Manager server from a location other than the installation directory
If you attempt to run the Tivoli Storage Manager server from a directory other than the installation directory, this may result in the server failing to start. The following messages can be displayed: v ANR0109E Attempt number 1 Unable to load cryptography module from .. v ANR0109E Attempt number 2 Unable to load cryptography module from ..
224
v ANR9999D cryptog.c(253): ThreadId<0> ICC load failed. To run a Tivoli Storage Manager server from a directory other than the Tivoli Storage Manager server installation directory, the DSMSERV_DIR environment variable must be set to the path of the server installation directory. The Tivoli Storage Manager server uses the installation directory as a reference to locate executables and other files in the Tivoli Storage Manager server installation. Perform the following steps to run a Tivoli Storage Manager server from a directory other than the installation directory: 1. Set the DSMSERV_DIR environment variable to the base installation directory of the Tivoli Storage Manager server (typically: /usr/tivoli/tsm/server/bin). 2. Add the path of the installation directory (/usr/tivoli/tsm/server/bin) to your PATH environment variable. 3. Follow the steps on Running Multiple Servers on a Single Machine in the Tivoli Storage Manager for AIX Installation Guide.
Storage area network hints and tips

There are many possible causes or problems with a storage area network (SAN). SAN problems may be with software on the machine trying to use the device, connections to the device, or the device itself. Whenever a SAN problem is encountered, ask Has anything been changed? Any kind of changes, from the machine trying to use the device to the device itself may be suspect, especially if the device worked prior to a given change, then stopped working after that change. To better understand the following discussion on diagnosing problems with a SAN, review the following terminology and typical abbreviations that are used: Fibre channel Fibre channel (FC) denotes a fibre-optical connection to a device or component. Host bus adapter A host bus adapter (HBA) is used by a given machine to access a storage area network. An HBA is similar in function to a network adapter in how it provides access for a machine to a local area network or wide area network. Storage area network A storage area network (SAN) is a network of shared devices that can typically be accessed using fibre. Often, a storage area network is used to share devices between many different machines.
Knowing your SAN configuration

Understanding the storage area network (SAN) configuration is critical in SAN environments. Various SAN implementations have limitations or requirements on how the devices are configured and set up. The following are the three SAN configurations: Point to point The devices are connected directly to the HBA.
225
Arbitrated loop Arbitrated loop topologies are ring topologies and are limited in terms of the number of devices that are supported on the loop and the number of devices that can be in use at a given time. In an arbitrated loop, only two devices can communicate at the same time. Data being read from a device or written to a device is passed from one device on the loop to another until it reaches the target device. The main limiting factor in an arbitrated loop is that only two devices can be in use at a given time. Switched fabric In a switched fabric SAN, all devices in the fabric will be fibre native devices. This topology has the greatest bandwidth and flexibility because all devices are available to all HBAs through some fibre path.
Devices supported by the SAN

Many devices or combinations of devices may not be supported in a given storage area network (SAN). These limitations arise from the ability of a given vendor to certify their device using Fibre Channel Protocols. For a given device, verify with the device vendor that it is supported in a SAN. This includes whether or not it is supported by the HBAs used in your SAN environment. This means that you must verify with the vendors that this device is supported by the hubs, gateways, and switches that make up the SAN.
Considerations for an HBA

The HBA is a critical device for the proper functioning of a storage area network (SAN). The problems that might occur with HBAs range from improper configuration to outdated bios or device drivers. For a given HBA, check the following: BIOS HBAs have an embedded BIOS that can be updated. The vendor for the HBA has utilities to update the BIOS in an HBA. Periodically, the HBAs in use on your SAN should be checked to see if there are BIOS updates that should be applied.
Device driver HBAs use device drivers to work with the operating system to provide connectivity to the SAN. The vendor will typically provide a device driver for use with their HBA. Similarly, the vendor will provide instructions and any necessary tools or utilities for updating the device driver. Periodically the device driver level should be compared to what is available from the vendor and, if needed, should be updated to pick up the latest fixes and support. Configuration HBAs typically have a number of configurable settings. The settings typically affect how Tivoli Storage Manager functions with a SAN device. See HBA configuration issues.
HBA configuration issues

HBAs typically have many different configuration settings and options. The HBA vendor usually provides information about the settings for your HBA and the appropriate values for these settings. Similarly, the HBA vendor should
226
provide a utility and other instructions on how to configure your HBA. The following are settings that typically affect using Tivoli Storage Manager with a SAN: v Storage area network (SAN) topology The HBA should be set appropriately based on the currently-used SAN topology. For example, if your SAN is an arbitrated loop, the HBA should be set for this configuration. If the HBA connects to a switch, this HBA port should be set to point to point and not loop. With Tivoli Storage Manager SAN Device Mapping, you can perform SAN discovery on most of the platforms and the persistent binding of the devices are not required. A Tivoli Storage Manager server can find the device if the device path was changed due to a restart or other reason. Go to IBM Support to verify the platform/HBA vendor/driver level support for Tivoli Storage Manager SAN discovery. v FC link speed In many SAN topologies, the SAN is configured with a maximum speed. For example, if the FC switch maximum speed is 1 GB/sec, the HBA should also be set to this same value. Or the HBA should be set for automatic (AUTO) negotiation if the HBA supports this capability. v Is fibre channel tape support enabled? Tivoli Storage Manager requires that an HBA is configured with tape support. Tivoli Storage Manager typically uses SANs for access to tape drives and libraries. As such, the HBA setting to support tapes must be enabled.
FC switch configuration issues

An FC switch typically supports many different configurations. The ports on the switch need to be configured appropriately for the type of storage area network (SAN) that is set up, as well as the attributes of the SAN. The vendor for the switch usually provides information about the appropriate settings and configuration based upon the SAN topology being deployed. Similarly, the switch vendor should provide a utility and other instructions on how to configure it. The following are settings that typically affect how Tivoli Storage Manager uses a switched SAN: FC link speed In many SAN topologies, the SAN is configured with a maximum speed. For example, if the FC switch maximum speed is 1 GB/sec, the HBA should also be set to this same value. Or the HBA should be set for automatic (AUTO) negotiation if the HBA supports this capability. Port mode The ports on the switch must be configured appropriately for the type of SAN topology being implemented. For example, if the SAN is an arbitrated loop, the port should be set to FL_PORT. For another example, if the HBA is connected to a switch, the HBA options should be set to point-to-point and not loop.
Verifying the data gateway port settings

A data gateway in a storage area network (SAN) translates fibre channel to SCSI for SCSI devices attached to the gateway. Data gateways are popular in SANs because they allow the use of SCSI devices, therefore it is important that the port settings for a data gateway are correct.
227
The vendor for the data gateway usually provides information about the appropriate settings and configuration based upon the SAN topology being deployed and SCSI devices used. Similarly, the vendor might provide a utility and other instructions on how to configure it. The following settings can be used for the FC port mode on the connected port on a data gateway: Private target Only the SCSI devices attached to the data gateway are visible and usable from this port. For the available SCSI devices, the gateway simply passes the frames to a given target device. Private target port settings are typically used for arbitrated loops. Private target and initiator Only the SCSI devices attached to the data gateway are visible and usable from this port. For the available SCSI devices, the gateway simply passes the frames to a given target device. As an initiator, this data gateway may also initiate and manage data movement operations. Specifically, there are extended SCSI commands that allow for third-party data movement. By setting a given port as an initiator, it is eligible to be used for third-party data movement SCSI requests. Public target All SCSI devices attached to the data gateway, as well as other devices available from the fabric, are visible and usable from this port. Public target and initiator All SCSI devices attached to the data gateway, as well as other devices available from the fabric, are visible and usable from this port. As an initiator, this data gateway may also initiate and manage data movement operations. Specifically, there are extended SCSI commands that allow for third-party data movement. By setting a given port as an initiator, it is eligible to be used for third-party data movement SCSI requests.
Verifying the SAN configuration between devices

Devices in a storage area network (SAN), such as a data gateway or a switch, typically provide utilities that display what that device sees on the SAN. It is possible to use these utilities to better understand and troubleshoot the configuration of your SAN. The vendor for the data gateway or switch should provide a utility for configuration. As part of this configuration utility, typically there is information about how this device is configured and other information that this device sees in the SAN topology, of which it is a part. These vendor utilities can be used to verify the SAN configuration between devices: Data gateway A data gateway reports all the FC devices as well as the SCSI devices that are available in the SAN. Switch A switch reports information about the SAN fabric. Tivoli Storage Manager Management Console The Tivoli Storage Manager management console will display device names and the paths to those devices. This can be useful to help verify that the definitions for Tivoli Storage Manager match what is actually available.
228
Monitoring the fibre-channel-link error report

Most storage area network (SAN) devices provide monitoring tools that can be used to report information about errors and performance statistics. The vendor for the device should provide a utility for monitoring. If a monitoring tool is available, it will typically report errors. The following are the errors that are often seen: v CRC error, 8b/10b code error, and other similar symptoms These are recoverable errors where the error handling is usually provided by firmware or hardware. In most cases, the method to recover the device is to retransmit the failing frame. The FC link is still active when these errors are encountered. Applications using a SAN device that encounter this type of link error usually are not aware of the error unless it is a solid error. A solid error is one where the firmware and hardware recovery cannot successfully retransmit the data after repeated attempts. The recovery for these types of errors is typically very fast and will not cause system performance to degrade. v Link failure (loss of signal, loss of synchronization, NOS primitive received) This error indicates that a link is actually broken for a period of time. It is likely that a faulty gigabit interface connector (GBIC), media interface adapter (MIA), or cable. The recovery for this type of error is disruptive. This error will appear in the application using the SAN device that encountered this link failure. The recovery is at the command exchange level and involves the application and device driver having to perform a reset to the firmware and hardware. This will cause the system to run degraded until the link recovery is complete. These errors should be monitored closely, as they typically affect multiple SAN devices. Note: Often these errors are caused by a customer engineer (CE) action to replace a SAN device. As part of the maintenance performed by the CE to replace or repair a SAN device, the fibre cable might be temporarily disconnected. If this is the case, the time and duration of the error should correspond to when the service activity was performed.
NDMP filer-to-TSM server operations

During filer-to-server operations, the Tivoli Storage Manager uses up to two additional TCP/IP ports; a control port used internally by Tivoli Storage Manager during both backup and restore operations, and a data port used during NDMP backup operations to a Tivoli Storage Manager native storage pool. Tivoli Storage Manager defaults to the standard NDMP control port of 10000. If this port is in use by another application (such as a second Tivoli Storage Manager server), all filer-to-server operations fail. To avoid conflicts with other applications, use the NDMPCONTROLPORT server option to specify a different port for your server. The data port is an ephemeral port that is acquired at the beginning of NDMP backup operations to a Tivoli Storage Manager native storage pool. If a port is not available, an error message is issued and backup of NAS devices to Tivoli Storage Manager native pools is not possible. To avoid conflicts with other applications, you can control which port is acquired for use as the data port during NDMP backup operations by setting the NDMPPORTRANGELOW and
229
NDMPPORTRANGEHIGH server options. Note that a data port is not needed by the Tivoli Storage Manager server for NAS restores from Tivoli Storage Manager native pools.
Firewall considerations with NDMP Filer-to-TSM server backup and restore

A firewall may prevent the NAS file server from contacting the Tivoli Storage Manager server on the acquired data port during NAS backup operations to a native storage pool. If you must modify the data port selected by the Tivoli Storage Manager server, use the NDMPPORTRANGELOW and NDMPPORTRANGEHIGH server options. A firewall may prevent the Tivoli Storage Manager server from contacting the NAS file server on the configured data port during NAS restore operations from a native storage pool. If a firewall prevents Tivoli Storage Manager from accessing the NAS file server, the outbound connection from Tivoli Storage Manager fails. If you must modify the data port selected by NAS file server, consult the file servers documentation.
SAN devices
You may be experiencing problems with your storage agent SAN devices. For additional error messages related to SAN, see SAN device mapping problems on page 234.
ANR8302E I/O error on drive TSMDRIVE01 (/dev/mt9) (OP=WRITE, Error Number=5, CC=205, KEY=FF, ASC=FF, ASCQ=FF, SENSE=**NONE**, Description=General SCSI failure). Refer to Appendix D in the Messages manual for recommended action
For SAN device errors, this message is issued in many cases. The CC=205 reports that the device driver detects a SCSI adapter error. In the case of a SAN-attached device that encounters a link reset caused by link loss, it will be reported back to the device driver as a SCSI adapter error. The underlying cause of this error is the event that caused the link reset due to the link loss. The path for this device should be updated to ONLINE=NO by issuing the UPDATE PATH command. Do not set the path to ONLINE=YES until the cause for the link reset was isolated and corrected.
ANR8957E: command : Autodetect is OFF and the serial number reported by the library did not match the serial number in the library definition
The Tivoli Storage Manager SAN Device Mapping encountered a path for the library that reports a different serial number than the current Tivoli Storage Manager definition for the library. The AUTODETECT parameter was set to NO for the command which prevented the server from updating the serial number for the library. For the Tivoli Storage Manager 5.2 on Windows with systems using QLogic host bus adapters, reissue the command with AUTODETECT=ON. The newly-discovered serial number is automatically updated for this library. The following informational message is issued:
230
ANR8953I: Library libraryName with serial number serialNumber is updated with the newly-discovered serial number newSerialNumber For other Tivoli Storage Manager servers, determine the new path and issue the UPDATE PATH command to correct this.
ANR8958E: command : Autodetect is OFF and the serial number reported by the drive did not match the serial number in the drive definition
Tivoli Storage Manager SAN Device Mapping encountered a path for a drive that reports a different serial number than the current Tivoli Storage Manager definition for that drive. The AUTODETECT parameter was set to NO for the command, which prevents the server from updating the serial number for this drive. For Tivoli Storage Manager 5.2 on Windows with systems using QLogic host bus adapters, reissue the command with AUTODETECT=ON. The newly-discovered serial number is automatically updated for this drive. The following informational message is issued: ANR8955I: Drive driveName in library libraryName with serial number serialNumber is updated with the newly discovered serial number newSerialNumber. For other Tivoli Storage Manager servers, determine the new path and issue the UPDATE PATH command to correct this.
ANR8963E: Unable to find path to match the serial number defined for drive driveName in library libraryName
For the Tivoli Storage Manager 5.2 Windows servers and storage agents, the server automatically maps SAN devices. This SAN Device Mapping feature allows the server to compare its definitions for SAN devices to the actual definition for these devices. If a discrepancy is detected, the server definition for this device is updated to correct this. The SAN Device Mapping was not able to find a SAN device that was previously defined to the server. The most likely cause for this is that the device itself has been removed or replaced in the SAN. The following are the possible steps to resolve this: v Device Removed If the device was removed from the SAN, simply delete the server definitions that refer to this device. Issue the QUERY PATH F=D command to determine any paths that reference the device. Then issue the DELETE PATH command to remove these paths. v Device Replace A SAN Device was replaced with a new device as a result of maintenance or an upgrade. Perform the following procedures: Try not to delete the drive or drive path definition after you replace the drive. Issue one of the following server commands: - UPDate DRive <libraryName> <driveName> SERIAL=AUTODetect This command will force-record the new serial number into the server database. Because the drive was replaced, the element number stays the same.
231
- UPDate PATH <sourceName> <driveName> SRCT=SERVER DESTT=DRIVE LIBRary=<libraryName> DEVIce=xxxxx AUTODetect=Yes This command will force-record the new serial number into the database. Because the drive was replaced, the element number stays the same. If the drive or drive path was deleted, redefine this new, replaced drive. You must restart the Tivoli Storage Manager server so that the element number/serial number map for the library can be refreshed. This mapping only occurs at initialization. Issue the QUERY PATH F=D command to find any paths defined on the server that reference this device, then issue the following command to update the path information:
UPDATE PATH AUTODetect=Yes
ANR8965W: The server is unable to automatically determine the serial number for the device
For a Tivoli Storage Manager 5.2 non-Windows server or an Tivoli Storage Manager 5.3 server with the SANDISCOVERY option OFF, the Tivoli Storage Manager server does not automatically detect SAN device serial numbers. If the Tivoli Storage Manager server detects that a device serial number was changed, it marks the devices path as offline. Perform the following steps to correct this: 1. Determine the correct serial number for the device. 2. Issue the UPDATE DRIVE or UPDATE LIBRARY command to update the device serial number.
ANR8972E: Unable to find element number for drive driveName in library libraryName
If the ELEMent parameter was set to AUTODetect when defining the drive, the Tivoli Storage Manager tries to get the drives element number automatically. However, if the library does not provide an element number/serial number map, this message is issued. Perform the following steps to correct this: 1. Determine the element number for this tape drive. 2. Issue the UPDATE DRIVE command to update the device element number.
SAN device mapping hints and tips

Starting with Tivoli Storage Manager 5.3.0, Storage Area Network (SAN) device discovery and device mapping are supported on Windows 2000, Windows 2003 (32 bits), AIX, Solaris, and Linux (except Linux zSeries). The following are the advantages of Tivoli Storage Manager SAN device discovery and device mapping: Tivoli Storage Manager can display all the devices on the SAN The QUERY SAN server command shows all the devices seen by the Tivoli Storage Manager server via all the Fibre Channel host bus adapters (HBAs) installed on the system. The parameters shown are device type, vendor name, product model name, serial number, and the device name. If FORMAT=DETAIL is specified for the query, additional information such as
232
World Wide Name (WWN), port, bus, target, and LUN are displayed. This information will help identify all the tape, disk, and Data Mover devices on the SAN. For AIX, the data mover is transparent and is not shown. Tivoli Storage Manager can update the device path automatically when a devices path changes Tivoli Storage Manager does not require persistent binding for the devices it sees via HBA. Instead, the server uses the SNIA HBAAPI to discover and obtain the serial number for all the devices on the SAN. It can also determine each devices path. By comparing a devices serial number recorded in the Tivoli Storage Manager database with the serial number obtained from the device in real time, a change in a devices path is detected. If the path was changed, SAN discovery automatically performed to obtain the new path for the device. The Tivoli Storage Manager database is also updated with the new path information. The HBAAPI wrapper library is the wrapper used by the Tivoli Storage Manager server to communicate with the SNIA HBAAPI. The HBAAPI wrapper library is installed in the same directory as the Tivoli Storage Manager executable (unless the full path is given). The following is a list of the HBA wrapper files shipped with the Tivoli Storage Manager server package (except on AIX): v Windows: hbaapi.dll v AIX: /usr/lib/libhbaapi.a (This is provided by AIX with HBAAPI installation) v 32-bit Linux: libhbaapi32.so v 64-bit Linux: libhbaapi64.so v 32-bit Solaris: libhbaapi32.so v 64-bit Solaris: libhbaapi64.so If any of these files are missing, message ANR1791W HBAAPI wrapper library xxxxxxxxx failed to load or is missing. is displayed.
Disabling SAN device mapping

Occasionally you must disable SAN device mapping to circumvent a problem or to isolate a problem when you are troubleshooting device problems. Perform the following step to disable SAN device mapping and device discovery: Issue the setopt SANDISCOVER OFF server command. The setopt SANDISCOVERY commands can be issued as many times as needed. Note: Another way to disable/enable SAN discovery is to enter the following option in the dsmserv.opt file: SANDISCOVERY OFF disables SAN discovery. SANDISCOVERY ON enables SAN discovery. SANDISCOVERY ON is the default for the Windows platform. SANDISCOVERY OFF is the default for all non-Windows platforms.
Platform-specific information
When working on your SAN device mapping, it is important that you know your platform-specific information.
233
AIX
Query SAN command will NOT show any Gateway devices because Gateway devices are transparent to AIX.
Linux There are separate libraries, utilities, and other items for RHEL3U3. To run them you must also install an Emulex ioctl kernel module in addition to the Emulex driver. Make sure to load the Emulex driver before loading the ioctl module. Emulex has provided an Application kit for RHEL3. To find the Emulex application kit, go to the Emulex website and click on Support. Under Choose your supplier from the following list, select vendor IBM. A list of drivers and kits are available for you to download. Note: For a list of supported HBAs and required driver levels by operating system, go to IBM Support and search on keyword TSMHBA.
SAN device mapping problems

The SAN device mapping problems that are documented are the more common issues that you might have. If you need additional help, see SAN device mapping hints and tips on page 232.
ANR1745I Unable to discover SAN devices. Function is busy

This occurs if there is another active SAN discovery. RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery. Retry again after the other SAN discovery is completed.
ANR1786W, ANR1787W or ANR1788W messages were issued

The following three messages usually indicate that the HBAAPI library is not working: v ANR1786W HBAAPI not able to get adapter name v ANR1787W Not able to open adapter adaperName v ANR1788W Not able to get the adapter attributes for adapterName RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery. Go to Support to verify that the HBA driver is up-to-date and at a supported level.
ANR1789W Get HBA target mapping failed

This is the most common error on the SAN with HBAAPI. This means that the HBA encountered an error while gathering device mapping information by sending various SCSI commands. Verify that all SAN devices are working properly (e.g. a SAN Data Gateway could be hung and need to be rebooted). If all devices appear functional, verify the firmware of device on the SAN, as well as HBA driver, are at the appropriate levels. RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery. Go to Support to verify that the HBA driver is up-to-date and at a supported level.
234
For IBM tape devices, make sure the latest firmware is installed. Firmware prior to 4772 for IBM 3580 tape devices caused problems with Qlogic HBAAPI.
ANR1790W SAN discovery failed

This message is a general message indicating that the HBAAPI function failed and we cannot perform SAN discovery. Verify that all SAN devices are working properly (e.g. a SAN Data Gateway could be hung and need to be rebooted). If all devices appear functional, verify that the firmware of device on the SAN, as well as the HBA driver, are at the appropriate levels. For IBM tape devices, make sure the latest firmware is installed. Firmware prior to 4772 for IBM 3580 tape devices causes problems with Qlogic HBAAPI. RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery.
ANR1791W HBAAPI wrapper library xxxxx failed to load or is missing

The HBAAPI wrapper library is the wrapper used by the Tivoli Storage Manager server to communicate with the SNIA HBAAPI. The HBAAPI wrapper libraries in the same directory as the Tivoli Storage Manager executable (unless full path is given as shown below). The following is a list of the HBA wrapper files shipped with the Tivoli Storage Manager server package (except on AIX). This warning message indicates that the HBAAPI wrapper file is either missing or could not be loaded by the Tivoli Storage Manager. Verify that the wrapper file is in the same directory as the Tivoli Storage Manager executable. The following are the HBAAPI wrapper library files: v Windows : hbaapi.dll v AIX : /usr/lib/libhbaapi.a (This is provided by AIX with HBAAPI installation) v 32-bit Linux: libhbaapi32.so v 64-bit Linux: libhbaapi64.so v 32-bit Solaris: libhbaapi32.so v 64-bit Solaris: libhbaapi64.so RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery.
ANR1792W HBAAPI vendor library failed to load or is missing

This message indicates that the vendors library file failed to load. Verify validity of these files. UNIX systems store their HBAAPI libraries in the location specified by the /etc/hba.conf file, and Windows file are stored in the C:\winnt\system32 directory. The following are examples of vendor library files: v C:\winnt\system32\qlsdm.dll (QlLogics Windows file) v /usr/lib/libHBAAPI.a (Emulexs AIX file) v /usr/lib/libqlsdm.so (Qlogics Linux file) v /usr/lib/libemulexhbaapi.so (Emulexs Linux 32-bit file) v /usr/lib64/libemulexhbaapi.so (Emulexs Linux 64-bit file) v /usr/lib/libqlsdm.so (Qlogics Solaris file)
235
v /opt/JNIsnia/Solaris/Jni/64bit/JniHbaLib.so (JNIs Solaris file) RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery.
ANR1793W Tivoli Storage Manager SAN discovery is not supported on this platform or this version of OS
This message is only displayed if the Tivoli Storage Manager attempts a SAN device mapping or device discovery operation on an unsupported operating system. The following platforms are not currently supported by SAN device mapping or device discovery: v HP-UX v 64-bit Windows 2003 v Linux zSeries v AIX versions that are not 52L or 53A. Support for SAN device mapping and device discovery on AIX requires either version 52L (fileset level of 5.2.0.50) or 53A (fileset level of 5.3.0.10) or higher. RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery.
ANR1794W Tivoli Storage Manager SAN discovery is disabled by options

The message indicates that the SAN discovery on Tivoli Storage Manager server is disabled. RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery. The SAN discovery can be disabled by issuing the following server command: v setopt SANDISCOVERY OFF This command disables the SAN discovery. The Tivoli Storage Manager server is not able to correct the device path automatically if the path was changed. This only has to be issued once. v setopt SANDISCOVERY ON This enables the SAN discovery. The setopt SANDISCOVERY ON command can be issued as many times as necessary. Another way to disable/enable SAN discovery is to put the following option in the dsmserv.opt file: v SANDISCOVERY OFF This disables the SAN discovery. v SANDISCOVERY ON This enables the SAN discovery. SANDISCOVERY is defaulted to ON for Windows platform. SANDISCOVERY is defaulted to OFF for all non-Windows platforms. Go to Support to verify the platform/HBA vendor/driver level support level prior to setting SANDISCOVERY ON to enable the SAN discovery.
236
ANR2034E QUERY SAN: No match found using this criteria

The Tivoli Storage Manager server tries to collect configuration information for the SAN and finds nothing. RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery. There are a number of possible reasons for not being able to find information about the SAN. These possibilities are: v This is an unsupported platform or OS level. v This is not a SAN environment. v There may be a problem with the SAN. Perform the following tasks to find the SAN configuration information: v Check the FC HBA driver and make sure it is installed and enabled. v Check the HBA driver level to make sure that it is up-to-date. v Use the HBA vendors utility to check for any reported FC link problems. v Uninstall and then install the HBA driver. If there is an issue with the HBA configuration, device driver, or compatibility, sometimes uninstalling and re-installing it corrects the problem. v Check the FC cable connection to the HBA. v Check the FC cable connection from the HBA to the SAN device (switch, data gateway, or other device). v Check the GBIC. v On the SAN device (switch, data gateway, or other device) try a different target port. Sometimes the SAN devices may have a specific port failure. v Halt the Tivoli Storage Manager server, restart the machine, and restart the server. If there were configuration changes in the SAN, sometimes the operating system, device driver, or HBA requires a machine restart before they can communicate with the SAN. v Recycle the destination port on the SAN device. v Reseat the HBA card. v Replace the HBA.
ANR8226E Error detecting version of HBA-API library

This error message is only displayed for AIX. The server attempted to determine the level of the devices.common.IBM.fc.hba-api fileset and encountered an error. This message indicates that an error was encountered while trying to detect the HBA-API libraryFileset version on AIX. RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery.
ANR8227E Fileset devices.common.IBM.fc.hba-api is not at the required level

Due to problems in AIX HBAAPI code, the minimum fileset devices.common.IBM.fc.hba-api level needed for successful SAN discovery are the following: v AIX52 - Need 5.2.0.50 v AIX53 - Need 5.3.0.10
237
This error message is only displayed for AIX. The server specified that the file set devices.common.IBM.fc.hba-api is at a level that is incompatible with Tivoli Storage Manager operations. Install the latest maintenance for this file set if you use SAN devices. RESULT: The Tivoli Storage Manager server is not able to perform SAN discovery.
SAN devices are missing from the display of QUERY SAN server command
The possible reasons for the QUERY SAN server command not displaying all the devices can be due to configuration or vendor support issues.
Refreshing the SAN configuration

The QUERY SAN server command may not be displaying all the devices because of the SAN configuration. You might have to refresh the SAN because the configuration was changed (add/remove device) and the system configuration needs to be updated. Update configuration on AIX: For IBM devices: Issue the cfgmgr command to view the new configuration. The special file name for IBM tape devices (not the Tivoli Storage Manager devices) is added or removed (drv_add and drv_rem). Note: Special file name: /dev/rmt0, /dev/smc0 For the Tivoli Storage Manager devices: To update the special files, use smitty devices Tivoli Storage Manager Devices remove all defined devices, then discover devices supported by Tivoli Storage Manager. Note: Special file name: /dev/mt0, /dev/lb0 Note: Alternatively, reinstalling the IBM device driver/ Tivoli Storage Manager device driver updates all the current special file name. Update configuration on Windows: With the plug and play, the Windows registry is updated and the device name may change without the need to restart the machine or having the device drivers involvement. The Tivoli Storage Manager server detects the change in a special file name and updates the new special file name when it accesses the tape devices (during server initialization or normal operation). The correct device name is updated in the Tivoli Storage Manager database. Note: Special file name: mtx.x.x.x, lbx.x.x.x (can be used for both Tivoli Storage Manager devices and IBM devices) Tape0, Changer0 (only for IBM devices) Update configuration on Linux: The host bus adapter (HBA) gets the most up-to-date configuration information as a result of the RSCN. Sometimes, the machine must be restarted to be able to pick up the configuration changes. For IBM devices: Issue the IBMtapeconfig command to view the new configuration
238
information. The special file name for the IBM tape devices (not Tivoli Storage Manager devices) is added or removed. Note: Special file name: /dev/IBMTape0, /dev/IBMChanger0 For the Tivoli Storage Manager devices: Users can issue autoconf, the Tivoli Storage Manager device driver auto configure script. This resides in the /opt/tivoli/tsm/devices/ bin directory to be able to get all the current special file names. Note: Special file name: dev/tsmscsi/mt0, /dev/tsmscsi/lb0 Note: Alternatively, reinstalling the IBM device driver/Tivoli Storage Manager device driver updates all the current special file names. With the Linux pass-thru device driver for the Tivoli Storage Manager devices, the HBA driver and the generic driver must be reloaded to get all the current special file names. You have to run the autoconf script so that the Tivoli Storage Manager device driver can create configuration files (/dev/tsmscsi/lbinfo and /dev/tsmscsi/mtinfo). These files are used by the Tivoli Storage Manager server to create the special file names after each SAN discovery. 32 bits (Linux xSeries) Ensure that the HBAAPI wrapper library libhbaapi32.so is in the same directory as dsmserv.exe or in the /opt/tivoli/tsm/server/ bin directory. 64 bits (Linux pSeries) Ensure that the HBAAPI wrapper library libhbaapi64.so is in the same directory as dsmserv.exe or in the /opt/tivoli/tsm/server/ bin directory. Update the configuration on Solaris: The HBA gets the most up-to-date configuration information as result of the RSCN. Most of the time, the machine must be restarted to pick up the configuration changes. Reinstall the IBM device driver/Tivoli Storage Manager device driver or issue the following commands to have the special file name updated: For IBM devices: The IBM device driver has already done the rem_drv and add_drv and nothing needs to be done. Note: Special file name: /dev/rmt/0st, /dev/rmt/0smc For Tivoli Storage Manager devices: Modify /usr/kernel/drv/mt.conf and /usr/kernel/drv/lb.conf to add the following: mt.conf: (name=mt parent=pseudo instance=16383) lb.conf (name=lb parent=pseudo instance=16383) Issue the following commands: rem_drv lb add_drv lb rem_drv mt add_drv mt add_drv -m * 0666 bin bin mt
239
Alternatively, the autoconf command can do the above by itself. Ensure that the virtual device special file is in /devices/pseudo/mt@16383:tsmmtctl and /devices/pseudo/ lb@16383:tsmlbctl Issue the command twice
sync sync
Restart the machine The following files are used by the Tivoli Storage Manager server to create the special filenames after each SAN discovery: v 32 bits (Linux xSeries) Ensure that libhbaapi32.so is in the same directory as dsmserv.exe or in /opt/tivoli/tsm/server/bin v 64 bits (Linux pSeries) Ensure that libhbaapi64.so is in the same directory as dsmserv.exe or in /opt/tivoli/tsm/server/bin
Configuration problems causing SAN device absence

The possible reasons for the QUERY SAN server command not displaying all the devices can be due to a configuration problem with the HBA hardware, HBA driver level, or OS level. Perform the following steps to resolve your configuration issues: 1. Go to IBM Tivoli Storage Manager support to verify the platform/HBA vendor/driver level support level to make sure the HBA driver level and OS level are compatible and supported by IBM Tivoli for SAN discovery. 2. Use the HBA vendor utility to check to see if the device can be seen by the HBA. If the device is not seen by the HBA, the device may not be connected. Check the Fibre Channel or SCSI cable. If the device is seen by the HBA, check the HBA driver version. This driver version may have problems with the HBA API.
Vendor support for any particular device in the SAN

Many devices or combinations of devices may not be supported in a given SAN. These limitations arise from the ability of a given vendor to certify their device using Fibre Channel Protocol. For a given device, verify with the device vendor that it is supported in a SAN environment. This includes all hardware associated with the SAN, which means verifying that this device is supported with the vendors of the HBAs, hubs, gateways, and switches that make up the SAN environment.
Error message ANR1745I: Unable to discover SAN devices. Function is busy.

This error message appears if there is another active SAN discovery. The Tivoli Storage Manager server is not able to perform SAN discovery. Retry again after the other SAN discovery is completed.
240
Error messages ANR1786W, ANR1787W or ANR1788W were issued

You might see error messages ANR1786W, ANR1787W, or ANR1788W due to a problem with SAN discovery. The following three messages usually indicate that the HBAAPI library is not working in general: v ANR1786W HBAAPI not able to get adapter name v ANR1787W Not able to open adapter adaperName v ANR1788W Not able to get the adapter attributes for adapterName If the result is that the Tivoli Storage Manager server is unable to perform SAN discovery, go to IBM Tivoli Storage Manager support to verify that the host bus adapter (HBA) driver is up-to-date and at a supported level.
ANR1789W Get HBA target mapping failed

Error message ANR1789W is the most commo HBAAPIn error on the SAN. Get HBA target mapping failed means that the host bus adapter (HBA) encountered an error while gathering device mapping information by sending various SCSI commands. Verify that all SAN devices are working properly (e.g. a SAN Data Gateway could be hung and need to be rebooted). If all devices appear functional, verify the firmware of device on the SAN, as well as HBA driver, are at the appropriate levels. If the result is that the IBM Tivoli Storage Manager server is not able to perform SAN discovery, go to IBM Tivoli Storage Manager support to verify that the HBA driver is up-to-date and at a supported level. For IBM tape devices, make sure the latest firmware is installed. Firmware prior to 4772 for IBM 3580 tape devices caused problems with Qlogic HBAAPI.
ANR1790W SAN discovery failed

Error message ANR1790W is a general message indicating that the HBAAPI function failed and we cannot perform SAN discovery. Verify that all SAN devices are working properly (e.g. a SAN Data Gateway could be hung and need to be rebooted). If all devices appear functional, verify that the firmware of device on the SAN, as well as the HBA driver, are at the appropriate levels. For IBM tape devices, make sure the latest firmware is installed. Firmware prior to 4772 for IBM 3580 tape devices causes problems with Qlogic HBAAPI.
ANR1791W HBAAPI wrapper library xxxxx failed to load or is missing

The HBAAPI wrapper library is used by the Tivoli Storage Manager server to communicate with the SNIA HBAAPI. The HBAAPI wrapper libraries are in the same directory as the Tivoli Storage Manager executable (unless full path is given as shown below). The following is a
241
list of the HBA wrapper files shipped with the Tivoli Storage Manager server package (except on AIX). Error message ANR1791W indicates that the HBAAPI wrapper file is either missing or could not be loaded by the Tivoli Storage Manager. Verify that the wrapper file is in the same directory as the Tivoli Storage Manager executable. The following are the HBAAPI wrapper library files: v Windows : hbaapi.dll v AIX : /usr/lib/libhbaapi.a (This is provided by AIX with HBAAPI installation) v 32-bit Linux: libhbaapi32.so v 64-bit Linux: libhbaapi64.so v 32-bit Solaris: libhbaapi32.so v 64-bit Solaris: libhbaapi64.so The result is that the Tivoli Storage Manager server is not able to perform SAN discovery.
ANR1792W HBAAPI vendor library failed to load or is missing

Error message ANR1792W indicates that the vendors library file failed to load. Verify the validity of the library files. UNIX systems store their HBAAPI libraries in the location specified by the /etc/hba.conf file, and Windows file are stored in the C:\winnt\system32 directory. The following are examples of vendor library files: v C:\winnt\system32\qlsdm.dll (QlLogics Windows file) v /usr/lib/libHBAAPI.a (Emulexs AIX file) v /usr/lib/libqlsdm.so (Qlogics Linux file) v v v v /usr/lib/libemulexhbaapi.so (Emulexs Linux 32-bit file) /usr/lib64/libemulexhbaapi.so (Emulexs Linux 64-bit file) /usr/lib/libqlsdm.so (Qlogics Solaris file) /opt/JNIsnia/Solaris/Jni/64bit/JniHbaLib.so (JNIs Solaris file)
The result is that the Tivoli Storage Manager server is not able to perform SAN discovery.
Error message ANR1793W

ANR1793W Tivoli Storage Manager SAN discovery is not supported on this platform or this version of OS. Error message ANR1793W is only displayed if the Tivoli Storage Manager attempts a SAN device mapping or device discovery operation on an unsupported operating system. The following platforms are not currently supported by SAN device mapping or device discovery: v HP-UX v 64-bit Windows 2003 v Linux zSeries v AIX versions that are not 52L or 53A. Support for SAN device mapping and device discovery on AIX requires either version 52L (fileset level of 5.2.0.50) or 53A (fileset level of 5.3.0.10) or higher. The result is that the Tivoli Storage Manager server is not able to perform SAN discovery.
242
ANR1794W Tivoli Storage Manager SAN discovery is disabled by options

Error message ANR1794W indicates that the SAN discovery on the Tivoli Storage Manager server is disabled. The SAN discovery can be disabled or enabled by issuing the following server commands: v setopt SANDISCOVERY OFF This command disables the SAN discovery. The Tivoli Storage Manager server is not able to correct the device path automatically if the path was changed. This only has to be issued once. v setopt SANDISCOVERY ON This command enables the SAN discovery. The above setopt SANDISCOVERY command can be issued as many times as necessary. Another way to disable/enable SAN discovery is to put the following option in the dsmserv.opt file: v SANDISCOVERY OFF This disables the SAN discovery. v SANDISCOVERY ON This enables the SAN discovery. SANDISCOVERY is defaulted to ON for Windows platform. SANDISCOVERY is defaulted to OFF for all non-Windows platforms. Go to IBM Tivoli Storage Manager support to verify the platform/HBA vendor/driver level support level prior to setting SANDISCOVERY ON to enable the SAN discovery.
ANR2034E QUERY SAN: No match found using this criteria

Error message ANR2034E is issued when the Tivoli Storage Manager server tries to collect configuration information for the SAN and finds nothing. The result is that the Tivoli Storage Manager server is unable to perform SAN discovery. The following list displays a number of possible reasons for not finding information about the SAN: v This is an unsupported platform or OS level. v This is not a SAN environment. v There may be a problem with the SAN. v Check the FC HBA driver and make sure it is installed and enabled. v Check the HBA driver level to make sure that it is up-to-date. v Use the HBA vendors utility to check for any reported FC link problems. v Uninstall and then install the HBA driver. If there is an issue with the HBA configuration, device driver, or compatibility, sometimes uninstalling and re-installing it corrects the problem. v Check the FC cable connection to the HBA.
243
v Check the FC cable connection from the HBA to the SAN device (switch, data gateway, or other device). v Check the GBIC. v On the SAN device (switch, data gateway, or other device) try a different target port. Sometimes the SAN devices may have a specific port failure. v Halt the Tivoli Storage Manager server, restart the machine, and restart the server. If there were configuration changes in the SAN, sometimes the operating system, device driver, or HBA requires a machine restart before they can communicate with the SAN. v Recycle the destination port on the SAN device. v Reseat the HBA card. v Replace the HBA.
ANR8226E Error detecting version of HBA-API library

Error message ANR8226E is only displayed for AIX. The server attempted to determine the level of the devices.common.IBM.fc.hba-api fileset and encountered an error. Error message ANR8226E indicates that an error was encountered while trying to detect the HBA-API libraryFileset version on AIX. The result is that the Tivoli Storage Manager server is not able to perform SAN discovery.
ANR8227E Fileset devices.common.IBM.fc.hba-api is not at the required level

Error message ANR8227E is only displayed for AIX. Due to problems in AIX HBAAPI code, the minimum fileset devices.common.IBM.fc.hba-api level needed for successful SAN discovery are the following: v AIX52 - Need 5.2.0.50 v AIX53 - Need 5.3.0.10 The server specified that the file set devices.common.IBM.fc.hba-api is at a level that is incompatible with Tivoli Storage Manager operations. Install the latest maintenance for this file set if you use SAN devices. The result is that the Tivoli Storage Manager server is not able to perform SAN discovery.
244
Chapter 9. Tivoli Storage Manager ODBC driver

Tivoli Storage Manager provides an SQL interface which supports queries to its internal database. The interface includes an open database connectivity (ODBC) driver for Windows NT/2000 open database connectivity.
Supported ODBC applications and functions

The Tivoli Storage Manager open database connectivity (ODBC) driver is supported with any ODBC application, within the constraints specified here and in Known ODBC problems and limitations on page 248. Other constraints might also be specified in the README file that accompanies the Tivoli Storage Manager ODBC driver. Resolution of problems due to limitations or defects in the ODBC application is the responsibility of that applications vendor. The following ODBC 3.x functions are implemented in the Tivoli Storage Manager ODBC driver: v SQLAllocHandle v SQLBindCol v SQLBindParameter v SQLCancel v SQLCloseCursor v v v v v v v v v v v v v v v v v v v v v SQLColAttribute SQLColumns SQLConnect SQLDescribeCol SQLDisconnect SQLDriverConnect SQLExecDirect SQLExecute SQLFetch SQLFreeHandle SQLFreeStmt SQLGetConnectAttr SQLGetCursorName SQLGetData SQLGetDiagRec SQLGetEnvAttr SQLGetFunctions SQLGetInfo SQLGetStmtAttr SQLGetTypeInfo SQLMoreResults (requires ODBC driver version 5.4 or higher)
245
v v v v v v v v v v v v
SQLNumParams SQLNumResultCols SQLPrepare SQLPrimaryKeys SQLRowCount SQLSetConnectAttr SQLSetCursorName SQLSetEnvAttr SQLSetStmtAttr SQLSpecialColumns SQLStatistics SQLTables
The ODBC 3.x specification can be found in the Microsoft Developer Network (MSDN) library at the ODBC 3.x specification Web site. Go to Win32 and COM Development Data Access and Storage Microsoft Open Database Connectivity (ODBC) Microsoft Open Database Connectivity (ODBC) ODBC Programmers Reference to navigate to the ODBC specification.
Defining ODBC data sources

The open database connectivity (ODBC) Data Source Administrator in the Control Panel is used to add, remove, and configure data source entries. ODBC applications can then be used to connect to the Tivoli Storage Manager server. Note: For 64-bit Windows XP and 2003 users: The Tivoli Storage Manager ODBC driver is only available in a 32-bit version. Therefore you must use the 32-bit version of the ODBC Data Source Administrator (odbcad32.exe) to configure Tivoli Storage Manager ODBC driver data source names (DSNs). This executable is located in directory x:\WINDOWS\SysWOW64, where x: is the drive letter where Windows is installed. The Tivoli Storage Manager ODBC setup program creates a default user DSN called TSM ODBC Sample User DSN. You can either update this entry or remove it and create one or more new DSNs to connect to your Tivoli Storage Manager server or servers. When creating new DSNs, use the Tivoli Storage Manager ODBC Driver driver name. The ODBC data source setup panel for Tivoli Storage Manager has the following controls:
Control Data source name Type REQUIRED Description Specify the name that you want to use for the DSN. This is the name that will appear in the Name column when you run the ODBC Data Source Administrator. The maximum allowed length is 255 characters. This is a free-form text field where you may enter a comment or description about this DSN. The maximum allowed length is 255 characters.
Description
OPTIONAL
246
Control Language
Type REQUIRED
Description Select the language in which ANSxxxxx messages are to be displayed. The available languages are those that were selected at the time that the ODBC driver was installed. Note: Only the ANSxxxxx messages are available in other languages. The ODBC driver dialogs and other non-ANSxxxxx text strings are currently displayed in English only. Specify the Tivoli Storage Manager Administrator ID to use with this DSN. Specify the number of records that the ODBC driver requests from the Tivoli Storage Manager server per fetch request. Larger values may improve performance, especially when fetching a large number of records. This value must be between 1 and 1000000 inclusive. The default value is 50. Specify the TCP/IP address of the Tivoli Storage Manager server to which you are connecting. If you do not know which address to use, see your Tivoli Storage Manager administrator. Specify the TCP/IP port number of the Tivoli Storage Manager server to which you are connecting. If you do not know which port number to use, see your Tivoli Storage Manager administrator. Selecting this check box activates the Tivoli Storage Manager ODBC driver tracing. Clearing this check box deactivates tracing. This check box is normally not selected. Tracing should only be performed at the request of the Tivoli Storage Manager Service or Development. This field is available only when Enable trace is selected. Here you can specify a fully-qualified file name to which trace data is written. Tracing should only be performed at the request of the Tivoli Storage Manager Service or Development.
Administrator name Fetch size
OPTIONAL REQUIRED
TCP/IP address
REQUIRED
TCP/IP port
REQUIRED
Enable trace
OPTIONAL
Trace file name OPTIONAL
Table 7. Keywords to use when creating file DSNs Keyword DRIVER=driverName UID=administratorName TCPADDR=tcpipAddress TCPPORT=tcpipPortNumber Description driverName must be Tivoli Storage Manager ODBC Driver. administratorName is a valid Tivoli Storage Manager Administrator ID. tcpipAddress is the TCP/IP address for your Tivoli Storage Manager server. tcpipPortNumber is the TCP/IP port number used to connect to your Tivoli Storage Manager server. fetchSize is the number of records that the ODBC driver requests from the Tivoli Storage Manager server per fetch request. This value must be between 1 and 1000000 (inclusive). The default value is 50.
FETCHSIZE=fetchSize
For example: if your administrator ID is storman, your Tivoli Storage Manager server address is storman.tucson.ibm.com, and your Tivoli Storage Manager TCP/IP port number is 1500. The file DSN should look like the following entry:
247
[ODBC] DRIVER=Tivoli Storage Manager ODBC Driver UID=storman TCPADDR=storman.tucson.ibm.com TCPPORT=1500
Note: UID, TCPADDR, and TCPPORT are not required; if they are not specified, then you will be prompted for their values. The [ODBC] line is automatically added for you when you use the ODBC Data Source Administrator to create the file DSN. Perform the following steps to create a file DSN using the ODBC Data Source Administrator: 1. Start the ODBC Data Source Administrator from the Control Panel. Alternatively you can run odbcad32.exe. 2. Click the File DSN tab, and then click Add. 3. Select Tivoli Storage Manager ODBC Driver (located in the Name column) and click Next. 4. Enter the name that you want to use for this DSN. For example, Tivoli Storage Manager File DSN. Click Next. The dialog now shows you a summary of the options chosen thus far. 5. Click Finish. 6. You are now presented with the Connect to the Tivoli Storage Manager Server dialog. Fill in the TCP/IP address, TCP/IP port, administrator ID, password, and click Connect. An attempt is made to connect to the Tivoli Storage Manager server using the information that you supplied above. If the connection is successful, you will be returned to the ODBC Data Source Administrator dialog. If there was a problem connecting to the Tivoli Storage Manager server, you will be prompted again to enter the correct information. Note: For security purposes, the password is NOT saved. It is used in this step for the purpose of validating the connection information.
Known ODBC problems and limitations

Before attempting to troubleshoot your ODBC problem, you might save some time by looking at the known problems. The following are the known problems or limitations with the Tivoli Storage Manager open database connectivity (ODBC) driver: v SELECT statements containing the AS keyword in the column specification with a quoted column name do not work. For example, the following does not work:
select node_name as "Node Name" from nodes
Instead, use single word column names, which do not require quotes:
select node_name as NODENAME from nodes
v When using Microsoft Excel or Microsoft Query (msqry32.exe), statements containing the AS keyword in the column specification do not work at all. This is because Microsoft Query (which is what Microsoft Excel uses to import the data) always puts single quotes around the new column name, even if it does not contain blank spaces. The Tivoli Storage Manager server does not accept single quotes around the column name. Microsoft Knowledge Base article 298955 describes this issue. v If your ODBC application provides the option to select one or more columns as unique identifiers, do not choose any columns. For example, when linking tables in Microsoft Access, you might be presented with a Select Unique Record
248
Identifier dialog. In this case, just click OK without selecting any fields. This feature is usually associated with maintaining integrity for database updates, but since it is not possible to update the Tivoli Storage Manager database with the ODBC driver, there is no need to select any fields. Further, selecting one or more fields might result in erroneous output. One common symptom is seeing #DELETED in the result set, in which case the solution is to link the table again and not select any unique identifiers. v For 64-bit bit Microsoft Windows XP and 2003 users: The ODBC driver is available in a 32-bit version only, and thus is not compatible with 64-bit applications. The ODBC driver can be installed on 64-bit Windows XP and 2003 systems, but must be used with 32-bit applications. Also, the 32-bit version of the ODBC Data Source Administrator must be used to create and configure data source names (DSNs). See Defining ODBC Data Sources. v If the ODBC application is running in a different locale from that of the server, data containing characters that do not map into the ODBC application machines locale may appear unintelligible (garbled). v When opening the Tivoli Storage Manager database tables in Lotus Approach , the Open dialog includes a field labeled Files of type. Select ODBC Data Sources (*) for this field. Do not select the Tivoli Storage Manager ODBC Driver (*) because Approach will not open the tables. v If you use a file DSN to link Tivoli Storage Manager database tables into Microsoft Access, you will not be able to open the linked table after Access is shut down and restarted. This problem is discussed in Microsoft support article Q327268. v The ANSxxxxx messages can now be displayed in languages other than English. Available languages include Chinese (Simplified and Traditional), Czechoslovakian, English, French, German, Hungarian, Italian, Japanese, Korean, Polish, Portuguese (Brazilian), Russian, and Spanish. However, the Tivoli Storage Manager ODBC configuration dialogs and other non-ANSxxxxx text strings are currently displayed in English only. v In order to maintain the database integrity, the Tivoli Storage Manager ODBC Driver supports only SELECT statements. You will not be able to manipulate the database with any other statements. v If there are more than two tables specified in the FROM clause of a SELECT statement, the query results will not be optimized by the Tivoli Storage Manager server. You might see duplicate records in the results. Example: The following SQL statement will generate duplicate records:
SELECT * FROM NODES, FILESPACES, CONTENTS
Note: The SQL statement running time may be much longer due to preparing large amounts of query results. v The columns in the ORDER BY clause must be in the select list. The Tivoli Storage Manager ODBC driver provides the information for the application. Some applications do not retrieve the information before generating the SQL query. Example: The following SQL statement will fail because PWSET_TIME is not in the select list:
SELECT ADMIN_NAME FROM ADMINS ORDER BY PWSET_TIME
v Not all Tivoli Storage Manager database tables have a primary key. When selecting Link Tables using Microsoft Access, Microsoft Access prompts the users for primary key fields if the table does not have a primary key or if the table has a primary key that is comprised of multiple index fields. You can select primary key fields or click Cancel to ignore the prompt.
249
ODBC troubleshooting and diagnostics

The open database connectivity (ODBC) troubleshooting and diagnostic information can help you if you cannot find your issue in the Known problems and limitations. Before proceeding, be sure to check Known problems and limitations to see if your problem is already known.
Resolving ODBC installation problems

The Microsoft Software Installer (MSI) is used to install the Tivoli Storage Manager open database connectivity (ODBC) driver. If you are having trouble installing the Tivoli Storage Manager ODBC driver, contact IBM/Tivoli support. Provide the following information to IBM/Tivoli support: v Operating system version and service pack number (for example, Windows 2000 SP 3, Windows XP SP 1a) v Exact version/release/level/patch of the Tivoli Storage Manager ODBC driver that you are trying to install (for example, 5.2.0.0). The version information can be found at the top of the README file. v Recreate the scenario, step-by-step, including the last dialog box that you see before the failure occurs, and the exact text of any error messages. v Examine the system Event Log and report any event log errors that coincide with the time that the installation failure occurred. v An MSI setup.log file. To create this file, run setup.exe again as follows:
setup /v"/l*v setup.log"
Follow the prompts to recreate the failure. After the setup program ends, setup.log will contain detailed diagnostic information about the setup.
Resolving ODBC data source configuration problems

The open database connectivity (ODBC) data source configuration problems that you are experiencing may be related to your system settings, or your ODBC trace. If you are having trouble creating or configuring a data source for the Tivoli Storage Manager ODBC driver, see ODBC configuration details and compare your system settings with those described in that section. If you find any differences, and if you do not know how to repair them or are uncomfortable doing so, contact IBM support and provide the information described below the next paragraph (you do not have to provide trace data at this point). If the configuration looks correct, try to obtain a trace of the Tivoli Storage Manager ODBC driver. See Tivoli Storage Manager ODBC Driver Tracing for instructions on tracing. After you have the trace data, contact IBM support and provide the following information: v Operating system version and service pack number (for example, Windows 2000 Server SP 3, Windows XP SP 1) v Exact version/release/level/patch of the Tivoli Storage Manager ODBC driver that you are trying to install (for example 5.2.0.0). v Step-by-step recreation of the scenario, including the last dialog box that you see before the failure occurs, and the exact text of any error messages that you see. v Your ODBC configuration information. Include the following items: X:\WINNT\ODBC.INI X:\WINNT\ODBCINST.INI
250
The odbchkcu.reg and odbchklm.reg files created by the following commands:

regedit -e odbchkcu.reg hkey_current_user\software\odbc regedit -e odbchklm.reg hkey_local_machine\software\odbc
A directory listing of the files in the directory where the Tivoli Storage Manager ODBC driver is installed. For example, if the driver is installed to directory C:\Program Files\Tivoli\TSM\odbc, then open an OS command prompt, change into that directory, and issue the following command:
dir > odbcdir.out
Then you can send in the odbcdir.out file. Trace data.
Resolving ODBC application problems

It is suggested to perform the ODBC tasks if you can install the Tivoli Storage Manager open database connectivity (ODBC) driver and configure a data source, but cannot get your application to work. Perform the following to get your application to work: v Obtain a trace as described in ODBC Tracing. v Contact IBM support and provide the following information: Operating system version and service pack number (for example, Windows 2000 Server SP 3, Windows XP SP 1) Exact version/release/level/patch of the Tivoli Storage Manager ODBC driver that you are trying to install (for example, 5.2.0.0). Name and exact version/service pack of the ODBC application that you use. Recreate the scenario, step-by-step, including the last dialog box that you see before the failure occurs, and the exact text of any error messages. Your ODBC configuration information. Include the following items: - X:\WINNT\ODBC.INI - X:\WINNT\ODBCINST.INI - The odbchkcu.reg and odbchklm.reg files created by the following commands:
regedit -e odbchkcu.reg hkey_current_user\software\odbc regedit -e odbchklm.reg hkey_local_machine\software\odbc
v Obtain a directory listing of the files in the directory where the Tivoli Storage Manager ODBC driver is installed. For example, if the driver is installed to directory C:\Program Files\Tivoli\TSM\odbc, then open an OS command prompt, change into that directory, and issue the following command: dir > odbcdir.out You can then send in the odbcdir.out file. v Trace data.
ODBC configuration details

The ODBC configuration details include environmental variables, installation files, and registry settings. Starting with Version 5.3 of the Tivoli Storage Manager ODBC driver, the environment variables DSMG_DIR and DSMODBC_DIR are no longer supported and must not be used. The Tivoli Storage Manager ODBC driver installation program attempts to remove DSMG_DIR and DSMODBC_DIR if they exist on your system. However, if they still exist after the installation, then they should be removed manually.
251
Viewing ODBC installation files

The Tivoli Storage Manager ODBC driver files are located in a directory specified during installation. The default installation directory for the operating systems are listed below: 32-bit Windows x:\Program Files\Tivoli\TSM\odbc 64-bit Windows x:\Program Files (x86)\Tivoli\TSM\odbc where x: is the drive letter where the operating system is installed (usually C:). The installed files include the following: v dscenu.txt v dsmntapi.dll v odbcadsm.dll v tsmutil1.dll One or more of the following files may also be installed, depending on the language version of Microsoft Windows that you are running and any additional language support that you specified during installation. v dscchs.txt - Chinese (Simplified) v dsccht.txt - Chinese (Traditional) v dscdeu.txt - German v dscesp.txt - Spanish v dscfra.txt - French v v v v v v v dscita.txt - Italian dscjpn.txt - Japanese dsckor.txt - Korean dscptb.txt - Portuguese (Brazilian) dsccsy.txt - Czech dschun.txt - Hungarian dscplk.txt - Polish
v dscrus.txt - Russian Note: Prior versions of the Tivoli Storage Manager ODBC driver placed its files in the winnt\system32 directory. The following files may be deleted from that directory if they still exist: v dsgameng.txt v odbcadsm.dll v odbcadsm.txt If your machine is running the German version of Windows, then the dscdeu.txt file is installed by default. If your machine is running the German version of Windows and you also select French and Italian language support, then dscdeu.txt, dscfra.txt, and dscita.txt are installed.
252
If your machine is running the English version of Windows and you also select Spanish and Japanese support, then dscesp.txt and dscjpn.txt are installed.
Viewing ODBC registry settings

The ODBC registry settings are integral to the ODBC driver and language support. Important: Do NOT change any of the registry settings values unless you are directed by IBM Development or Service. The Tivoli Storage Manager ODBC driver installation settings are located in subkeys of the following registry key:
HKEY_LOCAL_MACHINE\SOFTWARE\IBM\ADSM\CurrentVersion
The following subkeys and associated values are pertinent to the Tivoli Storage Manager ODBC driver installation:
Table 8. ComponentsMsi subkey Name ODBC Driver Explanation A REG_SZ value that must be Odbc
Table 9. Odbc subkey Name Path Explanation This is the directory in which you installed the Tivoli Storage Manager ODBC driver. It is the parent directory for the odbc subdirectory which contains the odbcadsm.dll file. For example, if odbcadsm.dll is in directory C:\Program Files\Tivoli\TSM\odbc, then this value should be C:\Program Files\Tivoli\TSM. This is a REG_SZ value. A REG_SZ value indicating the 4-node version number of the Tivoli Storage Manager ODBC driver. For example, if the Tivoli Storage Manager 5.1.5.0 ODBC driver is installed, then this value should be 5.1.5.0 (without the quotation marks).
PtfLevel
Odbc\nls\xx_XX: where xx_XX is a language abbreviation as follows (with the language name shown to the right for your reference): v de_DE - German v v v v v v v v v v v v v en_US es_ES fr_FR it_IT ja_JP ko_KR pt_BR zh_CN zh_TW cs_CZ hu_HU pl_PL ru_RU U.S. English Spanish French Italian Japanese Korean Portuguese (Brazilian) Chinese (Simplified) Chinese (Traditiona Czech Hungarian Polish Russian
Note: You must always have the Odbc\nls\en_US subkey.
253
If you install the Tivoli Storage Manager ODBC driver on a non-English Windows system, then you may also have a subkey that corresponds to the language version of Windows that you are running. For example, if you are running the German language version of Windows, then you may also have the Odbc\nls\de_DE subkey. You may have other subkeys as well if, during the installation, you selected multiple language support.
Table 10. Language support settings Name FileName Explanation A REG_SZ value indicating the name of the Tivoli Storage Manager message repository that corresponds to the subkeys language setting. For example, in subkey Odbc\nls\de_DE, the FileName value should be dscdeu.txt. A REG_SZ value indicating the language of the message repository specified in the FileName value. For example, if FileName is dscdeu.txt, then the Language setting must be German.
Language
Table 11. Language value settings Subkey File name Language German English (US) Spanish French Italian Japanese Korean Portuguese (Brazilian) Chinese (Simplified) Chinese (Traditional) Czech Hungarian Polish Russian
Odbc\nls\de_DE dscdeu.txt Odbc\nls\en_US dscenu.txt Odbc\nls\es_ES Odbc\nls\fr_FR Odbc\nls\it_IT Odbc\nls\ja_JP dscesp.txt dscfra.txt dscita.txt dscjpn.txt
Odbc\nls\ko_KR dsckor.txt Odbc\nls\pt_BR dscptb.txt Odbc\nls\ zh_CN Odbc\nls\ zh_TW dscchs.txt dsccht.txt
Odbc\nls\cs_CZ dsccsy.txt Odbc\nls\ hu_HU Odbc\nls\pl_PL dschun.txt dscplk.txt
Odbc\nls\ru_RU dscrus.txt
Table 12. The Tivoli Storage Manager ODBC driver settings are located in subkeys of the following registry key: System 32-bit Windows 64-bit Windows Registry Key HKEY_LOCAL_MACHINE\SOFTWARE\ ODBC HKEY_LOCAL_MACHINE\SOFTWARE\ Wow6432Node\ODBC
The following subkeys and associated values are pertinent to the Tivoli Storage Manager ODBC driver (these are all located under HKEY_LOCAL_MACHINE\ SOFTWARE\ODBC):
254
Table 13. ODBC.INI\ODBCADSM Name ADSMTrace Explanation A REG_SZ value indicating whether the Tivoli Storage Manager ODBC driver tracing is enabled. Values may be either No or Yes. If tracing is enabled (ADSMTrace is set to Yes), trace output is written to the file specified by the TraceFileName value (see below). Language A REG_SZ value that indicates which language version of the Tivoli Storage Manager message repository is in use. Values may be one of the following: v German v English (US) v Spanish v French v Italian v Japanese v Korean v Portuguese (Brazilian) v Chinese (Simplified) v Chinese (Traditional) v Czech v Hungarian v Polish v Russian TraceFileName A REG_SZ value that indicates the name of the Tivoli Storage Manager ODBC driver trace file.
Table 14. ODBCINST.INI\ODBC Drivers Name Explanation
Tivoli Storage Manager ODBC A REG_SZ value indicating that the Tivoli Storage Manager Driver ODBC driver is installed. This value is always set to Installed. Table 15. ODBCINST.INI\TSM ODBC Driver Name APILevel Explanation A REG_SZ value that is always set to 0 (zero). This indicates that the Tivoli Storage Manager ODBC driver implements ODBC Core-level functions only. A REG_SZ value that is always set to YYN. This string indicates whether the driver implements the SQLConnect(), SQLDriverConnect(), and SQLBrowseConnect() functions, respectively. Y (yes) means that the function is implemented in the driver; N (no) means that the function is not implemented in the driver. Thus, the Tivoli Storage Manager ODBC driver currently implements SQLConnect() and SQLDriverConnect(). The driver does not implement SQLBrowseConnect(). A REG_SZ value that specifies the full path to the Tivoli Storage Manager ODBC driver. If the driver is installed to C:\Program Files\Tivoli\TSM\odbc, the value is: C:\Program Files\Tivoli\TSM\odbc\odbcadsm.dll
ConnectFunctions
Driver
255
Table 15. ODBCINST.INI\TSM ODBC Driver (continued) Name DriverODBCVer FileUsage Explanation A REG_SZ value that indicates which version of the ODBC specification the driver supports. This value is set to 03.51. A REG_SZ value that indicates how the driver behaves toward files in a data source. Because the Tivoli Storage Manager ODBC driver is not a file-based driver, this value is always 0 (zero). A REG_SZ value that specifies the full path to the drivers Data Source configuration program. The Tivoli Storage Manager ODBC drivers configuration program is in the Tivoli Storage Manager ODBC driver itself, so this value should always match that of the Driver setting (see above). A REG_SZ value that indicates the SQL-92 grammar supported by the driver. Because Tivoli Storage Manager supports a subset of the SQL-92 entry-level grammar, this value is always 0 (zero). A REG_DWORD value that indicates how many components use this driver. Because the driver is used by only one component (the Tivoli Storage Manager ODBC driver itself), this value should be 1.
Setup
SQLLevel
UsageCount
The Tivoli Storage Manager ODBC driver data source settings are located in the following registry key: HKEY_CURRENT_USER\Software\ODBC\ODBC.INI\<data source name> where <data source name> is a user-defined data source name. Note that the Tivoli Storage Manager ODBC driver installation program configures a sample data source called Tivoli Storage Manager ODBC Sample User DSN. Each instance of this subkey contains the following values:
Name CommMethod Explanation A REG_SZ value that indicates the communications method that the Tivoli Storage Manager ODBC driver uses to communicate with the Tivoli Storage Manager server. Because the only supported communications protocol is TCP/IP, this value is always set to TCP/IP. A REG_SZ value that specifies the full path to the Tivoli Storage Manager ODBC driver. Thus if the driver is installed to C:\Program Files\Tivoli\TSM\odbc, the value is: C:\Program Files\Tivoli\TSM\odbc\ odbcadsm.dll Note: This value always matches that of the driver value specified in registry subkey HKEY_LOCAL_MACHINE\SOFTWARE\ODBC\ODBCINST.INI\TSM ODBC Driver. A REG_SZ value containing a user-specified free-form comment (or description) for the data source definition.
Driver
DsnDescription
256
Name FetchSize
Explanation A REG_SZ value indicating the maximum number of rows that the ODBC driver requests (buffers) from the Tivoli Storage Manager server at a time. Note that this is not the same as the SQL_ATTR_ROW_ARRAY_SIZE ODBC statement attribute. The former governs the number of records that the ODBC driver will buffer from the server, while the latter governs the number of records that the application will request from the ODBC driver. The default setting is 50 (50 records). Larger settings may improve performance.
TcpPort
A REG_SZ value indicating the TCP/IP communications port that the Tivoli Storage Manager ODBC driver uses to connect to the Tivoli Storage Manager server. A REG_SZ value indicating the TCP/IP address that the Tivoli Storage Manager ODBC driver uses to connect to the Tivoli Storage Manager server. A REG_SZ value indicating the name of the Tivoli Storage Manager Administrator ID uses to connect to the Tivoli Storage Manager server.
TcpSrvAddr
UID
The operating system logs information from ODBC registry keys in two files located in the x:\winnt directory (x:\windows for Windows XP users): v ODBC.INI v ODBCINST.INI Important: Do NOT rename, delete, or otherwise modify these files or their contents unless it is at the direction of IBM Development or Service.
257
258
Chapter 10. Help facilities

The IBM Support Assistant product plug-in for the Tivoli Storage Manager Backup-Archive Client provides additional product specific web search tools as well as a data collector. The data collector can be used to gather various documentation when reporting a problem to IBM technical support. Note: The Tivoli Storage Manager has a built-in help facility within the client command line. Issue the dsmc help command to access the command line clients help facility. The help facility is a menu-driven interface with information that includes the following: v Command reference v Option reference v Extended information about client messages
Backup-archive client help

The command-line client interface includes a help facility that provides descriptions and syntax for client commands and options, and a full description of client messages, including Explanation, System Action, and User Response.
Help for commands

You can access the command-line client help facility by issuing the help command: dsmc help. You are then presented with a menu of help topics. Each topic has a corresponding menu number. You may enter the menu number for a topic or use the u (up) and d (down) keys to scroll through the menu pages. Use the q key to quit the help facility. For instance, to locate help for the restore command, use the d key until the page containing the RESTORE topic is displayed:
The following help topics are available. Enter the number of the desired help topic or q to quit, d to scroll down, u to scroll up. 57 - QUERY SESSION 58 - QUERY SYSFILES 59 - QUERY SYSTEMINFO 60 - QUERY SYSTEMOBJECT 61 - QUERY SYSTEMSERVICES 62 - QUERY SYSTEMSTATE 63 - QUERY SYSVOL 64 - QUERY WAS 65 - QUERY WMI 66 - RESTART RESTORE 67 - RESTORE 68 - RESTORE ACTIVEDIRECTORY 69 - RESTORE ASR 70 - RESTORE BACKUPSET 71 - RESTORE CERTSERVERDB 72 - RESTORE CLUSTERDB 73 - RESTORE COMPLUSDB 74 - RESTORE EVENTLOG 75 - RESTORE FRS
In this example, you would enter 67 to view information on the restore command.
259
To view information for a message, use the d and u keys to navigate the menu until you see entries that appear like the following entries:
The following help topics are available. Enter the number of the desired help topic or q to quit, d to scroll down, u to scroll up. 266 - ANS1820 - ANS1839v 267 - ANS1840 - ANS1859 268 - ANS1860 - ANS1879 269 - ANS1880 - ANS1899 270 - ANS1900 - ANS1919 271 - ANS1920 - ANS1939 272 - ANS1940 - ANS1959 273 - ANS1960 - ANS1979 274 - ANS2000 - ANS2019 275 - ANS2020 - ANS2039 276 - ANS2040 - ANS2059 277 - ANS2600 - ANS2619 278 - ANS2620 - ANS2639 279 - ANS3000 - ANS3019 280 - ANS4000 - ANS4019 281 - ANS4020 - ANS4039 282 - ANS4040 - ANS4059 283 - ANS4900 - ANS4919 284 - ANS4980 - ANS4999
Notice that each topic covers a range of message numbers. For example, to look up help information for message ANS1895I, you would select menu item 269 because ANS1895I is within the range of messages ANS1880 - ANS1899. This will display help information for all messages within that range. Use the navigation keys to move forward and backward through the information until you find the following text for message ANS1895I:
local filesystem. -----------------------------------------------------------------------ANS1895I Highest macro return code was return code value. Explanation: This message is issued after all commands in a client macro have completed. The return code represents the highest return code that was issued during processing of the macro. In order of increasing severity, the return code meanings are: 0 - The command completed successfully 4 - One or more files were skipped 8 - One or more warning messages were issued 12 - One or more error messages (except for skipped files) were issued System Action: None. User Response: For return codes other than 0, the user may wish to verify the results of the client operations and take diagnostic and repair actions, as necessary. u to scroll up, or d to scroll down: <<
Help for the Windows NT service configuration utility (dsmcutil)

To obtain help information, issue the dsmcutil command as follows: dsmmcutil help You are prompted to enter C for a basic command summary or F for full help information. The basic command summary is a listing of available dsmcutil subcommands and options. This information is most useful if you are already familiar with using dsmcutil.
260
The full help information will display more comprehensive information within the Windows help utility.
Help for the GUI and Web GUI clients

Help information for the GUI and Web GUI clients is available through the Help menu item.
Server or storage agent help

The server and storage agent both provide a help facility, which provides descriptions and syntax for server commands, as well as a full description of server messages which includes Explanation, System Action, and User Response.
Accessing server or storage agent help for commands

You must issue the help command to access help for the server or storage agent. Issue the following command for help on a server command: HELP commandName where commandName is the server command for which you want information. For example, the HELP REGISTER NODE command displays the description of the REGISTER NODE command as well as the syntax and information about the command parameters.
Accessing help for messages

You must issue the help command to access help for messages. Issue the following command for help on a server message: HELP message number where message number is the message for which you want information. If you specify the message number without including the message prefix, for example HELP 2011, it assumes the message prefix ANR and reports the help information for ANR2011. If the message number is specified with the prefix, for example HELP ANR2011, it will report the help information for that message. Issue HELP ANR2011 to view the following example output for that message:
ANR2011W Server is out of DB space in adding entries to the Activity Log. Console messages will not be logged until DB space is available. Explanation: The process that monitors the default (console) output stream and maintains the activity log cannot update the activity log due to a lack of database space. System Action: The server does not update the activity log. User Response: An authorized administrator can issue the DEFINE DBVOLUME command to add volumes for use by the database, and can issue the EXTEND DB command to extend the size of the database so that the new volumes are used.
Command-line interface help for server or storage agent

The command-line client interface includes a help facility that provides descriptions and syntax for client commands and options as well as a full description of client messages, including Explanation, System Action, and User Response. Help information for the GUI and Web GUI clients is available through the Help menu item.
261
Accessing help for commands

The command-line client help facility is accessed with the help command (dsmc help). After issuing the dsmc help command, you are presented with a menu of help topics. Each topic has a corresponding menu number. You may enter the menu number for a topic or use the u (up) and d (down) keys to scroll through the menu pages. Use the q key to quit the help facility. For instance, to locate help for the restore command, use the d key until the page containing the RESTORE topic is displayed:
The following help topics are available. Enter the number of the desired help topic or q to quit, d to scroll down, u to scroll up. 57 - QUERY SESSION 58 - QUERY SYSFILES 59 - QUERY SYSTEMINFO 60 - QUERY SYSTEMOBJECT 61 - QUERY SYSTEMSERVICES 62 - QUERY SYSTEMSTATE 63 - QUERY SYSVOL 64 - QUERY WAS 65 - QUERY WMI 66 - RESTART RESTORE 67 - RESTORE 68 - RESTORE ACTIVEDIRECTORY 69 - RESTORE ASR 70 - RESTORE BACKUPSET 71 - RESTORE CERTSERVERDB 72 - RESTORE CLUSTERDB 73 - RESTORE COMPLUSDB 74 - RESTORE EVENTLOG 75 - RESTORE FRS
In this example, you would enter 67 to view information on the restore command. To view information for a message, use the d and u keys to navigate the menu until you see entries that appear like the following entries:
The following help topics are available. Enter the number of the desired help topic or q to quit, d to scroll down, u to scroll up. The following help topics are available. Enter the number of the desired help topic or q to quit, d to scroll down, u to scroll up. 266 - ANS1820 - ANS1839v 267 - ANS1840 - ANS1859 268 - ANS1860 - ANS1879 269 - ANS1880 - ANS1899 270 - ANS1900 - ANS1919 271 - ANS1920 - ANS1939 272 - ANS1940 - ANS1959 273 - ANS1960 - ANS1979 274 - ANS2000 - ANS2019 275 - ANS2020 - ANS2039 276 - ANS2040 - ANS2059 277 - ANS2600 - ANS2619 278 - ANS2620 - ANS2639 279 - ANS3000 - ANS3019 280 - ANS4000 - ANS4019 281 - ANS4020 - ANS4039 282 - ANS4040 - ANS4059 283 - ANS4900 - ANS4919 284 - ANS4980 - ANS4999
262
Notice that each topic covers a range of message numbers. For example, to look up help information for message ANS1895I, you would select menu item 269 because ANS1895I is within the range of messages ANS1880 - ANS1899. This will display help information for all messages within that range. Use the navigation keys to move forward and backward through the information until you find the following text for message ANS1895I:
local filesystem. -----------------------------------------------------------------------ANS1895I Highest macro return code was return code value. Explanation: This message is issued after all commands in a client macro have completed. The return code represents the highest return code that was issued during processing of the macro. In order of increasing severity, the return code meanings are: 0 - The command completed successfully 4 - One or more files were skipped 8 - One or more warning messages were issued 12 - One or more error messages (except for skipped files) were issued System Action: None. User Response: For return codes other than 0, the user may wish to verify the results of the client operations and take diagnostic and repair actions, as necessary. u to scroll up, or d to scroll down: <<
Accessing help for the Windows NT service configuration utility (dsmcutil)

You must issue the dsmcutil command to obtain help information. Issue dsmmcutil help to access help. You are prompted to enter C for a basic command summary or F for full help information. The basic command summary is a listing of available dsmcutil subcommands and options. This information is most useful if you are already familiar with using dsmcutil. The full help information will display more comprehensive information within the Windows help utility.
Help system does not display

You must verify that the help system is running and it is installed correctly.
Verifying that the help system is running

You cannot view the help when the help system is not running. First verify that the help server process is running. On Windows: Open the Task Manager and select the process tab. On UNIX: 1. Run a command similar to ps -ef | grep eclipse. 2. Look for either eclipse.exe or EclipseSrv.exe in the list of processes. If you do not see either process name in the list, the help system is not running. You must start the help system. 3. When there are multiple processes with the same names in the list, or if continued attempts to start the help server fail, further investigation is necessary. This situation arises if you are using the publications CD on the same system, or some other application is using the port. For
263
example, if the random port number that the publications CD chooses is 8423 (the port for the help system), then the Integrated Solutions Console help server will fail to start. 4. Try pointing your Web browser to http://<machine name>:8423. For example, http://workgroup.tucson.ibm.com:8423 would show the help system on workgroup.tucson.ibm.com. You should see the Integrated Solutions Console title on the top, left-hand corner of the screen. If you do not see that title, then some other help system or program is running on port 8423. Stop that application and restart the Integrated Solutions Console help server.
Resolving help system installation problems

The Integrated Solutions Console (ISC) uses Eclipse to provide a help system. The help systems content is stored in a number of plug-ins, which are directories with a defined structure. The Tivoli Storage Manager Administration Center uses two plug-ins for its help topics: com.tivoli.dsm.admincenter.help.doc.5.3.0.0 and com.tivoli.dsm.admincenter.demo. The first of these plug-ins contains the help content for the Tivoli Storage Manager Administration Center at Version 5.3.0. (Future versions of the Tivoli Storage Manager Administration Center might change the version number, the numerical characters in the plug-in name.) Verify that the help plug-in exists and contains a doc.zip file. Plug-ins are located in <isc root>/PortalServer/ISCEclipse/plugins where <isc root> is the ISC installation directory. For example, on Windows, look in C:\program files\IBM\ISC\PortalServer\ISCEclipse\plugins for the com.tivoli.dsm.admincenter.help.doc.5.3.0.0 directory. On UNIX, look in /opt/IBM/ISC/PortalServer/ISCEclipse/plugins for that directory. If you see the directory and it contains the doc.zip file, the plug-in is installed correctly. Otherwise, you must re-run the Administration Center installation to install the help topics.
Starting and stopping the help system

You might be able to resolve your help system problem by starting and stopping the service. On Windows The help server is a service on Windows. Open the service manager and locate the ISC Help Server service. To start the service, right-click on the server and select Start from the pop-up menu. To stop the service, right-click on the service name and select Stop from the pop-up menu. On UNIX To start or stop the help server on UNIX, change the directory to <isc root>/PortalServer/ISCEclipse. Two scripts, startEclipse.sh and stopEclipse.sh, are in the ISCEclipse directory. Run startEclipse.sh to start the help server. Run stopEclipse.sh to stop the help server. See Verifying that the help system is running on page 263 to determine if the server is running.
264
Adding the publications to the Integrated Solutions Console help server

The publication CD contains several Eclipse plug-ins that you can copy from the publication CD to your Integrated Solutions Console (ISC) help server. Perform the following steps to install the publications into the help server: 1. Copy any directory that starts with com.ibm from ibm_help_win\eclipse\ plugin on the publication CD to <isc root>/PortalServer/ISCEclipse/plugins, where <isc root> is the installation location of the Integrated Solutions Console. 2. Stop the help server. See Starting and stopping the help system on page 264 for more information. 3. Start the help server. See Starting and stopping the help system on page 264 for more information. The publications are now in the help system. If you run into a problem, verify that the directory structure on the publication CD matches the one in <isc root>/PortalServer/ISCEclipse/plugins. You might not need all publications; you can choose which plug-ins to copy. To avoid problems when attempting to access the publications in the help system, copy the following plugins: v com.ibm.help.doc v com.ibm.help.doc.nl1 v com.ibm.help.ic.doc v com.ibm.support.itsm.core.doc.nl1_1.0.0 v com.ibm.support.itsm.core.doc_1.0.0 v com.ibm.support.itsm.doc_1.0.0
Reporting a problem with a help topic

You must collect certain information before you report a problem with the help system. v Record what you clicked on to get the help. For example, if you clicked the question mark for a portal, record the name of the portal. v View the source of the help pop-up window. On most browsers, a right mouse-click will show you a menu with a View Source option. Select View Source and a window is displayed with the HTML source code for that window. Write down the title of that window, which will be the URL or the name of the file that the help system is attempting to display.
265
266
Chapter 11. Determining data storage problems

Data storage issues can be resolved through several methods.
Using data storage diagnostic tips

If you are experiencing a problem in storing or retrieving data, review the diagnostic tips to try to isolate or resolve the problem.

Check the server activity log for other messages 30 minutes before and 30 minutes after the time of the error. Issue the QUERY ACTLOG command to check the activity log. Often, other messages that are issued can offer additional information about the cause of the problem and how to resolve it.
Checking HELP for messages issued for the problem

Check HELP for any messages issued by the IBM Tivoli Storage Manager. The IBM Tivoli Storage Manager messages provide additional information beyond just the message itself. The Explanation, System Action, or User Response sections of the message may provide additional information about the problem. Often, this supplemental information about the message may provide the necessary steps necessary to resolve the problem.
Recreating the problem

If a problem can be easily or consistently recreated, it may be possible to isolate the cause of the problem to a specific sequence of events. Data read or write problems may be sequence-related, in terms of the operations being performed, or may be an underlying device error or failure. Typical problems relating to the sequence of events occur for sequential volumes. One example would be that a volume is in use for a client backup and that volume is preempted by a restore of another client nodes data. In this case, this may surface as an error to the client backup session that was preempted. However, that client backup session may succeed if it was retried or if it was not preempted in the first place.
Resolving error related to reading or writing to a device

If the problem is an error reading or writing data from a device, many systems and devices record information in a system error log. Examples of the system error log the are errpt for AIX, Event Log for Windows, and the System Log for z/OS. If a device or volume used by the IBM Tivoli Storage Manager is reporting an error to the system error log, it is likely a device issue. The error messages recorded in the system error log may provide enough information to resolve the problem.
267
Reviewing the hints and tips for drives and libraries

The hints and tips for drives and libraries offer many items to check to ensure that data read and write operations are successful. Review the Tape drives and libraries hints and tips on page 223.
Changing the storage hierarchy

The storage hierarchy includes the defined storage pools and the relationships between the storage pools on the server. These storage pool definitions are also used by the storage agent. If attributes of a storage pool were changed, this may affect data store and retrieve operations. Review any changes to the storage hierarchy and storage pool definitions. Use QUERY ACTLOG to see the history of commands or changes that may affect storage pools. Also, use the following QUERY commands to determine if any changes were made: v QUERY STGPOOL F=D Review the storage pool settings. If a storage pool is UNAVAILABLE, then data in that storage pool cannot be accessed. If a storage pool is READONLY, then data cannot be written to that pool. If either of these is the case, review why these values were set and consider issuing the UPDATE STGPOOL command to set the pool to READWRITE. Another consideration is to review the number of scratch volumes available for a sequential media storage pool. v QUERY DEVCLASS F=D The storage pools can be influenced by changes to device classes. Review the device class settings for the storage pools. This may also include checking the library, drive, and path definitions using the QUERY LIBRARY, QUERY DRIVE, and QUERY PATH commands for sequential media storage pools.
Changing the server policies

The server policy attributes that directly relate to data storage are the copy group destinations for backup and archive copy groups. Similarly, the management class, MIGDESTINATION, also impacts where data is stored. Review any changes to the server storage policies. Issue the QUERY ACTLOG command to view the history of commands or changes that may affect storage policies. Also, use the following QUERY commands to determine if any changes were made: v QUERY COPYGROUP F=D Review the DESTINATION settings for the TYPE=BACKUP and TYPE=ARCHIVE copy groups. Also review the Migration Destination for management classes used by HSM clients. If storage pool destinations were changed and resulting data read or write operations are now failing, either evaluate the changes made and correct the problem, or revert to the previous settings. v QUERY NODE F=D Assigning a node to a different domain may impact data read and write operations for that client. Specifically, the node may now be going to storage pool destinations that are not appropriate, based on the requirements of this
268
node. For example, it may be assigned to a domain that does not have any TYPE=ARCHIVE copy group destinations. If this node tries to archive data, it will fail.
Resolving a backup or copy problem that occurs only with a specific node
If you cannot backup or copy data to a specific node, you might not have an active data pool listed in your active destinations. These are specified in the nodes policy domain. Issue the QUERY NODE nodeName F=D command to verify that the node that is storing the data is authorized. The QUERY NODE command finds the policy domain name to which the node is assigned. Issue the QUERY DOMAIN domain_name where domain_name is the output gathered from the previous QUERY NODE command. Look in the ACTIVEDESTINATION parameter for the list of active data ports. If the active data pool into which you want to store data is not on the list, issue the UPDATE DOMAIN command to add the active data pool to the list.
Resolving a problem that occurs only for a specific volume

If problems occur only for a specific storage volume, there may be an error with the volume itself. This is true whether the volume is sequential media or DISK. If this is a data write operation, issue the UPDATE VOLUME volumeName ACCESS=READONLY command to set this volume to READONLY, then retry the operation. If the operation succeeds, try setting the original volume back to READWRITE by issuing the UPDATE VOLUME volumeName ACCESS=READWRITE command and retrying the operation. If the operation fails only when using this volume, consider issuing the AUDIT VOLUME command to evaluate this volume and issue the MOVE DATA command to move the data from this volume to other volumes in the storage pool. After the data is moved off of this volume, delete the volume by issuing the DELETE VOLUME command.
Resolving a problem that occurs only when you use a specific drive to read or write to a volume
For sequential media volumes, does the error occur only when using a specific drive? If the error occurs only when using a specific drive, update that drive to be offline by issuing the UPDATE DRIVE libraryName driveName ONLINE=NO command. Then take the appropriate steps to correct the drive error, or else have the vendor for the drive, service the drive. After the drive is corrected, issue the UPDATE DRIVE libraryName driveName ONLINE=YES command to set the drive online again.
Resolving SCSI device problems

Did the Tivoli Storage Manager issue message ANR8300, ANR8301, ANR8302, ANR8303, ANR8943, or ANR98944? Tape drives and libraries may report information back to the Tivoli Storage Manager about the error encountered. This information is reported in one or more of the messages listed. The data that the Tivoli Storage Manager reports from these devices may provide sufficient detail to determine the steps needed to resolve the problem. Generally, when the Tivoli Storage Manager server reports device sense
Chapter 11. Determining data storage problems
269
data using these messages, the problem is typically with the device, the connection to the device, or some other related issue outside of the Tivoli Storage Manager. Using the information reported in Tivoli Storage Manager message ANR8300, ANR8301, ANR8302, ANR8303, ANR8943, or ANR8944, refer to the Appendix C of the Tivoli Storage Manager Messages manual. This appendix documents information about standard errors that may be reported by any SCSI device. You can also use this information with documentation provided by the vendor for the hardware to help determine the cause and resolution for the problem.
Sequential media volume (tape)

You can determine your sequential media volumes problem through some helpful flags.
ANR0542W Retrieve or restore failed for session sessionNumber for node nodeName - storage media inaccessible
Error message ANR0542W is often related to an issue with the drive or connection to the drive that was selected to read this tape volume. Perform the following steps to verify that the Tivoli Storage Manager can access this volume: v Issue the QUERY LIBVOL libraryName volumeName command. v For a 349X library, issue the mtlib -l /dev/lmcp0 -qV volumeName command. The device is typically /dev/lmcp0, but if it is different, then substitute the correct library manager control point device. The following are the possible steps to resolve this problem: 1. If mtlib does not report this volume, then it appears that this volume is out of the library. In this case, put the volume back into the library. 2. If the volume is not reported by QUERY LIBVOL, then the server does not know about this volume in the library. Issue the CHECKIN LIBVOL command to synchronize the library inventory in the server with the volumes that are actually in the tape library. 3. If both commands successfully report this volume, then the cause is likely a permanent or intermittent hardware error. This may be an error with the drive itself or an error with the connection to the drive. In either case, review the system error logs and contact the vendor of the hardware to resolve the problem.
ANR8778W Scratch volume changed to private status to prevent re-access

Review the activity log messages to determine the cause of the problem involving this scratch volume. Also, review the system error logs and device error logs for an indication that there was a problem with the drive used to try to write to this scratch volume. If this was caused by a drive requiring cleaning, or some other hardware-specific issue that was resolved, any volumes that were set to private status as a result of this may be reset to scratch by issuing the AUDIT LIBRARY libraryName command.
270
Notices
This information was developed for products and services offered in the U.S.A. IBM may not offer the products, services, or features discussed in this document in other countries. Consult your local IBM representative for information on the products and services currently available in your area. Any reference to an IBM product, program, or service is not intended to state or imply that only that IBM product, program, or service may be used. Any functionally equivalent product, program, or service that does not infringe any IBM intellectual property right may be used instead. However, it is the users responsibility to evaluate and verify the operation of any non-IBM product, program, or service. IBM may have patents or pending patent applications covering subject matter described in this document. The furnishing of this document does not grant you any license to these patents. You can send license inquiries, in writing, to: IBM Director of Licensing IBM Corporation North Castle Drive Armonk, NY 10504-1785 U.S.A. For license inquiries regarding double-byte (DBCS) information, contact the IBM Intellectual Property Department in your country or send inquiries, in writing, to: IBM World Trade Asia Corporation Licensing 2-31 Roppongi 3-chome, Minato-ku Tokyo 106-0032, Japan The following paragraph does not apply to the United Kingdom or any other country where such provisions are inconsistent with local law: INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION AS IS WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of express or implied warranties in certain transactions, therefore, this statement may not apply to you. This information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in new editions of the publication. IBM may make improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time without notice. Any references in this information to non-IBM Web sites are provided for convenience only and do not in any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of the materials for this IBM product and use of those Web sites is at your own risk. IBM may use or distribute any of the information you supply in any way it believes appropriate without incurring any obligation to you.
271
Licensees of this program who wish to have information about it for the purpose of enabling: (i) the exchange of information between independently created programs and other programs (including this one) and (ii) the mutual use of the information which has been exchanged, should contact: IBM Corporation 2Z4A/101 11400 Burnet Road Austin, TX 78758 U.S.A. Such information may be available, subject to appropriate terms and conditions, including in some cases, payment of a fee. The licensed program described in this information and all licensed material available for it are provided by IBM under terms of the IBM Customer Agreement, IBM International Program License Agreement, or any equivalent agreement between us. Any performance data contained herein was determined in a controlled environment. Therefore, the results obtained in other operating environments may vary significantly. Some measurements may have been made on development-level systems and there is no guarantee that these measurements will be the same on generally available systems. Furthermore, some measurements may have been estimated through extrapolation. Actual results may vary. Users of this document should verify the applicable data for their specific environment. Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. All statements regarding IBMs future direction or intent are subject to change or withdrawal without notice, and represent goals and objectives only. This information is for planning purposes only. The information herein is subject to change before the products described become available. This information contains examples of data and reports used in daily business operations. To illustrate them as completely as possible, the examples include the names of individuals, companies, brands, and products. All of these names are fictitious and any similarity to the names and addresses used by an actual business enterprise is entirely coincidental. COPYRIGHT LICENSE: This information contains sample application programs in source language, which illustrate programming techniques on various operating platforms. You may copy, modify, and distribute these sample programs in any form without payment to IBM, for the purposes of developing, using, marketing or distributing application programs conforming to the application programming interface for the operating platform for which the sample programs are written. These examples have not been thoroughly tested under all conditions. IBM, therefore, cannot guarantee or imply reliability, serviceability, or function of these programs.
272
Each copy or any portion of these sample programs or any derivative work, must include a copyright notice as follows: IBM (2007). Portions of this code are derived from IBM Corp. Sample Programs. Copyright IBM Corp. 1993-2007. All rights reserved. If you are viewing this information softcopy, the photographs and color illustrations may not appear.
Programming interface information

This Problem Determination Guide contains information on intended programming interfaces that allow the customer to write programs to obtain the services of IBM Tivoli Storage Manager. This Problem Determination Guidecontains information that is not intended to be used as a programming interface of IBM Tivoli Storage Manager.
Notices
273
274
Glossary
The terms in this glossary are defined as they pertain to the IBM Tivoli Storage Manager library. If you do not find the term you need, refer to the IBM Software Glossary on the Web at this address: http://www.ibm.com/ibm/terminology/. This glossary may include terms and definitions from: v The American National Standard Dictionary for Information Systems, ANSI X3.172-1990, copyright (ANSI). Copies may be purchased from the American National Standards Institute, 11 West 42nd Street, New York 10036. v The Information Technology Vocabulary, developed by Subcommittee 1, Joint Technical Committee 1, of the International Organization for Standardization and the International Electrotechnical Commission (ISO/IEC JTC2/SC1). term definition
275
276
Trademarks
Tivoli Storage Manager, the Tivoli Storage Manager logo and the following terms are trademarks of the International Business Machines Corporation in the United States, other countries, or both: v AIX v v v v v v v v v v v v v v v v v v v v v v v v v v v v v v v v v v v AS/400 DB2 DFS DFSMS/MVS DFSMShsm DFSMSrmm DPI Enterprise Storage Server ESCON eServer FICON FlashCopy HACMP Informix iSeries Lotus Lotus 123 Lotus Approach Lotus Domino Lotus Notes Magstar MVS NetView OpenEdition OS/2 OS/390 OS/400 Passport Advantage pSeries RACF Rational Redbooks RS/6000 S/390 SANergy SecureWay
v StorageSmart v SystemView
277
v v v v v v v
Tivoli Tivoli Enterprise Console Tivoli Management Environment TotalStorage TME VTAM WebSphere
v z/OS v zSeries v Intel, Intel logo, Intel Inside, Intel Inside logo, Intel Centrino, Intel Centrino logo, Celeron, Intel Xeon, Intel SpeedStep, Itanium, and Pentium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. v Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both. v Microsoft, Windows Windows NT and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both. v UNIX is a registered trademark of The Open Group in the United States and other countries. v Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both. Other company, product, and service names may be trademarks or service marks of others.
278
Appendix A. Using gt script

#!/bin/ksh # this script uses gdb to extract thread info from a core file # input parameters are the path/name of the executable (default ./dsmserv) # and the path/name of the core file (default ./dsmcore) # output file is dsm_gdb.info # # note that files named dsm_gdb.cmd and dsm_gdb.info will be overwritten # # if you see the following error: # ./dsm_gdb.cmd:9: Error in source command file: # No symbol table is loaded. Use the "file" command. # then comment out the line that prints buildStringP # # if you see other errors, youre on your own ... exe=${1:-"./dsmserv"} # get parm 1 (executable file path/name), set default core=${2:-"./dsmcore"} # get parm 2 (core file path/name), set default echo " " # look for the executable file ... quit if not found if [[ -f $exe ]]; then echo "using executable file:" $exe else echo "didnt find executable file ("$exe") ... exiting" exit fi # look for the core file, if not found, look for ./core ... quit if not found if [[ -f $core ]]; then echo "using core file:" $core else if [[ -f ./core ]]; then echo "didnt find core file ("$core") but found ./core ... renaming to" $core mv ./core $core echo "using core file:" $core else echo "didnt find core file ("$core") ... exiting" exit fi fi echo " " # make gdb command file to get thread info nl="\0134\0156" # octal codes for \n (so echo wont think its \n) echo "# dsm gdb command file" >|dsm_gdb.cmd echo "define doit" >>dsm_gdb.cmd echo "info registers" >>dsm_gdb.cmd # show register values echo "echo" $nl >>dsm_gdb.cmd echo "where" >>dsm_gdb.cmd # show function traceback echo "echo" $nl"====================================="$nl >>dsm_gdb.cmd echo "end" >>dsm_gdb.cmd echo "echo" $nl"====================================="$nl$nl >>dsm_gdb.cmd echo "x/s buildStringP" >>dsm_gdb.cmd echo "echo" $nl"====================================="$nl$nl >>dsm_gdb.cmd echo "info threads" >>dsm_gdb.cmd # show thread info echo "echo" $nl"====================================="$nl >>dsm_gdb.cmd echo "thread apply all doit" >>dsm_gdb.cmd echo "quit" >>dsm_gdb.cmd echo "invoking gdb to get thread info (watch for errors) ..." echo "if you see:" echo ". warning: The shared libraries were not privately mapped; setting a" echo ". breakpoint in a shared library will not work until you rerun the program" echo "thats ok." echo "if you see:" echo ". ./dsm_gdb.cmd:x: Error in source command file:"
279
echo "then type quit, edit this script, and read the comments at the top" gdb -se $exe -c $core -x ./dsm_gdb.cmd >|dsm_gdb.info rm dsm_gdb.cmd # done with this now exit
280
Appendix B. Installing and running the tsmdiag utility

The purpose of the Tivoli Storage Manager Diagnostic utility (tsmdiag) is to ease and speed up the process for gathering data that is considered valuable to assist in diagnosing a problem caused by a server component. The tsmdiag utility collects a set of data by default. This data is specified in a configuration file called configfile.txt. If at any point more data is identified as valuable to diagnose any particular problem from a Tivoli Storage Manager component, users are instructed to modify the configuration file used by tsmdiag so that such data is collected by the tsmdiag utility. A strong suggestion is that users run tsmdiag on the system where the Tivoli Storage Manager component with a problem is installed. This is done before submitting a problem to Tivoli Storage Manager Customer Support. Submit the data collected by tsmdiag along with the problem report.
Table 16. Supported OS environments and components Operating System Environment AIX Tivoli Storage Manager Component Server Storage Agent Device Driver HP-UX Server Storage Agent Device Driver Linux Server Storage Agent Device Driver Solaris Server Storage Agent Device Driver Windows Server Storage Agent Device Driver Management Console
Note: The tsmdiag utility is not installed with the Tivoli Storage Manager device driver, or with the Tivoli Storage Manager Management Console, but if these components are installed on the system where tsmdiag is running, tsmdiag will collect data for these components.
281
Where to find and how to use the Tivoli Storage Manager diagnostic utility
Tivoli Storage Manager V5.3, 5.4, and 5.5 are equipped with the tsmdiag utility, with supported Tivoli Storage Manager components found under the components install directory. The tsmdiag utility can be found in the following locations: v AIX: /usr/tivoli/tsm/server/bin/tsmdiag v Windows: C:\Program Files\Tivoli\TSM\server\tsmdiag v Linux: /opt/tivoli/tsm/server/bin/tsmdiag Usage:
tsmdiag [options]
Where options can be any combination of the following:

Option -id adminName -pa adminPwd Description Tivoli Storage Manager server administrator id Tivoli Storage Manager server administrator password Tivoli Storage Manager server TPC/IP name or address Tivoli Storage Manager server TCP/IP port Directory for created files verbosely list activity processed Display usage information Default Value admin admin
-tcpserveraddress ipAddress
localhost
-tcpport portNumber -results resultsDir -v -?
1500 ($cwd)/results non-verbose N/A
Examples:
./tsmdiag ./tsmdiag -v ./tsmdiag -id tsmadmin -pa pwd4u -results /home/tsmdiag/results_oct29
Note: For the Tivoli Storage Manager server environments on Unix, if you are running multiple servers on the same system, you must set the DSMSERV_DIR and DSMSERV_CONFIG environment variables. Failure to do this prevents tsmdiag from finding data for the desired server.
Where to find the data collected by tsmdiag

After the tsmdiag utility has completed running, go to the either the default results directory, $CWD/results, or the directory that you specified through the -results command line option. There will be a file there called tsmdiag_results.tar (for Unix environments) and tsmdiag_results.zip (for Windows environments). Submit this file with the PMR.
Default data collected by tsmdiag

The default data collected by the tsmdiag utility is specified in the configfile.txt configuration file. More data can be added to this file as needed.
282
Each line in the configuration file consists of five fields, each separated by comma. A line that begins with an asterisk(*) is considered a comment and is ignored by the tsmdiag utility. The first field of a line in the configuration file denotes the operating system where tsmdiag will be run. Valid platforms are Linux, AIX, HP, SUN, and Windows. The second field in the line denotes an action. Valid actions are listed in the table below:
Action Name SYSCOMMAND COPY COMPCOMMAND REGISTRY Action Performed Run a command on the system. Copy a specified file. Run a command on a specified Tivoli Storage Manager component. Get the value for a given registry key. (Windows only.)
The third field denotes the Tivoli Storage Manager component (i.e. component for which tsmdiag will collect data). Valid components are: v v v v v v v SERVER STAGENT BACLIENT (Unix only) MGMTCONSOLE (Windows only) DEVDRIVER SYSTEM CLIENAPI (Unix only)
The fourth field denotes the data that the action will act upon. The data field can be a file name, a system command, a command for a Tivoli Storage Manager component, a registry key, and so on, depending on the action. The last field is optional and denotes the location of where the collected data is to be stored.
Appendix B. Installing and running the tsmdiag utility
283
284
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Appendix C. IBM Global Security Kit return codes

Version 5.5 of the IBM Tivoli Storage Manager uses the IBM Global Security Kit (GSKit) on AIX and Windows for Secure Sockets Layer (SSL) processing between the Tivoli Storage Manager server and the backup-archive client. Some messages issued for SSL processing include GSKit return codes. GSKit is automatically installed or updated during Tivoli Storage Manager installation on AIX and Windows and provides the following libraries: v GSKit SSL v GSKit Key Management API v IBM Crypto for C (ICC) The tsmdiag utility reports the GSKit level installed on your system, or you can use one of the following methods: v For Windows, issue the following commands:
regedit /e gskitinfo.txt "HKEY_LOCAL_MACHINE\software\ibm\gsk7\" notepad gskitinfo.txt
CAUTION: You can damage the system registry if you use regedit incorrectly. v For the 64-bit AIX server, issue the following from the command line: gsk7ver_64 See Table 17 for the GSKit SSL return codes. The Tivoli Storage Manager server uses the GSKit Key Management API to automatically create the key management database and Tivoli Storage Manager server private and public keys. Some messages issued for this processing might include GSKit Key Management return codes. See Table 18 on page 289 for the key management return codes.
Table 17. IBM Global Security Kit SSL general return codes Return code (hex) 0x00000000 Return code (decimal) Constant 0 GSK_OK Explanation The task completed successfully. Issued by every function call that completes successfully. The environment or Secure Sockets Layer (SSL) handle is not valid. The specified handle was not the result of a successful open() function call. The dynamic link library (DLL) was unloaded and is not available. (Occurs on Microsoft Windows systems only.) Internal error. Report this error to IBM Software Support. Insufficient memory is available to perform the operation.
0x00000001
GSK_INVALID_HANDLE
0x00000002
GSK_API_NOT_AVAILABLE
0x00000003 0x00000004
3 4
GSK_INTERNAL_ERROR GSK_INSUFFICIENT_STORAGE
285
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Table 17. IBM Global Security Kit SSL general return codes (continued) Return code (hex) 0x00000005 Return code (decimal) Constant 5 GSK_INVALID_STATE Explanation The handle is in an invalid state for operation, such as performing an init() operation on a handle twice. Specified key label not found in key file. Certificate not received from partner. Certificate validation error. Error processing cryptography. Error validating ASN fields in certificate. Error connecting to user registry. Internal error. Report this error to IBM Software Support. Internal error. Report this error to IBM Software Support. I/O error reading the key file. The key file has an invalid internal format. Recreate key file. The key file has two entries with the same key. The key file has two entries with the same label. The key file password is used as an integrity check. Either the key file has become corrupted or the password ID is incorrect. The default key in the key file has an expired certificate. An error occurred loading one of the GSK dynamic link libraries. Be sure GSK was installed correctly. Indicates that a connection is trying to be made in a GSK environment after the GSK_ENVIRONMENT_ CLOSE_OPTIONS was set to GSK_DELAYED_ ENVIRONMENT_CLOSE and gsk_environment_close() function was called. Neither the password nor the stash-file name was specified, so the key file could not be initialized. Unable to open the key file. Either the path was specified incorrectly or the file permissions did not allow the file to be opened. Unable to generate a temporary key pair. Report this error to IBM Software Support. A User Name object was specified that is not found.
0x00000006 0x00000007 0x00000008 0x00000009 0x0000000a 0x0000000b 0x0000000c 0x00000065 0x00000066 0x00000067 0x00000068 0x00000069 0x0000006a
6 7 8 9 10 11 12 101 102 103 104 105 106
GSK_KEY_LABEL_NOT_FOUND GSK_CERTIFICATE_NOT_AVAILABLE GSK_ERROR_CERT_VALIDATION GSK_ERROR_CRYPTO GSK_ERROR_ASN GSK_ERROR_LDAP GSK_ERROR_UNKNOWN_ERROR GSK_OPEN_CIPHER_ERROR GSK_KEYFILE_IO_ERROR GSK_KEYFILE_INVALID_FORMAT GSK_KEYFILE_DUPLICATE_KEY GSK_KEYFILE_DUPLICATE_LABEL GSK_BAD_FORMAT_OR_INVALID_PASSWORD
0x0000006b 0x0000006c
107 108
GSK_KEYFILE_CERT_EXPIRED GSK_ERROR_LOAD_GSKLIB
0x0000006d
109
GSK_PENDING_CLOSE_ERROR
0x000000c9
201
GSK_NO_KEYFILE_PASSWORD
0x000000ca
202
GSK_KEYRING_OPEN_ERROR
0x000000cb
203
GSK_RSA_TEMP_KEY_PAIR
0x000000cc
204
GSK_ERROR_LDAP_NO_SUCH_OBJECT
286
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Table 17. IBM Global Security Kit SSL general return codes (continued) Return code (hex) 0x000000cd 0x000000ce 0x000000cf 0x0000012d Return code (decimal) Constant 205 206 207 301 GSK_ERROR_LDAP_INVALID_CREDENTIALS GSK_ERROR_BAD_INDEX GSK_ERROR_FIPS_NOT_SUPPORTED GSK_CLOSE_FAILED Explanation A Password used for an LDAP query is not correct. An index into the Fail Over list of LDAP servers was not correct This installation of GSKit does not support FIPS mode of operation Indicates that the GSK environment close request was not properly handled. Cause is most likely due to a gsk_secure_socket*() command being attempted after a gsk_close_environment() call. The system date was set to an invalid value. Neither SSLv2 nor SSLv3 is enabled. The required certificate was not received from partner. The received certificate was formatted incorrectly. The received certificate type was not supported. An I/O error occurred on a data read or write operation. The specified label in the key file could not be found. The specified key file password is incorrect. The key file could not be used. The key file also might be corrupt. In a restricted cryptography environment, the key size is too long to be supported. An incorrectly formatted SSL message was received from the partner. The message authentication code (MAC) was not successfully verified. Unsupported SSL protocol or unsupported certificate type. The received certificate contained an incorrect signature. Incorrectly formatted certificate received from partner. Invalid SSL protocol received from partner. Report this error to IBM Software Support. The self-signed certificate is not valid. The read() failed. Report this error to IBM Software Support.
0x00000191 0x00000192 0x00000193 0x00000194 0x00000195 0x00000196 0x00000197 0x00000198
401 402 403 404 405 406 407 408
GSK_ERROR_BAD_DATE GSK_ERROR_NO_CIPHERS GSK_ERROR_NO_CERTIFICATE GSK_ERROR_BAD_CERTIFICATE GSK_ERROR_UNSUPPORTED_CERTIFICATE_TYPE GSK_ERROR_IO GSK_ERROR_BAD_KEYFILE_LABEL GSK_ERROR_BAD_KEYFILE_PASSWORD
0x00000199 0x0000019a 0x0000019b 0x0000019c 0x0000019d 0x0000019e 0x0000019f 0x000001a0 0x000001a1 0x000001a2
409 410 411 412 413 414 415 416 417 418
GSK_ERROR_BAD_KEY_LEN_FOR_EXPORT GSK_ERROR_BAD_MESSAGE GSK_ERROR_BAD_MAC GSK_ERROR_UNSUPPORTED GSK_ERROR_BAD_CERT_SIG GSK_ERROR_BAD_CERT GSK_ERROR_BAD_PEER GSK_ERROR_PERMISSION_DENIED GSK_ERROR_SELF_SIGNED GSK_ERROR_NO_READ_FUNCTION
287
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Table 17. IBM Global Security Kit SSL general return codes (continued) Return code (hex) 0x000001a3 0x000001a4 0x000001a5 0x000001a6 0x000001a7 0x000001a8 0x000001a9 0x000001aa 0x000001ab 0x000001ac 0x000001ad 0x000001ae 0x000001af 0x000001b0 0x000001b1 0x000001b2 Return code (decimal) Constant 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 GSK_ERROR_NO_WRITE_FUNCTION GSK_ERROR_SOCKET_CLOSED GSK_ERROR_BAD_V2_CIPHER GSK_ERROR_BAD_V3_CIPHER GSK_ERROR_BAD_SEC_TYPE GSK_ERROR_BAD_SEC_TYPE_COMBINATION GSK_ERROR_HANDLE_CREATION_FAILED GSK_ERROR_INITIALIZATION_FAILED GSK_ERROR_LDAP_NOT_AVAILABLE GSK_ERROR_NO_PRIVATE_KEY GSK_ERROR_PKCS11_LIBRARY_NOTLOADED GSK_ERROR_PKCS11_TOKEN_LABELMISMATCH GSK_ERROR_PKCS11_TOKEN_NOTPRESENT GSK_ERROR_PKCS11_TOKEN_BADPASSWORD GSK_ERROR_INVALID_V2_HEADER GSK_CSP_OPEN_ERROR Explanation The write() failed. Report this error to IBM Software Support. The partner closed the socket before the protocol completed. The specified V2 cipher is not valid. The specified V3 cipher is not valid. Report this error to IBM Software Support. Report this error to IBM Software Support. The handle could not be created. Report this error to IBM Software Support. Initialization failed. Report this internal error to service. When validating a certificate, unable to access the specified user registry. The specified key did not contain a private key. A failed attempt was made to load the specified PKCS11 shared library. The PKCS #11 driver failed to find the token specified by the caller. A PKCS #11 token is not present in the slot. The password/pin to access the PKCS #11 token is invalid. The SSL header received was not a properly SSLv2 formatted header. Could not open the hardware-based cryptographic service provider. Either the CSP name is not specified correctly or a failed attempt was made to access the specified CSPs certificate store. Some conflicting attributes for SSL operation were defined. The Microsoft Crypto API is only supported on Microsoft Windows 2000 with Service Pack 2 applied. System is running in IPv6 mode without setting a PEERID. The buffer size is negative or zero. Used with nonblocking I/O. Refer to the nonblocking section for usage. SSLv3 is required for reset_cipher(), and the connection uses SSLv2. An invalid ID was specified for the gsk_secure_soc_misc() function call.
0x000001b3 0x000001b4
435 436
GSK_CSP_OPEN_ERROR GSK_CSP_OPEN_ERROR
0x000001b5 0x000001f5 0x000001f6 0x00000259 0x0000025a
437 501 502 601 602
GSK_CSP_OPEN_ERROR GSK_INVALID_BUFFER_SIZE GSK_WOULD_BLOCK GSK_ERROR_NOT_SSLV3 GSK_MISC_INVALID_ID
288
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Table 17. IBM Global Security Kit SSL general return codes (continued) Return code (hex) 0x000002bd Return code (decimal) Constant 701 GSK_ATTRIBUTE_INVALID_ID Explanation The function call has an invalid ID. This also might be caused by specifying an environment handle when a handle for a SSL connection should be used. The attribute has a negative length, which is invalid. The enumeration value is invalid for the specified enumeration type. Invalid parameter list for replacing the SID cache routines. When setting a numeric attribute, the specified value is invalid for the specific attribute being set. Conflicting parameters were set for additional certificate validation The AES cryptographic algorithm is not supported. The PEERID does not have the correct length.
0x000002be 0x000002bf 0x000002c0 0x000002c1
702 703 704 705
GSK_ATTRIBUTE_INVALID_LENGTH GSK_ATTRIBUTE_INVALID_ENUMERATION GSK_ATTRIBUTE_INVALID_SID_CACHE GSK_ATTRIBUTE_INVALID_NUMERIC_VALUE
0x000002c2 0x000002c3 0x000002c4 0x000005dd 0x000005de 0x00000641 0x00000642 0x00000643 0x00000644 0x00000645
706 707 708 1501 1502 1601 1602 1603 1604 1605
GSK_CONFLICTING_VALIDATION_SETTING GSK_AES_UNSUPPORTED GSK_PEERID_LENGTH_ERROR GSK_SC_OK GSK_SC_CANCEL GSK_TRACE_STARTED GSK_TRACE_STOPPED GSK_TRACE_NOT_STARTED GSK_TRACE_ALREADY_STARTED GSK_TRACE_OPEN_FAILED
The trace started successfully. The trace stopped successfully. No trace file was previously started so it cannot be stopped. Trace file already started so it cannot be started again. Trace file cannot be opened. The first parameter of gsk_start_trace() must be a valid full path file name.
Table 18. IBM Global Security Kit key management return codes Return code (hex) 0x00000000 Return code (decimal) Constant 0 GSK_OK Explanation The task completed successfully. Issued by every function call that completes successfully. The environment or Secure Sockets Layer (SSL) handle is not valid. The specified handle was not the result of a successful open() function call. The dynamic link library (DLL) was unloaded and is not available. (Occurs on Microsoft Windows systems only.) Internal error. Report this error to IBM Software Support.
0x00000001
GSK_INVALID_HANDLE
0x00000002
GSK_API_NOT_AVAILABLE
0x00000003
GSK_INTERNAL_ERROR
289
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Table 18. IBM Global Security Kit key management return codes (continued) Return code (hex) 0x00000004 0x00000005 Return code (decimal) Constant 4 5 GSK_INSUFFICIENT_STORAGE GSK_INVALID_STATE Explanation Insufficient memory is available to perform the operation. The handle is in an invalid state for operation, such as performing an init() operation on a handle twice. Specified key label not found in key file. Certificate not received from partner. Certificate validation error. Error processing cryptography. Error validating ASN fields in certificate. Error connecting to user registry. Internal error. Report this error to IBM Software Support. Internal error. Report this error to IBM Software Support. I/O error reading the key file. The key file has an invalid internal format. Recreate key file. The key file has two entries with the same key. The key file has two entries with the same label. The key file password is used as an integrity check. Either the key file has become corrupted or the password ID is incorrect. The default key in the key file has an expired certificate. An error occurred loading one of the GSK dynamic link libraries. Be sure GSK was installed correctly. Indicates that a connection is trying to be made in a GSK environment after the GSK_ENVIRONMENT_ CLOSE_OPTIONS was set to GSK_DELAYED_ ENVIRONMENT_CLOSE and gsk_environment_close() function was called. Neither the password nor the stash-file name was specified, so the key file could not be initialized. Unable to open the key file. Either the path was specified incorrectly or the file permissions did not allow the file to be opened.
0x00000006 0x00000007 0x00000008 0x00000009 0x0000000a 0x0000000b 0x0000000c 0x00000065 0x00000066 0x00000067 0x00000068 0x00000069 0x0000006a
6 7 8 9 10 11 12 101 102 103 104 105 106
GSK_KEY_LABEL_NOT_FOUND GSK_CERTIFICATE_NOT_AVAILABLE GSK_ERROR_CERT_VALIDATION GSK_ERROR_CRYPTO GSK_ERROR_ASN GSK_ERROR_LDAP GSK_ERROR_UNKNOWN_ERROR GSK_OPEN_CIPHER_ERROR GSK_KEYFILE_IO_ERROR GSK_KEYFILE_INVALID_FORMAT GSK_KEYFILE_DUPLICATE_KEY GSK_KEYFILE_DUPLICATE_LABEL GSK_BAD_FORMAT_OR_INVALID_ PASSWORD
0x0000006b 0x0000006c
107 108
GSK_KEYFILE_CERT_EXPIRED GSK_ERROR_LOAD_GSKLIB
0x0000006d
109
GSK_PENDING_CLOSE_ERROR
0x000000c9
201
GSK_NO_KEYFILE_PASSWORD
0x000000ca
202
GSK_KEYRING_OPEN_ERROR
290
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Table 18. IBM Global Security Kit key management return codes (continued) Return code (hex) 0x000000cb Return code (decimal) Constant 203 GSK_RSA_TEMP_KEY_PAIR Explanation Unable to generate a temporary key pair. Report this error to IBM Software Support. A User Name object was specified that is not found. A Password used for an LDAP query is not correct. An index into the Fail Over list of LDAP servers was not correct This installation of GSKit does not support FIPS mode of operation Indicates that the GSK environment close request was not properly handled. Cause is most likely due to a gsk_secure_socket*() command being attempted after a gsk_close_environment() call. The system date was set to an invalid value. Neither SSLv2 nor SSLv3 is enabled. The required certificate was not received from partner. The received certificate was formatted incorrectly. The received certificate type was not supported. An I/O error occurred on a data read or write operation. The specified label in the key file could not be found.
0x000000cc 0x000000cd 0x000000ce 0x000000cf 0x0000012d
204 205 206 207 301
GSK_ERROR_LDAP_NO_SUCH_OBJECT GSK_ERROR_LDAP_INVALID_ CREDENTIALS GSK_ERROR_BAD_INDEX GSK_ERROR_FIPS_NOT_SUPPORTED GSK_CLOSE_FAILED
0x00000191 0x00000192 0x00000193 0x00000194 0x00000195 0x00000196 0x00000197 0x00000198
401 402 403 404 405 406 407 408
GSK_ERROR_BAD_DATE GSK_ERROR_NO_CIPHERS GSK_ERROR_NO_CERTIFICATE GSK_ERROR_BAD_CERTIFICATE GSK_ERROR_UNSUPPORTED_ CERTIFICATE_TYPE GSK_ERROR_IO GSK_ERROR_BAD_KEYFILE_LABEL
GSK_ERROR_BAD_KEYFILE_PASSWORD The specified key file password is incorrect. The key file could not be used. The key file also might be corrupt. GSK_ERROR_BAD_KEY_LEN_ FOR_EXPORT GSK_ERROR_BAD_MESSAGE GSK_ERROR_BAD_MAC GSK_ERROR_UNSUPPORTED GSK_ERROR_BAD_CERT_SIG GSK_ERROR_BAD_CERT GSK_ERROR_BAD_PEER In a restricted cryptography environment, the key size is too long to be supported. An incorrectly formatted SSL message was received from the partner. The message authentication code (MAC) was not successfully verified. Unsupported SSL protocol or unsupported certificate type. The received certificate contained an incorrect signature. Incorrectly formatted certificate received from partner. Invalid SSL protocol received from partner.
0x00000199 0x0000019a 0x0000019b 0x0000019c 0x0000019d 0x0000019e 0x0000019f
409 410 411 412 413 414 415
291
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Table 18. IBM Global Security Kit key management return codes (continued) Return code (hex) 0x000001a0 0x000001a1 0x000001a2 0x000001a3 0x000001a4 0x000001a5 0x000001a6 0x000001a7 0x000001a8 0x000001a9 0x000001aa 0x000001ab 0x000001ac 0x000001ad 0x000001ae 0x000001af 0x000001b0 0x000001b1 0x000001b2 Return code (decimal) Constant 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 GSK_ERROR_PERMISSION_DENIED GSK_ERROR_SELF_SIGNED GSK_ERROR_NO_READ_FUNCTION GSK_ERROR_NO_WRITE_FUNCTION GSK_ERROR_SOCKET_CLOSED GSK_ERROR_BAD_V2_CIPHER GSK_ERROR_BAD_V3_CIPHER GSK_ERROR_BAD_SEC_TYPE GSK_ERROR_BAD_SEC_TYPE_ COMBINATION GSK_ERROR_HANDLE_CREATION_ FAILED GSK_ERROR_INITIALIZATION_FAILED GSK_ERROR_LDAP_NOT_AVAILABLE GSK_ERROR_NO_PRIVATE_KEY GSK_ERROR_PKCS11_LIBRARY_ NOTLOADED GSK_ERROR_PKCS11_TOKEN_ LABELMISMATH GSK_ERROR_PKCS11_TOKEN_ NOTPRESENT GSK_ERROR_PKCS11_TOKEN_ BADPASSWORD GSK_ERROR_INVALID_V2_HEADER GSK_CSP_OPEN_ERROR Explanation Report this error to IBM Software Support. The self-signed certificate is not valid. The read() failed. Report this error to IBM Software Support. The write() failed. Report this error to IBM Software Support. The partner closed the socket before the protocol completed. The specified V2 cipher is not valid. The specified V3 cipher is not valid. Report this error to IBM Software Support. Report this error to IBM Software Support. The handle could not be created. Report this error to IBM Software Support. Initialization failed. Report this internal error to service. When validating a certificate, unable to access the specified user registry. The specified key did not contain a private key. A failed attempt was made to load the specified PKCS11 shared library. The PKCS #11 driver failed to find the token specified by the caller. A PKCS #11 token is not present in the slot. The password/pin to access the PKCS #11 token is invalid. The SSL header received was not a properly SSLv2 formatted header. Could not open the hardware-based cryptographic service provider. Either the CSP name is not specified correctly or a failed attempt was made to access the specified CSPs certificate store. Some conflicting attributes for SSL operation were defined. The Microsoft Crypto API is only supported on Microsoft Windows 2000 with Service Pack 2 applied. System is running in IPv6 mode without setting a PEERID. The buffer size is negative or zero.
0x000001b3 0x000001b4
435 436
GSK_CSP_OPEN_ERROR GSK_CSP_OPEN_ERROR
0x000001b5 0x000001f5
437 501
GSK_CSP_OPEN_ERROR GSK_INVALID_BUFFER_SIZE
292
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Table 18. IBM Global Security Kit key management return codes (continued) Return code (hex) 0x000001f6 0x00000259 0x0000025a 0x000002bd Return code (decimal) Constant 502 601 602 701 GSK_WOULD_BLOCK GSK_ERROR_NOT_SSLV3 GSK_MISC_INVALID_ID GSK_ATTRIBUTE_INVALID_ID Explanation Used with nonblocking I/O. Refer to the nonblocking section for usage. SSLv3 is required for reset_cipher(), and the connection uses SSLv2. An invalid ID was specified for the gsk_secure_soc_misc() function call. The function call has an invalid ID. This also might be caused by specifying an environment handle when a handle for a SSL connection should be used. The attribute has a negative length, which is invalid. The enumeration value is invalid for the specified enumeration type.
0x000002be 0x000002bf 0x000002c0 0x000002c1
702 703 704 705
GSK_ATTRIBUTE_INVALID_LENGTH GSK_ATTRIBUTE_INVALID_ ENUMERATION
GSK_ATTRIBUTE_INVALID_SID_CACHE Invalid parameter list for replacing the SID cache routines. GSK_ATTRIBUTE_INVALID_ NUMERIC_VALUE GSK_CONFLICTING_VALIDATION_ SETTING GSK_AES_UNSUPPORTED GSK_PEERID_LENGTH_ERROR GSK_SC_OK GSK_SC_CANCEL GSK_TRACE_STARTED GSK_TRACE_STOPPED GSK_TRACE_NOT_STARTED GSK_TRACE_ALREADY_STARTED GSK_TRACE_OPEN_FAILED The trace started successfully. The trace stopped successfully. No trace file was previously started so it cannot be stopped. Trace file already started so it cannot be started again. Trace file cannot be opened. The first parameter of gsk_start_trace() must be a valid full path file name. When setting a numeric attribute, the specified value is invalid for the specific attribute being set. Conflicting parameters were set for additional certificate validation The AES cryptographic algorithm is not supported. The PEERID does not have the correct length.
0x000002c2 0x000002c3 0x000002c4 0x000005dd 0x000005de 0x00000641 0x00000642 0x00000643 0x00000644 0x00000645
706 707 708 1501 1502 1601 1602 1603 1604 1605
293
294
Index A
accessibility features xviii administration center application error 97 blank pages 78 configuring the IP address 75 establishing a connection with the IBM Tivoli Storage Manager server 65 messages ANRW0089E 86 ANRW0090E 86 ANRW0126E 86 ANRW0127E 87 ANRW0128E 87 ANRW0129E 88 ANRW0130E 88 ANRW0131E 88 ANRW0132E 89 ANRW0209E 89 ANRW0212E 89 ANRW0213E 90 ANRW0221E 90 ANRW0257E 90 messages versus IBM Tivoli Storage Manager messages performance tuning suggestions 77 portlet is unavailable 97 problems with tutorials 75 support utility 99 task fails 91 check the server activity log 91 internal errors 92 portlets 91 wizard 91 task fails with message 85 trace 149 unable to access the server from a Web browser 80 advanced copy services DB2 creating DB2UDB transaction after backup 147 AIX JFS2 image backup 19 snapshot-based backup-archive 19 ALL_BACK client trace class 195 ALL_FILE client trace class 195 ALL_IMAGE client trace class 195 ALL_JBB client trace class 195 ALL_NAS client trace class 196 ALL_SESS client trace class 196 ALL_SNAPSHOT client trace class 196 ALL_WAS client trace class 196 ANRW0089E 86 ANRW0090E 86 ANRW0126E 86 Copyright IBM Corp. 1993, 2007 ANRW0127E 87 ANRW0128E 87 ANRW0129E 88 ANRW0130E 88 ANRW0131E 88 ANRW0132E 89 ANRW0209E 89 ANRW0212E 89 ANRW0213E 90 ANRW0221E 90 ANRW0257E 90 application programming interface (API) instrumentation 20 tracing 214 AUDIT client trace class 197
B
backup application files automatically excluded 12 files excluded by EXCLUDE DIR 13 files excluded by include/exclude statements 14 files excluded due to incremental copy frequency 15 include/exclude due to compression, encryption, and subfile backup statements 15 include/exclude statements coded wrong 16 platform-specific include/exclude statements 15 Windows files excluded 13 backup-archive client help 259 SHOW commands 33 CACHE 33 CLUSTER 33 DOMAIN 33 OPTIONS 33 OPTTABLE 34 PLUGINS 34 SESSION 34 SYSTEMSERVICES 34, 35 SYSTEMSTATE 35 TRACEFLAGS 35 VERSION 35 blank pages administration center 78
92
C
client authentication 7 can problem be reproduced 2 diagnostic tips 1 error messages examining 1 generating errors connected to the server 61 identifying when and where problems occur image backup 16 passwords 7 resolving problems 1
295
client (continued) scheduler 9 server activity log examining 1 trace classes 194, 205 ALL_BACK 195 ALL_FILE 195 ALL_IMAGE 195 ALL_JBB 195 ALL_NAS 196 ALL_SESS 196 ALL_SNAPSHOT 196 ALL_WAS 196 AUDIT 197 CLIENTTYPE 197 COMPRESS 197 DELTA 197 DIROPS 198 DOMAIN 198 ENCRYPT 198 ERROR 198 FILEOPS 199 FIOATTRIBS 199 INCLEXCL 199 INCR 199 MEMORY 200 NETWARE 200 OPTIONS 200 PASSWORD 200 PID 201 POLICY 201 SCHEDULER 201 SERVICE 202 SESSION 202 SESSVERB 202 SMS_DEBUG 203 SMS_DETAIL 202 STATS 203 THREAD_STATUS 203 TXN 203 VERBDETAIL 204 VERBINFO 204 WIN2K 204 client option sets 4 CLIENTTYPE client trace class 197 communication errors resolving 61 COMPRESS client trace class 197 compressed data during backup-archive customer support, contacting xvi
213
D
Daemon traceflags client and journal 194 data sent to the IBM Tivoli Storage Manager storage agent or server 22 data protection 101 Domino 101 DP for SAP files to gather before calling IBM 136 DP for Snapshot Devices for mySAP 103 files to gather before calling IBM 119 gathering information before calling IBM 118
data protection (continued) DP for Snapshot Devices for mySAP (continued) locating solutions 118 tracing 116 troubleshooting 103 Exchange 122 determining the problem 122 failing silent install 125 gathering files before calling IBM 124 locating solutions 123 tracing 123 Exchange with VSS backup/restore support 126 determining the issue 126 gathering files before calling IBM 129 gathering information before calling IBM 129 tracing when using VSS 128 troubleshooting 131 Oracle 120 gathering files 121 gathering information 121 locating solutions 120 tracing 120 SAP 132 information to gather before calling IBM 135 locating solutions 135 tracing 134 troubleshooting 132 SQL 136 gathering files before calling IBM 138 information to gather before calling IBM 137 items to gather if silent install fails 138 locating solutions 137 tracing the client 136 SQL with VSS backup/restore support 139 determining the issue 139 gathering files before calling IBM 143 gathering information before calling IBM 142 tracing when using VSS 141 troubleshooting 144 data storage diagnostic tips 267 debug print output obtaining 28 DELTA client trace class 197 device driver 64-bit/32-bit Linux kernel modules 218 Adaptec SCSI requirements 221 error messages in the system error log 218 HBA changes 218 HBA drivers on the Linux 2.6.x kernels 219 Linux server running on x86_64 architecture 219 loose cable connection 218 multiple LUN support on Linux kernels 219 operating system changes 217 performing ddtrace from version 5.3.2 on Linux 220 Qlogic fibre-channel HBA BIOS requirements 221 SCSI adapter changes 218 updating device information 220 device driver trace from a command shell - all platforms 193 from the server console/admin client 191 diagnostic tips client 1 data storage 267 DIROPS client trace class 198
296
documentation to resolve client problems 2 DOMAIN client trace class 198 Domino data protection client files to gather 102 information to gather 102 locating solutions 101 tracing 101 dsmc/dsmadmc/dsmj/ no start 3
E
education see Tivoli technical training xv ENCRYPT client trace class 198 encrypted data during backup-archive 213 ERROR client trace class 198 error messages examining 1 establishing a connection with the IBM Tivoli Storage Manager server 65
help system (continued) starting/stopping 264 Windows NT 263 hints and tips device driver 217 disk subsystems 221 hard disk drives 221 miscellaneous server information 224 NDMP filer-to-Tivoli Storage Manager server operations 229 SAN 225 SAN configuration 225 SAN device mapping 232 tape drives and libraries 223 adapter firmware changes 223 cabling between the computer and device changes device driver changes 223 device firmware changes 223 error messages in system error log 224 loose cable connections 224 operating system changes 223 other hardware changed or fixed 224 replaced adapter 224
223
I
IBM Global Security Kit key management return codes 289 return codes 285 IBM Software Support submitting a problem xvii IBM Support Assistant xv image backup client 16 error 17, 18 INCLEXCL client trace class 199 INCLEXCL option 4 INCR client trace class 199 installation failure IBM Tivoli Storage Manager 67 Integrated Solutions Console 67 Integrated Solutions Console authority problems 69 excessive memory consumption 73 installation failure 67 manually uninstalling 67 server crash 71 collector tool 71 log analyzer tool 72 Internet, search for problem resolution xv Internet, searching for problem resolution xvi IP address configuring 75
F
FILEOPS client trace class 199 FIOATTRIBS client trace class 199 fixes, obtaining xvi
G
GSKit return codes gt script 279 285
H
health monitor 81 how it works 81 re-synchronizing the ADMIN_CENTER administrator ID password 84 warning or critical database status 83 warning or critical storage status 83 help server or storage agent 261 system does not display 263 help facilities 259 help system adding publications to the ISC help server 265 CLI for server or storage agent 261 commands 262 dsmcutil 263 GUI and Web GUI clients 261 installation issues 264 reporting a problem 265 running 263 server or storage agent commands 261 messages 261
J
JBB 25, 27 database viewing utility 27 journal-based backup database viewing utility 27 determining 25 running in foreground 27
Index
297
K
knowledge bases, searching xv
L
LAN-free setup storage agent 175 Linux image backup error 17 Linux Snapshot image backup error 18 log files installation 73 LVSA examining Windows system event log forcing a dump 29 full memory dump 29 problem determination 28
28
PID client trace class 201 POLICY client trace class 201 portlets errors caused by starting or stopping 91 problem determination describing problem for IBM Software Support xvii determining business impact for IBM Software Support xvii submitting a problem to IBM Software xvii publications download xi order xi search xi Tivoli Storage Manager xi
M
MEMORY client trace class 200 message definitions IBM Tivoli Storage Manager 94 Microsoft diagnostic information VSS 31 Microsoft tuning recommendations VSS 31
R
red error messages administration center 97
S
SAN configuration 238 configuration between devices 228 configuration problems 240 devices supported 226 fibre channel switch configuration 227 gateway port settings 228 host bus adapter configuration 226 host bus adapters 226 monitoring fibre-channel link error report 229 vendor support 240 SAN device mapping 234 ANR1745I 234 disabling 233 discovery 240 error detecting version of HBA-API library 244 fileset not at required level 244 HBAAPI library 241 vendor library failed to load or is missing 242 wrapper library failed to load 241 missing from the display of QUERY SAN 238 no match found using this criteria 243 SAN discovery failed 241 SAN discovery is disabled 243 SAN discovery not supported 242 target mapping 241 SAN devices storage agent 230 SCHEDULER client trace class 201 SCSI devices 269 secure sockets layer 62 general return codes 285 sequential media volume tape 270 server database 42 diagnostic tips 37 change server options or the settings create errors 38 checking HELP messages 37 checking the server activity log 37
N
NETWARE client trace class 200 non-root user running applications using API ntbackup.exe 32 23
O
odbc application problems 251 data source configuration 250 registry settings 253 ODBC configuration details 251 data sources 246 driver tracing 215 installation files 252 known problems and limitations troubleshooting 250 ODBC driver 245 supported applications 245 supported functions 245 open file support best practices 30 problem determination 28 OPTIONS client trace class 200
248
P
PASSWORD client trace class 200 performance tuning suggestions administration center 77
298
server (continued) diagnostic tips (continued) failing a scheduled client operation 38 recreating the problem 37 resolving errors from reading or writing to a device 37 resolving failed connections by client or administrators 61 resolving server space issues 38 server error messages indicate code page conversion failure 38 hang or loop errors 44 process 46 process messages 51 processes AUDIT VOLUME 47 BACKUP DB 47 BACKUP STGPOOL 47 CHECKIN LIBVOLUME 48 CHECKOUT LIBVOLUME 48 EXPIRATION 48 IMPORT 48 LABEL LIBVOLUME 49 MIGRATION 49 MOVE DATA 49 MOVE DRMEDIA 49, 50 MOVE NODEDATA 50 PREPARE 50 RECLAMATION 50 RESTORE STGPOOL 51 RESTORE VOLUME 51 recovery log 45 storage pool 57 server activity log checking 91 server command definition file 96 server crash 39 server or storage agent SHOW commands AGGREGATE 178 ASQUEUED 179 ASVOL 179 BFOBJECT 179 BUFSTATS 180 BUFVARS 180 CONFIGURATION 180 DBTXNTABLE 181 DBVARS 181 DEVCLASS 181 GROUPLEADERS 181 GROUPMEMBERS 182 INVOBJECT 182 LANFREE 183 LIBINVENTORY 183 LIBRARY 183 LOCK 184 LOGPINNED 184 LOGVARS 184 MEMTREND 185 MP 185 NASDEV 185 NASFS 186 NASINFORMATION 186 NASWORKLOAD 186 RESQUEUE 187 SESSIONS 187 SLOTS 187 SSPOOL 188
server or storage agent (continued) SHOW commands (continued) THREADS 188, 189 TOCSETS 189 TXNTABLE 190 VALIDATE LANFREE 190 VERIFYEXPTABLE 190 VOLINUSE 191 trace class ACTIONS 150 ADDMSG 157 ADMCMD 158 AF 158 AFCREATE 158 AFMOVE 158 ALL 150 API 150 AS 158 ASALLOC 159 ASDEALLOC 159 ASMOUNT 159 ASRTRV 159 ASSD 160 ASTXN 160 ASVOL 160 BF 160 BFAGGR 161 BFCREATE 161 BFREMOTE 161 BFRTRV 161 BKSET/OBJSET 162 BLKDISK 162 BRNODE 162 COLLOCATE 162 CONTROL 150 CRC 163 CRCDATA 163 CRCPROTO 163 CRCVAL 163 CRYPTO 164 DBNETDB 164 DELTA/GROUP 164 DF 164 DFCREATE 165 DFMOVE 165 DFRTRV 165 DS 165 DSALLOC 166 DSDEALLOC 166 DSRTRV 166 DSVOL 166 HEALTH 151 ICVOLHST 166 IMFS 167 LANFREE 167 MMS 167 NA 167 PROXYNODE 168 PS 151 PVR 168 RETPROT 168 SCHED 169 SESSION 169 SESSREMOTE 169 SHRED 170 SNAPIN 151 SPI/SPID 170 Index
299
server or storage agent (continued) trace class (continued) SSLDATA 170 SSLINFO 171 TAGLIB 151 TAPE 171 TEC 172 TOC 172 TOCBUILD 172 TOCLOAD 173 TOCREAD 173 TOCUTIL 173 TPC 171 TPCDATA 171 TPCINFO 172 UNICODE 173 UTIL 151 VIEWS 152 XIDETAIL 174 XMLVALS 152 trace classes 154 SERVICE client trace class 202 SESSION client trace class 202 SESSVERB client trace class 202 SHOW commands server or storage agent 178 SMS_DEBUG client trace class 203 SMS_DETAIL client trace class 202 Software Support contacting xvi describing problem for IBM Software Support xvii determining business impact for IBM Software Support xvii SSL 62 stack trace server or storage agent 153 STATS client trace class 203 storage agent diagnostic tips 174 check HELP for messages 174 check the server activity log 174 error caused by reading or writing to a device 174 problems caused by changing server options 175 problems from changing storage agent options 175 LAN-free setup 175 data sent directly to server 175 SHOW LAN-free and VALIDATE LAN-free 177 storage pool configured for simultaneous write 176 testing LAN-free configuration 178 z/OS session failed 176 SAN devices 230 support for API 20 before calling IBM files to gather 21 information to gather 20 support information xv
THREAD_STATUS client trace class 203 Tivoli technical training xv trace administration center 149 client 205 device driver 191 enable client trace on command line 205 enable client trace while client is running 206 known problems and limitations 210 options 211 server or storage agent 152 trace classes administration center 150 client 194 server or storage agent 154 trace data is it compressed during backup-archive 213 is it encrypted during backup-archive 213 tracing application programming interface (API) 214 client 193 trademarks 277 training, Tivoli technical xv transient errors VSS 30 tsmdiag utility 281 TXN client trace class 203
V
VERBDETAIL client trace class 204 VERBINFO client trace class 204 Volume Shadow Copy Services Windows 30 vsreq.exe sample program 32 VSS Microsoft diagnostic information 31 Microsoft tuning recommendations 31 ntbackup.exe 32 test flags 30 transient errors 30 vsreq.exe sample program 32 Windows 30 Windows hot fixes 31
W
WIN2K client trace class 204 Windows VSS 30 Windows hot fixes VSS 31 wizards errors caused by starting or stopping
91
T
test flags VSS 30
300
Program Number: 5608-ISM 5608-ISX 5608-ACS 5608-APD 5608-APE 5608-APH 5608-APR 5608-APW 5608-CSS 5608-HSM 5608-SAN 5698-A11 5698-A12 5698-A13
5698-A25 5698-USS
Printed in USA
SC32-0142-01
Spine information:
IBM Tivoli Tivoli Storage Manager
Version 5.5

B PDG Master

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

B PDG Master

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

B PDG Master

Uploaded by

Copyright:

Available Formats

Tivoli Tivoli Storage Manager

Problem Determination Guide

Tivoli Tivoli Storage Manager

Problem Determination Guide

Chapter 1. Resolving client problems . . 1

Chapter 2. Resolving server problems

Copyright IBM Corp. 1993, 2007

Chapter 3. Resolving communication problems . . . . . . . . . . . . . . 61

Chapter 5. Data Protection . . . . . . 101

Chapter 4. Administration Center . . . 65

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

Chapter 8. Hints and tips . . . . . . 217

Chapter 7. Trace . . . . . . . . . . 149

. 261 . 261 . . . . . 261 263 263 264 264

Chapter 11. Determining data storage problems . . . . . . . . . . . . . 267

Glossary . . . . . . . . . . . . . 275 Trademarks . . . . . . . . . . . . 277

Chapter 9. Tivoli Storage Manager ODBC driver . . . . . . . . . . . . 245

Chapter 10. Help facilities . . . . . . 259

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

Copyright IBM Corp. 1993, 2007

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

Who should read this guide

Copyright IBM Corp. 1993, 2007

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

Tivoli Storage Manager publications

Copyright IBM Corp. 1993, 2007

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

Getting technical training

Searching knowledge bases

Searching the Internet

Using IBM Support Assistant

Finding product fixes

Getting E-mail notification of product fixes

Contacting IBM Software Support

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

Determine the business impact

Severity 2 Severity 3 Severity 4

Describe your problem and gather background information

Submit your problem to IBM Software Support

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

Chapter 1. Resolving client problems

Client diagnostic tips

Examining error messages

Examining the server activity log messages

Identifying when and where the problem can occur

Reproducing the problem

Collecting documentation to resolve problems with the client application

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

dsmc/dsmadmc/dsmj/... will not start

Chapter 1. Resolving client problems

Client option sets

What are the client option sets?

Where to find detailed information

IBM Tivoli Tivoli Storage Manager: Problem Determination Guide

The order of option processing

Client and client option traceflag settings

Examples of when to use client options sets