Release Notes for Tivoli Workload Scheduler, version 9.2

Content

The Release Notes for Tivoli Workload Scheduler, version 9.2 contain the following topics:

• Interoperability tables
• Fix packs
• Installation limitations and problems, and their workarounds
• Software limitations and workarounds
• APARS Fixed in this release
• Documentation updates

To download the appropriate package for your operating system, see the Tivoli® Workload Scheduler download page.

For detailed system requirements for all operating systems, see the Detailed System Requirements page.

To access the Tivoli Workload Scheduler documentation, see the online Information Center.

A complete list of new or changed functions in this release are documented in the "Overview->Summary of enhancements" section of the information center at the following link: Summary of enhancements.

The level of support provided for the products and components listed in the interoperability tables covers all the shown versions.

In the tables in this section the following acronyms are used:

Tivoli Workload Scheduler: compatibility

Dynamic Workload Console: compatibility

Tivoli Workload Scheduler Command-line client: compatibility

Tivoli Workload Scheduler for Applications: compatibility

Note:

It is recommended that you upgrade Tivoli Workload Scheduler for Applications to the latest fix pack version before upgrading the Tivoli Workload Scheduler engine to version 9.2. This applies to all Tivoli Workload Scheduler for Applications versions up to version 8.6. If you have already upgraded the Tivoli Workload Scheduler engine to version 9.2 and now need to upgrade earlier versions of Tivoli Workload Scheduler for Applications up to version 8.4, refer to Technote 1384969.

Tivoli Workload Scheduler database, version 9.2: compatibility

Note:

You can upgrade from V8.5.1 or V8.6 of the product, while maintaining a database version that is no longer supported by V9.2. However, as soon as you complete the upgrade process, you must upgrade the database to a supported level.

Fix packs released for this version of the product

To find the fix packs released for Tivoli Workload Scheduler, and the Dynamic Workload Console including links to the related readmes files, see the Fixes by version.

The following are software limitations that affect the installation of Tivoli Workload Scheduler:

Space notification issues in Installation Manager panels when installing the Integration Workbench component.

For some operating systems (Windows and Linux), the memory size values for the installation directory provided in the Disk space information section of the Package selection panel and in the following Installation panel (as Total Installation size) were found to be inconsistent. (94253)

A defect was opened to the Installation Manager development team.

Unable to start/stop broker after MDM installation with local system user in domain controller environment.

When you install a Master Domain Manager on a Windows system configured as a domain controller with localOS user, the installation completes with warnings and the dynamic workload broker does not start. (95772)

This is due to a problem of WebSphere Application Server. The problem is resolved by applying fix pack 8.5.0.2 on WebSphere Application Server 8.5.

Master and BKmaster hangs during uninstallation on AIX 7.1 (99926)

On AIX 7.1 you cannot uninstall a master domain manager or a backup master because the uninstallation process hangs on the Check user panel and also during the uninstall action.

This is due to an AIX problem with the fuser command. The workaround to this problem is to set fuser to 666 rights (not executable). Follow these steps:

1. Make sure all the Tivoli Workload Scheduler and all the WebSphere Application Server processes for the specific instance are stopped.
2. Start the uninstallation process.
3. Go to the User validation panel.
4. Enter:

   chmod 666 /usr/sbin/fuser

5. Click the Validate button.
6. Proceed with the uninstallation process.

A problem has been opened to AIX Development.

If you encounter this problem with earlier AIX versions (specifically 6.1.0.0-00, 6.1.0.0-01, and 6.1.0.0-02), either follow the steps listed for version 7.1, or apply the respective APAR.

e BASE: Full stack upgrade from 9.1 to 92 gives error VALIDATE_OLD_EWAS at IM DWC user validation panel. (100628)

On Installation Manager the process that migrates Tivoli Workload Scheduler and Dynamic Workload Console V9.1 to version 9.2 in a single action returns errors and does not complete.

You must migrate the two products singularly.

The following are software limitations that affect Tivoli Workload Scheduler, and workarounds (where available) tested to overcome them:

File monitor provider events: older event configurations might stay active on workstations after rule redeployment. (34103)

If you deployed rules containing event types from the FileMonitor event provider, and then you redeploy rules that no longer include these file monitoring events, you might find that, despite the new configurations, the local monitoring agents still forward the file monitoring events to the event processor. The event processor correctly ignores these events, accordingly with the new configuration specifications deployed on the server; however, a certain amount of CPU time and bandwidth are needlessly misused.

The status of the local monitoring configuration on the agent is corrected when one of the following occurs:

• The planman deploy -scratch command is issued
• The event processing server is restarted
• Another rule containing an event condition involving the FileMonitor provider is deployed to the agents.

Deploy (D) flag not set after ResetPlan command used. (36924)

The deploy (D) flag is not set on workstations after the ResetPlan command is used.

This is not a problem that affects the processing of events but just the visualization of the flag that indicates that the event configuration file has been received at the workstation.

Workaround

You can choose to do nothing, because the situation will be normalized the next time that the event processor sends an event configuration file to the workstation.

Alternatively, if you want to take a positive action to resolve the problem, do the following:

1. Create a dummy event rule that applies only to the affected workstations
2. Perform a planman deploy to send the configuration file
3. Monitor the receipt of the file on the agent
4. When the file is received, delete the dummy rule at the event processor.

Deploying large numbers of event rules. (37011)

The rule deployment process (run either automatically or with the planman deploy command) performs slowly when you deploy high numbers of new and changed rules (2000 and more).

Workaround

If you need to deploy large numbers of event rules, there are two actions you can take to improve performance:

Use planman deploy with the -scratch option

To deploy large numbers of rules collectively in an acceptable time limit, use planman deploy with the -scratch option.

Increase the Java™ heap size of the application server

Increase the Java heap size of the application server, as described in the Scalability section of the Performance chapter in the Administration Guide. The critical point at which you should increase your heap size is difficult to calculate, but consider a deployment of several thousands of rules as being at risk from an out of memory failure.

Some data is not migrated when you migrate database from DB2® to Oracle, or to Informix® Dynamic Server, or vice versa. (38017)

Neither of the two migration procedures migrate the following information from the source database:

• The pre-production plan.
• The history of job runs and job statistics.
• The state of running event rule instances. This means that any complex event rules, where part of the rule has been satisfied prior to the database migration, are generated after the migration as new rules. Even if the subsequent conditions of the event rule are satisfied, the record that the first part of the rule was satisfied is no longer available, so the rule will never be completely satisfied.

"Absolute" keyword required to resolve time zones correctly with enLegacyStartOfDayEvaluation set. (41192)

If the master domain manager of your network runs with the enLegacyStartOfDayEvaluation and enTimeZone options set to yes to convert the startOfDay time set on the master domain manager to the local time zone set on each workstation across the network, and you submit a job or job stream with the at keyword, you must also add the absolute keyword to make sure that the submission times are resolved correctly.

The absolute keyword specifies that the start date is based on the calendar day rather than on the production day.

Messages in the TWSMERGE.log file in English are corrupt. (51518)

Messages in the TWSMERGE.log file in English are corrupt.

Workaround:

On windows, you must set the TWS_TISDIR environment variable at system level and then restart the workstation.

The return code mapping for Tivoli Workload Scheduler jobs that are migrated to broker is ignored. (51582)

Return code mapping for Tivoli Workload Scheduler jobs that are migrated to broker using the JSDL template is ignored. Any script or command issued with a nonzero return code is considered abended.

Workaround:

Do not use return code mapping for static jobs that are migrated to dynamic job using the JSDL template.

Certificates for SSL communication defined at Job Brokering Definition Console startup cannot be modified. (52753)

In the Job Brokering Definition Console, the certificates for SSL communication defined at the Job Brokering Definition Console startup are used during the entire session and cannot be modified until the Job Brokering Definition Console is restarted.

If you want to set the SSL parameters on the dynamic workload broker component connection, you must include your own certificates in the keystore and truststore that are shown in the SSL section. Alternatively, you can change the keystore and truststore that appear on the SSL section with your own certificates and restart the Job Brokering Definition Console to save the updates.

File monitor provider events: if you specify a file path with forward slashes (/) in an event rule definition, on Windows workstations this results in the failed detection of the file. (53843)

This is due to the fact the SSM agent is not currently designed to automatically convert the slash sign to be Windows compatible when running the rule.

Workaround

Use backward slashes when you specify file paths in the definitions of rules that run on Windows systems.

Problems when International Date Line West time zone is used. (80881/96581)

Tivoli Workload Scheduler is unable to manage the Date Line West time zone (UTC -12).

Missing stop message in the stdlist when a Windows job is killed. (91191)

Following a kill action on a running job, no messages are issued in the standard list file to give notice of this event. Message should be logged.

Different behavior of UNIX and Windows® operating systems at springtime daylight savings. (94279)

During the springtime daylight savings time change (01:59:59-03:00:00), different operating systems behave differently.

For example,

Windows

The command: conman submit job at=02xx is set to 01xx

HP-UX

The command: conman submit job at=02xx is set to 03xx

Workaround

Avoid creating jobs that have a start time in the "lost" hour and are due to run on the night of the daylight savings change (a Saturday night in spring).

TWS/TDWC EWAS on HP crashes in groups authenticating/authorizing. (96389 - IV50242)

On HP-UX version 11iv3 operating systems, based on Itanium architecture, if you do a group authentication you must be sure that each line contained in the /etc/group file is not longer than 4096 characters.

Batchman fails on V8.6 and V9.1 fault-tolerant agents configured with the mailman server process. (96620)

The enAddWorkstation global option is set to "yes" on a V9.2 Master Domain Manager in a mixed version environment. After you successfully install a V9.2 dynamic agent, the batchman process fails on the V8.6 and V9.1 fault-tolerant agents that have the mailman server process configured.

The cause is that the dynamic agent workstation is automatically added to the plan of each agent in the Tivoli Workload Scheduler mixed version network. The batchman process on the V8.6 and V9.1 fault-tolerant agents that have the mailman server process configured is unable to manage the broadcast of the Tivoli Workload Scheduler event that communicates the addition of the dynamic agent to the plan.

To solve this problem, set enAddWorkstation to "no" on the master domain manager to prevent the automatic addition of the dynamic agent workstation in the plan.

Random core after several start/stop on Solaris machine. (98416)

The core dump is triggered after - following the execution of a Java job - the agent is stopped and restarted several times. The Java Virtual Machine is stopped abruptly and a core dump is generated. Experienced on Solaris Sparc platform only.

Incorrect dates retrieved after submission of mssql job. (99504)

The DATE values returned from the execution of database jobs running MS SQL stored procedures may at times be incorrect.

This problem comes from the Microsoft JDBC Driver for SQL Server that currently supports JRE 1.5 and JRE 1.6 (not JRE 1.7). As documented by Microsoft, this issue is solved by downloading the latest version of the Microsoft JDBC Driver for SQL Server.

On Windows 2012: all cli not supported on windows power shell. (100237)

On Windows 2012, the Tivoli Workload Scheduler command lines are not supported in the Windows Power shell. They work only in the traditional shell.

On AIX 6.1, batchman process does not correctly recognize the time zone of the local workstation (IZ70267)

On AIX 6.1, the batchman process does not correctly recognize the time zone of the local machine that is set to GMT, even if, in the Tivoli Workload Scheduler CPU definition, it is correctly set to the correct timezone. You see in the stdlist log the following message: "10:29:39 24.11.2008|BATCHMAN:AWSBHT126I Time in CPU TZ (America/Chicago): 2008/11/24 04:29 10:29:39 24.11.2008|BATCHMAN:AWSBHT127I Time in system TZ (America/Chicago): 2008/11/24 10:29 10:29:39 24.11.2008|BATCHMAN:+ 10:29:39 24.11.2008|BATCHMAN:+ AWSBHT128I Local time zone time differs from workstation time zone time by 360 minutes."

Batchman does not recognize the correct time zone because AIX 6.1 uses (International Components for Unicode) ICU libraries to manage the timezone of the system, and these ICU libraries are in conflict with the Tivoli Workload Scheduler libraries.

Workaround:

Export the TZ environment variable before starting Tivoli Workload Scheduler to the old POSIX format, for example, CST6CDT. This is an example of a POSIX name convention instead of an Olson name convention (for example America/Chicago). It avoids the new default TimeZone management through the ICU libraries in AIX 6.1, by switching to the old POSIX one (as in AIX 5.x).

• IV28201: PROBLEM WITH TCL USED IN MAKEPLAN.CMD.
  
  
• IV33962: AWSUI5017E UNABLE TO SAVE THE EVENT RULE IF MINLENGTH NOT 1.
  
  
• IV36450: CHANGEDATASOURCEPROPERTIES DOES NOT CHANGE SECURITYMECHANISM.
  
  
• IV36624: TIME RESTRICTIONS ARE NOT APPLIED AFTER MAKING CHANGES IN JSC UNDER JOB STREAM'S PROPERTIES.
  
  
• IV37443: AWSBHT228W MESSAGES IN TWSMERGE ALTHOUGH THE JOB ALREADY RAN.
  
  
• IV37453: PROBLEM WITH THE TCL USED IN CREATEPOSTREPORT.CMD.
  
  
• IV38003: DB2 TYPE2 DOES NOT ALLOW TO USE REPORTS UTILITIES FROM TDWC.
  
  
• IV38032: LIBICUUC.SO.44 & LIBICUDATA.SO.44 MISSING 8.6 FP2 DYNAMIC AGENT
  
  
• IV39263: AT/UNTIL TIME IS NOT CORRECT FOR EET ON THE DAY AFTER DST CHANGE.
  
  
• IV39528: UNIXSSH OR UNIXRSH JOBS SCHEDULE VIA ZCENTRIC FAILED.
  
  
• IV40159: COMPOSER DOES NOT ACCEPT " ETC/GMT+" TIMEZONES.
  
  
• IV40269: SLOW SSH TRANSFER OF FILE DUE TO PROCESSING LOCAL FILE.
  
  
• IV41568: PATCH LEVEL AND MEMORY ERROR WHEN INSTALLING TWS 8.6 FTA ON HPUX ITANIUM SYSTEM.
  
  
• IV41597: A RECOVERY INTERACTIVE JOB IS LAUNCHED AS NOT INTERACTIVE.
  
  
• IV41929: INTERACTIVE JOBS DOES NOT WORK IF THE STREAMLOGON HAS A DOMAIN.
  
  
• IV42194: BKM SILENT INSTALL SETS ITA:INI AND JOBMANAGER.INI PORTS TO 0.
  
  
• IV42438: PARAMETER DELETION PERFORMED IN TDWC NOT RECORDED IN AUDIT LOGS.
  
  
• IV42786: "LOGMESSAGEWRITTEN" EVENT COMPONENT NOT WORKING AFTER 8.6 FP 2.
  
  
• IV43150: HIGH CPU USAGE OF JOBMON DUE A LOCAL TCP SCAN ON JOBMON PORT.
  
  
• IV43483: CHANGE ELAPSED(SECONDS) TO ELAPASED(MINUTES) IN REP1.
  
  
• IV43547: IF A Z/OS JOB ENDS WITH JCL OR JCLI, THE RETURN CODE IS SET 0 AND THE STATUS IS SET TO SUCC.
  
  
• IV44485: ERROR SUBMITTING SCEHEDULES WITH ALIAS 16 CHARS LONG ON AIX PLATFORMS.
  
  
• IV44771: MAKEPLAN FAILED DUE TO AN ERROR IN THE CODE THAT CAN CORRUPT THE MEMORY.
  
  
• IV44771: eWAS CRASH DUE TO AN ERROR IN THE CODE THAT CAN CORRUPT THE MEMORY.
  
  
• IV45042: DATA COLLECTOR TWS_INST_PULL_INFO.CMD IS LAUNCHED AT FINAL OR JNEXTPLAN TIME.
  
  
• IV45206: AFTER CHANGING THE TEMP DIRECTORY FOR TWS 9.1 MASTER INSTALL, IT STILL USES THE /TMP FOLDER.
  
  
• IV45220: RMSTDLIST INTERMITTENTLY FAILS ON WINDOWS CLUSTERS ATTEMPTING TO REMOVE LOCKED FILE, CLU_OFFLINE.LOG.
  
  
• IV46103: JOBMAN IS NOT ABLE TO FIND THE LJUSER DEFINITION IN R3BATCH.OPTS.
  
  
• IV46504: DISCREPANCY BETWEEN THE OUTPUT OF REP1 AND TDWC.
  
  
• IV46837: TWS_INST_PULL_INFO.CMD FOR TWS 8.6 ON WINDOWS FAILS TO EXECUTE EVEN THOUGH CUSTOMER PROVIDED CORRECT ARGUMENTS.
  
  
• IV47074:STARTEVTPROC FAILS IF THISCPU LINE IN LOCALOPTS HAS A COMMENT.
  
  
• IV47336: OLD DM SENT OLD SYMPHONY TO THEIR FTAS AFTER UNPLANNED OUTAGE.
  
  
• IV48720: JOBMON MEMORY LEAKS FOR XA JOBS AND RANDOM CRASHES.
  
  
• IV49858: TWS 8.6 JOBMANAGER DOES NOT START AFTER REBOOT ON LINUX.
  
  
• IV50164: JOBMANAGER CAN NOT CREATE JM JOBLOG FOLDER, WHILE MANY JOBS RUN IN PARALLEL.
  
  
• IV50202: UNABLE TO PROCESS LOG MESSAGE WRITTEN EVENTS THAT CONTAIN AN EQUAL SIGN IN TEXT STRING.
  
  
• IV50484: DATECALC PLACEHOLDER DOES NOT RUN IN SAP EXTENDED VARIANT.
  
  
• IV50649: MONMAN DOES NOT REPORT A VALID MESSAGE IN CASE SSMAGENT SERVICE.
  
  
• IV50688: 910 CHECK PREREQ RETURN INCORRECT.
  
  
• IV50769: ALTPASS "USERNAME" CAUSES A CORE DUMP IF ENPLANAUDIT IS ENABLED.
  
  
• IV51242: CREATE SIMULTANEOUSLY TWO FORECAST WITH SAME NAME PRODUCES NO FILE AS RESULT.
  
  
• IV51793: WHEN UPGRADING BROKER FROM 851 TO 86 UNIQUE AGENT ID CHANGES.
  
  
• IV52091: MDM CRASHES IF TDWC QUERIES, MAKEPLAN, GC RUNS SIMULTANEOUSLY.
  
  
• IV52564: TIVOLI WORKLOAD SCHEDULER V8.6.0.02 JOBMANRC.CMD FILE DOES NOT SET USER'S VARIABLES ON WINDOWS WHEN LAUNCHING JOBS.
  
  
• IV52846: ADD JOBMON CAPABILITY TO RUN INTERACTIVE JOBS EVEN IF THERE AREN'T AVAILABLE SESSIONS.
  
  
• IV52870: NULLPOINTEREXCEPTION ON SBD COMMAND RUN AFTER A REPLACE.
  
  
• IV52973: PLANMAN BIND CAN CAUSE A DB DEADLOCK IF THERE ARE JOBS BINDING AND CONCURRENT SHADOW JOBS NOTIFICATIONS.
  
  

IBM Tivoli Workload Scheduler Planning and Installation, Version 9.2

In Chapter 7. "Upgrading", section "Upgrading from V8.5.0 or V8.5.1", sub-section "Upgrading a master domain manager instance or its backup", paragraph "Installing a new master domain manager or upgrading your old master domain manager", replace the following statement:

Upgrade the old master domain manager

You cannot make a direct upgrade of a version 8.5.0 or 8.5.1 master domain manager to a version 9.2 master domain manager. Either maintain the master domain manager at the original version or perform a direct upgrade to version 9.1.

with:

Upgrade the old master domain manager

A direct upgrade is possible if you upgrade a version 8.5.0 or 8.5.1 master domain manager to a version 9.2 Fix Pack 2 master domain manager.

Top

© Copyright International Business Machines Corporation 1991, 2016. All rights reserved. US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.