IBM Support

Collecting Data: Reporting high CPU or OutOfMemory issues with zRule Execution Server for z/OS

Question & Answer


Question

The Rule Execution Server component of Operational Decision Manager on z/OS is using high CPU or is encountering OutOfMemory issues. What documentation you must collect (MustGather) so that the IBM Operation Decision Manager Support team can diagnose your problem. If you gather this documentation before contacting support it will expedite the troubleshooting process, and save you time.

Answer

To ensure ODM z/OS L2 and L3 Support teams have a full set of documents to investigate your issue, please collect following documentation for your scenario:
OutOfMemory Condition
  1. If verbose GC isn't enabled, enable it now by adding the following to the server's JAVA_OPTIONS:
    -verbose:gc
    Note: a server restart is required before the JVM will start collecting verbose GC data.
  2. Terse the SVC dump generated when the OutOfMemory condition occurred and provide it to IBM. 
  3. Collect the following logs and provide them to IBM
    1. Collect the logs as specified here:  https://www.ibm.com/support/pages/node/544277
    2. Please provide all log files in the ++HBRWORKPATH++/logs directory
    3. Please provide the SYSOUT, SYSPRINT, HBRPRINT, SYS00001 and SYS00002 job output elements from the SDSF job logs for the Master address space
    4. Please provide all the members located in the ++HBRWORKDS++.SHBRPARM dataset.
High CPU 
  1. In order to ensure full data is collected, we recommend updating the default dump options by following the steps below:
    • Create a plain EBCDIC text file <filepath>/dump.option which contains:
      -Xdump:system:defaults:request=exclusive+compact+prepwalk
    • Update JAVA_OPTIONS to include:
      -Xoptionsfile=<filepath>/dump.option

      Note: A server restart is required for the new settings to take effect. 
  2. While the issue is occurring, collect a system dump. We recommend using the MODIFY command:
    MODIFY <zResStartedTaskName>,DUMP TYPE(SYSTEM)

    As outlined by the Commands Knowledge Center article.
  3. Collect the dump and logs to provide IBM:
    • Terse the dump collected in step above
    • Collect the logs as specified here:  https://www.ibm.com/support/pages/node/544277
    • Please provide all log files in the ++HBRWORKPATH++/logs directory
    • Please provide the SYSOUT, SYSPRINT, HBRPRINT, SYS00001 and SYS00002 job output elements from the SDSF job logs for the Master address space
    • Please provide all the members located in the ++HBRWORKDS++.SHBRPARM dataset.

[{"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SSQPJS","label":"IBM Operational Decision Manager for z\/OS"},"ARM Category":[{"code":"a8m50000000L1RZAA0","label":"zOS"}],"ARM Case Number":"","Platform":[{"code":"PF035","label":"z\/OS"}],"Version":"All Version(s)","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
16 June 2020

UID

ibm16212231

Page Feedback