PM97994: WMQ CLUSTER MAINTENANCE ROUTINE RECEIVES BAD RETURN CODE FROM GARBAGE COLLECTION AFTER WHICH THE REPOSITORY MANAGER STOPS

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• In this case cluster maintenance received a return code of
  rrcE_ACTIVE_CACHE_QUERY from garbage collection. Prior to
  version 7.1.0 WebSphere MQ would disregard bad return codes
  from maintenance, but in version 7.1.0 the repository manager
  will be stopped ( CSQX448E ). This is considered too drastic
  and instead the problem should only be reported by generating
  an FFST.
  .
  Verification steps:
  - The CHIN joblog has
    CSQX053E CSQXFFST Error information recorded in CSQSNAP data
             set
    CSQX448E CSQXREPO Repository manager stopping because of
             errors.  Restart in 600 seconds
  and
  - The FFST in CSQSNAP has eyecatchers of
      *X...XFFSrrmMaintenance..........*
    and
      *                        ....In-H*
      *and Maintenance Object..WDR ....*
    Keywords: Hand In-Hand rrmMaintenance XFFSrrmMaintenance
  and
  - The hex data in the FFST for XFFSrrmMaintenance has a word
    with 20009458 at +x'30'.  This is a code that indicates the
    error in garbage collection.
  

Local fix

• To set SYSTEM.CLUSTER.COMMAND.QUEUE to be get-enabled so that
  the repository manager will restart successfully, issue
   /cpf ALTER QLOCAL(SYSTEM.CLUSTER.COMMAND.QUEUE) GET(ENABLED)
  where "cpf" is the command prefix for the queue manager.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of WebSphere MQ for z/OS Version 7 *
  *                 Release 1 Modification 0.                    *
  ****************************************************************
  * PROBLEM DESCRIPTION: Cluster repository manager terminates   *
  *                      with FFST indicating a failure in       *
  *                      rrmMaintenance and a return code in the *
  *                      FFST of 20009458.                       *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  After rrmMaintenance has enumerated over all the cluster objects
  it calls rfxCheckRegister in order to do garbage collection.
  If there are other active queries of the cache at the time then
  it will return rc = rrcE_ACTIVE_CACHE_QUERY.
  Since V710 the caller, rrmRunTimers, responds to the rc by
  exiting early and percolating the rc back so that the repository
  manager terminates.
  The repository manager will start again after 10 minutes but
  only when the SYSTEM.CLUSTER.COMMAND.QUEUE is GET(ENABLED). The
  termination of the repository manager is rather excessive in
  this situation. Better to report the error and allow the
  repository manager to continue.
  

Problem conclusion

• CMQXRMNT has been changed to report the problem with an FFST if
  there are 5 consecutive failures and allow the repository
  manager to continue if garbage collection could not run.
  100Y
  CMQXRMNT
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

  IV50907 UI12030

Modules/Macros

• CMQXRMNT
  

Fix information

• Fixed component name

  WMQ Z/OS V7

• Fixed component ID

  5655R3600

Applicable component levels

• R100 PSY UI12030

     UP13/11/20 P F311

Fix is available

• Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.