PH03171: MQ: A CLUSTER REPOSITORY LOSES SIGHT OF SOME OTHER QUEUE MANAGERS AFTER A REFRESH CLUSTER IS DONE

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• A REFRESH CLUSTER(clusname) REPOS(NO) command was issued.
  Afterward, some other queue managers in the cluster were
  unreachable from the queue manager where the REFRESH was done.
  Applications received MQRC 2087 when attempting to send reply
  messages back to these queue managers.
  
  The issue occurs when a cluster sender channel is active at the
  time that the refresh is issued but then goes inactive before
  the full repository has republished that cluster-sender record
  in response to the refresh.
  
  Additional Symptom(s) Search Keyword(s):
  MQRC_UNKNOWN_REMOTE_Q_MGR CLUSSDR
  

Local fix

• Do another REFRESH CLUSTER command or simply alter the
  description of the CLUSRCVR channel on the target queue manager
  to cause the cluster subscription to be republished.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of IBM MQ for z/OS Version 9       *
  *                 Release 1 Modification 0                     *
  ****************************************************************
  * PROBLEM DESCRIPTION: A Partial Repository Queue Manager may  *
  *                      incorrectly lose sight of other Qmgrs   *
  *                      in a cluster, after a REFRESH CLUSTER   *
  *                      REPOS(NO) command is issued.            *
  ****************************************************************
  If auto-defined cluster sender channels are running when a
  REFRESH CLUSTER REPOS(NO) command is issued, but the channel
  stops prior to the REFRESH command processing completes, the
  republish of cluster Qmgr records may not be correctly added to
  the local cluster cache by reconcile processing. As a result the
  Qmgr may lose sight of the remote cluster Qmgr until the
  internal subscription for the Qmgr expires or the remote Qmgr's
  cluster receive definition is altered.
  

Problem conclusion

• Cluster refresh and reconcile processing has been corrected,
  to ensure the republished cluster Qmgr record is added to the
  cluster cache, when a running cluster channel stops prior to
  refresh processing completing.
  

Temporary fix

Comments

APAR Information

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

  PI98289

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

  UI59798 IT27834

Modules/Macros

• CMQXRECO CMQXRREF
  

Fix information

• Fixed component name

  IBM MQ Z/OS V9

• Fixed component ID

  5655MQ900

Applicable component levels

• R100 PSY UI59798

     UP18/12/15 P F812 ¢

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.