IV46533: WebSphere MQ Cluster full repository stops "REFRESH CLUSTER" command and generates error AMQ9511

WebSphere MQ V7.5 Fix Pack 7.5.0.3
WebSphere MQ V7.0 Fix Pack 7.0.1.12
Fix Pack 7.1.0.5 for WebSphere MQ V7.1

APAR status

• Closed as program error.

Error description

• When executing the "REFRESH CLUSTER" command for a WebSphere MQ
  7.0 cluster full repository, it is stopped before completion.
  
  This prevents some updates being sent to partial repository
  queue managers. Then, error message AMQ9511 (reason code 2041)
  is written in the full repository queue manager error log files.
  
  Later (some weeks later) on partial repository queue managers
  there were name lookup failures, causing applications to receive
  reason code 2085 from MQOPEN.
  

Local fix

Problem summary

• ****************************************************************
  USERS AFFECTED:
  Users running REFRESH CLUSTER on a Full Repository queue manager
  in a cluster.
  
  Platforms affected:
  All Distributed (iSeries, all Unix and Windows)
  ****************************************************************
  PROBLEM SUMMARY:
  During REFRESH CLUSTER on a full repository, the WebSphere MQ
  product code was mistakenly trying to send updates via channels
  that were not "in use".  Channels in this state will exist for a
  few weeks if the user has defined more than one queue manager
  with the same channel name.  This configuration might have
  happened deliberately or by a user mistake.
  
  The WebSphere MQ product code should not have tried to send an
  update to a non-"in use" channel, and checks later in the queue
  manager code were preventing this happening.  The error
  conditions being flowed internally to the queue manager were not
  being handled correctly.
  
  If the user introduces a queue manager with a channel name
  already in use in the cluster, it replaces the existing queue
  manager as the "in use" record.  This is as per design, to allow
  replacement queue managers to take over in the event of a
  hardware failure and software rebuild.  But the same
  configuration can also be the result of a mistake - example: the
  creation of a test queue manager with the same CLUSRCVR
  definition as a production queue manager.
  
  Regardless of the reason for the configuration change, the user
  might want to change their mind or correct their mistake.  If
  the user wants to ensure that the "correct" queue manager is
  in-use they should alter the CLUSRCVR definition on the correct
  queue manager - for example, by altering the DESCR attribute -
  which will ensure a new update is sent to the full repositories.
  On rare occasions (example: a duplicate qmgr was created by
  restoring a file-level backup to a new machine) it will also be
  necessary also to issue REFRESH CLUSTER on the correct queue
  manager before altering its CLUSRCVR definition, to ensure the
  internal sequence numbers are up to date.
  

Problem conclusion

• The WebSphere MQ product code will be corrected so that sending
  of updates will not be attempted via non-"in use" channels.
  
  Additions will be made to the WebSphere MQ Information Center to
  emphasize the need to alter the CLUSRCVR DESCR attribute.
  Additions will also be made to describe the matters referred-to
  in this APAR description.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v7.1       7.1.0.5
  v7.0       7.0.1.12
  v7.5       7.5.0.3
  
  The latest available maintenance can be obtained from
  'WebSphere MQ Recommended Fixes'
  http://www-1.ibm.com/support/docview.wss?rs=171&uid=swg27006037
  
  If the maintenance level is not yet available information on
  its planned availability can be found in 'WebSphere MQ
  Planned Maintenance Release Dates'
  http://www-1.ibm.com/support/docview.wss?rs=171&uid=swg27006309
  ---------------------------------------------------------------
  

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:

  PI04720

Fix information

• Fixed component name

  WMQ AIX V7

• Fixed component ID

  5724H7221

Applicable component levels

• R701 PSY

     UP