PM92897: 5C6-00C51045 ABENDS ISSUED BY ALL QUEUE MANAGERS IN A CFSTRUCT DUE TO A IXLRSNCODEINVALIDEVENT RETURN CODE FROM A IXLEERSP CALL

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• The recovery of QMGR connection in a CFSTRUCT, following
  the EEPLDISCFAILCONNECTION event, fails on each queue
  manager at phase 1, causing phase 2 recovery not to be
  attempted.
  IXLUSYNC calls are issued after phase 1 for synchronization
  of the recovery related to the EEPLDISCFAILCONNECTION
  event.
  When the user sync point has been confirmed by all queue
  managers, they decide not to attempt phase 2 recovery
  and then call IXLEERSP to respond to the
  EEPLDISCFAILCONNECTION event.
  At this point the event has been dealt with from a CF
  point of view, thus becoming invalid for any future
  notifications.
  The queue managers that have pending USYNC events
  queued will now go ahead and initiate another attempt
  to recover the connection.
  This fails again during phase 1 recovery, and when
  the IXLEERSP calls are issued for the same
  EEPLDISCFAILCONNECTION event, which is invalid,
  and IXLRSNCODEINVALIDEVENT is returned.
  The structure task in the queue manager issues
  the 00C51045 abend.
  .
  DUMP TITLE=XXXX,ABN=5C6-00C51045,U=SYSOPR,C=R3600.710.CFM
  -CSQESTE ,M=CSQGFRCV,LOC=CSQELPLM.CSQESTE +0000245C
  .
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of WebSphere MQ for z/OS Version 7 *
  *                 Release 1 Modification 0.                    *
  ****************************************************************
  * PROBLEM DESCRIPTION: EEPLLOSSCONN event, from loss of a      *
  *                      queue manager in a queue sharing group  *
  *                      may result in abend 5C6-00C51045, if    *
  *                      peer level recovery phase 2 is not      *
  *                      attempted as shown by message CSQE015I. *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  If during the recovery initiated by the EEPLLOSSCONN event,
  phase 1 of recovery fails on a queue manager, phase 2 recovery
  will not be attempted, as shown by message CSQE015I. This may
  result in the queue managers abending 5C6-00C51045 in location
  CSQELPLM.CSQESTE.
  
  When the recovery event is triggered, all queue managers in the
  queue sharing group are notified of the event. All the queue
  managers attempt to respond to the event with a usync to
  initiate peer level recovery. Only one queue manager will
  succeed and the other queue managers queue their usyncs. If
  phase 1 of recovery fails on any queue manager, phase 2 is not
  attempted and the EEPLDISCFAILCONNECTION event is responded to,
  and processed by the coupling facility. When this is complete
  the queued usyncs are then processed. This will initiate
  recovery again for the same event, which will fail due to the
  event already having been processed, resulting in the 00C51045
  abend.
  

Problem conclusion

• The recovery processing has been updated to clear any pending
  duplicate events, if peer level recovery fails during phase 1
  in this scenario.
  100Y
  CSQESTE
  

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:

  UK97348

Modules/Macros

• CSQESTE
  

Fix information

• Fixed component name

  WMQ Z/OS V7

• Fixed component ID

  5655R3600

Applicable component levels

• R100 PSY UK97348

     UP13/10/16 P F310

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.