PI86167: IBM MQ V8: ABEND 0C4-00000004,C=W9700.800.CHIN,M=CSQXDISP, LOC=CSQXADPM.????????+0118E WHEN STOPPING QUEUE MANAGER.

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• During shutdown, the queue manager tries to put an event message
  to a cluster queue defined on another queue manager. The cluster
  channel was in a get-wait on the SYSTEM.CLUSTER.TRANSMIT.QUEUE
  at the time of the STOP QMGR command and there was an attempt to
  put the message directly into the getter's buffer. However CHIN
  shutdown processing posts the channel out of the get-wait at the
  same time. This results in abend 0C4 in CSQM2PGB and the
  subsequent abend 0C4 in CSQXADPM.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of IBM MQ for z/OS Version 8       *
  *                 Release 0 Modification 0.                    *
  ****************************************************************
  * PROBLEM DESCRIPTION: A storage overlay can occur in the      *
  *                      channel initiator address space if a    *
  *                      message is put to a transmission queue  *
  *                      at the same time as the queue manager   *
  *                      is quiescing.                           *
  *                      Symptoms of the overlay will depend on  *
  *                      how the overlaid storage is used, and   *
  *                      can include abend 0C4 in CSQXADPM.      *
  ****************************************************************
  A channel with NPMSPEED(FAST) was waiting for a message to
  arrive on the transmission queue when the queue manager was
  quiesced.
  CSQXDSPT posted the channel with MQEC_CONNECTION_QUIESCING,
  allowing the channel to leave CSQXADPM and  start termination
  processing.
  However, until the channel closes the transmission queue, the
  handle remains on the get wait chain. If a non-persistent
  message is put out of syncpoint to the queue, it is eligible to
  be put directly to the getter's buffer - this was in CSQXADPM's
  working storage, which could have been reused by other routines
  called by the channel, leading to unpredictable behaviour.
  

Problem conclusion

• CSQXADPM is changed to cancel the getwait request, removing the
  handle from the getwait chain, prior to returning when posted
  with MQEC_CONNECTION_QUIESCING.
  

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:

  PH05919 UI60518

Modules/Macros

• CSQXADPM
  

Fix information

• Fixed component name

  WMQ Z/OS 8

• Fixed component ID

  5655W9700

Applicable component levels

• R000 PSY UI60518

     UP19/02/02 P F901

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.