PI29480: WMQ V710 - ABEND5C6-00D1032A OCCURS IF A UNIT OF WORK HAS BEEN LEFT 'ORPHANED' AND THE ARCHIVE LOGS FOR IT ARE NOT AVAILABLE.

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• A WebSphere Application Server ( WAS ) region received an
  ABENDS13E directly after commit phase 1 was started for a UOW.
  This unit-of-work was left 'stranded'.
  Normally, it would be expected for the queue-manager to shunt
  long running units of work such as these, but it will not do
  so if only ACECMT1 is set (it will only be expected that
  this state is short-lived). This therefore resulted in the
  queue-manager abending during shutdown (when it tried to backout
  the unit of work, but the archive log was not available) and
  when trying to be restarted.
  
  CSQJ113E (ssid RBA 14F19DB94000 NOT IN ANY ACTIVE OR ARCHIVE
           LOG DATA SET
  CSQY291E CSQWDSDM SDUMPX FAILED, RC=00000B08,ABN=5C6-00D1032A,
           LOC=CSQJL002.CSQJR003+00000ACA
  *CSQV086E (ssid QUEUE MANAGER ABNORMAL TERMINATION
           REASON=00D96031  - ABEND=S6C6 U0000 REASON=00D96031
  

Local fix

• n/a
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of WebSphere MQ for z/OS Version 7 *
  *                 Release 1 Modification 0.                    *
  ****************************************************************
  * PROBLEM DESCRIPTION: If an application bound with an RRS     *
  *                      stub abends or terminates during a      *
  *                      commit a unit of work may be 'stranded' *
  *                      in a state similar to being indoubt,    *
  *                      but will not be backed out and logs     *
  *                      will not be shunted as expected for     *
  *                      long running UOW.                       *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  A timing window exists for applications bound using a RRS batch
  adapter, when an abend occurs during commit processing. If an
  abend occurs in the application after phase 1 commit processing
  has been started, but not completed for a persistent message,
  the associated unit of work may become stranded. This results in
  the UOW being left around for the rest of the active life of the
  queue manager but does not show the UOW as long running, message
  CSQJ160I is not output. The log records for the UOW are not
  shunted in this case, which can result in the queue manager
  abending during shutdown, and being unable to be restarted, if
  the archive logs for the UOW are no longer available, as the UOW
  cannot be backed out.
  

Problem conclusion

• The processing during an abend in this scenario has been updated
  to treat the unit of work, where only phase 1 of commit
  processing has been started, as inflight. This results in the
  unit of work being backed out in this scenario.
  100Y
  CSQARIB
  CSQ3RRSM
  CSQ3RRSX
  

Temporary fix

• *********
  * HIPER *
  *********
  

Comments

APAR Information

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

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

  PI30414 UI24447

Modules/Macros

• CSQARIB  CSQ3RRSM CSQ3RRSX
  

Fix information

• Fixed component name

  WMQ Z/OS V7

• Fixed component ID

  5655R3600

Applicable component levels

• R100 PSY UI24447

     UP15/02/04 P F502 ¢

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.