PH71513: ALLOW MQ TO DETECT THAT THE APPLICATION ADDRESS SPACE HAS GONE EOM AND REACT APPROPRAITELY

APAR status

• Closed as program error.

Error description

• If an application address space (such as WAS) is forced into EO
  the MQ RRS EXIT code may get stuck in a loop waiting for the
  task to complete processing (which does not happen due to the
  EOM). The solution is for MQ to detect that the application
  address space has gone EOM and react appropriately.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of IBM MQ for z/OS Version 9       *
  *                 Release 3 Modification 0 and Release 4       *
  *                 Modification 0.                              *
  ****************************************************************
  * PROBLEM DESCRIPTION: The MQ Backout exit hangs when called   *
  *                      by RRS during End of Memory (EOM)       *
  *                      processing.                             *
  *                      After several minutes RRS abends the    *
  *                      task S5C4-082C0019                      *
  *                      The UOW is left in-flight in MQ,        *
  *                      leading to repeated CSQJ160I and        *
  *                      CSQR026I messages during checkpoint     *
  *                      processing.                             *
  ****************************************************************
  An application was memtermed while an EB associated with an RRS
  context was still active in MQ (EBACTL=OFF). During EOM
  processing RRS called MQ's backout exit, however this hung
  because the context EB was still marked as active in MQ. After
  several minutes RRS abends the task 5C4-082C0019, and marks the
  in-backout RRS UR as damaged, however the UR still shows as
  inflight in MQ, resulting in repeated CSQJ160I (long running uow
  found) and CSQR026I (long running uow shunted) messages being
  issued during MQ checkpoint processing until the queue manager
  next restarts or resynchronizes with RRS.
  

Problem conclusion

• CSQ3RRSX is updated to detect when the context EB associated
  with the Backout Exit call is still marked active due to EOM
  processing, and to correctly back out the unit of work.
  

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:

  PH72186 UO08769 UO08770

Modules/Macros

• CSQ3RRSX CSQARIB
  

Fix information

• Fixed component name

  IBM MQ Z/OS V9

• Fixed component ID

  5655MQ900

Applicable component levels