PM77017: ACE AND EB CONTROL BLOCKS REMAIN ALLOCATED EVEN AFTER ASSOCIATEDWORKER TASKS NO LONGER EXIST 12/11/15 PTF PECHANGE

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• ACE and EB control blocks remain allocated after the context
  associated with them have already ended. The orphaned control
  blocks are allocated from subpools of storage used by WebSphere
  MQ and so can still be freed if the QMGR is shutdown.
  

Local fix

• Recycle queue manager
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of WebSphere MQ for z/OS Version 7 *
  *                 Release 0 Modification 1 and Release 1       *
  *                 Modification 0.                              *
  ****************************************************************
  * PROBLEM DESCRIPTION: After applying PTF UK79818 (for Version *
  *                      701) or UK81712 (for Version 710), a    *
  *                      leak of ACE/EB control blocks can occur *
  *                      when an application task using WMQ      *
  *                      private contexts ends without first     *
  *                      ending all the contexts.  This can      *
  *                      happen due to a normal end without      *
  *                      appropriate cleanup, or due to an       *
  *                      application abend.                      *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  Various changes were made in PM64200 (for V701) and PM67446 (for
  V710), corresponding to the above PTFs, helping to ensure that
  RRS threads do not go through deferred end-of-task if
  end-context has taken place or deferred end-context has been
  scheduled (or vice versa).  This code added extra serialization
  using the CCBDENDS flag to ensure that ACE/EBs were not cleaned
  up by one process while queued for cleanup by another.
  
  Following APAR PM64200 and PM67446 a problem can be seen during
  an abend or a normal end (without ending the context) of an
  application that creates a private context.  In this case a
  private context is scheduled for deferred end-context
  processing, which sets CCBDENDS to ensure that no-one else tries
  to deal with it.  When RRS is called to end the context during
  processing of the deferred end-context, RRS drives CSQ3RRSX exit
  for CTX_END_CONTEXT_EXIT.  CSQ3RRSX finds that CCBDENDS is set
  on and does not free the ACE/EB for the context and the ACE/EB
  gets orphaned.
  

Problem conclusion

• The code has been changed to reset CCBDENDS prior to calling RRS
  to end-context.
  010Y
  100Y
  CSQMCTXE
  

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:

  UK90568 UK90569

Modules/Macros

• CSQMCTXE
  

Fix information

• Fixed component name

  WMQ Z/OS V7

• Fixed component ID

  5655R3600

Applicable component levels

• R010 PSY UK90568

     UP13/02/02 P F302 Ø

• R100 PSY UK90569

     UP13/02/02 P F302 Ø

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.