PI40053: HANG IN DB2 THREAD USING MQ WITH RRS AS SYNCPOINT COORDINATOR.

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• An issue occurs when an application repeatedly calls a
  DB2 stored procedure that puts a message to MQ.  This is
  background information:
  .
  1. The customer recreated this using a distributed application,
     but for ease of doc-gathering, the customer used a batch job
     this time.  So, they started batch job BATCHPGM1.
  .
  2. Batch job BATCHPGM1 executes package PACKAGE1, which
     connects to DB2 and executes stored procedure
     STORED_PROCEDURE.
  .
  3. STORED_PROCEDURE executes package PACKAGE2, which puts a
     message to MQ series.
  .
  4. The batch job calls the stored procedure in step #3 5000
     times
  The issue with MQ support is around the fact that performing
  queue-manager abnormal termination processing, CSQVATRM, goes
  through all ASCEs looking for any threads which are in the
  middle of processing in the queue-manager (such as the backout
  exit thread mentioned above). However, CSQVATRM does not
  process the ASCE for MQ Master as ASCEACT is not set. This
  means that the thread is left suspended indefinitely following
  queue-manager termination.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of WebSphere MQ for z/OS Version 8 *
  *                 Release 0 Modification 0                     *
  ****************************************************************
  * PROBLEM DESCRIPTION: Following abnormal termination of a     *
  *                      queue manager any threads which were    *
  *                      acting as Server Distributed Syncpoint  *
  *                      Resource Manager SDSRM may hang         *
  *                      indefinitely.                           *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  When an address space is acting as a Server Distributed
  Syncpoint Resource Manager and a commit or abort is
  requested, a temporary dynamic ASCE is created. When a queue
  manager terminates abnormally CSQVATRM will terminate all the
  threads associated with allied ASCEs, however it overlooks
  temporary dynamic ones. If the commit or abort hadn't completed
  then the temporary dynamic ASCE will still exist and the thread
  will hang indefinitely.
  

Problem conclusion

• CSQVATRM will now terminate EBs which are associated with
  temporary dynamic ASCEs.
  000Y
  CSQVATRM
  

Temporary fix

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

Comments

APAR Information

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

  PI35954

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

  UI27739

Modules/Macros

• CSQVATRM
  

Fix information

• Fixed component name

  WMQ Z/OS 8

• Fixed component ID

  5655W9700

Applicable component levels

• R000 PSY UI27739

     UP15/06/03 P F506 ¢

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.