PH47613: INTERMITTENT ABEND0C4 - 10 AND ABEND202 WHEN USING THE CICS ASYNCHRONOUS API

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• When using the CICS asynchronous API the customer encounters
  abend0C4 - 10 and an abend202. The systrace shows :
  
  00837688 SVC     2 00000000_23A088A6 00000000 29343EA0  Post
                     07850001 80000000
  00837688 SSRV 12A  A38D5CB4  2380AF28 00000000 00000000 Post
  
  00837688 PGM   010 00000000_015F5344  00040010
                     07842000 80000000           273EC400
  
  
  
  
  00837688 *RCVY  PROG 940C4000 00000010
  
  
  
  The SVC 2 is issued out of DFHASRM and the SSRV 12A out of
  DFHDSAUT. DFHDSAUT is attempting
  to post an ECB (2380AF28) that is part of an ECBLIST :
  
  238F6000   2380AF28   000478B8   00036548   273EC26C   |
  ..............B% |
  238F6010   0004468C   800445E4   800445E4   800445E4   |
  .......U...U...U |
  238F6020   800445E4   00000000   00000000   00000000   |
  ...U...............|
  
  This ECB is the wakeup ECB for the QR TCB. As part of the POST
  process the other ECBs in the list
  need to be unwaited. The fourth ECB in this list, 273EC26C,
  belongs to another user transaction
  and this resides in a different subspace. It is not permissable
  to touch an ECB in another task's
  subspace and this causes the abend0C4-10 and the subsequent
  abend202.
  
  Additional info :
  
  The last entry in the CICS internal trace was :
  
  67558 L808W AS 0500 ASRM  ENTRY PERFORM_COMMIT
  27F44090,NO,NO,NO,NO,NO,FORWARD,019A22EB,NOTHING,UNNECESSARY
  
  The POST is issued as part of the commit processing
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All CICS users                               *
  ****************************************************************
  * PROBLEM DESCRIPTION: Intermittent abend202 when using the    *
  *                      CICS asynchronous API.                  *
  ****************************************************************
  A transaction defined with ISOLATE issued an EXEC CICS WAITCICS
  command on an ECB from under the QR TCB.
    Another QR TCB transaction also defined as ISOLATE issued an
  EXEC CICS RUN TRANSID of a third transaction. This ran a
  program defined as Required, so was run under an open TCB.
  The QR TCB had no work to do so an MVS WAIT on all ECBs
  occurred in partition exit, and this set the top bit of the
  ECBs to x'80'.
    When the open TCB child transaction ended, DFHASRM was
  called to commit the asynchronous API parts of its UOW. This
  has to issue an MVS POST to wake up its suspended parent task.
  The MVS POST goes through all the ECBs that the QR TCB is
  waiting on and resets the top bits. When an attempt was made to
  reset the top bit of the ECB for the ISOLATE=YES WAITCICS
  application, running in its own subspace, an 0C4-10 occurred
  followed by an abend 0202 AKEB.
  

Problem conclusion

• DFHASRM has been modified.  Around the MVS POST call, if the
  transaction is running on an open TCB in its own subspace,
  CICS will switch to basespace before the POST, and restore the
  transaction isolation subspace afterwards.
  

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:

  PH47649 UI82598

Modules/Macros

• DFHASAS  DFHASDM  DFHASDUF DFHASNT  DFHASRM  DFHASST  DFHASSY
  DFHASTRI DFHASXM  DFHSAXDF
  

Fix information

• Fixed component name

  CICS TS Z/OS V5

• Fixed component ID

  5655Y0400

Applicable component levels

• R300 PSY UI82598

     UP22/09/29 P F209

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.