PH38985: DFHTS0001 ABEND (CODE 0C4/AKEA) IN DFHTSMB OFFSET X'156C'

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• CICS region went SOS above the line and then abended shortly
  afterward following program checks in DFHTSMB and DFHDSTCB:
  
  +DFHTS0001 An abend (code 0C4/AKEA) has occurred at offset
  X'156C' in module DFHTSMB.
  
  +DFHDS0001 An abend (code 0C4/AKEA) has occurred at offset
  X'0732' in module DFHDSTCB.
  
  The 0C4 in DFHDSTCB is what causes CICS to abend and occurred
  because two pages of storage had been overlaid -- a page of
  Dispatcher DTA control blocks and a prior page allocated to
  Lock Manager.  The 0C4s in DFHTSMB occur after GETMAINs for a
  temporary storage QAB fail to get a QAB, but the code proceeds
  as if it got one.  DFHTSMB first does a conditional GETMAIN with
  no suspend, and that will get an exception response with reason
  smgf_insufficient_storage.  DFHTSMB releases the TSLOCK, then
  does a 2nd unconditional GETMAIN for the QAB and suspends
  because storage is not available. The transaction has a DTIMOUT
  time, so after that time expires, the 2nd GETMAIN will fail.
  Before dealing with the GETMAIN response, DFHTSMB gets the
  TSLOCK again but the DFHLMLM parameters are built in the same
  place that the DFHSMGF parms are built ( XXYY_PARMS ). So the
  GETMAIN response from DFHSMGF is lost and the code proceeds as
  if it got a QAB, which includes overlaying storage which results
  in the 0C4 in DFHDSTCB.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All CICS users                               *
  ****************************************************************
  * PROBLEM DESCRIPTION: CICS transactions abend with message    *
  *                      "DFHTS0001 An abend (code 0C4/AKEA)     *
  *                      has occurred at offset                  *
  *                      X'156C' in module DFHTSMB" in a region  *
  *                      which is SOS and processing a heavy     *
  *                      Temporary Storage workload.             *
  ****************************************************************
  CICS is short on storage (SOS) and is issuing messages:
  "DFHSM0133 CICS is under stress" and
  "DFHSM0142 CICS is still under stress"
  Applications issue Temporary Storage API commands.
  The TS domain attempts to obtain storage for QAB control blocks.
  The SM domain suspends tasks, waiting for storage.
  If no storage becomes available within the DTIMOUT time-out
  interval, the task is purged, and SM returns to TS with
  SMGF_RESPONSE of SMGF_PURGED.
  Code in module DFHTSMB does not process the SMGF_PURGED response
  correctly - it is corrupted by a call to the LM domain because
  DFHTSMB uses the same area of storage for SMGF_PARMS and
  LMLM_PARMS.
  DFHTSMB continues processing, and attempts to clear the
  new QAB to nulls - but suffers an 0C4 program check because
  storage was not available and the pointer to the QAB is null.
  Additional keywords: ATSC AbendATSC MVCL TSQAB S0C4 abendS0C4
  

Problem conclusion

• DFHTSMB has been modified to use separate areas of storage
  for LMLM_PARMS and SMGF_PARMS.
  

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:

  UI76700 UI76701 UI76702

Modules/Macros

• DFHTSMB
  

Fix information

• Fixed component name

  CICS TS Z/OS V5

• Fixed component ID

  5655Y0400

Applicable component levels

• R100 PSY UI76702

     UP21/08/13 P F108

• R200 PSY UI76701

     UP21/08/13 P F108

• R300 PSY UI76700

     UP21/08/13 P F108

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.