PH39877: DFHAP0001 ABEND (CODE 0C4/AKEA) IN MODULE DFHAPCR +X'09D6' DURING SOS

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• During a CICS SOS, module DFHAPCR +x'9D6' takes multiple 0C4
  abends. The SOS occurred after a flood of pipeline transactions
  started. Here are corresponding error messages:
  
  +DFHSM0142 CICS is still under stress (short on storage above
  16MB).
  
  +DFHSM0002 A severe error (code X'0315') has occurred in module
  DFHSMGF.
  
  +DFHAP0001 An abend (code 0C4/AKEA) has occurred at offset
  X'09D6' in module DFHAPCR.
  
  The 0C4s occur after a task suspends during a GETMAIN request
  from DFHAPCR, trying to obtain a channel buffer. The task is
  eventually purged, however DFHAPCR does not check the response
  but instead proceeds as if the GETMAIN was successful. The 0C4s
  occur because the address of the buffer is not valid.
  
  After the 0C4s in DFHAPCR, some tasks may loop in DFHRMRS and
  eventually get abended as runaway, message DFHRM0001 abend AKEC.
  The DFHRM0001 AKECs seem to be due to residual data following
  the 0C4s in DFHAPCR.
  
  Additional Symptom(s) Search Keyword(s):
  EXPORT_ALL  dfhapcr_obtain_channel_buffer CHANBUFF
  CheckIterators  abend0C4 S0C4 abend S0C4
  DFHSM0133 CICS is under stress (short on storage above 16MB)
  DFHSM0134 CICS is no longer short on storage above 16MB
  DFHSM0002 A severe error (code X'0305') has occurred in
            module DFHSMGF.
  You see the following exception trace entries:
  SM 1002 SMSQ  EXIT  - FUNCTION(SUSPEND_REQUEST)
                     RESPONSE(PURGED) REASON(INSUFFICIENT_STORAGE)
  SM 0302 SMGF  EXIT  - FUNCTION(GETMAIN) RESPONSE(PURGED)
                    REASON(INSUFFICIENT_STORAGE) ADDRESS(0011B0E9)
  Even though the getmain was not successful, the code continues
  as if it was successful, using a residual address above. In this
  case the address belonged to another task, so CICS moved data
  into this area causing storage violations as well as CICS areas
  like QPH's getting overlaid.  The next trace following the purge
  was:
  PG 1900 PGCR ENTRY - FUNCTION(GET_CONTAINER_INTO)
              CONTAINER_TOKEN(1F59A2D0) CALLER(SYSTEM) CONVERT(NO)
              ITEM_BUFFER(0011B418 ,00000000 , 0011B1DA)
  
  DFHSM0102 INVALID QPH
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All CICS users.                              *
  ****************************************************************
  * PROBLEM DESCRIPTION: Abend S0C4 in DFHAPCR during a short on *
  *                      storage in CICS.                        *
  ****************************************************************
  CICS is acting as a Web Services provider and, during a workload
  spike, goes in and out of SOS. Module DFHAPCR's EXPORT_ALL
  issues a getmain while CICS is SOS and suspends, waiting for
  storage to become available.
  After a while, the task is purged.
  DFHAPCR's GETMAIN procedure regains control, with a response of
  SMGF_PURGED, but continues as though an exception hadn't
  occurred.
  When DFHAPCR tries to initialise the storage, it suffers an 0C4
  abend because it has a null pointer.
  

Problem conclusion

• DFHAPCR, DFHPIIS and DFHPISN have been changed to act on the
  purged response rather than trying to access storage using a
  null pointer and percolating the response.
  DFHRZST has been changed to prevent a short on storage
  situation caused by the DFHAPCR IMPORT_ALL code trying to
  receive an 8 byte channel header and only getting the 4 byte
  trailer.
  

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:

Modules/Macros

• DFHAPCR  DFHPIA1  DFHPIAD  DFHPIAT  DFHPICA  DFHPICC  DFHPIDM
  DFHPIDSH DFHPIDUF DFHPIII  DFHPIIW  DFHPIJG  DFHPIJP  DFHPIJP1
  DFHPIJS  DFHPIJW  DFHPILN  DFHPIMM  DFHPIPL  DFHPIPM  DFHPIPM2
  DFHPIPS  DFHPIRE  DFHPIRI  DFHPIRL  DFHPIRM  DFHPIRN  DFHPIRS
  DFHPISC  DFHPISF  DFHPISN  DFHPIST  DFHPITC  DFHPITC1 DFHPITC2
  DFHPITC3 DFHPITE  DFHPITH  DFHPITJ  DFHPITL  DFHPITQ  DFHPITRI
  DFHPITS  DFHPIUE  DFHPIWR  DFHPIWT  DFHPIXI  DFHPIXM  DFHPIXO
  DFHRZTA  DFHWSADH DFHWSATR DFHWSATX DFHXCSCR
  

Fix information

• Fixed component name

  CICS TS Z/OS V5

• Fixed component ID

  5655Y0400

Applicable component levels

• R200 PSY UI78985

     UP22/01/21 P F201

• R300 PSY UI78984

     UP22/01/22 P F201

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.