PH26596: CICS GETS AN ABEND 0C4 IN PROGRAM DFHSJSA AT X'656'

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• A CICS region gets an abend 0C4 in program DFHSJSA at x'656'.
  The following event appears immediately after that and the
  JVMServer stops working.
  
  'JVMSERVER' Event:
  A Posix signal was received by the JVM server. SIGQUIT may be
  used to obtain a Java dump. Signals causing termination
  (SIGKILL, SIGTERM, SIGABRT etc) should NEVER be issued. CICS may
  become unstable and the JVM server will be terminated in an
  uncontrolled fashion.
  
  
  
  The investigation showed the following:
  
  2 T8 TCBs are terminating at the same time.
  
  The first T8, executing in DFHSJSA, posts KTCB_TERMINATE_ECB and
  then loses control of the processor.
  
  The second T8, also executing in DFHSJSA, posts
  KTCB_TERMINATE_ECB, turns on KTCB_DAUGHTER_TERMINATED in the
  mother KTCB and then post the WAKE_UP_ECB of the parent TCB (the
  TP in this case).
  
  The systrace shows the following (0095C590 is the T8, 00998E88
  is the TP TCB)
  
  When the WAKE_UP_ECB of the TP gets posted, that drives the
  DETACH_TERMINATED_OWN_TCBS processing. This scans the active
  KTCB chain and will detach any daughter KTCB that is found to
  have the KTCB_TERMINATE_ECB posted. This causes the KTCB for
  second T8 to be detached as requested (it was this TCB that
  woke the TP up). The TP also finds the KTCB for first T8 with
  KTCB_TERMINATE_ECB posted so it also gets detached.
  
  The first T8 TCB resumes execution in DFHSJSA.  It suffers an
  0C4 when turning on KTCB_DAUGHTER_TERMINATED in the parent KTCB
  because KTCB_MOTHER_KTCB is now zero.
  

Local fix

• not available
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All CICS users.                              *
  ****************************************************************
  * PROBLEM DESCRIPTION: ABEND0C4 in DFHSJSA causes JVM server   *
  *                      to terminate.                           *
  ****************************************************************
  A JVM server is running in CICS.  Two T8 TCBs within that JVM
  are being detached at the same time.
  
  The first T8 TCB posts KTCB_TERMINATE_ECB in its KTCB and then
  loses control of the processor.  The second T8 TCB also posts
  its copy of the ECB but continues execution and posts the
  WAKE_UP_ECB for the parent TP TCB.
  
  The TP TCB is resumed.  It detaches the second T8 TCB and puts
  the KTCB on the free chain.  The first T8 TCB is also found to
  have the KTCB_TERMINATE_ECB posted so it gets detached and the
  KTCB placed on the free chain.
  
  The first T8 TCB gets control of the processor again and
  continues executing in DFHSJSA.  It loads field KTCB_MOTHER_KTCB
  to access the KTCB for the TP TCB.  This field is now zero
  because the KTCB is on the free chain.  DFHSJSA tries to turn on
  flag KTCB_DAUGHTER_TERMINATED in the parent KTCB which fails
  with an abend 0C4.  This abend may lead to an unexpected
  termination of the JVM server.
  
  Additional Keywords:
  SIGSEGV  SIGABRT  ABORT  LINKETCB DFHSJ0007 msgDFHSJ0007 SJ0007
  

Problem conclusion

• DFHSJSA has been changed to access field KTCB_MOTHER_KTCB before
  the current KTCB gets freed.
  

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:

  UI70396 UI70397 UI70398 UI70399

Modules/Macros

• DFHSJSA  DFJ@H356 DFJ@H427 DFJ@H467 DFJ@H468
  

Fix information

• Fixed component name

  CICS TS Z/OS V5

• Fixed component ID

  5655Y0400

Applicable component levels

• R000 PSY UI70396

     UP20/07/08 P F007 ¢

• R100 PSY UI70397

     UP20/07/21 P F007 ¢

• R200 PSY UI70398

     UP20/08/03 P F007 ¢

• R300 PSY UI70399

     UP20/07/09 P F007 ¢

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.