PH25265: AFTER AN ABEND IN A JVMSERVER PURGE ESCALATION DOES NOT HAPPEN

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• After an unexpected abend in a JVM server it will become stuck
  in DISABLING state and require a KILL to get it to a DISABLED
  state.
  
  After this sort of failure CICS should have moved on to issue
  PURGE, FORCEPURGE and eventually KILL without the need for
  operator involvement.
  CICS would wait for 60 seconds before issuing PURGE because this
  is a Liberty server.
  FORCEPURGE and KILL would have been issued at 15 second
  intervals afterwards.
  
  There does appear to be a problem in the CICS code that means
  that the flag controlling whether the escalation process is
  started or not is not initialized correctly.
  This means that the escalation process may not be started in
  some cases.
  

Local fix

• NA
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All.                                         *
  ****************************************************************
  * PROBLEM DESCRIPTION: Automatic purge escalation does not     *
  *                      happen after an unexpected abend in a   *
  *                      JVM server.                             *
  ****************************************************************
  A CICS thread running in a JVM server suffer an unexpected
  abend.  This causes the JVM server to be disabled and then
  automatically re-enabled.
  
  CICS initially disables the JVM server using the PHASEOUT option
  but should initiate an automatic escalation of the purge level
  used (PURGE, FORCEPURGE and finally KILL) to ensure the JVM
  server reaches a DISABLED state without operator involvement.
  
  A flag controlling the purge escalation process was not being
  correctly initialised.  This means that the purge escalation
  process may not take effect when it should.  This could leave
  JVM servers stuck in a DISABLING state for an extended period.
  

Problem conclusion

• DFHSJTH has been changed to ensure the purge escalation process
  is always used when required.
  

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:

  UI71678 UI71679

Modules/Macros

• DFHSJTH
  

Fix information

• Fixed component name

  CICS TS Z/OS V5

• Fixed component ID

  5655Y0400

Applicable component levels

• R000 PSY UI71679

     UP20/09/24 P F009

• R100 PSY UI71678

     UP20/09/23 P F009

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.