PH10479: CMAS NOT SHOWING SOME ACTIVE MAS TASKS

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• There is a problem trying to extract info from MASes in a
  specific CICSPLEX. When it uses a batch job using CPSM API, or
  uses WUI, the CONNECT command works but fails on the GET
  OBJECT. CPSM trace shows the following error:
  OUT
     *RSP RESPONSE( OK )
     *RSN REASON( *             )
     *LID LIST_ID( 0000000000000000 )
     *BIN ELEMENT_CNT( 00000000 )
     *BIN INV_OFFSET( 00000000 )
     *STA STATUS( NONE_ACTIVE )
  And Batch job shows:
  CONTEXT=plexname,SCOPE=scopename,OBJECT=PROGRAM....
   CONNECT    command successfully executed
   FUNCTION: GET          RESPONSE: NOTAVAILABLE   REASON: SCOPE
  This occurs after message EYUTS0002E was issued to report
  a failure to join a MAS to a CMAS.
  

Local fix

• Restart the CMAS (warm start).
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of CICSPlex SM V5R3M0, V5R4M0,     *
  *                 and V5R5M0.                                  *
  ****************************************************************
  * PROBLEM DESCRIPTION: When a MAS is terminated shortly after  *
  *                      it is started or is terminated while a  *
  *                      CMAS is in the process of starting, it  *
  *                      is possible for different phases of the *
  *                      Topology Connect or Join processing     *
  *                      to interleave with the Termination.     *
  *                                                              *
  *                      When this occurs, Topology messages     *
  *                      EYUTS0004I/EYUTS0002E may be issued     *
  *                      to record the failure. Subsequent       *
  *                      attempts to restart the MAS will        *
  *                      produce the same failure.               *
  *                                                              *
  *                      If this occurs in the Connecting CMAS   *
  *                      then the MAS will be be unavailable to  *
  *                      every CMAS in the CICSplex. If instead  *
  *                      the failure occurs in any CMAS          *
  *                      attempting to Join to the MAS then the  *
  *                      MAS remains available to other CMASes   *
  *                      which can complete join processing      *
  *                      correctly.                              *
  *                                                              *
  *                      API commands for this MAS directed to   *
  *                      a CMAS where this failure occurred      *
  *                      will report failures such as            *
  *                      RESPONSE: NOTAVAILABLE   REASON: SCOPE  *
  ****************************************************************
  * RECOMMENDATION: After applying the PTF which solves PH10479, *
  *                 all CMASes must be restarted.                *
  *                 The restarts need not be performed at the    *
  *                 same time, however if systems are not        *
  *                 restarted at the same time, the following    *
  *                 rules apply:                                 *
  *                                                            . *
  *                 - Maintenance Point (MP) CMASes must be      *
  *                   restarted on the updated code before       *
  *                   non-MP CMASes.                             *
  *                                                            . *
  *                 - If you have more than one MP CMAS and any  *
  *                   of those MP CMASes are connected directly  *
  *                   or indirectly, then those MP CMASes        *
  *                   must be restarted at the same time.        *
  *                                                            . *
  *                 - Before a MAS, regardless of whether it is  *
  *                   running as a WUI server, is restarted      *
  *                   with the updated code, the CMAS to which   *
  *                   the MAS connects must be running with the  *
  *                   updated code.                              *
  *                                                            . *
  *                 - This fix is being provided across the      *
  *                   following releases of CICSPlex SM:         *
  *                                                            . *
  *                   -  CPSM V5R3M0 - APAR PH10479              *
  *                   -  CPSM V5R4M0 - APAR PH10479              *
  *                   -  CPSM V5R5M0 - APAR PH10479              *
  *                                                            . *
  *                   Before a CMAS running with the PTF that    *
  *                   resolves this APAR for its release         *
  *                   connects directly or indirectly to a CMAS  *
  *                   running a higher release of CPSM, the      *
  *                   higher release CMAS must be restarted so   *
  *                   that it is running with the appropriate    *
  *                   PTF for its release.                       *
  ****************************************************************
  When a CMAS learns of a MAS start or termination, an event is
  sent to the processing queue for method EYU0TIST (TIST), which
  runs under the long running task TIST.  When TIST receives a
  connect (EENDCONN) or join (EENDJOIN) event, it requests that
  a TSSC (connect) or TSSJ (join) task be started to handle the
  request.  When a MAS disconnects, the CSDB descriptor block
  for the MAS will immediately be updated to move the MAS to
  TERMINATING state. Later, the TSST task will be notified
  through another event and will call method EYU0TSST directly
  to handle the request and complete termination processing.
  
  In the reported problem, a CMAS region began initialisation,
  and EENDJOIN records were generated for all of the MAS regions
  connected to other CMASes.  These EENDJOIN records were placed
  on the TIST task work queue.
  Meanwhile, one of those MAS regions terminated.  The MAS was
  immediately moved to TERMINATING state, and this happened
  before the TIST task could read the EENDJOIN relating to this
  MAS.  This allowed elements of the termination processing to
  interleave with the TSSJ task, causing not only that join, but
  also all future joins of that MAS to this CMAS, to fail.
  
  The CMAS will issues the following messages to the EYULOG:
  date time EYUTS0001I applid Topology Connect for name Initiated
            - APPLID(name) CICSplex(plexname).
  date time EYUTS0004I applid Topology Found name already Active.
  date time EYUTS0002E applid Topology Connect for name Failed
            - APPLID(name) CICSplex(plexname).
  

Problem conclusion

• CPSM has been modified so that a timestamp is taken when an
  EENDCONN or EENDJOIN event is generated. This timestamp is
  embedded in that record, and passed though to the resulting
  TSSC or TSSJ task. This allows CPSM to detect the case where
  a MAS terminates after the EENDJOIN event is generated, but
  before that event can be processed by the TIST task.
  
  To facilitate this, the following changes have been made:
  
    - The EENDCONN and EENDJOIN resource tables have been
      updated to include a new timestamp value.
      This will be set when the record is generated.
  
    - Methods EYU0CPAM ( CPAM ), EYU0CPEJ ( CPEJ ), and
      EYU0TSSA ( TSSA ) have all been updated to capture
      a new timestamp when deciding to initiate a new
      connect or join event.
  
    - Method EYU0TIST ( TIST ) is updated to pass the
      new event start time to the TSSC and TSSJ tasks.
  
    - Methods EYU0TSSC ( TSSC ) and EYU0TSSJ ( TSSJ )
      have been updated to use the new event time stamp
      when determining whether or not the MAS has
      terminated since the connect/join was initiated.
  

Temporary fix

Comments

• ×**** PE20/01/28 FIX IN ERROR. SEE APAR PH21495  FOR DESCRIPTION
  

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

  UI63618 UI63619 UI63620

Modules/Macros

• EYU0CPAM EYU0CPEJ EYU0CPLT EYU0TIST EYU0TSSA EYU0TSSC EYU0TSSJ
  EYU9TSR3 EYU9TSR4 EYU9TSRU EYUT2459 EYUT2461 EYUY2459 EYUY2461
  

Fix information

• Fixed component name

  CICS TS Z/OS V5

• Fixed component ID

  5655Y0400

Applicable component levels

• R00M PSY UI63618

     UP19/06/18 P F906

• R10M PSY UI63619

     UP19/06/18 P F906

• R20M PSY UI63620

     UP19/06/18 P F906

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.