IBM Support

PM68981: REGION REMAINS IN STOPPING-FORCE CONNECTION STATUS AFTER STOP FORCE TO DISCONNECT FROM MQ

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • After an MQ crash CICS is trying to disconnect from MQ, however
    the connection status remains in Stopping-Force. There are a
    number of transactions currently active to MQ, and trying to
    execute MQ requests. As long as these tasks are still active the
    disconnect from MQ will not complete and will remain in the
    Stopping-Force state. x'7C' into the DFHMQLOC (CTTASKS) shows
    that there are currently a number of active CICS-MQ tasks. The
    disconnect will not complete until these tasks are resolved and
    purged.
    .
    Looking at those tasks it appears that these tasks were attached
    after a STOP FORCE was issued, so they could not be the reason
    that the STOP FORCE did not complete and why the connection is
    in the Stopping-Force status. New tasks should not be able to
    obtain an MQLOT and try to issue MQ requests because we are
    trying to disconnect at this point. So all new tasks should be
    abended when they try to do this In order for the disconnect to
    complete, the CTTASKS value has to reach 0. However, we are not
    reaching that value because we are still allowing new tasks to
    be attached and carry out MQ function, thus raising the CTTASKS
    value.
    .
    In addition, since the STOP FORCE is issued, all tasks currently
    in MQ are supposed to be forcepurged immiediately which would
    bring that CTTASKSvalue down to 0. However, we see in the trace
    that even after the STOP FORCE is issued, tasks are still
    issuing MQ requests and they are purged eventually, not
    immediately. So in essence we have a situation where:
    .
    - MQ crashes
    - CKAM issues STOP FORCE in order to disconnect then reconnect
      to MQ
    - Tasks that should have been purged for the STOP FORCE still
      run and issue MQ requests
    
    - New waves of tasks arrive and issue MQ requests that fail
      (these should have been abended and never allowed to get
      on the MQ task chain)
    - Tasks that were around when the STOP FORCE is issued are
      finally forcepurged
    - The New tasks continue to run and attempt MQ requests
    .
    So in this situation we will never get an CTTASKS value of 0 and
    in turn the disconnect will not complete and we will remain in
    the Stopping-Force status. Logic should be added to issue a
    connection status check prior to creating a new task LOT in
    order to prevent this kind of error.
    Additional Keywords: Forcepurge , Disconnect , Stopping , Force
    KIXREVSLY
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All                                          *
    ****************************************************************
    * PROBLEM DESCRIPTION: The CICS/WMQ adapter remains in an      *
    *                      indefinite "Stopping Force" state       *
    *                      after the WMQ address space abends.     *
    ****************************************************************
    * RECOMMENDATION:                                              *
    ****************************************************************
    A WMQ Queue Manager address space abends S0C6, after which the
    CICS/WMQ adapter in a CICS region attached to the Queue Manager
    remains in a "Stopping Force" state indefinitely.
    Additional keywords: abends0c6 abend0c6 msgdfhmq0341 msgmq0341
    msg0341 mq0341
    

Problem conclusion

  • DFHMQTRU has been modified so that new tasks entering the system
    when the CICS/WMQ adapter is not available are prevented from
    being added to the LOT chain and thus halting the successful
    shutdown of the adapter.
    

Temporary fix

  •             *********
                * HIPER *
                *********
    FIX AVAILABLE BY PTF ONLY
    

Comments

  • ž**** PE13/05/22 FIX IN ERROR. SEE APAR PM88203  FOR DESCRIPTION
    

APAR Information

  • APAR number

    PM68981

  • Reported component name

    CICS TS Z/OS V4

  • Reported component ID

    5655S9700

  • Reported release

    600

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    YesHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2012-07-16

  • Closed date

    2012-10-24

  • Last modified date

    2013-06-18

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

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

    UK82953 UK82954

Modules/Macros

  •    DFHMQTRU
    

Fix information

  • Fixed component name

    CICS TS Z/OS V4

  • Fixed component ID

    5655S9700

Applicable component levels

  • R600 PSY UK82953

       UP12/11/03 P F211

  • R700 PSY UK82954

       UP12/11/03 P F211

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.

[{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSGMGV","label":"CICS Transaction Server"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"4.1","Edition":"","Line of Business":{"code":"LOB35","label":"Mainframe SW"}},{"Business Unit":{"code":"BU054","label":"Systems w\/TPS"},"Product":{"code":"SG19M","label":"APARs - z\/OS environment"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"4.1","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
18 June 2013