IBM Support

PI53551: MQ Z/OS V8: SMF 116 CLASS(3) DATA MAY BE INCORRECT FOR CICS AND CHIN THREAD CONTROL BLOCKS

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • You are finding SMF data indicating queue access for CICS
    tranids that do not access those queues. Those tranids do run
    in the same CICS regions, but they access different queue names.
    
    Other issues with validity of SMF116 data include:
    - 0 latency times (time on queue)
    - 0 pageset times and counts
    - 0 Persistent messages counts
    - 0 JWN (Journal writes)
    (when those values are expected to be non-zero)
    - application queue data saying the queue is on PSID(0) when
    you do not have application queues on PSID 0
    
    Counts appear valid at the queue level for other information
    such as  max qdepth, messages sizes, Gets/Puts....
    
    Tivoli Omegamon XE for Messaging and other monitoring tools are
    affected by this incorrect data.
    
    Additional Symptom(s) Search Keyword(s):
    WTID zero
    

Local fix

  • Issue the MQ command
      RECOVER QMGR(TUNE CICSFastDealloc OFF)
    (prefix it with "/cpf" without the quotes in SDSF, where "cpf"
    is the command prefix for the queue manager).
    
    This setting will not survive a queue manager restart, so
    include the command in a member in the CSQINP2 concatenation.
    
    Setting CICSFastDealloc OFF will disable a performance
    optimization introduced in MQ V8 which comes into play when
    CICS tasks and channels start and end.
    
    Once the command has been issued it will have immediate effect.
    Issuing the following command will revert back to the V8 way of
    doing things:
      RECOVER QMGR(TUNE CICSFastDealloc ON)
    
    The following command will show the current setting:
      RECOVER QMGR(TUNE CICSFastDealloc)
    

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All users of WebSphere MQ for z/OS Version 8 *
    *                 Release 0 Modification 0.                    *
    ****************************************************************
    * PROBLEM DESCRIPTION: SMF 116 accounting records generated by *
    *                      channel initiator and/or cics tasks     *
    *                      contain incorrect values in the WTID    *
    *                      control block, leading to an incorrect  *
    *                      channel name or transaction being       *
    *                      recorded against accounting data for    *
    *                      queues.                                 *
    ****************************************************************
    * RECOMMENDATION:                                              *
    ****************************************************************
    When CICS and chinit tasks end, a QRPL_END_THREAD request is
    processed by the queue manager. While processing this request
    the WTID and WTAS control blocks are passed to the SMF writing
    task CSQ7PLRT prematurely. When the end thread request
    completes, this causes a new WTID and WTAS to be allocated to
    the task prior to the ACE representing the task being added to a
    free chain.
    When a subsequent task in the same address space starts, the ACE
    is reused, and because it already has a WTID and WTAS associated
    with it, these control blocks are also reused. However they are
    not reinitialised, and continue to incorrectly represent the
    previous task that owned the ACE control block.
    This leads to SMF data for the new task being incorrectly
    recorded as having been generated by the earlier task, which
    could have been a different channel or transaction.
    

Problem conclusion

  • QRPL_END_THREAD processing is changed to pass the WTID/WTAS to
    CSQ7PLRT after all recording for the task is complete, ensuring
    that when the ACE is subsequently reused, a new WTID and WTAS
    block is allocated and initialised correctly for the new task.
    000Y
    CSQMCETH
    CSQMCPRH
    CSQ7ADAL
    CSQ7LEPL
    

Temporary fix

Comments

APAR Information

  • APAR number

    PI53551

  • Reported component name

    WMQ Z/OS 8

  • Reported component ID

    5655W9700

  • Reported release

    000

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2015-12-03

  • Closed date

    2016-03-17

  • Last modified date

    2016-05-04

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

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

    UI35846

Modules/Macros

  • CSQMCETH CSQMCPRH CSQ7ADAL CSQ7LEPL
    

Fix information

  • Fixed component name

    WMQ Z/OS 8

  • Fixed component ID

    5655W9700

Applicable component levels

  • R000 PSY UI35846

       UP16/04/15 P F604

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":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SSYHRD","label":"IBM MQ"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"8.0","Edition":"","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
04 May 2016