IBM Support

PK62939: NO ABENDAICA DURING LOOP IN APPLICATION PROGRAM. THIS HAPPENS AFTER AN AFCY CAUSED BY REACHING DTIMOUT DURING FILE OPEN HANG.

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • An application program is looping getting an INVREQ on a
    STARTBR over and over but the task never abends with AICA as it
    should.  Earlier in the life of the looping task, before it
    entered the loop, the task handled an abendAFCY that happened
    because of being purged by DTIMOUT during an implicit file open
    that was hanging.
       The Kernel Error Table entries for the looping task look
    like this:
    ERR_NUM ERR_TIME ERROR TYPE           ERR_CODE MODULE  OFFSET
    ======= ======== ==========           ======== ======  ======
    000000D 07:31:10 TRAN_ABEND_PERCOLATE ---/AKCS DFHPCP  0005E0
    000000E 07:31:10 TRAN_ABEND_PERCOLATE ---/AKCS DFHXCP  000464
    000000F 07:31:10 TRAN_ABEND_PERCOLATE ---/AKCS -noheda 01BB48
    0000010 07:31:10 TRAN_ABEND_PERCOLATE ---/AKCS DFHFCFR 001B26
    0000017 07:33:20 TRAN_ABEND_PERCOLATE ---/AFCY DFHPCP  0005E0
    0000018 07:33:20 TRAN_ABEND_PERCOLATE ---/AFCY DFHEIFC 001104
    0000019 07:33:20 TRAN_ABEND_PERCOLATE ---/AFCY DFHEIP  00089C
    .
    Here is a slightly different example of Kernel Error entries for
    a task that lost its runaway task protection:
    ERR_NUM ERR_TIME ERROR TYPE           ERR_CODE MODULE  OFFSET
    ======= ======== ==========           ======== ======  ======
    0000001 17:54:20 TRAN_ABEND_PERCOLATE ---/AKCS DFHPCP  0005E0
    0000002 17:54:20 TRAN_ABEND_PERCOLATE ---/AKCS DFHXCP  000464
    0000003 17:54:20 TRAN_ABEND_PERCOLATE ---/AKCS DFHFCFS 0016D8
    0000004 17:54:20 TRAN_ABEND_PERCOLATE ---/AFCY DFHPCP  0005E0
    0000005 17:54:20 TRAN_ABEND_PERCOLATE ---/AFCY DFHEIFC 001104
    0000006 17:54:20 TRAN_ABEND_PERCOLATE ---/AFCY DFHEIP  0008A4
    .
    The TASENTRY for the looping task shows that
    TAS_RUNAWAY_STOPPED is set.
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All.                                         *
    ****************************************************************
    * PROBLEM DESCRIPTION: Application fails to timeout with abend *
    *                      AICA.                                   *
    ****************************************************************
    * RECOMMENDATION:                                              *
    ****************************************************************
    An application program that had a HANDLE ABEND statement has
    handled an abend due to a file control request. It then
    entered an invalid loop and the task is never abended via AICA
    as expected.
    
    In the customer's case the application abended with AKCS in
    DFHFCFS during an implicit open request. The initial abend
    processing turned off runaway task detection while also turning
    on TCZABREC to reflect an abend was in progress. DFHFCFS and
    DFHFCFR both transform the abendAKCS into a purged response,
    they then fail to reestablish runaway task detection at this
    point. The purge response was passed back to the EI layer. At
    this point the AKCS was converted into an ABENDAFCY, where
    runaway task detection was turned off again. This abend was
    processed by the EI layer, which issued a start_runaway_timer
    request, but because the number of stops was greater than the
    number of starts, runaway task detection was still turned off.
    The application then entered a loop but CICS was unable to abend
    it AICA.
    
    Keywords: AbendAICA TCAABREC
    

Problem conclusion

  • DFHFCFS has been updated so that if an AKCS abend occurs and
    TCZABREC is on during recovery processing, the runaway timer is
    started again before TCZABREC is turned off. This will ensure
    that the stop and start requests of the runaway timer are always
    in sync.
    
    This problem can also arise in DFHFCFR's recovery routine if
    this intercepts an AKCS abend during a file control request. So
    this recovery routine has also been amended.
    

Temporary fix

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

Comments

APAR Information

  • APAR number

    PK62939

  • Reported component name

    CICSTS V3 Z/OS

  • Reported component ID

    5655M1500

  • Reported release

    400

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    YesHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2008-03-18

  • Closed date

    2008-05-15

  • Last modified date

    2008-06-02

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

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

    UK36500 UK36501

Modules/Macros

  •    DFHFCFR  DFHFCFS
    

Fix information

  • Fixed component name

    CICSTS V3 Z/OS

  • Fixed component ID

    5655M1500

Applicable component levels

  • R400 PSY UK36500

       UP08/05/20 P F805

  • R500 PSY UK36501

       UP08/05/20 P F805

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":"3.1","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":"3.1","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
02 June 2008