IBM Support

PK21134: ABEND AD3K OCCURS DUE TO RECOVERY BACKOUT FAILING AFTER TASK IS PURGED

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • Customer has a THREADSAFE user program the performs about
    50000 DB2 SELECT request.  They then purge the transaction
    which results in an AD2R (normal) or AKC3 (deferred) abend.
    They also receive an AD3K abend which results in UOWs in a
    WAITING waitstate WAITRMI waitcause.  The AD3K abend is due
    to the purge of the CICS-DB2 task and results due to
    recovery backout failing for this task.  These UOWs are
    recovered only during the next CICS-DB2 reconnect.
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All.                                         *
    ****************************************************************
    * PROBLEM DESCRIPTION: A CICS-DB2 transaction suffers an abend *
    *                      AD3K and is left in a RMI WAIT state.   *
    ****************************************************************
    * RECOMMENDATION:                                              *
    ****************************************************************
    If a CICS-DB2 transaction is purged, it is possible, depending
    where in its processing it is, to suffer an abendAD3K and then
    the task is left in a RMI WAIT. This situation arises because
    when the task is purged the task enters DFHD2EX1's recovery
    routine on the QR TCB. The task then enters DFHERM's recovery
    routine and then attempts to BACKOUT the task. DFHERM calls
    DFHD2EX1 for BACKOUT, this drives DFHD2D2. DFHD2D2 calls DB2
    with a CSUB address from a previous SQL call, but because the
    task was purged before this CSUB could be associated to a TCB,
    DB2 gives a FRB reason of 00F30018, TCB is not connected, and so
    the backout fails.
    

Problem conclusion

  • DFHD2EX1's recovery routine has been changed to turn off
    interest in syncpoint for the failing task when the TCB
    has terminated abnormally.
    

Temporary fix

  • FIX AVAILABLE BY PTF ONLY
    

Comments

  • ž**** PE06/11/20 FIX IN ERROR. SEE APAR PK31859  FOR DESCRIPTION
    

APAR Information

  • APAR number

    PK21134

  • Reported component name

    CICSTS 3.1 Z/OS

  • Reported component ID

    5655M1500

  • Reported release

    400

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2006-03-10

  • Closed date

    2006-05-10

  • Last modified date

    2006-12-05

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

    PK17180

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

    UK14394

Modules/Macros

  •    DESD2EX1 DFHD2EX1
    

Fix information

  • Fixed component name

    CICSTS 3.1 Z/OS

  • Fixed component ID

    5655M1500

Applicable component levels

  • R400 PSY UK14394

       UP06/05/16 P F605

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:
05 December 2006