IBM Support

PI62507: CURRENTLY PRF DOESN?T RETRY DLI JOBS IF THEY ABENDED, REGARDLESSOF AUTOBKO VALUE AND WHETHER BACKOUT IS REQUIRED OR NOT

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • Currently PRF doesn?t retry DLI jobs if they abended, regardless
    of AUTOBKO value and whether backout is required or not.
    However, if BBO was disabled via a DD, abended DLI job is
    automatically retried.
    Help panel IRTZ8000 says that only BMPs should be retried.
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All Users of Program Restart Facility V2R2.  *
    ****************************************************************
    * PROBLEM DESCRIPTION: PRF-controlled DLI/DBB jobs should not  *
    *                      normally be allowed to go through       *
    *                      reattach processing because this would  *
    *                      result in invalid recovery logs. In the *
    *                      one case where PRF-controlled DLI/DBB   *
    *                      jobs are allowed to go through reattach *
    *                      processing, PRF does not perform the    *
    *                      reattach processing unless batch        *
    *                      backout functionality has been          *
    *                      disabled. If batch backout              *
    *                      functionality is disabled, PRF will     *
    *                      ignore the CMPCBKOK and CMPCBKER        *
    *                      specifications.                         *
    ****************************************************************
    * RECOMMENDATION:                                              *
    ****************************************************************
    When batch backout processing is disabled in a PRF-controlled
    DLI/DBB job, and abend retry is active for the job, PRF
    re-attach processing can take place if the PRF job abends with
    one of the abend codes that exist in the abend retry table for
    the job. Re-attach processing should never occur for a DLI/DBB
    job unless RDORETRY processing is active for the job. This is
    because the current architecture of PRF only allows for a single
    allocation of the IMS logs for the job step. Any attempt to
    perform re-attach processing would result in the re-attached
    process overlaying the IMS log from the previous process. In the
    one case where re-attach processing in a DLI/DBB environment can
    safely take place, PRF does not allow re-attach processing to
    occur unless batch backout processing has been explicitly
    disabled for the job step.
    
    When PRF is completely disabled by the EXCLUDE=YES parameter, or
    by the presence of a DD name from the exclusion DD name table,
    messages are issued so that the user can verify the presence of
    PRF and why the job was excluded from PRF processing. In the
    case where the IRT$IGNR DD statement is present in the job step,
    no indication is made that PRF is present, or that PRF was
    disabled.
    
    When batch backout processing is disabled in a PRF-controlled
    DLI/DBB job, PRF will ignore the job's CMPCBKOK/CMPCBKER
    specifications on abend, and will cause the job to abend with
    the originating abend code instead.
    

Problem conclusion

  • The Program Restart Facility restart logic has been modified
    to not allow DLI/DBB jobs to go through re-attach processing
    unless RDORETRY processing is active for the job.
    
    When the IRT$IGNR DD statement is present in a job step, an
    IRT316I message will be generated to verify that PRF was
    present, and that it was inactivated for the job step.
    
    PRF will no longer ignore the CMPCBKOK/CMPCBKER specifications
    in a DLI/DBB job unless the following situation arises: If a job
    abends in such a way that ESTAE recovery routines cannot attempt
    retry processing, an attempt to change the originating abend
    code would fail. This situation can arise, for instance, in the
    case of an OS CANCEL of the job in question.
    

Temporary fix

Comments

APAR Information

  • APAR number

    PI62507

  • Reported component name

    IMS PGM RESTART

  • Reported component ID

    5655E1400

  • Reported release

    220

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2016-05-16

  • Closed date

    2016-06-22

  • Last modified date

    2016-07-04

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

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

    UI38848

Modules/Macros

  • IRTOPG1  IRTOPJ1  IRTPRE00 IRTRCC00 IRTUPX00
    IRTZF113
    

Fix information

  • Fixed component name

    IMS PGM RESTART

  • Fixed component ID

    5655E1400

Applicable component levels

  • R220 PSY UI38848

       UP16/06/24 P F606

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":"BU054","label":"Systems w\/TPS"},"Product":{"code":"SG19M","label":"APARs - z\/OS environment"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"220","Edition":"","Line of Business":{"code":"","label":""}},{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSAVHR","label":"IMS Program Restart Facility for z\/OS"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"220","Edition":"","Line of Business":{"code":"LOB35","label":"Mainframe SW"}}]

Document Information

Modified date:
24 January 2022