IBM Support

OA69701: LOGICAL COPY GROUP GETS STUCK IN P-COMP IN SOME CASES ZDMF/K

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • If a Logical Copy group has completed and restarted it can get
    stuck in P-COMP. This can occur in multiple ways. Having lots of
    DEBUG flags set seems to prevent the issue.
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED: Users of IBM z/OS Data Set Mobility          *
    *                 Facility who run multi-system migration      *
    *                 groups defined with LOGICAL_COPY and         *
    *                 EARLY_DATA_SET_COMPLETION and re-activate    *
    *                 a group that has already completed.          *
    ****************************************************************
    * PROBLEM DESCRIPTION: A migration group can hang and never    *
    *                      reach the COMPLETE state after it is    *
    *                      re-activated.  The owning system stays  *
    *                      at CMP-ALLP and the other systems stay  *
    *                      at CMP-PND.                             *
    ****************************************************************
    * RECOMMENDATION: Apply the provided PTF.                      *
    ****************************************************************
    A system reaches a pending-terminate point that is externalized
    in one of two equivalent ways: CMP-ALLP, reached when a system
    completes its own processing, and CMP-PND, reached when a
    system that has no remaining processing follows another system
    that is already terminating.  Both mean the system is ready to
    terminate.  On a re-activate of an already completed
    LOGICAL_COPY group, a non-owning system that has no remaining
    work is driven to CMP-PND, while the owning system reaches
    CMP-ALLP.  The completion logic recognizes only CMP-ALLP as
    ready to terminate, so the owning system never issues the group
    termination and the non-owning system never completes.  The
    group is stuck and never reaches COMPLETE.
    

Problem conclusion

  • The completion logic is corrected to treat CMP-ALLP and
    CMP-PND as equivalent, in the two places that previously
    recognized only CMP-ALLP:
    * The owning system now counts systems in either CMP-ALLP or
       CMP-PND when deciding to terminate the group.
    * A non-owning system now completes to the terminated state
       from either CMP-ALLP or CMP-PND.
    With these changes the group reaches COMPLETE regardless of
    the order in which the systems reach the pending-terminate
    point.
    

Temporary fix

Comments

APAR Information

  • APAR number

    OA69701

  • Reported component name

    ZDMF

  • Reported component ID

    ZDMF00001

  • Reported release

    341

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2026-07-01

  • Closed date

    2026-07-15

  • Last modified date

    2026-08-03

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

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

    UJ10018

Modules/Macros

  • GZDDBHB
    

Fix information

  • Fixed component name

    ZDMF

  • Fixed component ID

    ZDMF00001

Applicable component levels

  • R341 PSY UJ10018

       UP26/07/21 P F607

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":"BU048","label":"IBM Software"},"Product":{"code":"SSRLWE","label":"IBM z\/OS Data Set Mobility Facility (zDMF)"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"341","Line of Business":{"code":"LOB69","label":"Storage TPS"}}]

Document Information

Modified date:
03 August 2026