IBM Support

PM72331: ABENDS0C4 IN DFSORP50 DUE TO THE AWE POINTED TO BY PSTORAWE IS INVALID. 13/01/24 PTF PECHANGE

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • IMS V12 HALDB users running OLR initiated
    using a type-2 command in an OM environment are receiving a S0C4
    abend in DFSORP50.  When DFSORP50 is called, PSTOLRW ischecked
    to determine if an ISWITCH TO=CTL,STACK=YES should bedone.
    When PSTOLRW is zero, the iswitch is not performed because
    DFSORP50 should have been called when already running in the
    CTL region.  When DFSORP50 is called after label P0XE00Z
    (error processing) in module DFSORP00, PSTOLRW is not cleared,
    so the ISWITCH TO=CTL,STACK=YES is incorrectly done at entry
    to DFSORP50 and ISWITCH TO=UNSTACK is done incorrectly at exit.
    The ISWITCH TO=UNSTACK causes a switch to the wrong address
    space, causing subsequent OLRs for the next partitions to fail
    because the PSTORAWE and DFSOCMD addresses are invalid in the
    DLISAS address space.
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All IMS V12 HALDB users running multiple     *
    *                 OLRs in parallel initiated using type-2      *
    *                 INIT OLR command in an OM environment with   *
    *                 apar PM57827 / ptf UK76743 applied.          *
    ****************************************************************
    * PROBLEM DESCRIPTION: When running multiple OLRs in parallel  *
    *                      initiated using type-2 INIT OLR command *
    *                      in an OM environment, an ABEND0C4 can   *
    *                      occur in module DFSORP50 because the    *
    *                      OCMD ( DFSOCMD ) address in R7 is       *
    *                      invalid.                                *
    ****************************************************************
    * RECOMMENDATION: INSTALL CORRECTIVE SERVICE FOR APAR/PTF      *
    ****************************************************************
    An ABENDS0C4 can occur in module DFSORP50 because the DFSOCMD
    address is invalid.  The release of the INIT OLR command AWE by
    DFSCST00 was handled incorrectly.  After the INIT OLR command
    processes, DFSCST00 releases the INIT OLR AWE using the EPLOC
    (entry point location) instead of the EPAD (entry point
    address).  The FREE= parameter on the DFSSQPP macro call in
    DFSORC00 to queue the AWE to DFSCST00 specified
    (IMODULE,=CL8'DFSORC00',241), but should have specified
    (IMODULE_EPAD,(Rxx),241), where Rxx is the register containing
    the length of the AWE storage to be released.
    EPLOC=CL8'DFSORC00' was used on the IMODULE GETMAIN when the AWE
    storage was gotten in DFSORC00, but if multiple INIT OLR
    commands are processed in parallel, using the EPLOC in the FREE=
    parameter on the DFSSQPP macro call to queue the AWE to DFSCST00
    can free the wrong AWE when an INIT OLR command completes,
    eventualy causing an ABEND0C4 if an AWE storage is released too
    soon and reused before the INIT OLR command associated with that
    AWE completes.
    
    Additional Symptoms: ABENDS602.
                         DFSISERW hang in DFSOCMD0 called from
                         DFSORP50.
    

Problem conclusion

  • GEN:
    KEYWORDS:
    
    *** END IMS KEYWORDS ***
    The following modules have been modified to correct the reported
    problem:
    
    ************
    * DFSORC00 *
    ************
    Code has been modified to use FREE=(IMODULE_EPAD,(R14),241) on
    the DFSSQPP macro call when the AWE is queued to DFSCST00. When
    the AWE release is done by DFSCST00, it will be released based
    on the entry point address (EPAD) instead of the entry point
    location (EPLOC).
    
    ************
    * DFSORP50 *
    ************
    Code has been modified to use DFSKFUNC FUNC=CHKDLAD to
    determine if an ISWITCH to the CTL region is needed.
    

Temporary fix

  • *********
    * HIPER *
    *********
    

Comments

APAR Information

  • APAR number

    PM72331

  • Reported component name

    IMS V12

  • Reported component ID

    5635A0300

  • Reported release

    200

  • Status

    CLOSED PER

  • PE

    YesPE

  • HIPER

    YesHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2012-09-06

  • Closed date

    2013-02-05

  • Last modified date

    2013-03-04

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

    PM71481

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

    UK91409

Modules/Macros

  •    DFSORC00 DFSORP50
    

Fix information

  • Fixed component name

    IMS V12

  • Fixed component ID

    5635A0300

Applicable component levels

  • R200 PSY UK91409

       UP13/02/14 P F302 «

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"},"Platform":[{"code":"PF054","label":"z Systems"}],"Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
14 December 2020