Copy Plus V11.2 SYSTEM ABEND CODE = 04E, BMC30594I HPGCLRSN RC = 0013E000

Version 4
    Share This:

    This document contains official content from the BMC Software Knowledge Base. It is automatically updated when the knowledge article is modified.


    PRODUCT:

    COPY PLUS for DB2


    COMPONENT:

    COPY PLUS for DB2



    PROBLEM:

    Copy Plus V11.2 various problems when updating the CLRSN field

    From the manual:
     
    For table spaces, COPY PLUS resets the COPYUPDATEDPAGES column every time
    a copy is made and HPGCLRSN is updated, except for CUMULATIVE YES
    incremental copies. For CUMULATIVE YES incremental copies,
    COPYUPDATEDPAGES represents the total number of pages updated since the last
    full copy. For index spaces, COPY PLUS resets the COPYUPDATEDPAGES column
    every time a full copy is made and HPGCLRSN is updated. HPGCLRSN is updated
    for all spaces when RESETMOD YES is specified. HPGCLRSN is only updated for
    large spaces when RESETMOD NO is specified.

    The HPGCLRSN field is updated when copying a tablespace that is greater than 200M.

    If a customer asks, why did we start updating HPGCLRSN, it is so that RTS statistics will be more accurate.
     
    The COPYUPDATEDPAGES column in RTS is supposed to be the number of distinct pages changed since the last copy.  In previous releases, since we did not update HPGCLRSN, COPYUPDATEDPAGES was really the number of distinct pages changed since the customer ran an IBM COPY or a SHRLEVEL REFERENCE RESETMOD YES copy.  Since many customers never use IBM COPY or RESETMOD YES, COPYUPDATEDPAGES could be very inaccurate.
     
    As more customers start using RTS to feed automation, it is becoming more important for RTS to be accurate.  Also, in 11.2 we started using RTS for CHANGELIMIT and FULL AUTO.  If the statistics aren’t accurate, CHANGELIMIT and FULL AUTO may make copies more often than necessary.  In previous releases, we used mod bits for these functions, but they weren’t very accurate if people always used RESETMOD NO.
     
    We also must update HPGCLRSN for SHRLEVEL CHANGE RESETMOD YES copies, because if you don’t update the HPGCLRSN, DB2 won’t turn on the mod bits when pages are updated, and if you reset the mod bits and don’t turn them on for new updates, you will get incorrect incremental copies.
     
    In ACP V11.2.00 the ACPAGENT is now also required when making SHRLEVEL REFERENCE copies in a data sharing system in order to connect to the remove DB2's when updating the CLRSN field.
     


    SOLUTION:

    BPU8387
    BMC30231E DSNUTILB TASK NUMBER 1 ABNORMALLY TERMINATED WITH
    SYSTEM ABEND CODE = 0004E000
    LOCK CONTENTION BETWEEN COPY PLUS AND LONG RUNNING TRANSACTIONS
    HOLDING MASS DELETE LOCK
    QM001887471 – Copy Plus V11.2 SYSTEM=0C4 USER=0000 REASON=00000010 while updating CLRSN
     
    BPU8570 - QM001890847
    Copy Plus abend SYSTEM COMPLETION CODE=04E  REASON CODE=00E40070 with BPU8387 applied      
    BMC30531E UNABLE TO OPEN DATASET DDNAME = SYS00003
    BMC30536E UNABLE TO READ JFCB, DDNAME = SYS00003, RC = 4
    BMC47307E ERROR UNALLOCATING DATA SET SYS00003, ECODE=0438, ICODE=0000
     
    It turns out when DSNUTILB finishes, it unallocates all dynamically allocated data sets.  We were using the same dummy output data set for all tasks that called DSNUTILB to update the header page.  When one DSNUTILB terminated, if another was running, it had it's SYSPRINT pulled out from under it, causing DSNUTILB to abend.
     
    BPU8583
    BMC30138E UNEXPECTED ERROR = 8, FROM MODULE DSNUTILB
    IKJ56220I MAXIMUM NUMBER OF DATA SET ALLOCATIONS ALLOWED BY YOUR
       SESSION HAS BEEN REACHED,
    QM001891133 - BMC30138E UNEXPECTED ERROR = 8, FROM MODULE DSNUTILB
    QM001891137 - IKJ56220I MAXIMUM NUMBER OF DATA SET ALLOCATIONS ALLOWED
     
    BPU8612
    DSN3201I  *DJV3 ABNORMAL EOT IN PROGRESS FOR  259
        BMC30594I HPGCLRSN RC = 0013E000
    QM001894897 - Reduce contention when updating HPGCLRSN (Changed lock mode for repair to share.)
     
    BPU8702
    BMC30594I HPGCLRSN RC = 0013E000 when MAKING DSNUM ALL COPIES OF PARTITIONED SPACES WITH A
    LARGE NUMBER OF PARTITIONS
    QM001921336 - ABND=04E-00E40085
    Updating the HPGCLRSN for every partition each time we start a new partition at once takes a long time and is causing us to time out the REPAIR.  It is also a performance problem because we are doing lots of unnecessary updates.
     
    BPU8791
    QM001949238- Make UPDATECLRSN=NO work with SHRLEVEL REFERENCE
     
    BPU8803
    SYSTEM COMPLETION CODE=0C4  REASON CODE=00000011
        In ACPMAIN OFFSET=001F5
    QM001955913 – S0C4-11 in ACPMAIN OFFSET=001F5
     
    BPU8878
    BMC30594I HPGCLRSN RC = 0013E000
    DSNV086E :P012 DB2 ABNORMAL TERMINATION REASON=00F30429
    ACPC0002 TASK NUMBER 3 ABNORMALLY TERMINATED SYSTEM ABEND CODE = 0C4
    ACPC0002 TASK NUMBER 1 ABNORMALLY TERMINATED SYSTEM ABEND CODE = 0D5
    SYSTEM COMPLETION CODE=0C4  REASON CODE=00000004
    IEF402I B485RI22 FAILED IN ADDRESS SPACE 0146 468  SYSTEM ABEND S55F -
     
     
     
     


    Article Number:

    000113146


    Article Type:

    Solutions to a Product Problem



      Looking for additional information?    Search BMC Support  or  Browse Knowledge Articles