IBM Support

PI82129: DR SITE PREPARATION TAKES TOO LONG

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • DR SITE PREPARARION TAKES TOO LONG
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED:                                              *
    * All IMS 15 DRF users that take image copies along with the   *
    * recovery.                                                    *
    ****************************************************************
    * PROBLEM DESCRIPTION:                                         *
    * Performance improvement for recovery of large number of data *
    * sets using DRF along with taking image copies.               *
    ****************************************************************
    * RECOMMENDATION:                                              *
    * INSTALL CORRECTIVE SERVICE FOR APAR/PTF                      *
    ****************************************************************
    The recording of image copies during the DRF process takes too
    long.
    

Problem conclusion

  • When image copies are taken as part of the DRF process, all ICs
    will now be recorded in bulk as part of the end of job process
    instead of recording them separately by HPIC.
    
    DFSBRLSB
    --------
    Field BRLBRWCC is added and will contain the address of a list
    of image copies, mapped by DFSRWCC, to be recorded when the
    recovery completes.
    
    DFSRWCC
    -------
    rwcc_noclean - caller sets this flag to indicate that they do
    not want DBRC to perform IC cleanup when each IC is recorded for
    the recovery.
    
    Flags are added indicating the various types of ICs that can be
    recorded by the recovery process
    
    rwcc_entry_dual_dbname - DB name of the DBDS
    rwcc_entry_dual_ddname - DD name of the DBDS
    rwcc_entry_dual_ictime - IC run time
    rwcc_entry_dual_cictime - IC stop time for non-batch ICs
    rwcc_entry_icrcdcnt fix - IC record count
    
    rwcc_ic_err1 - flag indicating IC1 in error
    rwcc_ic_err2 - flag indicating IC2 in error
    
    DFSRWCR
    -------
    Added flag rwcr_NOTIFY_IC which is set by DBRC during the
    Recovery Start call indicating that this new support is present
    for recording all ICs in a batch at the end of Recovery
    processing.
    
    Field rwcr_cagrpname is set by DBRC on the Recovery Add call.
    This is the name of the Change Accumulation group in which the
    DBDS/Area belongs.
    
    DSPCCB
    ------
    Flag ccbnocln is set from flag rwcc_noclean during the Recovery
    End processing when the batch of ICs is recorded, indicating
    whether DBRC should bypass IC cleanup.
    
    DSPRWA00
    --------
    For Recovery Add processing, internal field RWA_CAGRPNM is set
    with the CA group name in which the DBDS/Area belongs.  This
    field is copied into rwcr_cagrpname when the RWCR is built for
    the caller.
    
    DSPRWV00
    --------
    Set flag rwcr_NOTIFY_IC during the Recovery Start call
    indicating that this new support is present for recording all
    ICs in a batch at the end of Recovery processing.
    
    All updates must be in the same commit scope for parallel
    access. A logical open must be done before starting the loop
    recording the batch of ICs.
    
    A logical open is done inside DSPURT00.  To avoid multiple
    enqueues (one per URT00 call), this logical open is also done
    for serial access.
    
    If BRLBRWCC is non-zero, the caller is indicating that they
    passed a batch of ICs to be recorded at the end of recovery
    processing.  DSPURT00 is called processing an internal NOTIFY.IC
    call for each IC that is to be recorded.  Before the call to
    DSPURT00, flag ccbnocln is set from flag rwcc_noclean indicating
    whether DSPURT00 should bypass IC cleanup.
    
    DSPRWV40
    --------
    This module is called during Recovery Start to build the list of
    ICs to be used by the recovery process.  This was changed to
    also return the number of records for the IC in
    rwcc_entry_icrcdcnt.
    
    DSPURT00
    --------
    Flag ccbnocln is set by internal NOTIFY.IC command callers that
    wish to bypass IC cleanup processing for performance reasons.
    DSPURT00 is changed to check this flag and bypass the call to
    DSPURCMH if indicated.
    

Temporary fix

Comments

APAR Information

  • APAR number

    PI82129

  • Reported component name

    IMS V15

  • Reported component ID

    5635A0600

  • Reported release

    500

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2017-05-24

  • Closed date

    2018-05-29

  • Last modified date

    2018-07-02

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

    PI73990

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

    UI56161

Fix information

  • Fixed component name

    IMS V15

  • Fixed component ID

    5635A0600

Applicable component levels

  • R500 PSY UI56161

       UP18/06/09 P F806

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":"SSEPH2","label":"IMS"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"15","Line of Business":{"code":"LOB35","label":"Mainframe SW"}}]

Document Information

Modified date:
22 December 2023