IBM Support

JR41540: OPERATIONS CONSOLE HAS PERFORMANCE AND SCALABILITY PROBLEMS WHEN, FOR EXAMPLE, THERE ARE VERY LARGE NUMBERS OF JOB RUNS

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • Operations Console, as shipped at 8.7 has some performance and
    scalability problems.
    
    If the Operations Console database contains a very large number
    of job runs, for example (c. 200,000), then response time from
    the UI is very slow and as the number of job runs grows it will
    eventually fail due to memory usage problems. This is true of
    both job runs and job log entries.
    
    The time for the UI to retrieve the service status is also very
    slow, irrespective of number of records in any table.
    
    
    The first problem occurs because all rows are returned to the
    client, before any selection or sorting is performed. The
    solution to this is to perform all selection and pagination on
    the server side via the database. Only the required rows are
    then returned to the client.
    
    The second problem occurs due to an internal design problem
    which means the status information is retrieved twice each time
    before it is displayed. The retrieval time can be halved by only
    retrieving the data once.
    

Local fix

Problem summary

  • ****************************************************************
    USERS AFFECTED:
    Users of Operations Console with high numbers of job runs.
    ****************************************************************
    PROBLEM DESCRIPTION:
    If the Operations Console database contains very large numbers
    of job runs, then the UI can have memory and performance issues,
    even when getting single pages of rows.
    ****************************************************************
    RECOMMENDATION:
    Apply patch or fix pack.
    ****************************************************************
    

Problem conclusion

  • Re-work selection of job run data so that only the a page worth
    at a time of data is returned to the client, when data is being
    paged.
    
    Re-work selection of job run data so that only the a page worth
    at a time of data is returned to the client, when data is being
    paged.
    
    IMPORTANT
    This fix changes the structure of the DSODB database.
    
    IF YOU HAVE CREATED A DSODB DATABASE AND HAVE NOT PREVIOUSLY
    INSTALLED THE OPERATIONS CONSOLE ROLLUP PATCH FOR 8.7:
    
    After installing the fix pack, there is a manual step that must
    be performed before the patch is usable. The patch
    requires an update to the operations database schema content,
    where this update adds some new database functions
    that are required by this patch.
    
    The instructions to follow are described in the readme.txt file
    in the path
          Server/DSODB/scripts/{database}/readme.txt
    where {database} is the database type for the DSODB database,
    for example DB2_LUW_9_7.
    Note: For DB2 v10, follow the instructions for DB2 v9.7
    
    Step A in the readme.txt must always be followed even if a DSODB
    database already exists, as this step will
    generate the scripts to either create a new database to include
    these additions, or to add the required schema
    additions to an existing DSODB database. When running Step A,
    the 'upgrade' option will only need to be specified
    if the DSODB database already exists. If a DSODB database
    already exists then the other arguments to the command in
    Step A must be the same as were specified when the DSODB
    database was created. The actions for Step B are dependant
    on whether a DSODB database already existed before installing
    this patch, and are as described in the readme.txt
    file.
    
    If a DSODB database already existed then after running this
    manual upgrade to the database, the AppWatcher process
    will need to be restarted.
    
    For Windows, this can be done from the Control Panel services
    menu or from a command window:
    
        net stop "DataStage AppWatcher Service"
        net start "DataStage AppWatcher Service"
    
    For Unix/Linux, this can be done by running the following
    command as the datastage administration user (e.g. dsadm)
    
      IBM/InformationServer/Server/DSODB/bin/DSAppWatcher.sh -stop
      IBM/InformationServer/Server/DSODB/bin/DSAppWatcher.sh -start
    
    If the Operations Console Rollup Patch for 8.7 has previously
    been installed, or if no DSODB database has been created, these
    instructions can be ignored.
    

Temporary fix

Comments

APAR Information

  • APAR number

    JR41540

  • Reported component name

    WIS DATASTAGE

  • Reported component ID

    5724Q36DS

  • Reported release

    870

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2011-12-22

  • Closed date

    2012-01-30

  • Last modified date

    2012-10-12

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

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

Modules/Macros

  • SERVER
    

Fix information

  • Fixed component name

    WIS DATASTAGE

  • Fixed component ID

    5724Q36DS

Applicable component levels

  • R870 PSY

       UP

[{"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SSVSEF","label":"InfoSphere DataStage"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"8.7","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Document Information

Modified date:
11 October 2021