Creating batch jobs to specify the log mark and log scan jobs


After performing the initial full migration of data from the production subsystem, perform the daily migration. This topic describes the daily migration.

For more information about performing an initial full migration, see Performance-of-an-initial-full-migration.

To perform the daily migration

  1. Define a BMC AMI Log Master batch job to create a log mark with a point of consistency (a quiesce point) for the table spaces involved. Give that point a log mark name of ENDBATCH.
  2. Run the batch job.
  3. Create a second (repeatable) BMC AMI Log Master batch job that contains the following job steps:
    • A step to define a log mark, create a quiesce point, and assign a log mark name to that point, also named ENDBATCH

      By repeating the same log mark name, you create two relative versions of the log mark. For more information about versions of log mark names, see the section about MARK logMarkName.

    • A step to define a log scan with the time frame of ENDBATCH(-1) to ENDBATCH(0)

      At this point, ENDBATCH(-1) refers to the log mark version that was created during the first job, and ENDBATCH(0) refers to the log mark version that was created during the first step of the current job. The log scan produces a MIGRATE SQL file and a Summary report. This output captures the data changes for migration.

  4. Run the second batch job one day after the initial full migration.

    Tip

    You can run the repeatable job daily to capture the data changes for migration to the decision-support database. The relative versions of the log marks always define a time frame from the log mark created during the previous run, to the log mark created at the beginning of the current run.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*