KBI 310059 Migrating Jobs To Additional Argent Queue Engines

Version

All

Date

30 Apr 2008

Summary

N/A

Technical Background

The Argent Job Scheduler product support up multiple Argent Queue Engines. If your production requirements change

and you need to add additional Argent Queue Engines, these new Argent Queue Engines can be added to any existing

jobs to achieve optimal load balancing.

If you have a large number of jobs, doing this using the Argent Job Control Panel GUI is inefficient.

This knowledge base article documents a SQL script that will allow existing jobs to be updated with the new

Argent Queue Engines being implemented.

The below SQL script assumes the following:

Argent Database Name = Argent_Job_Scheduler

Original Argent Queue Engine Name = QueueEngine1

Original Queue Being Used = *

New Argent Queue Engine Names = QueueEngine2, QueueEngine3, QueueEngine4

New Queue Being Used = *

SQL Script:

UPDATE [Argent_Job_Scheduler].[dbo].[AJS_JOB_TEMPLATE] SET queues=’QueueEngine1\*’ + CHAR(13) + CHAR(10) +

‘ QueueEngine2\*’ + CHAR(13) + CHAR(10) + ‘ QueueEngine3\*’ + CHAR(13) + CHAR(10) + ‘ QueueEngine4\*’ +

CHAR(13) + CHAR(10) WHERE queues LIKE ‘ QueueEngine1\*’


Note:  The above script will need to be modified for your environment to use appropriate

values for the Argent Queue Engine servers being used, the name of the actual Argent Job Scheduler database and

the value for the string specified for LIKE.

Steps:

  1. Before applying the above script, create a full SQL backup of

    the Argent Job Scheduler database.

  2. Run the SQL syntax:

    Select count (*) from AJS_JOB_TEMPLATE.

  3. Record the number of rows that were returned from step 2.
  4. Run the SQL script provided in this article to update the existing jobs in the AJS_JOB_TEMPLATE table.

    Record the number of rows that get updated.

  5. Compare the number of row that were updated in step 4 to the number of rows recorded in step 3.

    The number of rows that were updated in step 4, should match the number of rows recorded in step 3.

  6. If the rows that were updated do not match, determine where the differences are.

    One way to do this would be to use a simple SQL command that returns all of the rows in the

    AJS_JOB_TEMPLATE table so that you can scroll through all values to see what row(s) may not have been updated.

    The SQL syntax to do this would be:

    Select * from AJS_JOB_TEMPLATE

Resolution

N/A