Task 9: Set up the DB2 environment

 

You need to establish an environment in which WebSphere MQ can access and execute the DB2 plans that are used for queue-sharing groups.

The following steps must be performed for each new DB2 data-sharing group. All the sample JCL is in thlqual.SCSQPROC. (See Migrating queue-sharing groups to V6 for information on how to migrate an existing data-sharing group.)

  1. Customize and execute sample JCL CSQ45CSG to create the storage group that is to be used for the WebSphere MQ database, tablespaces, and tables.

  2. Customize and execute sample JCL CSQ45CDB to create the database to be used by all queue managers that will connect to this DB2 data-sharing group.

  3. Customize and execute sample JCL CSQ45CTS to create the tablespaces that will contain the queue manager and channel initiator tables used for queue-sharing groups (to be created in step 4).

  4. Customize and execute sample JCL CSQ45CTB to create the 12 DB2 tables and associated indexes. Do not change any of the row names or attributes.

  5. Customize and execute sample JCL CSQ45BPL to bind the DB2 plans for the queue manager, utilities, and channel initiator.

  6. Customize and execute sample JCL CSQ45GEX to grant execute authority to the respective plans for the user IDs that will be used by the queue manager, utilities, and channel initiator. The user IDs for the queue manager and channel initiator are the user IDs under which their started task procedures run. The user IDs for the utilities are the user IDs under which the batch jobs can be submitted. The names of the appropriate plans are:
    User Plans
    Queue manager CSQ5A600, CSQ5C600, CSQ5D600, CSQ5K600, CSQ5L600, CSQ5M600, CSQ5R600, CSQ5S600, CSQ5T600, CSQ5U600, CSQ5W600
    SDEFS function of the CSQUTIL batch utility CSQ52600
    CSQ5PQSG batch utility CSQ5B600
    CSQUZAP service utility CSQ5Z600

In the event of a failure during DB2 setup, the following jobs can be customized and executed:

Note:
If these jobs fail because of a DB2 locking problem it is probably due to contention for a DB2 resource, especially if the system is being heavily used. Resubmit the jobs later. It is preferable to run these jobs when the system is lightly used or quiesced.

See the DB2 for OS/390 Administration Guide for more information about setting up DB2.

See the WebSphere MQ for z/OS Concepts and Planning Guide for information about DB2 table sizes.