IBM SC34-7012-01 manual Forward recovery, No recovery

Physically recoverable TD queue (before failure)

Item 1

Item 2

Item 3

Next READ pointer

Item 4

CICS abends, causing UOW to fail in-flight

Next WRITE pointer

State of physically recoverable TD queue (after emergency restart)

Item 1

Item 2

Next READ pointer

Item 3

Item 4

CICS abends, causing UOW to fail in-flight

Next WRITE pointer

Figure 12. Illustration of recovery of a physically recoverable TD queue

Making intrapartition TD physically recoverable can be useful in the case of some CICS queues. For example, after a CICS failure, you might choose to restart CICS as quickly as possible, and then look for the cause of the failure. By specifying queues such as CSMT as intrapartition and physically recoverable, the messages produced just before the failure can be recovered and are therefore available to help you diagnose the problem.

No recovery

Recovery is not performed if you specify NO on the recovery attribute of an intrapartition transient data definition.

Forward recovery

CICS does not provide forward recovery support for transient data.

If you want forward recovery of intrapartition transient data, provide application programs to record the changes made to the contents of your intrapartition transient data queues while CICS is running. Changes are recorded in a user journal. The information journaled must include:

vEach WRITEQ, including the data that is written

vEach READQ

vEach DELETEQ of a queue

vFor logically recoverable queues, each backout, syncpoint, or syncpoint rollback

You must provide the application program to rebuild the data by reading the journaled information and applying that information to the transient data queue. Your application program could run in the program list table (PLT) phase or after emergency restart. Until the data set is fully recovered, do not write to the queue, because that would probably result in wrongly-ordered data, and a read might not provide valid data (or any data at all). For these reasons, running the recovery program in the PLT phase is probably preferable to running it after the restart.

If you do not have such a recovery strategy and you perform a cold start with a corrupted intrapartition data set, you will lose the contents of the intrapartition data set. You also lose the contents of the intrapartition data set if you specify TDINTRA=EMPTY as a system initialization parameter.

Chapter 12. Defining recoverability for CICS-managed resources 133