IBM SC34-7012-01 Defining recoverability for CICS-managed resources, Recovery for transactions

Chapter 12. Defining recoverability for CICS-managed resources

This section describes what to do to ensure that you can recover the resources controlled by CICS on behalf of your application programs.

About this task

It covers the recoverability aspects of the various resources as follows:

v“Recovery for transactions”

v“Recovery for files” on page 125

v“Recovery for intrapartition transient data” on page 131

v“Recovery for extrapartition transient data” on page 134

v“Recovery for temporary storage” on page 135

Recovery for transactions

CICS recoverability options are limited for local and remote transactions.

The following options apply for local transactions:

vAutomatic restart when a transaction abends

vTime-out when a transaction has been waiting for resources for longer than a specified time interval, usually because the transaction is deadlocked with another transaction for the same resources

vAllowing you to purge transactions that “hang” for reasons that cannot otherwise be resolved

For remote transactions, in addition to the above, CICS provides indoubt options to specify the recovery action in the event of indoubt failures.

Defining transaction recovery attributes

When defining user transactions that update local resources, you can specify the following options on the TRANSACTION resource definition for recovery purposes:

About this task

RESTART({NOYES})

This option defines whether, in some circumstances, the transaction is eligible to be restarted automatically by CICS under the control of the DFHREST user replaceable module.

The default is RESTART(NO).

DTIMOUT({NO1–6800})

If the task remains suspended (inactive) for the specified interval (and SPURGE(YES) is also specified), CICS initiates an abnormal termination of the task. (Specifying SPURGE(NO) overrides the DTIMOUT option.)

The default is DTIMOUT(NO).

© Copyright IBM Corp. 1982, 2010 123