IBM SC34-7012-01 Unit of work recovery and abend processing, In-flight-failed, Commit-failed

Chapter 8. Unit of work recovery and abend processing

A number of different events can cause the abnormal termination of transactions in CICS.

These events include:

vA transaction ABEND request issued by a CICS management module.

vA program check or operating system abend (this is trapped by CICS and converted into an ASRA or ASRB transaction abend).

vAn ABEND request issued by a user application program.

vA CEMT, or EXEC CICS, command such as SET TASK PURGE or FORCEPURGE.

Note: Unlike the EXEC CICS ABEND command above, these EXEC CICS commands cause other tasks to abend, not the one issuing the command.

vA transaction abend request issued by DFHZNEP or DFHTEP following a communication error. This includes the abnormal termination of a remote CICS during processing of in-flight distributed UOWs on the local CICS.

vAn abnormal termination of CICS, in which all in-flight transactions are effectively abended as a result of the CICS region failing.

In-flight transactions are recovered during a subsequent emergency restart to enable CICS to complete the necessary backout of recoverable resources, which is performed in the same way as if the task abended while CICS was running.

Unit of work recovery

A unit of work in CICS is also the unit of recovery - that is, it is the atomic component of the transaction in which any changes made either must all be committed, or must all be backed out.

A transaction can be composed of a single unit of work or multiple units of work. In CICS, recovery is managed at the unit of work level.

For recovery purposes, CICS recovery manager is concerned only with the units of work that have not yet completed a syncpoint because of some failure. This topic discusses how CICS handles these failed units of work.

The CICS recovery manager has to manage the recovery of the following types of unit of work failure:

In-flight-failed

The transaction fails before the current unit of work reaches a syncpoint, as a result either of a task abend, or the abnormal termination of CICS. The transaction is abnormally terminated, and recovery manager initiates backout of any changes made by the unit of work.

See “Transaction backout” on page 74.

Commit-failed

A unit of work fails during commit processing while taking a syncpoint. A partial copy of the unit of work is shunted to await retry of the commit process when the problem is resolved.

This does not cause the transaction to terminate abnormally.

© Copyright IBM Corp. 1982, 2010 73