IBM SC34-7012-01 - Manual (Page 73)

Chapter 6. CICS emergency restart
This section describes the CICS startup processing specific to an emergency restart.
If you specify START=AUTO, CICS determines what type of start to perform using
information retrieved from the recovery manager’s control record in the global
catalog. If the type-of-restart indicator in the control record indicates “emergency
restart needed”, CICS performs an emergency restart.
See Chapter 5, “CICS warm restart,” on page 53 for the restart processing
performed if the type-of-restart indicates “warm start possible”.
Recovering after a CICS failure
CICS initialization for an emergency restart after a CICS failure is the same as
initialization for a warm restart, with some additional processing.
The additional processing performed for an emergency restart is mainly related to
the recovery of in-flight transactions. There are two aspects to the recovery
operation:
1.
Recovering information from the system log
2. Driving backout processing for in-flight units of work
Recovering information from the system log
At some point during initialization (and before CICS performs program list table
post-initialization (PLTPI) processing), the recovery manager scans the system log
backwards. CICS uses the information retrieved to restore the region to its state at
the time of the abnormal termination.
For non-RLS data sets and other recoverable resources, any locks (ENQUEUES)
that were held before the CICS failure are re-acquired during this initial phase.
For data sets accessed in RLS mode, the locks that were held by SMSVSAM for
in-flight tasks are converted into retained locks at the point of abnormal
termination.
Driving backout processing for in-flight units of work
When initialization is almost complete, and after the completion of PLTPI
processing, the recovery manager starts backout processing for any units of work
that were in-flight at the time of the failure of the previous run.
Starting recovery processing at the end of initialization means that it occurs
concurrently with new work.
Concurrent processing of new work and backout
The backout of units of work that occurs after an emergency restart is the same
process as dynamic backout of a failed transaction. Backing out in-flight
transactions continues after “control is given to CICS”, which means that the
process takes place concurrently with new work arriving in the region.
© Copyright IBM Corp. 1982, 2012 61