Restarting CICS after a system log failure
When you restart CICS with START=AUTO after a failure following DFHLG0736, DFHLG0738, or DFHLG0740, CICS initializes for a diagnostic run only.
About this task
On a diagnostic run, CICS produces a dump of the CICS region state, retrieved from the CICS system log and then terminates. On a diagnostic run, CICS performs no recovery work and no new work. This situation persists until you start the region with an initial start.
A diagnostic run produces
diagnostics for investigation by IBM Service. For example, if DFHLG0772 reported
return code 8 and reason code IxgRsnCodeNoBlock (00000804),
but the associated system dump is lost, the diagnostic run reproduces the
dump (assuming the 00000804 condition is a solid failure).
For information about the AUTODIAG type-of-start override record, see Specifying parameters for DFHRMUTL. For more details of a diagnostic run, see Dealing with a corrupt system log.
When you have obtained the required diagnostics and are ready to restart the region with the broken system log, you can do so only with an initial start. You can do this either by running the DFHRMUTL utility with the SET_AUTO_START=AUTOINIT parameter, or by specifying START=INITIAL as a system initialization parameter.
Procedure
- Run the DFHRMUTL utility with the SET_AUTO_START=AUTOINIT parameter.
- Alternatively, specify the START=INITIAL as a system initialization parameter.