IBM SC34-7012-01 manual Connection failure to a coupling facility cache structure

simultaneously all data sets in use when the lock structure fails, each data set can be restored to service individually as soon as all its sharing CICS regions have completed lost locks recovery.

Connection failure to a coupling facility cache structure

If connection to a coupling facility cache structure is lost, DFSMS™ attempts to rebuild the cache in a structure to which all the SMSVSAM servers have connectivity. If the rebuild is successful, the failure is transparent to CICS.

If DFSMS is unable to recover transparently from a connectivity failure to a coupling facility cache structure, CICS issues message DFHFC0163 (followed by DFHFC0158) on detecting the condition. The recovery process from this failure is the same as for a cache structure failure:

vErrors are returned to CICS when it attempts to access data sets bound to the cache, as a result of which:

–CICS attempts to back out the unit of work, fails, and shunts the unit of work so that backout can be retried later.

–CICS converts any locks held by the unit of work against records in the data set to retained locks.

–CICS is informed by the quiesce protocols when the connectivity is restored.

–CICS retries backout of the shunted units of work.

Connection failure to a coupling facility lock structure

If an SMSVSAM server loses connectivity to the coupling facility lock structure, and it is not possible to rebuild locks in another lock structure to which all the SMSVSAM servers in the sysplex have access, the SMSVSAM server abends and restarts itself.

CICS issues message DFHFC0153 when it detects that the server is not available for service.

The restarted SMSVSAM is not available for service until it is successfully connected to its coupling facility lock structure. When it does become available, recovery follows dynamic RLS restart in the same way as for any other server failure, because no lock information has been lost.

MVS system recovery and sysplex recovery

When an MVS image fails, all CICS regions and the SMSVSAM server in that image also fail. All RLS locks belonging to CICS regions in the image are retained by SMSVSAM servers in the other MVS systems.

When the MVS image restarts, recovery for all resources is through CICS emergency restart. If any CICS region completes emergency restart before the SMSVSAM server becomes available, it performs dynamic RLS restart as soon as the server is available.

The surviving MVS images should be affected by the failure only to the extent that more work is routed to them. Also, tasks that attempt to access records that are locked by CICS regions in the failed MVS image receive the LOCKED response.

If all the MVS images in a sysplex fail, the first SMSVSAM server to restart reconnects to the lock structure in the coupling facility and converts all the locks into retained locks for the whole sysplex.

Chapter 8. Unit of work recovery and abend processing 91