CA IDMS DDS System Operations › Backup and Recovery

Backup and Recovery

Each node has available the full range of DC/UCF backup and recovery facilities (journaling, warmstart, and rollback). In the event of system or program failure, each node is responsible for the recovery of the database(s) under its control, including databases that have been accessed on behalf of other nodes.

The action taken by CA IDMS DDS in the event of system or application program failure varies, depending on whether the failure is a system or program failure:

• System failure—When a node is disconnected, either because the system crashed or a connection failed, the node is responsible for recovery of all active request units under its control. Additionally, if the node provides a path for other nodes, that path is no longer available. If alternative paths are available, CA IDMS DDS automatically selects the best alternative path and continues processing. Restart requires only that the failed node be restarted; when restart occurs, communication between all nodes in the network is automatically reestablished.
• Program failure—When an application program fails, recovery takes place in exactly the same way that it does in a standard DC/UCF system. All database recovery is handled using standard CA IDMS recovery features, and the applications are restarted as appropriate.

Note: For a complete discussion of CA IDMS backup and recovery facilities, see the CA IDMS System Operations Guide.

The following illustrations show how database recovery is handled in the CA IDMS DDS environment following an application program failure, and a system failure in the same CA IDMS DDS configuration.

Recovery after application program failure

Recovery after system failure