Restoring from a disaster

If one side of the replicated IBM® Runbook Automation setup is lost, the other side stays operational. As the remaining side is no longer redundant, it is recommended to restore the second side as soon as possible.

If a single side experiences a temporary outage, or a network partition that is no longer than a few minutes, RBA resumes replication automatically. If the outage persists for longer, it is recommended to rerun the Replication configuration to ensure that replications are resumed in time.

If an entire cluster is lost and cannot be recovered, a new installation can be connected to the replication. It is important to ensure that the old cluster is not replicating data to the remaining side, and that the new side is running the same code version before resuming the replication. Furthermore, side-specific data as described in Replicated data needs to be manually added again.