Point-in-time (incremental) data recovery refers to recovery of data changes up to a given point in time. Typically, you perform this type of recovery after restoring a full backup (adabck dump/restore). The full backup brings the cluster node to its state at the time the backup was created. Point-in-time recovery then updates the node incrementally from the time of the the full backup to a more recent time.

In an Adabas cluster, you can recover files or data on top of the last backup via the Adabas data protection log (PLOG).

Each node in a cluster writes its own PLOG data. You can rebuild a node by restoring the backup of the primary node and applying the PLOGs created by each node. The backup run by the SST does not generate a PLOG switch.

Whenever the primary node starts a new PLOG, all other nodes also start a new PLOG. These PLOGs have different numbers. When a new PLOG is started on a secondary node by adaopr, adabck or via restart, the other nodes don’t start a new PLOG. As a result, the content of the PLOGs that were created on different nodes are not necessarily identical. Either the session number is not common to all nodes in a cluster or several PLOGs on a secondary correspond to one PLOG on the primary node.

When you add a node to a cluster, back up the primary node so that the backup copy matches the local PLOG.

For more information about the Adabas data protection log, see the Adabas for Linux and Cloud documentation.