Restores for MongoDB

You can restore and recover the entire MongoDB cluster to its current location (in place) or to a different cluster (out of place). You can also restore only the data.

The restore operation does not require any manual steps. It automatically shuts down the destination MongoDB servers, deletes all data from the destination paths, restores the data to the servers in the destination clusters, initiates the replica sets, and then adds the secondary nodes to the corresponding replica set.

Use Cases for Full Restores

Use case	Type of restore	Restore process
Rebuild the source cluster	In place	Rebuild the source MongoDB cluster from the backup data by using the MongoDB nodes, data paths, and port numbers on the source.
Rebuild the source cluster Use this method in the following cases: Any of the existing nodes are unavailable because of a hardware failure. The data path volume is unavailable because of a disk failure.	Out of place	Rebuild the source MongoDB cluster from the backup data either by replacing with new nodes, or by changing the data paths or port numbers on an existing node.
Create a new MongoDB cluster Use this method to create a new MongoDB cluster for development and test environments. During the restore, you can specify the nodes, port numbers, and data paths to restore the data. Add the new MongoDB cluster in the Command Center, and specify it as the destination cluster for the restore operation.	Out of place	Create a new MongoDB cluster from the backup data when the source cluster is intact.
Restore to a different MongoDB cluster Use this method to restore the backup data to a different MongoDB cluster for development and test environments. In sharded clusters, the destination must have the same number of shards as the source.	Out of place	Restore the backup data to different MongoDB cluster that is configured and discovered by the Commvault software when the source cluster is intact.