Move Jobs from a Lost Backup Daemon to another Backup Daemon¶
If the server running a Backup Daemon fails, and if
you run multiple Backup Daemons, then an administrator
global owner or
global backup admin role can move all the daemon’s jobs to another
Backup Daemon. The new daemon takes over the responsibility to back
up the associated shards and replica sets.
When you move jobs, the destination daemon reconstructs the data using a snapshot and the oplog from the Backup Data Storage. Reconstruction of data takes time, depending on the size of the databases on the source.
During the time it takes to reconstruct the data and reassign the backups to the new Backup Daemon:
- Ops Manager does not take new snapshots of the jobs that are moving until the move is complete. Jobs that are not moving are not affected.
- Ops Manager does save incoming oplog data. Once the jobs are on the new Backup Daemon’s server, Ops Manager takes the missed snapshots at the regular snapshot intervals.
- Restores of previous snapshots are still available.
- Ops Manager can produce restore artifacts using existing snapshots with point-in-time restore for replica sets or checkpoints for sharded clusters.
With administrative privileges, you can move jobs between Backup daemons using the following procedure:
Move the jobs to the new daemon.¶
Select the destination daemon and click the Move all heads button.