If a node in a Windows cluster is corrupted or fails, you can recover the node.
Each cluster node must be backed up using an Image or Windows Agent Bare Metal Restore (BMR) job on the node. See Job B in Add backup jobs for a Windows cluster.
To recover a node in a Windows cluster:
1. On a replacement machine that has similar hardware to the original cluster node, launch the System Restore application.
For detailed procedures and information, see the System Restore User Guide.
2. Restore system volumes from a BMR backup of the original cluster node (Job B in Add backup jobs for a Windows cluster).
3. If required at the end of the restore, repair drivers on the system.
4. Reboot and then log in to the replacement node.
5. Configure network adaptors on the replacement node with the same network settings as the original node (i.e., same IP addresses and DNS entries).
6. If required, re-connect all cluster disks.
7. Open the Failover Cluster Manager and ensure that the restored node is up and running.
8. Fail over the SQL Server and SQL Server Agent cluster resources to the restored node to ensure that the restored node is functioning correctly.