12.2.2 Procedure for Restoring Both Nodes

This section describes the procedure to restore both nodes.

The procedure depends on a period when the backup is acquired in each node in the cluster system.

If a period when the backup is acquired by using the backup related services depends on a node

Stop all nodes in the cluster system.
Restore one node and start it.
For the node restoration, refer to the official NIFCLOUD documentation.
Stop RMS.
```
# hvshut -l
```
Restore and start the other node.
When using the mirroring among servers of GDS, perform the following procedure.
1. Check the status of the source slice for the synchronization copying.
  If the source slice for the synchronization copying is INVALID, restore the status of the slice.
  For the -d option of the sdxfix command, specify the source disk of the synchronization copying.
  Perform this procedure on either node.
```
# sdxfix -V -c <class name> -v <volume name> -d <disk name> -x NoRdchk
```
  If multiple volumes exist, repeat this step for the number of volumes.
  Note
  If multiple volumes belong to the same netmirror group, specify the same <disk name> for the command.
2. Restore the disk.
  Perform this step on any one node.
  If multiple netmirror groups exist, execute the following commands for the number of groups.
```
# sdxswap -O -c <class name> -d <copy destination disk name>
# sdxswap -I -c <class name> -d <copy destination disk name> -e nowaitsync
```
Start RMS on the node restored in step 2.
```
# hvcm -a
```
When using the mirroring among servers of GDS, the entire synchronization copying is also performed.
Use GUI or the sdxinfo command to check if the synchronization copying is completed.
If the server is restored with the backup related services other than the One Day Snap Shot function, perform step 3 to 5 in "9.8.1.2 Setting up the Shutdown Facility" on all nodes.

If the backup is acquired on both nodes at the same time by using the backup related services with the OS stopped on both nodes

Stop all nodes in the cluster system.
Restore both nodes and start them.
For the node restoration, refer to the official NIFCLOUD documentation.
If the server is restored with the backup related services other than the One Day Snap Shot function, perform step 3 to 5 in "9.8.1.2 Setting up the Shutdown Facility" on all nodes.