Special Considerations

The following provides information about special considerations for clusters:

We do not recommend a partial shared disk configuration in which some disks are owned by one node and some disks are owned by another node.
To avoid complications when matching disks, shared disks should be the last disks and have the highest number when viewed from Administrative Tools, Computer Management, and Disk Management.
For Windows 2003, run the dumpcfg.exe utility (available on the Windows Server 2003 Resource Kit) to save the cluster quorum disk signature. It is good practice to preserve the important hard disk signatures if this information is not often used.
For remote backup jobs, run the utility from the cluster machine.

For local backup jobs, use the Global Options dialog to run dumpcfg.exe as a pre-job during a backup to ensure that up-to-date information about the critical hard disk is available. To configure the pre-job, perform the following steps:
For Windows Server 2008, Windows Server 2008 R2, and Windows Server 2012, run the DiskPart utility, to obtain and record the signature of the disk and to shift focus to a unique identifier (signature) of the disk.

To configure the pre-job

From the Global Options dialog, click the Pre/Post tab.
In the field Enter the name of the file/application to execute before the job starts, enter the following command:
```
c:\dumpcfg > C:\cluster\DR\[Server_Name]\[Machine_Name]\dumpcfg.txt
```

Configure the pre-job in the Global Options dialog

You can configure disaster recovery information to be saved to an alternate location on a different computer to further protect disaster recovery information
On most cluster computers, there is no need to stop the shared disks. The cluster can continue to function during disaster recovery. Check your hardware documentation for more information about how to avoid shutting down the hard disks.