What is the process used to restore an entire cluster in the event of a catastrophic failure?

A.
Restore the first node by doing a fresh install, restore the first node database from DRS, and then restore the subsequent nodes from the DRS backup location.

B.
Restore the first node from tape or SFTP and then restore each subsequent node.

C.
Restore the first node and then perform a database push to the subsequent nodes in the
cluster.

D.
Restore all the nodes in the cluster by means of a fresh install and then restore the features and subcomponents to each server by means of DRS.

Explanation:

In case of a major failure or a hardware upgrade, you may need to restore all nodes in the cluster. To restore a whole cluster, you must first restore the first node or publisher and then restore the subsequent nodes.
http://www.cisco.com/univercd/cc/td/doc/product/voice/c_callmg/5_0/drs/drf5_0.htm#wp1870