you want to test disk failover from the cluster perspective or from SAN connectivity ? From cluster perspective if you have everything setup correctly you should be able to move disk resources from one node to another by manually initiating failover (depends on your cluster software), or just simply shut one node down and see if your disk recourse come online on the other node. From the SAN perspective if you have PowerPath/DM-MPIO/Veritas DMP running ..simply unplug cable from one of the hba's ..you might see different things on the backend, depending on the array you are attached too ..but from the host perspective it should be transparent ..it will throw a bunch of message into /var/log/message ..but volume groups/disk should keep on running.