Connect to the SAN switches, identify the right port for cluster node which indicates this issue with event IDs 1209 or 118 (maybe both) and perform a port reset. If this will doesn't help, power down both cluster nodes. Next start-up 1st cluster node and the the 2nd cluster node. After this action, the Microsoft Windows Cluster Service will be again up and running fine on both nodes. At the end, don't forget to divide the cluster resources as per your policies.
This event occurs after you apply hotfix ME911030
because the implementation of the Microsoft cluster driver (Clusdisk.sys) resets the bus even for the success case SCSISTAT_GOOD of the SCSI status code. See ME923424
for a hotfix applicable to Microsoft Windows Server 2003.
for additional information about this event.
The cluster service was unable to mount a shared disk, possibly the quorum disk. Verify that the shared drive is active; try to shutdown all the cluster servers including the shared storage and restart the storage and servers.