Reboot one after the other node, never shutdown both at the same time.

Use Cluster Administrator to initiate failure so the other nodes becomes the owner of all resources. Never take...