I have a two-node NW6.5SP8 cluster. Available memory on the master node
dropped precipitously so I migrated resources to the other node and rebooted
the machine. It came up fine and joined the cluster. I attempted to migrate
its normal services back and nothing happened. I looked in NRM on that node
and it reports that it is joined but shows state as empty on all services,
including master IP. NRM on the new master shows node2 as joined. The
cluster event log on both nodes appears the same and shows the node
successfully joining with no errors. A fresh reboot has the same result. I
tried uldncs/ldncs with no better luck. I'm not in a position to down the
cluster which I assume will correct the situation. Ideas?

Erick Larson
Duke University Library