Having an issue since migrating from a NetWare 3 node cluster to and OES2sp2 3 node cluster. We are use to NetWare when the node had an issue it took a poison pill and did a hard ABEND causing the cluster services to roll to the other nodes. We are not seeing this on OES2sp2.

With OES2sp2 we will get an alert from servers alive that port 524 is down. Cluster Status on the bad node still show the cluster resources active but we can not access them. I have to issue a cluster leave on the bad node in order to get the resources to fail and roll to the other two nodes.

Do we need to configure the servers some how like with NetWare so the node takes a poison pill and leaves the cluster?

Also anyone know why port 524 would randomly close on an OES2sp2 box?