I currently have a two node OES sp2 (linux) ISCSI cluster with 4 virtual
servers. This is currently working ok, I am able to migrate resources and
performance has been acceptable. The problem is I am now trying to test
the auto failover of resources and am having problems. For a test I pulled
the network connection of our master node (half of the resources were on
each node). When I did this the other node gave an abend error and
rebooted (is this not a good way to test?). After that I plugged back in
the server nic and resources from the other server went comatose. After
several attempts at offlineing an onlineing resources and rebooting both
nodes I was able to get them back up. Obviously a timeout or something in
the load or unload scripts is not right. What would be the best way to
start troubleshooting this?

Any help is appreciated.