The backbone of the highavailability is the nodes in a cluster checking each other so you should look into heartbeats. They are usually implemented by sharing a disk space, like a concurrent...
Make sure you are pining the Persistent IP and NOT the Service IP, because Service IP will jump between the nodes, whereas persistent IP is hard bounded to the node.
Ok, there is the culprit. The problem is, that JFS does not support files larger than 2GB. The message from mksysb is confusing.
You should recreate the filesystem as JFS2. JFS is out of date and...