We have a 7 node cluster , Netware v6.5 with sp6. A couple of times we
have had a pool corruption cause a cascading failover. Don't ask me how
the pool corruption occured, but lets say it was probably not the
technology's fault. However I am wondering why we are having the cascading
failover situation (meaning the dud resource took out 3 servers before it
had no further node to fail to) , when it is designed to go comatose by
default and stay offline pending manual intervention (refer TID 3005414).
Has anyone else seen this, is it a bug or a feature ?