I have a customer running Netware 6.5 SP2 with QLA2340-E-SP dual-attached
HBAs in a clustered environment. HBA Driver is QL2300 V6.80.03 firmware
1.42, they are trying to access a CX700 EMC storage array running flare
v16 with Powerpath v3.0.5.

The cluster will take down a node if it loses connectivity for more than 8
seconds, but it is taking the host 30 seconds to recognize a downed path
before failover. Is this a limitation within Novell to recognize the bad
path? Or possibly the HBA driver/firmware it is using? I'm curious if
there are TOV's that need to be adjusted, etc. Any inside knowledge would
be appreciated.