We have a 2 node cluster with one sql instance.

I'm a little bit confused about the default values on the policys for
failover of a resource.


Period for restarts 15:00 (default)
Max. restarts in the specified period 1 (default)
If restart fails, failover all resources in this service or
application
f all the restart attempts fail, begin restarting again after the
specified period (hh:mm) 01:00


What is the best practice with these numbers i win 2008?


We had an error in our virtual connect switch (HP) getting issues on
both disk and network resource. The disk failed twice in the period of
2 seconds and then it stopped retrying.

Is it anyone out there that have something to share about this?

Our Windows 2003/SQL2K5 handled this issue OK.

/Peter