Look at the cluster service on the 'bad' node. If communication is good but
the node appears down, it may be the cluster service is having problems.

Yes, you can evict the node and re-add it later. You will need to uninstall
SQL and reinstall SQL for that node as part of the removal and re-add.
There are step by step instructions in BOL under the topic 'Maintaining a
Failover Cluster'.

--
Geoff N. Hiten
Microsoft SQL Server MVP
Senior Database Administrator
Careerbuilder.com

I support the Professional Association for SQL Server
www.sqlpass.org

"RunninMan" <rflot (AT) doe (DOT) k12.ga.us> wrote

working great. Suddenly, node 2 is not cooperating in the cluster anymore.
The cluster service on N2 won't run. All dependencies are good. In Cluster
Admininstrator on N1 or my desktop, N2 has a big red X on it and can't be
contacted. The heartbeat between the two nodes is fine. There are errors in
the Event Viewer of N2 related to the Fibre Channel HBA's, and I'm thinking
of reinstalling drivers, or un-installing devices, and then reinstalling the
devices and drivers to try to address that.
node? What does this operation involve? A white paper would be nice.

Hey Goeff

Thanks for the information dude. I hear what you're saying about the cluster service, and I do think it's the service as it relates to the hardware. That may sound crazy, but I've seen this on clustered Exchange servers, where I had continuous hardware issues that would cause problems with the services. I found the step-by-step you suggested, and I think I'm going to evict this node and set it up again. Have a great day and thank you again

Ren