I'm running into a strange issue that I'm having a hard time troubleshooting. Once in awhile, we find that one of our controllers becomes unresponsive to RPC commands. We use Powershell script extensively in our environment to automate a large number of tasks. These tasks become unresponsive when this happens. More worryingly, SnapDrive stops responding during these events as well.
The only clues I have are the following:
1) If we run a CIFS terminate, it warns us that several hundred CIFS sessions are open and cannot be closed.
2) If we run a cifs sessions command, it does not show any cifs sessions.
3) If we stop all of our scripts from talking to the controller for awhile, the CIFS sessions go away and the controller becomes responsive again.
It's evident to me that something with the cmdlet calls I'm making is hanging occasionally on the controller, and causing a backup of RPC requests. I doubt the Powershell cmdlet calls open an actual CIFS session, but I do know they use RPC. In our case, we always use RPC for Powershell.
Has anyone ever seen this in their environment? Also, are there any configuration options I can play with to give me more information or help me troubleshoot?