I have a Windows 2008R2 file server which has been running fine for quite some time. With no known changes, in the last few days it's been misbehaving, with intermittent slows when users open/save files through the network share.
I haven't been able to reproduce a slow when opening a file locally on the server when there are problems experienced by the clients through the network.
The network ports on the switches don't report any error.
On the server, disk queue length and other disk indicators don't indicate a saturation of disk access. Average disk idle time is about 80%.
On the server, net statistics server shows a constant increase of "Sessions errored-out". "Sessions timed-out" does not significantly increase (haven't seen it increase while I checked).
The server was rebooted and NIC and RAID firmware/drivers updated last night and today, it looked like things were better. However, after about 12 hours of being up, it started misbehaving again - not dramatically like it was yesterday.
It becomes worst when I have a lot of users - or at least, more users complain. Since I didn't put a finger yet on a metric that is not good on the server itself, it's hard to tell.
Any pointer of various tools to debug this issue. It seems like the net statistics server errored-out session counter is an indicator but I am not sure where to take that from there.
The network ports on the switches don't report any error- That doesn't mean it's not a network problem. It could be network congestion or any of a number of other network related problems. If the problem doesn't occur when opening files locally on the server then I'd strongly suspect a network problem.