[wplug] NFS mount problem

[Michael Skowvron]

Gentgeen wrote:
> Jun 27 13:22:20 linuxbox kernel: nfs: server 192.168.0.10 not
> responding, still trying
> Jun 27 13:22:20 linuxbox kernel: nfs: server 192.168.0.10 not
> responding, still trying
> Jun 27 13:24:04 linuxbox kernel: nfs: task 17523 can't get a request
> slot

Kernel NFS bug or maybe a network problem. The NFS server (kingpin) 
has stopped answering the client (linuxbox). Linux box keeps sending 
requests that timeout and eventually the NFS request is aborted. A 
hadware-related network problem might cause this if packets are 
getting lost or corrupted. Lost packets might leave the NFS server in 
a state where it's waiting for a response that the client believes 
that it already sent -- deadlocking them.

What distribtion is running on each system?
What is the exact kernel version on each system? 'uname -a'
Are you using a distribution kernel or did you build your own?
What's in the syslog of kingpin during the timeout?
Also check for kernel NFS/RPC messages (/var/log/kernel/ on 
RedHat-type system).
When the NFS gets stuck, does other network traffic work okay? I.E. 
Can you still ssh into the server?

If all 3 systems are running the exact same distribution and kernel 
and you are only seeing the problem between linuxbox and kingpin; then 
it is probably a hardware-specific problem. Either it's what I 
described earler, or linuxbox may much faster than the laptop and able 
to stress the server into this failure.

Michael