[CentOS] NFS problem

Thu Sep 20 06:50:06 UTC 2012
Tony Molloy <tony.molloy at ul.ie>

On Wednesday 19 September 2012 23:22:13 Mark LaPierre wrote:
> On 09/19/2012 03:00 PM, m.roth at 5-cent.us wrote:
> > We've been seeing what I gather is an old, traditional problem:
> > kernel: lockd: server<ip address>  not responding, timed out
> >
> > The things I've found, googling, mostly involve rebooting the NFS
> > server, and I can't do that, it's a home directory server for a
> > *bunch* of people, and this is only one person's workstation.
> >
> > Are there other solutions I haven't found yet?
> >
> >         mark

I see the same problem with our student labs every now and then. I 
always assumed it was caused by a student power recycling the client 
rather than doing a clean reboot to get to Windoze ( dual boot 
machines ;-( ) causing NFS to get confused, NFSv3.

It doesn't seem to cause any problems just fills up the server logs 
with useless messages. Rebooting the client to Linux usually fixes the 


> >
> > _______________________________________________
> > CentOS mailing list
> > CentOS at centos.org
> > http://lists.centos.org/mailman/listinfo/centos
> Have you checked your network connectivity from the problem machine
>  to the NFS server and back?  If other machines on this NFS server
>  have no issues then you can eliminate the NIC on the server from
>  suspicion, but the NIC on the problem machine might be having a
>  problem.  Check your cable connections while you're at it.
> How are you mounting the NFS share?  If you are using AutoFS your
>  mount may be timing out.  Does the problem machine access the
>  share actively or does it tend to sit for a while between access
>  attempts?