[TriLUG] NFS woes

Jon Carnes jonc at nc.rr.com
Wed Nov 13 13:33:37 EST 2002


Interesting.  I've had this problem before with RH7.x boxes. Once it was
caused by the NFS Lock daemon not running. Another time we fixed it by
upgrading the NFS to latest version.

This kind of thing can be very frustrating to troubleshoot.  I highly
recommend that you load ethereal or some other packet sniffer on your
network so you can actually see the NFS exchanges.  It really helps to
shine a light on what is exactly going on.

One problem that we documented in one of the early RH 7.x clients was
that it would not re-initiate a new connection sequence with the NFS
master if it lost its initial communication. The client would happy send
and resend information to a process that was no longer listening to it.
It would constantly time-out, but that time-out would never trigger it
to establish a new connection.

Good Luck

On Wed, 2002-11-13 at 11:53, Jason Tower wrote:
> the 3Com problems are frequently related to what kind of hub/switch they're 
> connected to.  i know for a fact that intel switches and 3Com cards tend to 
> get along very poorly.  just for grins, try switching to a different nic and 
> see if that makes any difference.  you can also try some generic file copy 
> speed tests across the network to gage network performance.
> 
> and my RH73 NFS problems usually occured at the server end, but i didn't spend 
> much time troubleshooting to find the root cause.
> 
> jason
> 
> On Wednesday 13 November 2002 11:45, Sinner from the Prairy wrote:
> > On Wed, 2002-11-13 at 10:52, Jason Tower wrote:
> > > the redhat press book says that message occurs when client attempts to
> > > contact the server timeout.
> >
> > Yes, it looks like that kind of problem.
> >
> > > it suggests that the root problem is usually network related.
> >
> > Mmmmm
> >
> > > my hunch is that it may be a ethernet issue, incorrect speed and/or
> > > duplex settings.
> >
> > I had been using that NFS client for a few months now. It looks OK. I
> > had been working OK for a while.
> >
> > > are you using 3Com nics?
> >
> > Let me see:
> >
> > 00:09.0 Ethernet controller: 3Com Corporation 3c905C-TX/TX-M [Tornado]
> > (rev 78)
> >
> > > they have frequently caused me
> > > grief by having weird/difficult to troubleshoot problems.
> >
> > And I always have liked 3Com
> >
> > > furthermore, i had
> > > problems with NFS on RH73 like frequent timeouts,
> >
> > RH 7.3 acting as a client or as a server?
> >
> > I have RHL 7.3 on the client *and* on the server.
> >
> > IIRC, it was yesterday when I began to notice having trouble with my
> > Evolution-through-ssh-tunnel-using-NFS-exported-home-for-local-imap-mail
> >
> > I know, it can be done easier but it won't be fun otherwise.
> >
> > > very slow start/stop times
> > > (like ten minutes to execute 'service nfs start') and so on.
> >
> > service nfs restart takes about 2 seconds. NFS at the server seems to
> > work fine. The problems shows up at the client (also a RHL 7.3)
> >
> > > RH72 and RH80
> > > have been flawless with NFS though.  maybe try the latest RPM package?
> >
> > I'll try. Thank you.
> >
> > > jason
> >
> > Salut,
> > Sinner
> 
> _______________________________________________
> TriLUG mailing list
>     http://www.trilug.org/mailman/listinfo/trilug
> TriLUG Organizational FAQ:
>     http://www.trilug.org/~lovelace/faq/TriLUG-faq.html





More information about the TriLUG mailing list