SUMMARY: NFS problem

From: Matthias Heinitz (mhei@tet.uni-hannover.de)
Date: Fri Jun 02 1995 - 18:01:00 CDT


Dear colleagues,

first of all much thanks to all who responded:

jjr@edi-nola.com (Jack Reiner)
Jesse Adam <jaa@pollux.GEOG.UCSB.EDU>
epl@caps.kodak.com (Gene Loriot (epl@kodak.com))
michael@ide.com (Michael K. Glass)
curt@mrbill.is.lmsc.lockheed.com. (Curt Peterson)
mike-l@hero-uk.demon.co.uk
John.Murray@Germany.EU.net
Jim Rosetti <rosetti@rl.af.mil>
raoul@MIT.EDU
tcgny!dinghy!berghold@uunet.uu.net (Peter L. Berghold)
Ross.Stocks.INSDRS01@nt.com
Chris Swanson <cds@SSDS.com>
Michel Pilon <pilonm@CCG.RNCan.gc.ca>

Some of you stated that it *has* to be a hardware
problem. Unfortunately, I forgot to mention that
we used a sniffer to analyse our network, and I was
absolutely sure that the problem resided logically in the
NFS servers.
The majority suggested to check system settings such as
mounting parameters, routing etc., netstat and snoop were
recommended.

Finally, the solution of the problem was as simple as hard
find (for me): The number of NFS-Daemons of one NFS server
was too low: I had still the default value nfsd -a 16. Changing
the value to 64 fixed our NFS problem !!

Once again: Thanx a lot for your support !

Best regards,

Matthias

_________________________________________________________________________

matthias@tet.uni-hannover.de Dipl.-Ing. Matthias Heinitz
                                Institut fuer Theoretische Elektrotechnik
                                Universitaet Hannover
                                Appelstr. 9A _/_/_/_/_/_/_/_/_/_/_/
Tel.: +49 511 762 3250 30167 Hannover _/ _/_/ _/
Fax : +49 511 762 3204 Germany _/ _/_/_/ _/
_________________________________________________________________________

My views and opinions do not reflect the same for my employer or my country
and definitely not the chancelor! Your solution maybe someone else's problem.

Original Posting:
-----------------

Dear colleagues,

I have got a strange and strong problem with NFS: The messages

   "NFS server <host> not responding still trying
    NFS server ... ok"

are displayed.

We have 7 Suns (OS Solaris 2.3), and two of them are NFS
servers (user data, E-mail, s/w). The access via NFS is
almost impossible. Only cd and ls work for directories
mounted via NFS, but even (sometimes) ls -la fails, and
mv, cp, compress, etc. via NFS are impossible ! Although
the ok-message is displayed, the problem remains.

The NFS- and Mount-daemons seem to run correctly. The network
has already been analyzed exhaustively (transceiver, repeater,
resistance 50ohm look fine). In addition, one NFS server has been
replaced by another machine, but the observed effect remains
and concerns *all* NFS servers. Thus, I exclude a physical problem,
and believe that something is wrong in the configuration. However,
I have not modified anything for a long time before the problem
occured.

Any help will be appreciated. I will summarize !

Matthias



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:10:26 CDT