[4285] in linux-net channel archive
NFS problem with 2.0.17
daemon@ATHENA.MIT.EDU (Jacques Gelinas)
Tue Sep 3 19:07:38 1996
Date: Tue, 3 Sep 1996 14:52:36 -0400 (EDT)
From: Jacques Gelinas <jack@solucorp.qc.ca>
To: Linux net list <linux-net@vger.rutgers.edu>
I have a strange problem with 2.0.17. I will try to gather more info. The
problem is new: 2.0.15 and before were flawless (I am not running apache
:-) ).
I have seen the problem on 2 machines. Once in a while one NFS connection
just become unusable. One special thing is that sometime the hung process
(trying to read) can be killed, sometime not (The NFS is hard mount using
amd, the automounter). I would assume that those process should be never
killable. This may be related to the automounter. The problem is difficult
to pinpoint as all the machines here (3 linux and 1 SunOS) are
automounting each others.
I have seen this problem multiple time this morning. There is no relevant
message on the server and the only message I have gather so far on a hung
client is this one
amd[103]: Error reading RPC reply: Connection refused
I have also seen this problem with a SunOS 4.1.3 nfs server and linux
2.0.17 nfs servers.
One other thing, when one mount is hung, it is still possible to do
another mount to the same server without problem (from the same client).
Anyone has seen something like this ?
--------------------------------------------------------
Jacques Gelinas (jacques@solucorp.qc.ca)
Linuxconf: The ultimate administration system for Linux.
see http://www.solucorp.qc.ca:/linuxconf
