nfs to solaris hangs -- why

We are running linux as our nfs server. It is serving mostly solaris boxes.  On a pretty regular basis nfs
hangs and we need to reboot the server.  Any ideas why?
mxfAsked:
Who is Participating?
 
sherwoodConnect With a Mentor Commented:
On the Solaris box try explictly setting rsize and wsize to 1024.
(See the bottom of hte man page for mount for details.)
Not sure if this will be *the* answer, but it's worth a try,
and is quick & easy.  If it works you can try increasing
the size in increments of 1024. I had one network that I could
use 7K packets reliabley, but 8K caused a dumb router to
crash on a frequent basis.)

If this doesn't work, send me the output of nfsstat run
on the solaris box.
0
 
df020797Commented:
If it is the classic problem its an incomaptibility with the nfs lockd daemons. The perfomance is improved by achieveing a patch from Sun for their lockd.

0
 
mxfAuthor Commented:

0
 
Hal9kCommented:
We have a similar problem the other way round:
A Solaris (2.5.1) sparc is running our NFS server and the
Linux boxes (among other Solaris clients) mount these dirs.
Occasionally (well pretty random, on one day 3 times, then 2
weeks without fault) the server daemon (rpc.nfsd) hangs and
the clients cry: NFS server xxx not responding...
The process can only be killed with signal 9 and the system does
not release the ports, so it cannot be restarted.
Only rebooting the server helps.
Maybe these problems are correlated in a way.

0
 
e_chadCommented:
It may be the same problem that I am having.
We are running nfs and we are getting an error message "kernel: svc: unknown version 3" after a short time the "rpc.kmountd --no nfs version 3" daemon crashes

try recompiling the kernel
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.