Red Hat

Hi Forum
I am trying to mount /NFS as nfs mountpoint on two servers ( A & B ).
After mounting the nfs filesystem, both of them behave normally for around 10 mins and after that the NFS file handle become stale and the
mountpoints dont respond. While executing df -kh, the output hang out and the /var/log/messages show the following message:

Code:

 May 27 15:48:56 earth mountd[3078]: Cannot export /NFS, possibly unsupported filesystem or fsid= required
 May 28 04:04:20 earth kernel: nfs: server nas not responding, still trying
 May 28 10:11:51 earth kernel: nfs: server nas not responding, still trying

The fstab enteries for /NFS mountpoint on both servers is :

Code:

nas:/NFS               /NFS                 nfs   rw,bg,hard,nointr,tcp,nfsvers=3,timeo=600,rsize=32768,wsize=32768,acti
meo=0 1 2

/etc/export enteries on both server is :

Code:

/NFS *(rw,sync,no_wdelay,insecure_locks,no_root_squash)

Please let me know how can I fix this issue.
Thanks in advance,
Raj.

what happens when you leave this out of your conf line in fstab?:

Code:

timeo=600

Also, just a tip: df -h is human readable and -k is for KB. Using both seems to not make much sense. It will use the last switch, in your case, -h.

No, it didnt help out.

Does your NAS device have any logs?

Check and ensure that you are not using jumbo frames on some of the devices, and that all the network MTU sizes are the same.

Nas is the first server itself. There are no separate logs for this server.

The jumbo frames are not enabled and mtu is same on all boxes.

Please let me know if you need any further information.

Thanks,
Raj.

You are re-exporting /NFS on both servers which is an NFS volume mounted from the NAS. You can't do that. You need to remove it from /etc/exports..