Rikard Johnels wrote:
>>> I had my fileserver on a reboot, and the clients mounting extra areas via
>>> NFS lost track of the mounts.
>>>
>>> As i try to remount i get:
>>>
>>> mount /mnt/network/server/sda
>>> mount: Stale NFS file handle

> I have a "standard" NFS server with 
> /mnt/local/sda/ *(rw,sync,no_root_squash)
> /mnt/local/sdb/ *(rw,sync,no_root_squash)
> 
> and the client has in the fstab:
> 
> 192.168.1.10:/mnt/local/sda  /mnt/network/server/sda nfs  defaults 0 0
> 192.168.1.10:/mnt/local/sdb  /mnt/network/server/sdb nfs  defaults 0 0
> 
> I did "umount /mnt/network/server/sda" and i complained about not bein
> mounted,

If it was originally mounted via fstab, I don't understand at the moment
how it would report not mounted but also a stale NFS handle.

Please could you post the actual command you used and the actual error
message followed by the mount command and its output. Oh, and a df. Just
cut and paste the sequence from a terminal.

FWIW, I use "rsize=8192,wsize=8192,intr,bg,noatime" for the client's
fstab options. I find running without one of 'soft' or 'intr' is a pain
when problems arise. YMMV.

> and then as i try to mount "/mnt/network/server/sda" it again, i get
> the stale error.
> "lsof" doesnt show anything except an error about stale NFS...
>
> So i cant kill something that i dont know what it is...

Perhaps do a full ps and look at each process to see what it's doing?
Presumably you know which applications use that filesystem.

Cheers, Dave
-- 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to