[PLUG] How to umount NFS hung mouts

Mark Martin mmartin at u.washington.edu
Wed Aug 13 11:05:02 UTC 2003


On Wednesday 13 August 2003 10:55, Paul Heinlein wrote:
> On Wed, 13 Aug 2003, Steven A. Adams wrote:
> > If you can use the umount -f option to force the operation I would
> > suggest rewriting the fstab entry for this mount to include the soft
> > and timeo= nfs options (see man mount). This way if the nfs server
> > ever does this again it will allow the workstation to gracefully
> > recover from the event.
>
> I would highly suggest that you investigate the soft and timeo mount
> options very, very carefully before implementing either of them. It's
> been repeatedly stated on the NFS mailing list that you leave yourself
> open to silent data corruption when using soft mounts. Here's a
> representative opinion:
>
>   Having NFS timeout underneath you is exactly the same as having a
>   hard drive fail underneath you. Most if not all applications are
>   utterly unprepared to deal with the resulting I/O errors, thus
>   guaranteeing data corruption.
>
>   The 'soft' mounting is only acceptable for read-only mounts, and
>   even then only if you really don't care about your input data
>   disappearing at random. That's AT RANDOM -- it's not a typo, you'll
>   get RANDOM failures, depending on how loaded your network and/or
>   server are at times.
>
> --Paul Heinlein <heinlein at madboa.com>

Based on the info in the HOWTO, I strongly agree with Paul.  The HOWTO 
recommends using the "hard" and "intr" options rather than "soft" since 
"soft" is likely to lead to corruption.

Mark
-- 
---------------------------------------------------------------------
Mark A. Martin, Ph.D.
Applied Mathematics -- Software Development -- Systems Administration
Currently available for employment.
See http://www.amath.washington.edu/~mmartin/resume/ for details.
---------------------------------------------------------------------





More information about the PLUG mailing list