Re: An nfs/automount Oops

From: H. Peter Anvin (hpa@transmeta.com)
Date: Wed Apr 19 2000 - 15:56:07 EST


Jeremy Fitzhardinge wrote:
>
> On 19-Apr-2000 H. Peter Anvin wrote:
> > Either way, autofs doesn't pass superblocks anywhere... it calls
> > /bin/mount and /bin/umount in user space.
>
> I'm wondering if autofs is circumstantial, and the problem is an NFS bug
> caused by repeated mount/umount. Certainly an NFS filesystem gets a lot
> more mount/umount activity than one statically mounted at boot.
>

Wouldn't surprise me one bit. Over the years I have had a number of NFS
or VFS problems reported as "autofs bugs", mostly because it makes
mount/umount a significantly more common operation. It also makes it
possible for a filesystem to get called for expiry and by the time the
daemon calls umount it is no longer unmountable. This means the code in
the kernel to reject busy umounts has to work. This has been the
problem on several occations.

        -hpa

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sun Apr 23 2000 - 21:00:16 EST