This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--------------enigB7F2473E8C947F976D4D0D90
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Mark Davies wrote:
> One of our file servers (running 3.99.11) has started locking up every =
couple=20
> of days.  It gets into a state where any process will run fine until it=
 tries=20
> to access the disk at which point it stops responding.  Updating the ke=
rnel=20
> to a current from a couple of weeks ago makes no difference.
> I've had zero luck in tracking down whats causing this.
> However I set up an external-mode watchdog to panic the machine if a lo=
op of
> 	sleep 20; ls -l /a/local/directory > /dev/null ; wdogctl -t
> failed to tickle the watchdog for a minute, so I now have a core dump f=
rom=20
> such a panic.  I'd like some suggestions on what to look for/how to pok=
e at=20
> this core dump to try to find whats happening.
>
> cheers
> mark
>  =20
Are you using filesystem snapshots? I've had this problem on FreeBSD
when I used filesystem snapshots with certain RAID controllers (Dell
PERC 3/di and Mylex AcceleRAID). I never tracked it down, and ended up
giving up on snapshots altogether.

--=20
-- Skylar Thompson (skylar@cs.earlham.edu)
-- http://www.cs.earlham.edu/~skylar/



--------------enigB7F2473E8C947F976D4D0D90
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFErSJcsc4yyULgN4YRAoxGAKCwbV1BBn7y7VOwihMBRoS40wl0AgCgs5LW
WqealyOTboHW77yO1Tu2IYY=
=5dhV
-----END PGP SIGNATURE-----

--------------enigB7F2473E8C947F976D4D0D90--