Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 19 Feb 2005 16:18:30 -0800
From:      Kris Kennaway <kris@obsecurity.org>
To:        Daniel Eriksson <daniel_k_eriksson@telia.com>
Cc:        'FreeBSD Current' <freebsd-current@freebsd.org>
Subject:   Re: Crash on CURRENT from today
Message-ID:  <20050220001830.GA48352@xor.obsecurity.org>
In-Reply-To: <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA0VcX9IoJqUaXPS8MjT1PdsKAAAAQAAAAI6TAYRJQdkKhMiFnEgfiCQEAAAAA@telia.com>
References:  <20050219150513.J69556@carver.gumbysoft.com> <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA0VcX9IoJqUaXPS8MjT1PdsKAAAAQAAAAI6TAYRJQdkKhMiFnEgfiCQEAAAAA@telia.com>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help

--5mCyUwZo2JvN/JJP
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sun, Feb 20, 2005 at 01:14:20AM +0100, Daniel Eriksson wrote:
> Doug White wrote:
>=20
> > > Fatal trap 9: general protection fault while in kernel mode
> >=20
> > these are bad news.  The only times I've seen this is with=20
> > horrifically
> > broken code or hardware problems.  Mostly hardware problems. =20
> > You might
> > check the environmentals on your system, and check the event=20
> > log for any
> > ECC correction events or other abnormal behavior.  Normal code isn't
> > likely to trigger GPFs.
> >=20
> > Did you compile the kernel with any non-standard options?
>=20
> The machine has been running pretty solid for the last couple of months.
> This of course doesn't mean it's not a hardware problem. I'll know for su=
re
> in a couple of days I guess (if it keeps crashing with strange errors).
>=20
> Kernel/world is compiled only with standard flags. Nothing in make.conf
> except "CPUTYPE?=3Dathlon-xp" (which means -O2 is used). I reverted back =
to a
> non-debug kernel a few weeks ago when the makefiles were changed to only =
use
> -O for debug-kernels. However, after switching mpsafenet back on (replaced
> NFS (which locked up for me with mpsafenet) with geom_gate) I have some
> spare cycles to burn on the machine. I might as well use a debug-kernel in
> case this isn't a hardware problem.

FYI, a patch for a deadlock condition in NFS was committed yesterday.
It would be great if you can test this out.

Kris

--5mCyUwZo2JvN/JJP
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (FreeBSD)

iD8DBQFCF9dWWry0BWjoQKURAgI5AJ4k/kvhmTQAQv4fzXIejdODsyqf+gCg1TVr
wgRQUj8fqVytGuyWlovVy1c=
=FiV+
-----END PGP SIGNATURE-----

--5mCyUwZo2JvN/JJP--



Want to link to this message? Use this URL: <http://docs.FreeBSD.org/cgi/mid.cgi?20050220001830.GA48352>