Date: Wed, 25 Jan 2006 01:54:01 +0000
From: Chris Lightfoot <chris@ex-parrot.com>
To: Willy Tarreau <willy@w.ods.org>
Cc: linux-kernel@vger.kernel.org
Subject: Re: kernel freeze on 2.4.32, apparently in cached_lookup
Message-ID: <vZtReWfk3Hr2.aYFsDRcqPWlIvTnz7jlvlw@sphinx.mythic-beasts.com>
References: <tQjo3cDibk0S.MeaWngl+j1QiYR1U5+Ih3w@sphinx.mythic-beasts.com> <20060124211311.GV7142@w.ods.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20060124211311.GV7142@w.ods.org>
User-Agent: Mutt/1.4i
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1211
Lines: 27

On Tue, Jan 24, 2006 at 10:13:12PM +0100, Willy Tarreau wrote:
    [...]
> What seems strange in your report is that the kernel freezes.
> The only part in cached_lookup() which could freeze IMHO is
> when it calls d_lookup(), but for this, you should have a
> closed loop instead of a linked list. It could happen with
> some memory corruption, but you would get far more oopses
> and panics than freezes. For this reason, I believe you
> might have some random problem on your filesystem. Could
> you run a full fsck on it ?

fsck finds the filesystem is clean; I ran memtest
overnight when I built the machine and it didn't find
anything. Nick's suggestion that it could be a temperature
problem is also interesting; I've added another fan to the
machine and I'll see if that helps matters; if not I'll
try memtest again.

-- 
``It's not a bomb. It's a device that explodes.''
  (possibly-apocryphal statement by French spokesman,
  before the 1995 nuclear tests)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/