Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 3 Jun 2009 13:40:23 -0400
From:      Wesley Shields <wxs@FreeBSD.org>
To:        Dmitry Marakasov <amdmi3@amdmi3.ru>
Cc:        Michael Moll <kvedulv@kvedulv.de>, bz@FreeBSD.org, jamie@FreeBSD.org, freebsd-current@freebsd.org
Subject:   Re: Kernel panic when accessing ZFS-Filesystem via NFS
Message-ID:  <20090603174023.GA23095@atarininja.org>
In-Reply-To: <20090603172033.GC27282@hades.panopticon>
References:  <20090601182012.GA21543@darkthrone.kvedulv.de> <20090603121307.GA15659@hades.panopticon> <20090603152810.GA21014@atarininja.org> <20090603160945.GC21014@atarininja.org> <20090603172033.GC27282@hades.panopticon>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Jun 03, 2009 at 09:20:33PM +0400, Dmitry Marakasov wrote:
> * Wesley Shields (wxs@FreeBSD.org) wrote:
> 
> Well, cred->cr_prison was the only thing to panic on in prison_priv_check,
> so that was obvious without the backtrace. I'll try to addd checking
> for null pointer here for now (as I'm more concerned in bringing
> my NAS back up).

Adding a check for null to jailed() in kern_jail.c will most likely work
around the problem, but I was under the impression that cred->cr_prison
should be &prison0 and never null with the new way of doing things. I'd
still like to know why cr_prison is null.

Of course, I have no idea what the implications of a null check in
jailed() will be outside of this one scenario.

-- WXS



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090603174023.GA23095>