From owner-freebsd-current@FreeBSD.ORG  Mon Oct 26 15:00:32 2009
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Delivered-To: freebsd-current@freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 9DF9A106568B
	for <freebsd-current@freebsd.org>; Mon, 26 Oct 2009 15:00:32 +0000 (UTC)
	(envelope-from doug@polands.org)
Received: from atlmtaow03.cingularme.com (atlmtaow03.cingularme.com
	[66.102.165.8]) by mx1.freebsd.org (Postfix) with ESMTP id 1DDBF8FC20
	for <freebsd-current@freebsd.org>; Mon, 26 Oct 2009 15:00:31 +0000 (UTC)
Received: from [10.215.16.57] (really [166.137.135.185])
	by atlmtaow02.cingularme.com
	(InterMail vM.6.01.04.00 201-2131-118-20041027) with ESMTP
	id <20091026143632.OMTW27583.atlmtaow02.cingularme.com@[10.215.16.57]>; 
	Mon, 26 Oct 2009 10:36:32 -0400
References: <46e678eb1eeca98033b15979e034a765.squirrel@email.polands.org>
	<2D80E5CB-71E5-4F4F-A84B-A95C18B50F47@exscape.org>
Message-Id: <67BF95AE-2294-4249-9F22-487E6652E53E@polands.org>
From: Doug Poland <doug@polands.org>
To: Thomas Backman <serenity@exscape.org>
In-Reply-To: <2D80E5CB-71E5-4F4F-A84B-A95C18B50F47@exscape.org>
Content-Type: text/plain;
	charset=us-ascii;
	format=flowed;
	delsp=yes
Content-Transfer-Encoding: 7bit
X-Mailer: iPhone Mail (7D11)
Mime-Version: 1.0 (iPhone Mail 7D11)
Date: Mon, 26 Oct 2009 09:36:52 -0500
X-Cloudmark-Analysis: v=1.0 c=1 a=rr91BCKx9YgA:10
	a=cgSqiB6BYjiz3vxIx17AKQ==:17 a=r-KksaJNAAAA:8 a=6I5d2MoRAAAA:8
	a=nSrfBVD6MWPTn3FBk80A:9 a=ihJiw0rkUu-XBz7QCMr-_fLf_OwA:4
	a=XSp_zfTBMSoA:10
Cc: "freebsd-current@freebsd.org" <freebsd-current@freebsd.org>
Subject: Re: Fatal trap 12 on ZFS in 8.0-RC1
X-BeenThere: freebsd-current@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussions about the use of FreeBSD-current
	<freebsd-current.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>, 
	<mailto:freebsd-current-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-current>
List-Post: <mailto:freebsd-current@freebsd.org>
List-Help: <mailto:freebsd-current-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>,
	<mailto:freebsd-current-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Mon, 26 Oct 2009 15:00:32 -0000


On Oct 25, 2009, at 1:07, Thomas Backman <serenity@exscape.org> wrote:

> On Oct 25, 2009, at 3:10 AM, Doug Poland wrote:
>
>> Hello,
>>
>> I'm experimenting with ZFS on 8-0-RC1 (amd64) in a VMWare virtual
>> machine w/2GB RAM.  In installed ZFS on GPT root using the excellent
>> instructions at http://wiki.freebsd.org/RootOnZFS/GPTZFSBoot/RAIDZ1.
>>
>> All was well until I tried some benchmarking with
>> benchmarks/unixbench.  I kept getting kmem_map too small panics on  
>> the
>> filesystem tests so starting playing with vm.kmem_size.
>>
>> I finally got unixbench through the first 6 Filesystem Throughput
>> tests with:
>> vm.kmem_size="1296M"
>> vm.kmem_size_max="1296M"
>> vfs.zfs.arc_max="128M"
>> vfs.zfs.vdev.cache.size="24M" (i think)
>>
>> Unfortunately, ZFS paniced with:
>> Fatal trap 12: page fault while in kernel mode. Now, when I attempt  
>> to
>> restart the VM, as soon as the "boot menu" is gone, I immediately get
>> the same Fatal trap 12 panics again and cannot get past this point.
>>
>> I've got the 8.0-RC1 DVD and can run Fixit, but have no idea how to
>> repair this broken ZFS installation.  Suggestions welcome.
>
> You should be able to break into the loader prompt (choice #6 IIRC)  
> and do
> unset vm.kmem_size
> unset vm...
> ... etc.
> boot
>
> That way, you'll be back to the earlier panics, but at least it'll  
> boot. Oh, and the settings aren't permanent, so make editing /boot/ 
> loader.conf your first prority.
>
> Regards,
> Thomas

Thank you, that worked nicely.