Unplugging disk under ZFS yield panic

[Charles Sprickman]

On Jan 11, 2012, at 7:15 PM, Jeremy Chadwick wrote:

> On Wed, Jan 11, 2012 at 06:58:57PM -0500, Charles Sprickman wrote:
>> 
>> On Jan 11, 2012, at 4:43 PM, Fabian Keil wrote:
>> 
>>> Jeremy Chadwick <freebsd at jdc.parodius.com> wrote:
>>> 
>>>> On Wed, Jan 11, 2012 at 09:07:08PM +0100, Fabian Keil wrote:
>>>>> Gergely CZUCZY <phoemix at harmless.hu> wrote:
>>>>> 
>>>>>> I'd like to ask, whether it is normal behaviour when we're unplugging a
>>>>>> disk under a ZFS system then on the first write a kernel panic happened.
>>>>> 
>>>>> Sounds familiar. I currently have two PRs open for
>>>>> reproducible kernel panics after a vdev gets lost:
>>>>> http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/162010
>>>>> http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/162036
>>>>> 
>>>>> Note that the pool layouts are different, though.
>>>> 
>>>> Is this problem truly ZFS-specific?  I'd been tracking this problem for
>>>> years, and was told it was fixed:
>>> 
>>> I'm not saying that my problems are ZFS-specific.
>>> The backtraces mainly contain geom functions and no ZFS code,
>>> so ZFS might be the victim here.
>> 
>> Is there any relation between this issue and the "log_sysevent:
>> type 19 is not implemented" problem that happens on device
>> insertion/removal on 8.2?
>> 
>> http://lists.freebsd.org/pipermail/freebsd-fs/2011-June/011855.html
>> 
>> I still see that one on 8.2-STABLE from around 6/2011.  I initially
>> thought it was triggered by device failure or removal (even with
>> proper hotplug support), but I got hit by it last night when
>> inserting a new drive in a chassis that supports ahci/sata hotplug.
>> IIRC it's not a ZFS issue, ZFS just gets more spammy (well,
>> extremely spammy to the point the system can't do much else) about
>> reporting an issue with a device going away or being inserted.
> 
> This was fixed in RELENG_8 on 2011/06/14, r222343:
> 
> http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/cddl/compat/opensolaris/kern/opensolaris_sysevent.c#rev1.2.2.3
> 
> The code that would print that message is #if 0'd out now.  I can
> confirm the message is gone on all our systems running recent RELENG_8.

Wow, how unlucky:

FreeBSD 8.2-STABLE (BL8-64) #0 r222897: Sun Jun 12 16:35:52 EDT 2011

So that's totally cosmetic?  There was no underlying GEOM issue?
Sorry for the noise on that then...

Charles 

> 
> -- 
> | Jeremy Chadwick                                jdc at parodius.com |
> | Parodius Networking                       http://www.parodius.com/ |
> | UNIX Systems Administrator                   Mountain View, CA, US |
> | Making life hard for others since 1977.               PGP 4BD6C0CB |
>