From mboxrd@z Thu Jan  1 00:00:00 1970
From: Anthony Liguori <anthony@codemonkey.ws>
Subject: Re: qemu and qemu.git -> Migration + disk stress
 introduces qcow2 corruptions
Date: Wed, 09 Nov 2011 10:39:11 -0600
Message-ID: <4EBAACAF.4080407@codemonkey.ws>
References: <4EBAAA68.10801@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: Kevin Wolf <kwolf@redhat.com>, KVM mailing list <kvm@vger.kernel.org>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Marcelo Tosatti <mtosatti@redhat.com>, QEMU devel <qemu-devel@nongnu.org>,
	Juan Jose Quintela Carreira <quintela@redhat.com>,
	Avi Kivity <avi@redhat.com>
To: Lucas Meneghel Rodrigues <lmr@redhat.com>
Return-path: <qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org>
In-Reply-To: <4EBAAA68.10801@redhat.com>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
Errors-To: qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org
Sender: qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org
List-Id: kvm.vger.kernel.org

On 11/09/2011 10:29 AM, Lucas Meneghel Rodrigues wrote:
> Hi guys, here I am, reporting yet another issue with qemu. This time, it's
> something that was first reported in January, and Juan proposed a patch for it:
>
> http://comments.gmane.org/gmane.comp.emulators.qemu/89009

Migration with qcow2 is not a supported feature for 1.0.  Migration is only 
supported with raw images using coherent shared storage[1].

[1] NFS is only coherent with close-to-open which right now is not good enough 
for migration.

Regards,

Anthony Liguori

>
> [PATCH 4/5] Reopen files after migration
>
> The symptom is, when running disk stress or any intense IO operation in guest
> while migrating it causes a qcow2 corruption. We've seen this consistently on
> the daily test jobs, both for qemu and qemu-kvm. The test that triggers it is
> autotest stress test running on a VM with ping-pong background migration.
>
> The fix proposed by Juan is on our RHEL branch and such a problem does not
> happen on the RHEL branch. So, what about re-considering Juan's patch, or maybe
> work out a solution that is satisfactory for the upstream maintainers?
>
> I'll open a launchpad bug with this report.
>
> Thanks,
>
> Lucas
>