Re: [Qemu-devel] qemu and qemu.git -> Migration + disk stress introduces qcow2 corruptions

[Anthony Liguori <anthony@codemonkey.ws>]

On 11/10/2011 04:41 AM, Kevin Wolf wrote:
> Am 09.11.2011 22:01, schrieb Anthony Liguori:
>> On 11/09/2011 03:00 PM, Michael S. Tsirkin wrote:
>>> On Wed, Nov 09, 2011 at 02:22:02PM -0600, Anthony Liguori wrote:
>>>> On 11/09/2011 02:18 PM, Michael S. Tsirkin wrote:
>>>>> On Wed, Nov 09, 2011 at 11:35:54AM -0600, Anthony Liguori wrote:
>>>>>> On 11/09/2011 11:02 AM, Avi Kivity wrote:
>>>>>>> On 11/09/2011 06:39 PM, Anthony Liguori wrote:
>>>>>>>>
>>>>>>>> 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.
>>>>>>>
>>>>>>> Say what?
>>>>>>
>>>>>> Due to block format probing, we read at least the first sector of
>>>>>> the disk during start up.
>>>>>
>>>>> A simple solution is not to do any probing before the VM is first
>>>>> started on the incoming path.
>>>>>
>>>>> Any issues with this?
>>>>>
>>>>
>>>> http://mid.gmane.org/1284213896-12705-4-git-send-email-aliguori@us.ibm.com
>>>> I think Kevin wanted open to get delayed.
>>>>
>>>> Regards,
>>>>
>>>> Anthony Liguori
>>>
>>> So, this patchset just needs to be revived and polished up?
>>
>> What I took from the feedback was that Kevin wanted to defer open until the
>> device model started.  That eliminates the need to reopen or have a invalidation
>> callback.
>>
>> I think it would be good for Kevin to comment here though because I might have
>> misunderstood his feedback.
>
> Your approach was to delay reads, but still keep the image open. I think
> I worried that we might have additional reads somewhere that we don't
> know about, and this is why I proposed delaying the open as well, so
> that any read would always fail.
>
> I believe just reopening the image is (almost?) as good and it's way
> easier to do, so I would be inclined to do that for 1.0.

I don't think reopen is good enough without delaying CHS probing too.  That 
information is still potentially out of date.  I don't think you can fix this 
problem without delaying CHS probing at least.

Regards,

Anthony Liguori

>
> I'm not 100% sure about cases like iscsi, where reopening doesn't help.
> I think delaying the open doesn't help there either if you migrate from
> A to B and then back from B to A, you could still get old data. So for
> iscsi probably cache=none remains the only safe choice, whatever we do.
>
> Kevin
>