[one-users] Disk image type not propagated by saveas

[Robert Parrott]

Hi Tino,

I've discovered both of this points, so have a workaround:

(1) default disk type "qcow2"
(2) modify template for any raw disk images

However, I think I would still consider this a defect in ONE.  The
disk type is vital metadata for using the disk image, and needs to be
propagated along with location and other data from the running VM
instance to to the image manager when the "saveas" command is issued.
Otherwise the "DRIVER" metadata is missing from the image template,
and there's no information about how to use the image; one would have
to guess at the image type without prior knowledge ( as I did).

Thanks,
Rob


On Wed, Aug 17, 2011 at 7:09 AM, Tino Vazquez <tinova at opennebula.org> wrote:
> Dear Rob,
>
> The "raw" disk type attribute is being pulled from the default, which
> is set in /etc/one/vmm_exec/vmm_exec_kvm.conf  (in OpenNebula 3.0
> Beta1, in OpenNebula 2.2 is /etc/one/vmm_kvm/vmm_kvm.conf IIRC).
>
> Which version of OpenNebula are you using? In 3.0, you can update the
> image template to add the type parameter, using "oneimage update".
>
> Hope it helps,
>
> -Tino
>
> --
> Constantino Vázquez Blanco, MSc
> OpenNebula Major Contributor
> www.OpenNebula.org | @tinova79
>
>
>
> On Sat, Aug 13, 2011 at 12:36 AM, Robert Parrott
> <parrott at seas.harvard.edu> wrote:
>> Hi Folks,
>>
>> After fixing an older issue with the "onevm saveas" command, I'm
>> seeing the following behavior.
>>
>> If I do a "onevm saveas" on a running VM with a qcow2 disk image, and
>> then shutdown the VM, I see that it successfully shutdowns and then
>> copies over the VM image.
>>
>> However, I can't start VMs from these images.  The root cause appears
>> to be that the driver is propagated to the new image; i.e. the disk
>> tempalte for the new image doesn't contain a "DRIVER=qcow2" line. The
>> resulting libvirt deployment file has "type='raw'" for the disk entry,
>> and not the needed "type='qcow2'".
>>
>> So I expect this is a bug, and a bit of a showstopper for us. Where
>> should I look to fix this?
>>
>> Thanks!
>> Rob
>>
>>
>>
>> --
>> Robert E. Parrott, Ph.D. (Phys. '06)
>> Director, Academic Computing
>> Harvard University Sch. of Eng. and App. Sci.
>> Maxwell-Dworkin  211,
>> 33 Oxford St.
>> Cambridge, MA 02138
>> (617)-496-1520
>> _______________________________________________
>> Users mailing list
>> Users at lists.opennebula.org
>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>
>



-- 
Robert E. Parrott, Ph.D. (Phys. '06)
Director, Academic Computing
Harvard University Sch. of Eng. and App. Sci.
Maxwell-Dworkin  211,
33 Oxford St.
Cambridge, MA 02138
(617)-496-1520