[libvirt-users] Can I move the disk image of the guest while it is running?

[Kashyap Chamarthy]

On Wed, Feb 05, 2014 at 10:54:15PM +0100, Gergely Horváth wrote:
> Thank you Eric,
> 
> On 2014-02-05 17:23, Eric Blake wrote:
> > Yes, live storage migration is possible; although at the moment, qemu is
> > lacking a way to restart the operation if it fails midstream, so libvirt
> > only allows the operation if you are willing to temporarily make your
> > guest transient.
> 
> What does this mean? Will I loose anything if - for example - there is
> not enough space on the target device?

As I understand it, no, you'll not lose anything.

Here's some old notes for live backup with 'virsh blockcopy', this was
after discussing with Eric

  http://kashyapc.fedorapeople.org/virt/lc-2012/live-backup-with-blockcopy.txt

  Currently due to QEMU's limitation, libvirt expects a domain to be
  trasient. Because, blockcopy jobs last forever until canceled, which
  implies that they should last across domain restarts if the domain
  were persistent. But qemu doesn't yet provide a way to restart a copy
  job on domain restart. So the trick is to temporarily make the domain
  transient


> Or it will still use the original
> disk image?
> 
> AFAIK, a transient guest only means it will disappear after the
> virtualisation session ends.

The disk image stays intact, just it'll disappear from libvirt's view,
hence you take a copy of your XML file below ('virsh dumpxml') does that

> 
> > virsh dumpxml $dom > file
> > virsh undefine $dom
> > virsh blockcopy $dom /ssd/image.raw /hdd/image.raw \
> >   --wait --verbose --pivot
> > virsh define file
> 
> I could not find anything about "pivot" or "pivoting"? What does --pivot
> do in this case?

When you "pivot", libvirt will use the new copy (i.e. /hdd/image.raw
in this case), and this is done while the guest is running.

Eric, please correct if I said something wrong.

-- 
/kashyap