<p>The qcow2 tm driver takes advantage of qcow2 images. If you are not getting epilog state it means your vm isn't shutting down properly. Do you have acpi packages on your vm?</p>
<div class="gmail_quote">On May 7, 2012 4:48 AM, "Andreas Calvo" <<a href="mailto:andreas.calvo@scytl.com">andreas.calvo@scytl.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I does not get to the EPILOG state, so it's is not saving changes.<br>
Do I have to enable the qcow driver in the datastore drivers?<br>
--<br>
Andreas Calvo Gómez<br>
Systems Engineer<br>
Scytl Secure Electronic Voting<br>
Plaça Gal·la Placidia, 1-3, 1st floor · 08006 Barcelona<br>
Phone: <a href="tel:%2B%2034%20934%20230%20324" value="+34934230324">+ 34 934 230 324</a><br>
Fax: <a href="tel:%2B%2034%20933%20251%20028" value="+34933251028">+ 34 933 251 028</a><br>
<a href="http://www.scytl.com" target="_blank">http://www.scytl.com</a><br>
<br>
NOTICE: The information in this e-mail and in any of its attachments is<br>
confidential and intended solely for the attention and use of the named<br>
addressee(s). If you are not the intended recipient, any disclosure,<br>
copying,<br>
distribution or retaining of this message or any part of it, without the<br>
prior<br>
written consent of Scytl Secure Electronic Voting, SA is prohibited and<br>
may be<br>
unlawful. If you have received this in error, please contact the sender<br>
and<br>
delete the material from any computer.<br>
<br>
On Fri, 2012-05-04 at 14:03 -0400, Shankhadeep Shome wrote:<br>
> No you don't have to change image to persistent, it wouldn't change<br>
> the running vm anyways. After the shutdown there are two more states,<br>
> EPILOG and DONE. EPILOG is reached after the VM is completely shutdown<br>
> and reported back, DONE is after the VM files are deleted from the<br>
> system datastore, you should see both steps in the vm log. Maybe you<br>
> can post the last few lines of the vm log, just go<br>
> to /var/log/one/<vm#>.log on the opennebula server and post the last<br>
> 20 lines.<br>
><br>
> On Fri, May 4, 2012 at 5:07 AM, Andreas Calvo<br>
> <<a href="mailto:andreas.calvo@scytl.com">andreas.calvo@scytl.com</a>> wrote:<br>
> What I see when I run the SHUTDOWN command is:<br>
> Fri May 4 11:03:11 2012 [LCM][I]: New VM state is SHUTDOWN<br>
><br>
> Is there something else I should do?<br>
> Image is not marked as persistent, should it be changed?<br>
><br>
> On Thu, 2012-05-03 at 23:51 -0400, Shankhadeep Shome wrote:<br>
> > If you use a qcow backing store this is what happens in the<br>
> background<br>
> ><br>
> > qemu-img create -backing_store (your original image) -f<br>
> qcow2 (running<br>
> > image)<br>
> ><br>
> > when you save as its<br>
> ><br>
> > qemu-img convert (running image + backing store) -O qcow2<br>
> (new base<br>
> > image)<br>
> ><br>
> > The vm log should look like this... check to see if you have<br>
> errors.<br>
> > How big is your source image? Conversion can take a while<br>
> depending on<br>
> > your image size and backing store.<br>
> ><br>
> > Thu May 3 23:41:28 2012 [LCM][I]: New VM state is SHUTDOWN<br>
> > Thu May 3 23:41:48 2012 [VMM][I]: ExitCode: 0<br>
> > Thu May 3 23:41:48 2012 [VMM][I]: Successfully execute<br>
> virtualization<br>
> > driver operation: shutdown.<br>
> > Thu May 3 23:41:48 2012 [VMM][I]: ExitCode: 0<br>
> > Thu May 3 23:41:48 2012 [VMM][I]: Successfully execute<br>
> network driver<br>
> > operation: clean.<br>
> > Thu May 3 23:41:49 2012 [LCM][I]: New VM state is EPILOG<br>
> > Thu May 3 23:42:42 2012 [TM][I]: mvds:<br>
> > Moving /var/lib/one/datastores/0/88/disk.0 to datastore<br>
> ><br>
> as /var/lib/one/datastores/101/4f062daaf6ad2f47fd36c6b35a0bd56c<br>
> > Thu May 3 23:42:42 2012 [TM][I]: ExitCode: 0<br>
> > Thu May 3 23:42:43 2012 [TM][I]: delete:<br>
> > Deleting /var/lib/one/datastores/0/88<br>
> > Thu May 3 23:42:43 2012 [TM][I]: ExitCode: 0<br>
> > Thu May 3 23:42:43 2012 [DiM][I]: New VM state is DONE<br>
> ><br>
> > On Thu, May 3, 2012 at 2:44 PM, Andreas Calvo<br>
> > <<a href="mailto:andreas.calvo@scytl.com">andreas.calvo@scytl.com</a>> wrote:<br>
> > ><br>
> > > Hi all,<br>
> > > What is the procedure to save a running VM as an image if<br>
> it's<br>
> > running<br>
> > > with QCOW2?<br>
> > ><br>
> > > As per the documentation, I've used shutdown after issuing<br>
> save as,<br>
> > but<br>
> > > the image gets in a LOCKED stated and do not progress.<br>
> > > If the VM gets deleted, the image changes to READY but I<br>
> think it's<br>
> > just<br>
> > > a link to the old image used by the VM. Trying to start a<br>
> new<br>
> > template<br>
> > > using this new image throws an error.<br>
> > ><br>
> > > Tried with SHUTDOWN and STOP without luck.<br>
> > ><br>
> > > I guess I'm missing something, does anyone how to do it?<br>
> > ><br>
> > > Thanks in advance<br>
> > ><br>
> > > _______________________________________________<br>
> > > Users mailing list<br>
> > > <a href="mailto:Users@lists.opennebula.org">Users@lists.opennebula.org</a><br>
> > ><br>
> <a href="http://lists.opennebula.org/listinfo.cgi/users-opennebula.org" target="_blank">http://lists.opennebula.org/listinfo.cgi/users-opennebula.org</a><br>
> ><br>
><br>
><br>
><br>
<br>
</blockquote></div>