<p>I have solved this by upgrading to libvirt version 0.9.2 as it seems this behaviour was related to libvirt 0.8.3 which is the version installed on Ubuntu 10.10. I have also updated nodes' OS to ubuntu version 11.10.</p>
<div class="gmail_quote">On Aug 30, 2011 10:42 AM, "Alexandru - Florian Antonescu" <<a href="mailto:florinantonescu@gmail.com">florinantonescu@gmail.com</a>> wrote:<br type="attribution">> Hello,<br>> <br>
> I am having some problems while trying to migrate a virtual machine using<br>> OpenNebula 3.0 Beta 1 with NFS and KVM on Ubuntu 10.10. The error seems to<br>> be related to virsh restore (cannot close file: Bad file descriptor).<br>
> Does anyone have any ideas about what could be causing this?<br>> <br>> Thank you,<br>> Florian<br>> <br>> The error log follows:<br>> Tue Aug 30 06:31:20 2011 [LCM][I]: New VM state is RUNNING<br>> Tue Aug 30 06:33:54 2011 [LCM][I]: New VM state is SAVE_MIGRATE<br>
> Tue Aug 30 06:34:02 2011 [VMM][I]: save: Executed "virsh --connect<br>> qemu:///system save one-67 /srv/cloud/one/var/67/images/checkpoint".<br>> Tue Aug 30 06:34:02 2011 [VMM][I]: ExitCode: 0<br>> Tue Aug 30 06:34:02 2011 [LCM][I]: New VM state is PROLOG_MIGRATE<br>
> Tue Aug 30 06:34:02 2011 [TM][I]: tm_mv.sh: Will not move, source and<br>> destination are equal<br>> Tue Aug 30 06:34:02 2011 [TM][I]: ExitCode: 0<br>> Tue Aug 30 06:34:02 2011 [LCM][I]: New VM state is BOOT<br>
> Tue Aug 30 06:34:33 2011 [VMM][I]: Command execution fail: 'if [ -x<br>> "/var/tmp/one/vmm/kvm/restore" ]; then /var/tmp/one/vmm/kvm/restore<br>> /srv/cloud/one/var/67/images/checkpoint zrhsa 67 zrhsa; else<br>
> exit 42; fi'<br>> Tue Aug 30 06:34:33 2011 [VMM][E]: restore: Command "virsh --connect<br>> qemu:///system restore /srv/cloud/one/var/67/images/checkpoint" failed.<br>> Tue Aug 30 06:34:33 2011 [VMM][E]: restore: error: Failed to restore domain<br>
> from /srv/cloud/one/var/67/images/checkpoint<br>> Tue Aug 30 06:34:33 2011 [VMM][I]: error: cannot close file: Bad file<br>> descriptor<br>> Tue Aug 30 06:34:33 2011 [VMM][E]: Could not restore from<br>> /srv/cloud/one/var/67/images/checkpoint<br>
> Tue Aug 30 06:34:33 2011 [VMM][I]: ExitCode: 1<br>> Tue Aug 30 06:34:33 2011 [VMM][E]: Error restoring VM: Could not restore<br>> from /srv/cloud/one/var/67/images/checkpoint<br>> Tue Aug 30 06:34:33 2011 [DiM][I]: New VM state is FAILED<br>
> Tue Aug 30 06:34:33 2011 [TM][W]: Ignored: LOG I 67 tm_delete.sh: Deleting<br>> /srv/cloud/one/var/67/images<br>> Tue Aug 30 06:34:33 2011 [TM][W]: Ignored: LOG I 67 tm_delete.sh: Executed<br>> "rm -rf /srv/cloud/one/var/67/images".<br>
> Tue Aug 30 06:34:33 2011 [TM][W]: Ignored: LOG I 67 ExitCode: 0<br>> Tue Aug 30 06:34:33 2011 [TM][W]: Ignored: TRANSFER SUCCESS 67 -<br></div>