I'm setting up to do live migrations. Running KVM and libvirt on Ubuntu
10.04.1
Guest vhsot23 has qcow2 on NFS share, accessible to both vhost23 and 24
/net/fs11/srv/customer/nfs/kvm/mem52.img
root at vhost23:~# virsh migrate mem52 qemu+ssh://vhost24/system
At this point, it just sits there. virsh list on both vhost systems show
mem52 in a paused state.
I check /var/log/libvirt/qemu/mem52.log on the target system vhost24 it
states...
qemu: warning: error while loading state for instance 0x0 of device
'ram'
Further, if I do virsh destroy mem52 on the target it cleans up fine,
but after Ctrl-C on the source, mem52 cannot be managed.
root at vhost23:~# virsh destroy mem52
error: Failed to destroy domain mem52
error: Timed out during operation: cannot acquire state change lock
Only way to recover is manually kill the kvm process id for that guest.
Please help me get this to work.
--
Mark D. Foster <mark at foster.cc>
http://mark.foster.cc/ | http://www.freegeekseattle.org/
