The current default behaviour of libvirt, courtesy of the libvirt-guests service, is to suspend any running VMs when the host is shutdown.
The only problem with this is that qemu doesn't seem to guarantee that a VM suspended with one version of qemu can be restored by a later version, so if you have a running VM but qemu has been updated since it was started then you may find that rebooting the host leaves you with an unbootable guest. My specific example was that this morning qemu-kvm updated from 0.15.0-1.fc15 to 0.15.0-0.3.201108040af4922.fc15 and shortly after that I restarted my host only to find that the VM that had been running for some time wouldn't start. Looking at the logs showed: Unknown savevm section or instance 'kvmclock' 0 load of migration failed The actual error in virt-manager or virsh was worse than useless, just saying that it had lost contact with the monitor: virsh # start dove error: Failed to start domain dove error: Unable to read from monitor: Connection reset by peer There also doesn't seem to be any way to tell libvirt to ignore the saved image and do a new boot. The only way I was able to get things running again was to find the saved memory image on disk and move it out of the way so that libvirt would boot the VM from the disk image instead of trying to restore it. Tom -- Tom Hughes ([email protected]) http://compton.nu/ _______________________________________________ virt mailing list [email protected] https://admin.fedoraproject.org/mailman/listinfo/virt
