<div class="gmail_quote">On Feb 12, 2008 10:35 PM, Daniel P. Berrange <<a href="mailto:berrange@redhat.com">berrange@redhat.com</a>> wrote: <blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <div class="Ih2E3d"> </div>It won't care what type of storage the original VM is using - it'll simply read the data off the original disk whether its a file or block device or LVM vol. If you want to use LVM volumes in the destination VM, then you currently need to manually create the LVM volume ahead of time - virt-clone itself will not create new LVM volumes for you. This is in the future feature list though, once libvirt storage management APIs are merged. </blockquote> <div> thanks, one more question what will be the virt-clone command to clone existing vm which FS is based on loop back files and using more then one file for root. var, and swap eg here is the entry from /etc/xen/vm.cfg which we are using kernel = "/boot/vmlinuz-2.6.21-2952.fc8xen" ramdisk = "/boot/initrd-2.6.21-2952.fc8xen-domU.img" memory = 256 name = "vm1" vif = [ 'ip=xx.xx.xxx.xxx' ] disk = ['tap:aio:/var/uml/vm1/root,sda1,w', 'tap:aio:/var/uml/vm1/var,sda2,w', 'tap:aio:/var/uml/vm1/swap,sda3,w'] root = "/dev/sda1 ro" on_reboot = 'restart' on_crash = 'restart' all our our virtual hosts we are using separate loop back files for root, var and swap. thanks. </div> </div>