On Mon, Jan 28, 2013 at 12:27:15PM +0800, Osier Yang wrote:
On 2013年01月26日 03:12, Laine Stump wrote:
On 01/24/2013 10:44 PM, Osier Yang wrote:
So you agreed with just using the "pool name and volume name"?
I think so. Unless you can think of a situation where the pool or volume name legitimately wouldn't be known, or would be required to be different from one machine to another in spite of the path/key/etc being the same.
Hum, this inspires me thinking about the migration. The source and dest host could have pool && vol with the same name. However, the vol's content could be different. I.E, in this case, we will need the globally unique $IDs (pool UUID, and/or vol key).
IMHO you are describing an administrative mis-configuration. You could just as easily provide a storage pool with the same UUID on two hosts, which has different storage. Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|