On Fri, Sep 16, 2011 at 08:35:15AM -0600, Eric Blake wrote:
On 09/16/2011 03:51 AM, Michal Privoznik wrote:
If a domain has inactive XML we want to transfer it to destination when migrating with VIR_MIGRATE_PERSIST_DEST. In order to harm the migration protocol as least as possible, a optional cookie was chosen. --- diff to v2: -Fix substitution on wrong place src/qemu/qemu_migration.c | 81 ++++++++++++++++++++++++++++++++++++++++++--- 1 files changed, 76 insertions(+), 5 deletions(-)
Looks right to me, but I'd feel more comfortable with a second ACK from danpb. I'm also hesitant on whether we should apply this now, or wait to post-0.9.5. While it is fixing a design flaw, it is not a regression fix, and feels more like a feature addition - if Dan could run this through his TCK suite to ensure no migration regressions, I'd feel a bit better about taking it now.
I tend to agree. The code looks reasonable, but I think we should wait till post 0.9.5, to give more time for testing. In particular the interoperability beteen different versions of libvirt when doing the migration 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 :|