[libvirt] [PATCH 5/6] allow non-zero success values from a domain's MigratePerform

Daniel Veillard veillard at redhat.com
Wed Oct 7 14:05:53 UTC 2009


On Wed, Oct 07, 2009 at 03:09:17PM +0200, Paolo Bonzini wrote:
>
>>    Sounds fine by me, but I wonder about the change of semantic for
>>    client server not at the same version level. Could you explain what
>> might happen in those case ?
>
> Here is the result of my testing (source libvirtd is always new):
>
> source virsh     source method      destination libvirtd      result
> -----------------------------------------------------------------------
>    old              qemu://                 old              runs (bad)
>    old              qemu://                 new              runs (bad)
>    old            qemu+tcp://               old              runs (bad)
>    old            qemu+tcp://               new              runs (bad)
>    new              qemu://                 old              runs (bad)
>    new              qemu://                 new              paused (ok)
>    new            qemu+tcp://               old              runs (bad)
>    new            qemu+tcp://               new              paused (ok)
>
> So you do need not just a new daemon, but also a new virsh to fix the  
> bug.  However, the good news is that in no case the migration fails,  
> even when using an old virsh and connecting to the daemon via a remote  
> connection.

  Excellent, sounds fine then, but we need the rebased version once
Dan patches are applied.

   thanks !

Daniel

-- 
Daniel Veillard      | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
daniel at veillard.com  | Rpmfind RPM search engine http://rpmfind.net/
http://veillard.com/ | virtualization library  http://libvirt.org/




More information about the libvir-list mailing list