[Openstack-operators] Anyone using live-migration?

Joe Topjian joe at topjian.net
Thu Feb 6 17:53:03 UTC 2014


Did you mean use_local?

I wasn't aware that a use_local option existed. Does this require a
nova-conductor process locally with each nova-compute process? If so,
doesn't that defeat the purpose of nova-conductor?


On Thu, Feb 6, 2014 at 6:18 PM, Jonathan Proulx <jon at jonproulx.com> wrote:

> Using local conductors fixes my issue (though clearly there's one
> under this)  I have (had) 12 conductor nodes for 60 physical systems
> plus 4 conductor workers running on the controller and so not sure why
> switching to conductor/force_local=true on the computes would fix
> it...
>
> On Tue, Feb 4, 2014 at 10:16 AM, Jonathan Proulx <jon at jonproulx.com>
> wrote:
> > On Tue, Feb 4, 2014 at 2:23 AM, Joe Topjian <joe at topjian.net> wrote:
> >> Your log in the bug report shows "network_migrate_instance_finish" as
> the
> >> point where migration is failing. I wonder if this is local to Neutron
> only?
> >
> > That may be the case.  I am using Neutron & I wouldn't be too
> > surprised if the set of people trying to use both Neutron and live
> > migration only had 2 members...
> >
> > -Jon
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20140206/846e06d8/attachment.html>


More information about the OpenStack-operators mailing list