[openstack-dev] [nova] Devstack, cells and conductor

Andrew Laski andrew.laski at rackspace.com
Thu Mar 7 14:48:03 UTC 2013


On 03/08/13 at 01:10am, Kieran Spear wrote:
>Hi all,
>
>I'm working on cells support for devstack based on a review that was
>abandoned last year [1]. I hit an InstanceActionNotFound bug [2] but
>found that it's fixed by an in-progress review [3] (by the same author
>- thanks Andrew!). Has the review been forgotten? With the patch I was
>able to create an instance successfully after a stack.sh run.

I wouldn't say it's been forgotten, more a lack of prioritization on my 
part.  So if you're tackling this I am very thankful and happy to help 
in any way I can.  The review is essentially the same code as what I 
have at https://github.com/alaski/devstack/tree/cells_service but my 
github branch might have a few updates so you may want to look at that 
instead.

>
>This conductor stuff in Grizzly is still new to me. How does it fit in
>with cells? I take it we need one conductor service per cell/db, even
>at the top level cell? Any gotchas I need to be aware of?

We've only really run cells with the conductor local api, not running it 
separately, but there shouldn't be any issues running it separately.  
And you are correct on the setup, since each cell has its own db it 
would have its own conductor, and one at the top level.

>
>Cheers,
>Kieran
>
>[1] https://review.openstack.org/#/c/14978/
>[2] https://bugs.launchpad.net/nova/+bug/1132935
>[3] https://review.openstack.org/#/c/22886/
>
>_______________________________________________
>OpenStack-dev mailing list
>OpenStack-dev at lists.openstack.org
>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev



More information about the OpenStack-dev mailing list