[openstack-dev] [Nova] New API requirements, review of GCE

Steve Baker sbaker at redhat.com
Wed Nov 20 03:09:29 UTC 2013


On 11/20/2013 03:54 PM, Clint Byrum wrote:
> Excerpts from Robert Collins's message of 2013-11-19 16:22:41 -0800:
>> On 20 November 2013 13:00, Sean Dague <sean at dague.net> wrote:
>>>> As long as the metadataservice doesn't move out :) - that one I think
>>>> is pretty core and we have no native replacement [configdrive is not a
>>>> replacement :P].
>>>
>>> Slightly off tangent thread.
>>>
>>> So we recently moved devstack gate to do con fig drive instead of metadata
>>> service, and life was good (no one really noticed). In what ways is
>>> configdrive insufficient compared to metadata service? And is that something
>>> that we should be tackling?
>> * The metadata service can be trivially updated live - and Heat wants
>> to use this to get rid of it's own metadata service... whereas config
>> drive requires unplugging the device, updating the data and replugging
>> - and thats a bit more invasive.
>>
> This one is key. Both Trove and Savanna have run into the same
> limitation as Heat: networks that cannot reach the API endpoints don't
> get to have their API specific Metadata that updates over time. By
> putting it in the EC2 metadata service, we can access it via the Neutron
> proxy and then Heat/Trove/Savanna can update it later to provide a
> control bus for in-instance tools.
>
>
I've got some nova changes which I will resurrect when I get the chance.

They allow nova instance userdata to be updated via the nova API:
https://review.openstack.org/#/c/53732/
https://review.openstack.org/#/c/49971/
https://review.openstack.org/#/c/49973/


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20131120/48844821/attachment.html>


More information about the OpenStack-dev mailing list