Open Stack

On Fri, Dec 13, 2013 at 03:04:09PM +0100, Imre Farkas wrote:
> On 12/13/2013 11:36 AM, Jaromir Coufal wrote:
> >
> >*VERSION 0*
> >===========
> >Enable user to deploy OpenStack with the simpliest TripleO way, no
> >difference between hardware.
> >
> >Target:
> >- end of icehouse-2
> >
> >Features we need to get in:
> >- Enable manual nodes registration (Ironic)
> >- Get images available for user (Glance)
> >- Node roles (hardcode): Controller, Compute, Object Storage, Block Storage
> >- Design deployment (number of nodes per role)
> >- Deploy (Heat + Nova)
> 
> One note to deploy: It's not done only by Heat and Nova. If we
> expect a fully functional OpenStack installation as a result, we are
> missing a few steps like creating users, initializing and
> registering the service endpoints with Keystone. In TripleO this is
> done by the init-keystone and setup-endpoints scripts. Check devtest
> for more details: http://docs.openstack.org/developer/tripleo-incubator/devtest_undercloud.html

Excellent point Imre, as the deployment isn't really useable until those steps
are done.  The link to the overcloud setup steps is actually:
http://docs.openstack.org/developer/tripleo-incubator/devtest_overcloud.html
Very similar to what is done for the undercloud.

I think most of that logic could be reimplemented to be done via direct calls
to the API using the client libs vs using a CLI.  Not sure about
"keystone-manage pki_setup" though, would need to look into that.

--
-- James Slagle
--