Open Stack

If you have a pretty simple schema that isn't tightly integrated into
the broader neutron schema, you could could always just resort to
detecting of the schema during your ml2 driver initialization and just
create the schema with SQL statements right then.

The advantage to that approach is that you don't have to worry about
messing with the deployed alembic scripts.
The disadvantage is that it's not managed by alembic, so you would
have to manually handle upgrades yourself.

On Tue, May 12, 2015 at 10:42 AM, Alexey I. Froloff <raorn at raorn.name> wrote:
> Greetings!
>
> I am developing ML2 type/mech plugin for some very special
> environment.  Because this environment is very special (using
> addressing based on physical hypervisor location), it will be
> separate package, I don't plan to alter Neutron code.  And this
> plugin needs to store some information in Neutron database.  Not
> altering existing tables, but adding couple.
>
> I'm not very good with alembic (better say not good at all),
> maybe someone can enlight me, if this possible at all and how to
> make it with minimal damage.  I am writing this plugin for Kilo
> release, and this installation is supposed to be further updated
> to next OpenStack releases.
>
> --
> Regards,    --
> Sir Raorn.   --- https://plus.google.com/+AlexeyFroloff
>
> __________________________________________________________________________
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>



-- 
Kevin Benton