Open Stack

On 9 May 2016 at 15:27, Jean-Philippe Evrard <jean-philippe at evrard.me>
wrote:

> I am using ansible for some time now, and I think the current default
> Ansible inventory system is lacking a few features, at least when working
> on an OpenStack cloud - whether it's for its deployment, its support or its
> usage.
> I'm thinking of developing a new inventory, based on (openstack-)ansible
> experience.
>
There were discussions at the summit around the implementation of an
inventory system to cater for many of the things you're outlining. I wasn't
party to them but as I recall there were both presentations and design
discussions and the effort is being led by a team of OSIC developers.

My view is that all of this most certainly does not belong in Ansible and
most of it does not belong in an Ansible dynamic inventory script either.
It should be handled somewhere else and Ansible should simply consume it.

> I think all the results from this etherpad could be either used for
> ansible upstream or for the openstack-ansible project, so that's why I'm
> spamming here. Thank for your help!
>
Specifically for OpenStack-Ansible we need something that's easily
implemented for small or dev/test environments, but ideally if a deployment
is using a complex production inventory system or a non-complex development
system the interactions should be the same. This is one of the reasons
we've discussed (in theory) the use of a dynamic inventory with the tooz
library - it can do file back-ends or more distributed back-ends.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20160509/f873ba63/attachment.html>