[openstack-dev] [kolla] How to handle data containers

Steven Dake (stdake) stdake at cisco.com
Thu Aug 13 14:50:11 UTC 2015


Hi,

There has been a lot of debate in the Kolla community about how to handle data containers but it has happened all over the place (irc, various reviews, etc).  I’d like to have one record of the various tribal knowledge on this subject.

The latest proposal is to create one data container which is very small in nature and can be backed up with no VOLUME in the Dockerfile.  Next start that container image with different names for all of the various data containers we operate.  This should provide a good backup solution for the data containers, and removes the current extra stuff hiding in the various API containers that we use as data containers now in the ansible deployment.

I have created a blueprint:
https://blueprints.launchpad.net/kolla/+spec/one-data-container

It is in the discussion phase.  If there is any technical reason not to do this, lets get it out on the table now, otherwise I will approve the blueprint early next week and someone (perhaps me?) will begin the implementation.  The goal is just to simplify the code base and get rid of the data container bloat. ;)

Regards
-steve
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150813/6dd2f552/attachment.html>


More information about the OpenStack-dev mailing list