[openstack-dev] Continuous deployment - significant process change

Robert Collins robertc at robertcollins.net
Sun Aug 18 20:00:40 UTC 2013

On 1 May 2013 08:12, Sean Dague <sean at dague.net> wrote:

> I think the reality is that openstack tree and process is very close to what
> the CD folks want. The only thing that's causing some allergy among the core
> teams is the idea of features coming in in an off state, versus coming in
> default on so they get exercised by our normal system.

But they don't get exercised by our normal system today, and CD folk
/want them to be exercised/! Our normal system doesn't, for instance,
exercise qpid and 0mq [0mq is tested once a day]. Nor does it exercise
kvm, nor hyper-V. Smokestack checks Xen. We don't check 5000 node
swift cluster scaling and performance.

Can we test more of these? Yes, and we should. Features coming in
disabled has *no* impact on that : to use a feature in the gate just
enable it as part of the relevant test, /once/ it's mature enough to
enable at all. It will be just unit tested before that.


Robert Collins <rbtcollins at hp.com>
Distinguished Technologist
HP Converged Cloud

More information about the OpenStack-dev mailing list