[openstack-dev] [tripleo] Plan to switch the undercloud to be containerized by default

Emilien Macchi emilien at redhat.com
Sun Jul 15 20:37:01 UTC 2018

On Tue, Jul 10, 2018, 7:57 PM Emilien Macchi, <emilien at redhat.com> wrote:

> with [tripleo] tag...
> On Tue, Jul 10, 2018 at 7:56 PM Emilien Macchi <emilien at redhat.com> wrote:
>> This is an update on where things are regarding $topic, based on feedback
>> I've got from the work done recently:
>> 1) Switch --use-heat to take a boolean and deprecate it
>> We still want to allow users to deploy non containerized underclouds, so
>> we made this patch so they can use --use-heat=False:
>> https://review.openstack.org/#/c/581467/
>> Also https://review.openstack.org/#/c/581468 and
>> https://review.openstack.org/581180 as dependencies
>> 2) Configure CI jobs for containerized undercloud, except scenario001,
>> 002 for timeout reasons (and figure out this problem in a parallel effort)
>> https://review.openstack.org/#/c/575330
>> https://review.openstack.org/#/c/579755
>> 3) Switch tripleoclient to deploy by default a containerized undercloud
>> https://review.openstack.org/576218
It merged today, hopefully all CI jobs (including promotion) will continue
to run smoothly. Thanks everyone involved in this big effort!

>> 4) Improve performances in general so scenario001/002 doesn't timeout
>> when containerized undercloud is enabled
>> https://review.openstack.org/#/c/581183 is the patch that'll enable the
>> containerized undercloud
>> https://review.openstack.org/#/c/577889/ is a patch that enables
>> pipelining in ansible/quickstart, but more is about to come, I'll update
>> the patches tonight.
These scenarios are still on the edge of a timeout, we'll keep working on

>> 5) Cleanup quickstart to stop using use-heat except for fs003 (needed to
>> disable containers, and keep coverage for non containerized undercloud)
>> https://review.openstack.org/#/c/581534/
>> Reviews are welcome, we aim to merge this work by milestone 3, in less
>> than 2 weeks from now.
Since we merged the majority of the work, I think we can close the
blueprint and not require FFE.
Any feedback on that statement is welcome.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20180715/0604f738/attachment.html>

More information about the OpenStack-dev mailing list