As always, if these topics interest you and you want to contribute to the discussion, feel free to join the next meeting:

Time: Thursdays, 15:30-16:30 UTC

Full minutes:

We had only about half the usual attendance on our Thursday meeting as people had conflicts and other hindrances. I joined it from an airport lobby. We still managed to do some good work.

== Task prioritization ==

Our main focus was on prioritizing the remaining tasks for Quickstart upstream transition. There are a few high priority items which we put in the Next column on the RDO Infra board. See all the outstanding "Q to U" (Quickstart to Upstream) cards here[1].

Some of these are simple and quick low hanging fruits, a few are bigger chunks of work that need a good attention, like making sure that our multinode workflow can be reproduced over libvirt for easier debugging.

== Quickstart extra roles ==

We pulled in all useful roles into the quickstart-extras repo when we created it, and it seems it might be better if a few very specialized ones would live outside of it.

One example is Raul's validate-ha role, which we will split off to speed up development, as most cores are not involved in this and gates are not testing it.

== Update on transitioning to the new Quickstart jobs ==

We will use the job type field from the upstream jobs to figure out which quickstart job config we have to use for gate jobs (not the job name).

In addition to this, Gabrielle will tackle the issue of mixing the old and new jobs, and run them in parallel, letting us transition them one by one. Details in the trello card[2].

== Gating improvement ==

I was part of a meeting last week where we tried to identify problem areas for our testing and came to the conclusion that the ungated openstack common repo[3] is sometimes the cause for gating breaks.

We should start gating it to improve upstream quickstart job stability.

Best regards,


OpenStack Development Mailing List (not for usage questions)

Reply via email to