16:07:22 <priteau> #startmeeting blazar
16:07:23 <openstack> Meeting started Thu Jul  2 16:07:22 2020 UTC and is due to finish in 60 minutes.  The chair is priteau. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:07:25 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:07:27 <openstack> The meeting name has been set to 'blazar'
16:07:49 <priteau> #topic Roll call
16:08:05 <diurnalist> o7
16:08:38 <priteau> Hi diurnalist
16:09:37 <priteau> #topic Code contributions
16:10:02 <priteau> Sorry for starting late
16:10:36 <diurnalist> no worries!
16:10:59 <priteau> So good news, the Zuul checks for Blazar which had been broken for a while are not back to green
16:11:32 <diurnalist> s/not/now? :D
16:11:42 <priteau> *now* indeed
16:11:47 <priteau> I took a first look at the code that jakecoll submitted for usage enforcement: https://review.opendev.org/#/c/736993/
16:12:10 <priteau> Looks like there is still some work needed as it fails with TypeError: pickup_hosts() takes 3 positional arguments but 4 were given
16:12:12 <priteau> See https://49000a15860736fb0af7-30077d6e59c1ab878410714bbe057b1e.ssl.cf2.rackcdn.com/736993/1/check/blazar-devstack-dsvm/3480391/logs/screen-blazar-m.txt
16:13:01 <jakecoll> o/
16:13:21 <priteau> I wanted to leave a few comments on the code and then realised diurnalist had similar comments already made on GitHub @ https://github.com/ChameleonCloud/blazar/pull/31
16:13:47 <jakecoll> Yea, I set the workflow to -1. Still a work in progress.
16:13:53 <priteau> So it would make sense to solve pending comments there and resubmit to gerrit afterwards
16:14:31 <jakecoll> I think so. We're working on our external service filter so that should provide the best test for how that filter will work.
16:15:52 <priteau> The overall design looked OK, my comments were more about specific bits of the implementation, coding style, etc.
16:16:23 <priteau> I've also set https://review.opendev.org/#/c/731586/ as review priority but haven't been able to look into it yet
16:17:09 <diurnalist> thanks
16:17:31 <priteau> Anything else that we should highlight?
16:18:54 <diurnalist> I don't think there are any other patches we have in flight
16:19:24 <diurnalist> it might be necessary to soon propose a spec for out-of-tree blazar plugins
16:19:53 <diurnalist> if you have initial opinions on that design, especially if they are strongly held, it would be helpful to know them
16:20:33 <diurnalist> (by out-of-tree, for any who might not know what that means, i just mean "plugins not kept in the blazar repo")
16:21:11 <priteau> No strong opinions. I would suggest looking at how Neutron did it for its stadium extensions
16:23:13 <diurnalist> ok, i will do that
16:23:24 <diurnalist> thanks!
16:23:34 <priteau> One of the neutron-*aas repos might be a good example
16:23:56 <priteau> It depends how flexible we want to be with extending the API
16:24:32 <priteau> We could go with allowing plugins to extend the API freely, or keep them to well-defined layout
16:25:18 <priteau> e.g. you can register new resources but they will follow the same pattern as /os-hosts or /floatingips
16:25:44 <diurnalist> yeah, the API surface implications are particularly a bit tricky to consider
16:29:01 <diurnalist> that's all I have, still have much to kick around w/ design on that
16:29:11 <priteau> Thanks
16:29:26 <priteau> Nothing special from me this week
16:29:44 <priteau> I will likely be off in two weeks time so I suggest we cancel the next meeting
16:30:42 <priteau> Next meeting would be on July 30
16:31:36 <priteau> Let's wrap up then
16:31:50 <priteau> Thanks for joining, let's keep in touch via Gerrit in the meantime
16:31:53 <priteau> #endmeeting