16:01:03 <priteau> #startmeeting blazar
16:01:04 <openstack> Meeting started Thu Sep 10 16:01:03 2020 UTC and is due to finish in 60 minutes.  The chair is priteau. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:05 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:07 <openstack> The meeting name has been set to 'blazar'
16:01:33 <priteau> #topic Roll call
16:06:03 <tosky> (just saying hi, nothing to contribute :)
16:06:52 <priteau> Hi tosky
16:09:38 <diurnalist> o/
16:09:40 <priteau> Hi diurnalist
16:09:57 <diurnalist> just us today, I take?
16:10:33 <priteau> there's tosky too, he worked on zuulv3 migration
16:11:04 <priteau> #topic Victoria release
16:11:17 <priteau> So today is Victoria feature freeze day already :/
16:11:50 <priteau> Unless there's a push in the next couple of weeks, any new feature would have to wait for Wallaby
16:12:24 <diurnalist> sheesh, how time flies
16:12:57 <priteau> did you complete the usage enforcement code in Chameleon in the end?
16:13:26 <diurnalist> yes, we did actually. We are in the middle of testing it now (in production), but it's working well thusfar.
16:13:42 <diurnalist> before jake went on leave, he was working on buttoning up a changeset to propose
16:13:47 <diurnalist> which mostly entailed writing tests
16:14:14 <diurnalist> I'm not 100% sure if there was more needed on https://review.opendev.org/#/c/736993/
16:14:33 <diurnalist> from a cursory look it looks like more tests are needed at least
16:15:09 <priteau> At the very least fixing the failing CI checks
16:15:25 <priteau> That might be why Jake set the Workflow-1 flag on it
16:15:35 <priteau> The unit tests are passing, but no the tempest ones
16:16:10 <diurnalist> I see that jake had an updated changeset on our github repo that i don't think has made it up to gerrit
16:16:20 <priteau> And since no tempest scenario was added, I can only guess that it's caused by some code error in this patch
16:16:41 <priteau> If you think it can be submitted, please do
16:17:08 <diurnalist> ok
16:17:58 <priteau> It's also final release for client libs
16:18:00 <diurnalist> just looking through our github for any other things we've added to blazar...
16:18:22 <priteau> I reviewed and tested https://review.opendev.org/#/c/748540/
16:18:33 <priteau> it seems to work fine
16:18:44 <diurnalist> priteau: thank you for fixing the lint errors on that, and thanks for testing
16:19:01 <priteau> But tetsuro hasn't replied to my request for review
16:19:49 <priteau> I think I'll make an exception and approve it
16:20:20 <diurnalist> there is also the oslo.context patch i submitted to blazar, that one i am a bit more concerned about, how it may impact other deployments
16:20:32 <diurnalist> it's something that's difficult to have good test coverage for
16:22:11 <priteau> That one is more difficult
16:22:36 <priteau> But I think it can be considered as "not a feature" and merged in the coming weeks
16:23:45 <priteau> Are you running with this patch in prod?
16:23:58 <priteau> (the oslo context one)
16:28:15 <priteau> We've lost diunalist due to hardware issues
16:33:33 <priteau> Welcome back diurnalist!
16:33:38 <diurnalist> i am back, finally. my mac crashed spectacularly. to answer your question: yes, we run the oslo context patch in prod
16:33:54 <priteau> Well, that gives me more confidence in it
16:34:11 <priteau> I'll take a closer look at it
16:34:12 <diurnalist> i found one bug with it, which i fixed and i believe pushed as a new patchset
16:35:13 <diurnalist> yes, i did
16:35:43 <diurnalist> i may be misinterpreting the purpose of oslo.context
16:36:24 <diurnalist> but the changeset doesn't really alter _how_ the context is used, just how it is constructed
16:37:24 <priteau> I admit I don't really understand how oslo.context works :/
16:38:32 <priteau> Anything else that we should look into merging for Victoria?
16:38:50 <diurnalist> looking through our blazar-dashboard history now
16:40:00 <diurnalist> i think we have the resource API integration in the dashboard but don't have a patchset for that
16:40:30 <diurnalist> that feels like it's going to be a stretch. i think we should focus on the existing stuff in review and i'll try to get the enforcement stuff up
16:41:43 <priteau> At this point in the release cycle only minor changes will get approved, but you can still submit more patches, they'll be ready for Wallaby
16:42:14 <diurnalist> that was going to be my question--so feature freeze means freeze on anything that needs to be merged, even if it's in the review queue
16:42:46 <priteau> Well, any new feature that isn't yet merged has to wait for the next release
16:42:56 <priteau> Even if it was submitted a long time ago
16:43:10 <priteau> That's to focus on stabilising the software in the next few weeks
16:43:42 <priteau> But we can still start reviewing code for the next release already
16:44:01 <diurnalist> makes sense
16:44:02 <priteau> Talking about the PTG
16:44:07 <priteau> #topic Wallaby PTG
16:44:42 <priteau> I probably won't be able to make it due to our second child showing up just at the same time, so I've proposed not to meet (there's still a long to do list from the last PTG anyway)
16:44:59 <diurnalist> saw your email about that. yes, I think that makes sense
16:45:00 <priteau> Of course, if you would like to organise something, feel free to do so
16:45:17 <diurnalist> it sounds like it would be organising myself :)
16:45:50 <priteau> Yeah Blazar is too quiet at the moment :/
16:45:54 <diurnalist> are you still looking in to doing preemptible instances w/ blazar? it feels that at this point Chameleon is perhaps the only active contributor to the project
16:46:22 <priteau> I did submit the preemptibles patches just a few days ago, but they need more work
16:46:45 <priteau> https://review.opendev.org/#/c/749151/ and https://review.opendev.org/#/c/749152/
16:47:19 <diurnalist> thanks
16:47:35 <priteau> The work project that might have used this code was delayed so I haven't been able to spend much time on it yet
16:48:11 <diurnalist> do you have any idea of the new timeline?
16:48:36 <diurnalist> it's possible we could leverage this new ability to run htc computations on spare cycles
16:48:41 <priteau> No, but I'll try to get this in shape for an inclusion early in Wallaby
16:48:49 <diurnalist> ok
16:48:56 <priteau> The patches are very small so you should be able to cherry-pick them easily
16:49:06 <priteau> I would be glad to have your feedback
16:50:05 <priteau> Should we wrap up for today?
16:50:17 <diurnalist> yes, sounds good. lots to do!!
16:50:48 <priteau> Thanks diurnalist for joining, good talking with you
16:50:52 <priteau> #endmeeting