16:00:26 <gthiemonge> #startmeeting Octavia
16:00:26 <opendevmeet> Meeting started Wed Sep 20 16:00:26 2023 UTC and is due to finish in 60 minutes.  The chair is gthiemonge. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:26 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:26 <opendevmeet> The meeting name has been set to 'octavia'
16:00:31 <gthiemonge> \o
16:00:39 <johnsom> o/
16:00:42 <tweining> o/
16:01:39 <gthiemonge> #topic Announcements
16:01:47 <gthiemonge> * 2023.2 Bobcat Release Schedule: R-2
16:02:04 <gthiemonge> release patches for RC1 are merged
16:02:13 <gthiemonge> I think we're good to go for Bobcat ;-)
16:02:21 <gthiemonge> \o/
16:02:25 <johnsom> Nice
16:03:29 <gthiemonge> * 2024.1 Caracal vPTG
16:03:40 <gthiemonge> reminder: PTG will be held virtually October 23-27
16:03:49 <gthiemonge> I haven't booked our session yet
16:03:55 <gthiemonge> but I set up the etherpad for Octavia
16:04:02 <gthiemonge> https://etherpad.opendev.org/p/caracal-ptg-octavia
16:04:08 <gthiemonge> feel free to add your topics there!
16:05:04 <gthiemonge> any other announcements folks?
16:05:49 <tweining> no
16:07:51 <gthiemonge> I'm going to skip the CI status section, no update this week
16:07:56 <gthiemonge> #topic Brief progress reports / bugs needing review
16:08:11 <gthiemonge> FYI I found a bug with the member batch update API
16:08:23 <gthiemonge> there's race condition, it can make load balancer stuck in PENDING_UPDATE if your server is really loaded
16:08:33 <gthiemonge> https://review.opendev.org/c/openstack/octavia/+/895251
16:09:30 <tweining> I hate race conditions. good job finding it.
16:10:46 <gthiemonge> here's the launchpad
16:10:49 <gthiemonge> https://bugs.launchpad.net/octavia/+bug/2036156
16:12:57 <johnsom> I remember talking about that one. +2
16:13:54 <gthiemonge> I have an action item: check if other PUT methods have similar issues
16:15:25 <gthiemonge> #topic Open Discussion
16:15:43 <gthiemonge> maybe you read this email on the ML: "[octavia] state of taskflow jobboards"
16:16:18 <gthiemonge> they are asking clarification on the status of jobboard
16:16:31 <gthiemonge> it's marked as experimental in the doc but not in the config
16:16:44 <gthiemonge> my 2 cents:
16:17:05 <gthiemonge> we had a few bugs with the jobboard backends when it was added
16:17:24 <gthiemonge> I don't remember that we had any new jobboard-specific bug in the recent cycles (maybe 2 or 3 cycles?)
16:17:36 <gthiemonge> so for me, it's no longer experimental
16:17:42 <gthiemonge> (and we know that a few folks here are using it)
16:18:24 <gthiemonge> any opinions?
16:19:05 <johnsom> I haven't run with it a lot so I can't really comment much.
16:19:07 <tweining> I can't comment because I haven't even touched jobboard
16:19:15 <johnsom> I would like to hear from people that are running it
16:19:32 <gthiemonge> hmm I don't see them on IRC
16:19:40 <gthiemonge> (OVH Folks)
16:21:09 <gthiemonge> alright, anyway, we can discuss it at the PTG if they join
16:21:52 <gthiemonge> I will reply to this thread, I'll tell them that it will be discussed with operators
16:22:16 <gthiemonge> thanks guys
16:22:23 <tweining> I just saw https://review.opendev.org/c/openstack/octavia/+/873489 and that it has 2 CR+2s but no W+1 yet. are we still waiting for something?
16:22:37 <gthiemonge> there's a test commit I think
16:23:01 <gthiemonge> oh no there's a question from johnsom
16:23:06 <gthiemonge> johnsom: can you check my answer?
16:25:09 <johnsom> Yeah, that is fine.
16:25:55 <gthiemonge> thanks
16:26:03 <gthiemonge> ok, do you have any other topics?
16:26:40 <tweining> only one
16:27:32 <tweining> I had to rebase https://review.opendev.org/c/openstack/octavia/+/879749 again because of merge conflicts. it would be nice if you could review it. thanks gthiemonge for your review
16:28:06 <tweining> it looks huge but it isn't really.
16:29:40 <johnsom> Is this part of the PTI in any way?
16:29:57 <tweining> what is PTI?
16:30:13 <johnsom> https://governance.openstack.org/tc/reference/project-testing-interface.html
16:30:39 <tweining> well, I know lots of other Openstack projects use mypy
16:30:44 <johnsom> I just worry about adding some new thing that is more of a surface to break if we are the only project using it, etc.
16:32:06 <tweining> https://codesearch.opendev.org/?q=mypy&i=nope&literal=nope&files=&excludeFiles=&repos=
16:33:49 <tweining> as a bonus that patch fixes even a little bug
16:34:35 <tweining> https://review.opendev.org/c/openstack/octavia/+/879749/34/octavia/cmd/prometheus_proxy.py#297
16:35:52 <tweining> honestly, I find it frustrating that octavia tends to be reluctant to adopt new technologies
16:36:12 <gthiemonge> https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_998/879749/34/check/octavia-mypy/9980559/mypy-report/index.html
16:36:15 <gthiemonge> what about those reports
16:36:19 <gthiemonge> what would be the next step?
16:36:32 <johnsom> I think this patch is making a lot of unrelated changes too, like setup.cfg, etc.
16:36:54 <tweining> those reports only show how little typing information is in our code
16:36:56 <johnsom> This change is wrong: https://review.opendev.org/c/openstack/octavia/+/879749/34/octavia/cmd/prometheus_proxy.py#b742
16:38:01 <tweining> how is it wrong, "" is falsy
16:38:34 <tweining> line.replace(False, map_tuple[0]) would be wrong, which is why mypy complained
16:38:55 <johnsom> Before it was using False, for a boolean check. Changing it to an empty string is wrong.
16:39:19 <johnsom> 746 is the conditional that would block the line.replace from running
16:40:47 <johnsom> It looks like only nova, cinder, and a few oslo libraries use mypy
16:41:00 <tweining> the alternative is that I ignore that line using an ignore comment
16:41:08 <johnsom> Which is better than I thought frankly
16:41:20 <gthiemonge> maybe that should be discussed in the review
16:41:34 <tweining> I agree
16:42:08 <johnsom> Yeah. It wasn't a priority to get into the release, so I haven't looked at it
16:43:53 <gthiemonge> anything else?
16:45:21 <gthiemonge> ok I guess that's it for today
16:45:29 <gthiemonge> thank you guys!
16:45:33 <gthiemonge> #endmeeting