16:00:19 <gthiemonge> #startmeeting Octavia
16:00:19 <opendevmeet> Meeting started Wed Oct 30 16:00:19 2024 UTC and is due to finish in 60 minutes.  The chair is gthiemonge. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:19 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:19 <opendevmeet> The meeting name has been set to 'octavia'
16:00:21 <gthiemonge> hi
16:00:30 <tweining> o/
16:00:56 <johnsom> o/
16:01:18 <gthiemonge> #topic Announcements
16:01:25 <gthiemonge> It's our first meeting post-PTG!
16:01:39 <gthiemonge> so welcome to the Epoxy Release Cycle
16:01:57 <gthiemonge> reminder: you can find the release schedule at https://releases.openstack.org/epoxy/schedule.html
16:02:28 <gthiemonge> you may have noticed that Epoxy-1 is in 2 weeks, we don't have any specific goals for this milestone, but it's a reminder that time flies :D
16:02:55 <gthiemonge> do we have any other announcement folks?
16:03:01 <johnsom> Nope
16:03:21 <tweining> most important is feature freeze end of feb, but afair new features in openstack client need to get in even earlier?
16:03:27 <gthiemonge> I'm going to skip CI status, no update since last meeting or PTG
16:03:50 <gthiemonge> client freeze is at the same date
16:03:56 <gthiemonge> but octavia-lib freeze is one week before
16:04:34 <tweining> ah, ok I confused the two then. it will be important for rate limiting probably
16:05:29 <gthiemonge> yeah and if you need some code in octavia-lib, we need to propose a new release so it's pulled by all the CI jobs
16:05:39 <johnsom> I think we have at least three features that need to land. SG for vips, SR-IOV for members, and rate limiting
16:06:03 <gthiemonge> it means a lot of reviews
16:06:29 <johnsom> Yeah, I hit a number of tempest reviews in the last week
16:07:28 <johnsom> I updated the review etherpad for those
16:07:33 <gthiemonge> thanks
16:08:59 <gthiemonge> #topic Brief progress reports / bugs needing review
16:10:14 <tweining> thanks for reviewing the typing patches. I've updated them
16:10:17 <johnsom> I have the RBAC switch to using keystone default roles by default (instead of advanced RBAC we have had) patch done. I think this is going to be one where we need to disable gate jobs temporarily to complete the switch due to the dependency chain.
16:10:37 <tweining> https://review.opendev.org/q/topic:%22mypy%22+project:openstack/octavia
16:10:56 <tweining> it did hurt a bit to loose the CR+2s though
16:11:04 <gthiemonge> johnsom: ack we can sync for this
16:11:49 <tweining> I can try to get rid of more "#type: ignore" comments if you prefer. I can probably do that using the typing.cast() trick.
16:12:46 <johnsom> I just worry that the code will be littered with those
16:12:52 <tweining> https://review.opendev.org/c/openstack/octavia/+/880986 also, I worked on integrating ruff into pre-commit, at least in our main repo
16:13:04 <johnsom> Over time we can improve that
16:13:10 <gthiemonge> FYI there are a few backports waiting on stable branches:
16:13:17 <gthiemonge> https://review.opendev.org/q/project:openstack/octavia+status:open+branch:%5Estable/.*
16:13:29 <gthiemonge> I'll take a look at the V-{1,2}
16:13:54 <johnsom> I really hope we can stop adding linting tools. They really put the releases at risk. It's fine to run them locally, but adding them to the gate is a huge risk
16:16:04 <tweining> I'd suggest we merge the pre-commit updates soon, so we have time to test it. There isn't any risk.
16:17:31 <gthiemonge> ack
16:17:43 <gthiemonge> #topic Open Discussion
16:18:05 <gthiemonge> do you have anything else for this week folks?
16:18:25 <johnsom> I don't think I have anything else
16:18:33 <tweining> btw. I configured it to use not only the ruff lint tool, but also the formatter. I noticed the PyCharm formatter sometimes formats in a way that our flake8 config doesn't like. I hope ruff does it better
16:19:01 <gthiemonge> does it mean that pre-commit reformats/updates our code?
16:19:16 <tweining> the code you commit
16:19:33 <gthiemonge> the file? or just the patch?
16:19:58 <johnsom> This is why I think these are a bad idea and unnecessary.
16:20:14 <tweining> that's a good question. I need to check
16:21:36 <tweining> chatgpt says it runs on whole files
16:22:11 <gthiemonge> lol
16:22:28 <gthiemonge> anyway we'll test it when we will review the patch
16:23:12 <tweining> golang has the "go fmt" tool. I think such official tool would be the optimal solution.
16:23:40 <johnsom> Like their crypto library?
16:24:23 <tweining> well, i mean the concept of having an official formatter for a language is a good idea.
16:25:05 <tweining> because I also think that many different linters,checker,formatter is a bad thing
16:25:56 <tweining> "go fmt" isn't great either. in fact, I am pretty disappointed by it tbh. but we digest...
16:26:17 <tweining> *digress
16:26:35 <gthiemonge> hehe
16:26:40 <gthiemonge> anything else guys?
16:26:45 <tweining> yes
16:26:49 <johnsom> I can't think of anything
16:27:06 <tweining> who will take care of abandoning old patches?
16:27:50 <gthiemonge> the PTL? or a volunteer?
16:28:28 <tweining> I think I can do it. I already figured out a way to automate it, but haven't tested it.
16:28:35 <tweining> one more thing
16:28:52 <tweining> I forgot to mention that I've updated the failover cb test patch
16:28:55 <tweining> https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/932061
16:29:12 <gthiemonge> ack
16:29:16 <johnsom> Please review them and don't just auto-abandon. There may be some that are valid and/or duplicates to other fixes that should be noted.
16:29:55 <gthiemonge> I think we need a quick glance at the list of patches before abandoning them
16:30:17 <tweining> I set health_check_interval to 300, but the test still fails. I should update the timeouts in the test as well, but I didn't have time to investigate
16:30:30 <gthiemonge> tweining: let's talk offline about the method we can use
16:30:41 <tweining> ok
16:33:18 <gthiemonge> ok I guess that's all for this week
16:34:31 <gthiemonge> thank you guys
16:34:36 <gthiemonge> #endmeeting