14:00:30 <haleyb> #startmeeting networking 14:00:30 <opendevmeet> Meeting started Tue May 21 14:00:30 2024 UTC and is due to finish in 60 minutes. The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:30 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:30 <opendevmeet> The meeting name has been set to 'networking' 14:00:37 <haleyb> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, obondarev, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, amotoki, haleyb, ralonsoh 14:00:38 <mlavalle> \o 14:00:42 <slaweq> o/ 14:00:43 <frickler> \o 14:00:45 <obondarev> o/ 14:00:46 <ykarel> o/ 14:00:46 <mtomaska> o/ 14:00:57 <fkr> o/ 14:01:25 <lajoskatona> o/ 14:01:42 <haleyb> #topic announcements 14:01:50 <ihrachys> o/ 14:02:13 <haleyb> thanks for taking care of things last week, i had very little extra time 14:02:27 <haleyb> and thanks mlavalle for running the meetings 14:02:49 <mlavalle> :-) 14:03:16 <haleyb> Last week was the Dalmatian-1 milestone, so you've probably seen the releases for libraries 14:03:43 <haleyb> i didn't see one for neutron itself, but assume it will be soon 14:04:02 <haleyb> this week is R - 19 14:04:15 <haleyb> #link https://releases.openstack.org/dalmatian/schedule.html 14:05:18 <opendevreview> OpenStack Release Bot proposed openstack/networking-bgpvpn master: reno: Update master for unmaintained/wallaby https://review.opendev.org/c/openstack/networking-bgpvpn/+/911523 14:05:21 <haleyb> Reminder if you have a topic for the drivers meeting on Fridays, please add it to the wiki @ https://wiki.openstack.org/wiki/Meetings/NeutronDrivers 14:05:42 <haleyb> there will be a meeting this week pending we have a topic 14:06:00 <lajoskatona> ack 14:06:01 <mlavalle> haleyb: a lot of us RH people will be off this Friday 14:06:14 <mlavalle> that might impact quorum 14:06:31 <haleyb> mlavalle: thanks for letting me know, i might just cancel then 14:07:17 <haleyb> and depending on my schedule i might be out (again) next Tuesday, will let people know since that impacts this meeting 14:08:35 <haleyb> last week neutron had an issue with sqlalchemy 2.0 (conf file url bug), i think i've seen related changes in other network-* repos recently, please prioritize those since i'm assuming those gates are red 14:09:51 <haleyb> i didn't have any other announcements, did anyone else? 14:09:53 <lajoskatona> yes that one hit neutron-tempest-plugin also, but most of them are merged I beleive 14:10:32 <haleyb> lajoskatona: ack, thanks, if you know of others needing a +2 feel free to ping me 14:10:40 <lajoskatona> we can imagine how an eventlet change will look like if we have such hours with such small thing like SQLalchemy :P 14:10:53 <frickler> also the new ovsdbapp release seems to have some issue which ykarel is investigating? 14:10:55 <mlavalle> haleyb: I'll be glad to run the meeting again next Tuesday, if needed 14:11:15 <ihrachys> frickler test only and there's a neutron fix https://review.opendev.org/c/openstack/neutron/+/920101 14:11:15 <ykarel> yes reported https://bugs.launchpad.net/neutron/+bug/2066263 for ovsdbapp 14:11:19 <haleyb> mlavalle: thanks, i will let you know 14:13:11 <haleyb> ykarel: thanks for the bug 14:13:17 * haleyb is slow today 14:13:38 <haleyb> let's move onto bugs then 14:13:51 <haleyb> #topic bugs 14:14:37 <haleyb> i realize there was some coverage issues the past two weeks, thanks to mlavalle and rubasov for moving things around 14:15:27 <haleyb> what that means is the current schedule on the wiki needs to change 14:16:32 <haleyb> if it makes it easier, I can just do this week, which would leave the next few months the same, instead of moving everyone 14:16:57 <mlavalle> thanks! 14:17:21 <haleyb> since we did not hear from amotoki i will try and follow-up, but it might mean that we will lose one deputy 14:18:06 <lajoskatona> thanks 14:18:19 <haleyb> so i will update the schedule after this meeting to reflect that 14:18:48 <haleyb> so for next week, deputy will be lucasagomes - does that work for you lucas? 14:19:09 <haleyb> he might not be here 14:19:24 <haleyb> lets talk about the bugs 14:19:25 <mlavalle> haleyb: I'll ping him internally 14:19:28 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/QYRWPDRUP2ONHJXEH6GJPLT4EV524AWM/ 14:19:34 <haleyb> email from Bence 14:20:11 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2065821 14:20:17 <haleyb> cover job failures 14:20:48 <haleyb> from the bug, this is a 50% failure rate at the moment 14:21:48 <haleyb> sigh, what is the opensearch user/pass ? 14:21:57 <haleyb> #link https://opensearch.logs.openstack.org/_dashboards/app/login?nextUrl=%2F_dashboards%2Fapp%2Fdata-explorer%2Fdiscover%2F#?_a=(discover:(columns:!(_source),isDirty:!t,sort:!()),metadata:(indexPattern:'94869730-aea8-11ec-9e6a-83741af3fdcd',view:discover))&_g=(filters:!(),refreshInterval:(pause:!t,value:0),time:(from:now-10d,to:now))&_q=(filters:!(),query:(language:lucene,query:'message:%22Killed%22%20AND%20build_name:%22op 14:21:57 <haleyb> enstack-tox-cover%22')) 14:22:06 <slaweq> openstack/openstack I guess 14:22:10 <lajoskatona> openstack-openstack I think 14:22:23 <lajoskatona> at least this morning that allowed me in :-) 14:22:35 <haleyb> thanks, will need to save that 14:23:04 <haleyb> i have not looked, has anyone else? 14:23:41 <haleyb> maybe something for CI meeting, but is critical as affecting the gate 14:23:51 <haleyb> not a useful failure, just "Killed" 14:24:37 <haleyb> i will look after meeting 14:25:13 <haleyb> next bug 14:25:18 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2065743 14:25:25 <haleyb> CLI arguments for rbac create are misleading and possibly incorrect 14:26:16 <haleyb> from triage looks like we might need both a neutron and client change 14:27:28 <haleyb> can anyone take that one? 14:27:44 <slaweq> I may try but not this week probably 14:28:03 <haleyb> slaweq: ack, thanks 14:28:13 <slaweq> I will assign it to myself 14:29:06 <haleyb> the rest of the bugs seem to have owners and patches 14:29:15 <haleyb> any other bugs to talk about? 14:29:33 <lajoskatona> another one from last week not in the report: 14:29:53 <lajoskatona> https://bugs.launchpad.net/neutron/+bug/2065823 (neutron.tests.functional.agent.l3.test_legacy_router.L3AgentTestCase.test_legacy_router_lifecycle failed in _assert_internal_devices ) 14:30:27 <lajoskatona> it happened once as I seen so perhaps not high prio at the moment 14:31:37 <opendevreview> yatin proposed openstack/neutron master: [functional tests] compatibility with ovsdbapp>=2.7.0 https://review.opendev.org/c/openstack/neutron/+/920101 14:32:07 <haleyb> lajoskatona: i see your last comment there, maybe this is some sql race condition, will need to keep an eye on it 14:33:09 <haleyb> alright, next topic 14:33:13 <haleyb> #topic specs 14:33:19 <haleyb> #link https://review.opendev.org/q/project:openstack%252Fneutron-specs+status:open 14:33:21 <lajoskatona> +1, I attached the logs of the failing test 14:34:17 <haleyb> there is just one spec left, other change there is pre-commit hook 14:34:50 <haleyb> and i think we just had nits 14:36:02 <haleyb> maybe we can ping rbartzen to update? 14:36:46 <mlavalle> +1 14:37:15 <haleyb> racosta: can you update your remaining BGP spec? ^^ 14:37:50 <haleyb> moving on 14:38:03 <haleyb> #topic community-goals 14:38:40 <haleyb> slaweq or lajoskatona - any update? 14:38:40 <opendevreview> Ihar Hrachyshka proposed openstack/neutron master: tests: Use `del_fake_chassis` when deleting a chassis https://review.opendev.org/c/openstack/neutron/+/920103 14:39:19 <slaweq> Nothing from me this week 14:39:21 <lajoskatona> no news from me 14:39:39 <haleyb> ack 14:39:41 <racosta> Hi haleyb, sure. Sorry for the dalay in updating the spec, I'll update asap. 14:40:01 <haleyb> racosta: no problem, thanks for doing it 14:40:11 <haleyb> #topic on-demand 14:40:29 <haleyb> i don't see anything on the wiki, does anyone have a topic? 14:40:30 <mtomaska> I have one quick thing 14:40:59 <mtomaska> In the last PTG, we discussed python type annotation 14:41:11 <mtomaska> I started with getting some tooling in for neutron-lib 14:41:49 <mtomaska> and I wondering if we ever discussed(maybe TCs?) which type annotation tool we want to use 14:42:20 <mtomaska> mypy or pyright. Mypy is what I see other repos selected but I am also wondering if we ever considered pyright 14:42:30 <mtomaska> it has some Cons and Pros... 14:42:43 <haleyb> mtomaska: that's a good question and I don't know the answer, but have seen mypy mostly 14:42:51 <ihrachys> I think mypy was discussed before. I have no idea what pyright is and why it's better. perhaps you have and it justifies going the other way than other projects? 14:43:29 <ihrachys> patches are here btw in case others would like to see: https://review.opendev.org/q/topic:%22type_tooling%22 14:44:08 <mtomaska> Right, I have two patches just for tooling. They are basically similar. There are some pros and cons for Pyright 14:44:32 <ihrachys> what I'm missing in the patches (regardless of the tool to be used) is I don't see how the tooling is actually used. can we pick a module and convert it to the typing and show how the new tools warn / fail on violations in tests or wherever? 14:44:56 <mtomaska> yes ^, that is my next step 14:45:09 <ihrachys> mtomaska perhaps you could write up for others a comparison and then we can discuss after reading it? 14:45:18 <mtomaska> I can propose two patches. One against pyright and another mypy. Then we can decide which route to go 14:45:38 <haleyb> mtomaska: do other repos have voting/non-voting jobs? 14:45:45 <mtomaska> Excellent article here: https://github.com/microsoft/pyright/blob/main/docs/mypy-comparison.md 14:45:51 <haleyb> i would just use mypy if noone else is using pyright 14:46:11 <mtomaska> I am biased towards pyright but I dont want to go against what I see already in other openstack repos(mypy) 14:46:41 <ihrachys> right. there should be a good reason to not follow the herd 14:46:48 <ihrachys> a REALLY good reason 14:47:34 <mtomaska> the best I have is that it is "faster" and works better with IDEs 14:47:37 <mtomaska> thats is all 14:47:46 <ihrachys> mtomaska do you have an idea on the first guinea module for conversion and tool demo? 14:47:47 <slaweq> FWIW there is no any guidance about it from the TC. Projects can choose tools they want AFAIK 14:47:49 <mtomaska> so not very STRONG reasons 14:47:51 <haleyb> codesearch is spinning at the moment, but searching for pyright will find all the copyright strings 14:48:36 <mtomaska> I already searched, and I dont see any pyright mentions in the code base 14:49:51 <mtomaska> one disadvantage to pyright. Under the hood it is written in Javascript/TypeScript. We would still use the python module they provide but we will implicitly depend on npm... probably dont want that 14:50:19 <lajoskatona> +1 against npm in our deps 14:50:23 <mtomaska> mypy is pure python 14:50:30 <haleyb> mtomaska: seems mypy is the way... and thanks for working on this :) 14:50:31 <ihrachys> haha js hate ensues 14:51:25 <mtomaska> ok then seems like mypy then. I will propose some "demo" patches... 14:51:36 <haleyb> i have stayed away from js my entire career, no need to get closer now 14:51:44 <mlavalle> lol 14:51:44 <mtomaska> and I will keep pyright for personal use / future dreams :-) 14:52:00 <mtomaska> haleyb: you dont need to touch JS 14:52:30 <haleyb> ----------------------------------------------------------------| JS 14:52:33 <mtomaska> from the use perspective you dont realize its JS under the hood. 14:53:03 <mtomaska> ok thanks all. I will be pushing new patches to neutron-lib and making them Active 14:53:05 <haleyb> ok, again thanks for the work 14:53:10 <haleyb> any other topics? 14:53:12 <mlavalle> can I get this reviewd, please: https://review.opendev.org/c/openstack/neutron/+/918151? 14:53:12 <mtomaska> Ill update you on the next meeting 14:53:32 <ihrachys> I raised it before, but... we still have to resolve netaddr issues after 1.0.0 bump. I have patches that I raised before that would be nice to get some reviews on. also lajoskatona seems to've stepped in too with some. all can be found here: https://review.opendev.org/q/topic:%22netaddr1%22 14:54:12 <haleyb> ihrachys: ack, will look 14:54:39 <haleyb> ykarel: CI meeting online or in-person? 14:54:48 <ykarel> irc 14:55:04 <haleyb> great 14:55:32 <haleyb> ok, thanks for all the discussions and have a good week everyone 14:55:36 <haleyb> #endmeeting