14:00:50 <haleyb> #startmeeting networking
14:00:50 <opendevmeet> Meeting started Tue Apr 16 14:00:50 2024 UTC and is due to finish in 60 minutes.  The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:50 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:50 <opendevmeet> The meeting name has been set to 'networking'
14:00:52 <ihrachys> o/
14:00:53 <haleyb> Ping list: bcafarel, elvira, frickler, mlavalle, mtomaska, obondarev, slaweq, tobias-urdin, ykarel, lajoskatona, jlibosva, averdagu, amotoki, haleyb, ralonsoh
14:00:55 <mlavalle> \o
14:00:58 <jlibosva> o/
14:00:58 <ykarel> o/
14:01:06 <rubasov> o/
14:01:09 <elvira> o/
14:01:22 <obondarev> o/
14:01:53 <frickler> \o
14:02:16 <haleyb> ok, i'll get started
14:02:18 <lajoskatona> o/
14:02:19 <haleyb> #topic announcements
14:03:02 <haleyb> this will probably be a short meeting as i've been out a few days and still going through my backlog, if there are fires somewhere please let me know
14:03:18 <haleyb> We are now in Dalmatian release week (R - 24)
14:03:40 <haleyb> thanks again for attending the PTG last week, I will work on getting a write-up done for the ML soon
14:04:36 <lajoskatona> thanks for moderating and driving it :-)
14:04:36 <haleyb> Reminder: If you have a topic for the drivers meeting on Friday, please add it to the wiki @ https://wiki.openstack.org/wiki/Meetings/NeutronDrivers
14:05:04 <haleyb> for now i don't see anything new on the RFE list, but will check thursday
14:05:29 <haleyb> lajoskatona: np, hopefully people found it productive
14:06:04 <bcafarel> late o/
14:06:14 <haleyb> i did not have any other announcements
14:06:27 <haleyb> bcafarel: just in time
14:06:31 <haleyb> #topic bugs
14:06:53 <bcafarel> :)
14:06:55 <haleyb> bcafarel was bug deputy last week, his report is at
14:06:59 <haleyb> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/YBZCITD44VLW2LGSHTOJUJSBLARL4SFQ/
14:07:29 <haleyb> there were a couple of unassigned that might need owners
14:07:46 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2059405
14:07:54 <haleyb> OVN DNS not working as documented
14:08:48 <haleyb> the best I could do is boot a devstack and try some of the commands, i saw similar failures
14:09:18 <haleyb> but it needs an owner
14:10:36 <haleyb> it could just be a config issue...
14:10:51 <ykarel> haleyb, so you also saw that REFUSED responses?
14:10:57 <ykarel> in devstack tests
14:11:52 <haleyb> ykarel: i believe so, i had to use nslookup since it was cirros but responses were similar
14:11:58 <ykarel> and i recall ovn responses only when there was some nameserver defined in /etc/resolv.conf
14:12:19 <ykarel> haleyb, ack, so may be there some issue then
14:12:38 <ykarel> i can try to look at it
14:12:56 <frickler> the .local domain could also be treated special
14:13:28 <haleyb> maybe, the bug was a kolla deploy so if we could reproduce on a "real" deployment would be good
14:15:03 <haleyb> ykarel: thanks, i'm watching it and can respond to any questions on my setup once it's up again
14:16:00 <haleyb> frickler: that's a good question, not sure about .local exactly
14:16:36 <haleyb> next bug is
14:16:39 <ykarel> haleyb, ack
14:16:40 <haleyb> #link https://bugs.launchpad.net/neutron/+bug/2060828
14:16:49 <haleyb> [OVN] Changing public network MTU does not update router port gateway_mtu option
14:19:26 <haleyb> bug has steps to reproduce, just need to verify behavior
14:20:33 <haleyb> i can try and look but might be a few days
14:21:57 <haleyb> the only other new bugs were related to floating ip API behavior
14:22:11 <haleyb> we talked about this at PTG, just need to wait for nova to respond
14:22:37 <haleyb> not sure there is anything broken in the neutron API at least
14:23:28 <haleyb> any other bugs for discussion?
14:24:13 <haleyb> this week elvira is the deputy, next week is slaweq
14:24:49 <haleyb> Current bug count this week: 720, same as from last week (i will work on getting better metrics :)
14:25:24 <haleyb> #topic specs
14:25:38 <haleyb> #link https://review.opendev.org/q/project:openstack%252Fneutron-specs+status:open
14:25:50 <elvira> ack :)
14:26:20 <haleyb> we merged one spec last week, just one remaining, is also BGP-related
14:26:35 <haleyb> if anyone has cycles
14:26:39 <haleyb> #link https://review.opendev.org/c/openstack/neutron-specs/+/899210
14:26:43 <haleyb> elvira: thanks!
14:27:14 <haleyb> ok, moving on
14:27:20 <haleyb> #topic community goals
14:28:04 <haleyb> i did see slaweq's email regarding Manager S-RBAC role
14:29:11 <haleyb> lajoskatona: is there a new sdk version such that you can continue horizon work?
14:29:16 <lajoskatona> I will propose a release patch for openstackSDK to have a release for https://review.opendev.org/c/openstack/openstacksdk/+/909656 and push Horizon work forward
14:29:28 <lajoskatona> no new version
14:29:53 <haleyb> ack, thanks
14:30:13 <haleyb> #topic on_demand
14:30:24 <haleyb> does anyone have a topic?
14:30:29 <jlibosva> I have one if I may :)
14:30:48 <haleyb> sure, go ahead
14:31:22 <jlibosva> IIRC it was mentioned at the PTG and a few minutes ago too, that we're at 720 open bugs atm and at the last PTG time frame the bug count was roughly the same
14:31:55 <jlibosva> so I thought about it and quickly scrubbed through the bug list and I see the oldest bug is from 2014 and roughly 450 bugs are 2 years old and older
14:32:22 <haleyb> jlibosva: yes, the 720 number was the same as last cycle
14:33:00 <jlibosva> I know what some projects like Fedora close bugs that were reported against a release version that is EOL - so that made me thinking - we are most likely not gonna solve all those 700+ bugs and lot of them lose attention over time
14:33:09 <haleyb> there are a lot of bugs that old, i'm not sure we came to a conclusion as what to do with those as it will take a lot of resources to just reproduce each one
14:33:38 <jlibosva> so I thought what if we just brute-force close bugs that are X-years old or older, and do some periodic bug scrubbing of low priority/wishlist bugs
14:34:04 <jlibosva> it seems we did good job with triaging so we have usually the importance set
14:34:07 <lajoskatona> isn't there some launchpad macro or setting for that?
14:34:37 <haleyb> jlibosva: i think that is a good idea
14:34:39 <ykarel> +1 sounds good, like we do auto abandon for reviews which are stale
14:34:39 <jlibosva> I don't know that, I wanted to bring up this idea as it seems if the bug count is steady over cycles then it is not gonna change and we have to do something differently :)
14:35:02 <jlibosva> there is also the "heat" of bugs which seems to be mostly low (6) in most cases
14:35:04 <lajoskatona> +1
14:35:24 <jlibosva> and if some bug we close was actually important and bothered people, it's eventually gonna come back to us
14:35:33 <haleyb> i know there is an abandon script for reviews at least, and i've seen bugs get closed, maybe depends on state?
14:36:01 <jlibosva> I haven't inspected the tooling :) sorry
14:36:12 <haleyb> jlibosva: right, and we can close with that type of comment "please re-open if necessary..."
14:36:21 <jlibosva> yep
14:36:36 <ykarel> i think incomplete bugs get cloes automatically after some time
14:36:43 <jlibosva> so my idea was having like two buckets - old bugs and low/whishlist bugs
14:36:46 <haleyb> we only have tools/abandon_old_reviews.sh
14:37:03 <jlibosva> the old bugs closed automatically - that would put us to roughly 270 bugs to go over manually
14:37:22 <ihrachys> this is a good idea; close and ask to reopen if can reproduce on latest. for scrubbing - will need space allocated for this somewhere.
14:38:32 <jlibosva> right, so for example this meeting got to open agenda half way through. maybe we could use the extra half hour or do some dedicated meeting - like CI meeting
14:38:34 <ihrachys> maybe PTL could declare a bug fest maybe and we would block the time for this as a group
14:39:00 <haleyb> ihrachys: bug fest is every day! :)
14:39:18 <ihrachys> it's every day, then it's never the day :)
14:39:51 <haleyb> last cycle we did have review day(s), i am fine doing a bug fest if people find it useful
14:40:50 <haleyb> the other option if have the weekly bug deputy close XX old bugs during their cycle
14:42:11 <lajoskatona> for me the later one can also work
14:43:05 <haleyb> ihrachys: i do like the bug fest idea, just have to make sure to not duplicate effort, like two people looking at same bug (if even for 10 minutes), then closing
14:43:11 <jlibosva> or every contributor can check "Assigned bugs" - there is a filter prepared - like I just looked and I have some, maybe I'm the only one, dunno - but maybe we could just get back to those that are assigned to us
14:43:36 <jlibosva> in my case it's just that I forgot and never got back, or had a patch that got merged but LP was not updated, or patch abandoned but LP remained in In Progress
14:45:05 <haleyb> jlibosva: ack, that would be a good idea for everyone, especially with caracal release the bot should have closed anything that merged
14:46:16 <haleyb> i know rodolfo had a list of bugs he closed in an etherpad, i can create one or a query with "stale" bugs
14:47:00 <haleyb> and then can pick a day/days for people to look and close
14:48:39 <haleyb> jlibosva: thanks for the suggestion, I will work on something official, at minimum we should have a >5 years type query
14:48:58 <jlibosva> ack, thank you
14:49:14 <lajoskatona> thanks for the discussion and ideas
14:50:03 <haleyb> any other topics?
14:50:15 <haleyb> if not there is CI meeting in :10
14:50:38 <haleyb> ykarel: video or irc?
14:50:45 <ykarel> video
14:51:18 <haleyb> ok, thanks everyone for coming
14:51:23 <haleyb> #endmeeting