Monday, 2026-02-16

opendevreviewMerged openstack/masakari master: Detect missing [taskflow] connection  https://review.opendev.org/c/openstack/masakari/+/97300103:22
opendevreviewMerged openstack/masakari master: docs: update information about cinder endpoint in verify  https://review.opendev.org/c/openstack/masakari/+/97465604:14
opendevreviewMerged openstack/masakari-monitors master: Migrate setup configuration to pyproject.toml  https://review.opendev.org/c/openstack/masakari-monitors/+/97333504:34
say3no_hi07:00
masahitohi07:01
say3no_thank you masahito 3, let's start meeting07:01
say3no_#startmeeting masakari07:01
opendevmeetMeeting started Mon Feb 16 07:01:54 2026 UTC and is due to finish in 60 minutes.  The chair is say3no_. Information about MeetBot at http://wiki.debian.org/MeetBot.07:01
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.07:01
opendevmeetThe meeting name has been set to 'masakari'07:01
say3no_#info https://etherpad.opendev.org/p/masakari-meeting-2026.107:02
say3no_#topic Elections07:03
say3no_As I was going through the upcoming schedule, I noticed the election was coming up, so I thought I'd raise it as a topic.07:04
noonedeadpunko/07:04
noonedeadpunksorry for being late and not keeping in line with meetings07:04
say3no_hi!07:05
noonedeadpunknomination period is coming to it;s end actually07:05
say3no_I was hoping to hear from the current PTL if they had any thoughts, but I'm wondering — what happens if no one ends up nominating?07:05
noonedeadpunkso it;s high time to get some candidate out for the project07:05
noonedeadpunkI can elaborate on this :)07:06
say3no_+107:07
noonedeadpunkso if project has no nominations, it appears on a TC table, where TC gives a call to gein community attention and a couple of weeks for potential volunteer to step in07:07
noonedeadpunkthought it's really better not to get to this point, and have a nomination right away07:07
noonedeadpunkand better to have an election, rather then a project without any07:07
noonedeadpunkand if nobody steps in within these couple of weeks, project goes to inactive state and then getting retired for the next cycle07:08
say3no_I see, thanks for the explanation.07:09
masahitoright. we need slot in the next ptg. someone should raise their hand.07:10
say3no_Just to confirm — PTLs are elected every six months, right?07:12
noonedeadpunkyes, correct07:12
noonedeadpunkand also there\s a possibility to move project to DPL model07:13
noonedeadpunkto spread an administrative workload between different people07:13
noonedeadpunkthough, I guess for moving PTL->DPL it either should be a current PTL initiative, or project should be grapped by community without active nominations for elections07:14
noonedeadpunkand I'd really discourage to get to the point of no nominations, as then things can be really unpredictable07:15
say3no_I see. I do have some interest in taking on the Masakari PTL role, but I'd like to get a bit more experience with how the community works before stepping up. I haven't even been through a full cycle yet.07:16
noonedeadpunkwell, then DPL can be smth to consider...07:17
noonedeadpunk#link https://governance.openstack.org/tc/reference/distributed-project-leadership.html07:17
noonedeadpunkor, we can do PTL with liasons instead07:17
noonedeadpunk#link https://docs.openstack.org/project-team-guide/ptl.html07:17
say3no_reading now ...07:19
noonedeadpunkI likely can help out with things like releasing, if needed07:19
say3no_thank you07:20
noonedeadpunkand likely some other things as well07:21
noonedeadpunkI just having troubes to follow the bi-weekly meeting schedule, but I just need to be more organized I guess :D07:22
noonedeadpunkas if I have to skip one week - then I'm completely lost07:22
say3no_Given the risk that Masakari could be retired if no one nominates, let me take a little time to think about the Masakari PTL role — though I know time is short. I also need to consider some internal matters at my company.07:25
noonedeadpunkbut I\d guess I\d encourage to take part in elections, given that youre eligible candidate (have some patches merged), and then sorting out liason part07:26
noonedeadpunkyeah, internal matters are important to consider, shure07:26
noonedeadpunk*sure07:26
say3no_okay07:26
say3no_#topic [FYI] Change Masakari Team meeting Agenda URL07:27
say3no_Next is just an announcement/FYI. Thanks in advance.07:28
say3no_#topic Need review open MRs for masaki-monitors 937674 and 90213907:29
say3no_This topic was probably added by Ivan Anfimov. It's about a review request for the introspectiveinstancemonitor in masakari-monitor.07:31
say3no_Does anyone here have experience with or knowledge about the introspective monitor?07:33
say3no_In preparation for this review, I've been looking into the current state of Masakari's test implementation and DevStack integration, and I'm setting up an environment in my homelab.07:33
say3no_This might be slightly off-topic from the review itself, but...07:34
noonedeadpunkI have never used introspective monitor tbh07:34
masahitome neither.07:35
say3no_Currently, I feel frustrated that Masakari lacks tempest tests and functional tests for masakari-monitor, which makes it difficult to do proper code-level reviews. I believe contributing these functional tests and improving the DevStack integration would benefit the community. What do you all think?07:36
noonedeadpunk937674 does make sense to me at least07:37
masahitothere was less activity in the last few cycles, so let's improve everything one by one, imo.07:37
masahitoyes sure.07:37
noonedeadpunkwell, I'd say it's hard to test this in current devstack implementation07:37
noonedeadpunkas that would kinda require multi-node setup07:38
noonedeadpunkso we could shut it down and see if things are get executed as expected07:38
noonedeadpunkas how would you really test anything on single node devstack...07:38
noonedeadpunkand unit tests is smth we can base review on...07:39
say3no_Maybe the real issue is that the unit test coverage for introspective monitor seems insufficient? As for testing on DevStack, it should be possible if you prepare an image with qemu-ga and craft the userdata, but it's still not easy either way.07:41
noonedeadpunkoh, yes07:42
noonedeadpunkand that is smth which should be possible to fix with not-so-high effort with AI...07:42
noonedeadpunkbut again, somebody need to understand code paths for the monitor beforehands...07:43
say3no_As a core reviewer, I'd like to give it a +2 since the code looks fine, but I also want to make sure it actually works without issues — since I'd be taking responsibility for the merge. That said, verifying it in my homelab rather than in the community DevStack would probably be sufficient.07:44
noonedeadpunk++07:44
masahitoagree and understood your feelings07:45
say3no_:)07:45
say3no_And it's tough for an individual to bear that cost alone, so in the future I'd like to shift that burden toward DevStack and functional tests.07:45
noonedeadpunkas for 902139 - I\d request a unit test coverage for the method07:46
noonedeadpunkI kinda still not getting how you plan to test things like hostmonitor or instance monitor with single node devstack, but yeah...07:47
say3no_thank you Dmitriy-san07:47
say3no_For the host monitor, I think we'd have no choice but to set up a three-node multi-node environment.07:48
noonedeadpunkI\d say 2 node should work?07:49
noonedeadpunkbut then the same setup should work for retrospective monitor as well07:49
noonedeadpunkso probably it does not make much sense to do smth unique for retrospective instance monitor which won't work for other monitors07:50
noonedeadpunkdunno though07:50
say3no_Well, if we don't worry about tricky stuff like split brain, 2 nodes should be fine.07:50
say3no_Well, I think I'll keep tinkering with my homelab a bit more.07:51
noonedeadpunkhm, is there anything in masakari actually dealing with split brains? As I was under impressions this is offloaded to clustering software, like consul or pacemaker07:51
say3no_That's a good point. I was unnecessarily expanding the scope there.07:53
noonedeadpunkI was thinking of optimizing utilization of CI resources :D07:54
say3no_Right, our resources are limited after all.07:56
say3no_Regarding this topic, I'll look into adding unit tests for 937674, and I'll re-review 902139 today and give it a +2 if everything looks good.07:56
say3no_I'd like to wrap up the meeting. Does anyone have anything else they'd like to bring up?07:57
masahitoone thing ptg entry07:58
masahitowe need ptg slot request until 1st May. https://openinfrafoundation.formstack.com/forms/apr2026_ptg_survey07:59
masahitos/May/March07:59
say3no_Is this something the PTL is supposed to submit? Or can we as a team go ahead and do it ourselves?08:00
masahitoPTL is supposed to submit it usually.08:01
noonedeadpunkideally - this is a PTL responsibility08:01
say3no_okay08:01
noonedeadpunkbut Id say we can ask for exception in registering the team after 1st of March08:02
noonedeadpunkgiven that somebody more active step in as a ptl :)08:02
say3no_Understood. As an action item, I'll check with the current PTL about their intentions. 08:04
masahitothanks08:05
say3no_any questions?08:05
say3no_okay, I close this masakari meeting 2026-02-16. thank you for comingOkay, I'm closing the Masakari meeting for 2026-02-16. Thank you all for joining.08:07
say3no_#endmeeting08:07
opendevmeetMeeting ended Mon Feb 16 08:07:23 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)08:07
opendevmeetMinutes:        https://meetings.opendev.org/meetings/masakari/2026/masakari.2026-02-16-07.01.html08:07
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/masakari/2026/masakari.2026-02-16-07.01.txt08:07
opendevmeetLog:            https://meetings.opendev.org/meetings/masakari/2026/masakari.2026-02-16-07.01.log.html08:07
masahitothank, bye08:07
opendevreviewSei Sano proposed openstack/masakari-monitors master: Handle exception from guest agent correctly  https://review.opendev.org/c/openstack/masakari-monitors/+/90213922:56
opendevreviewSei Sano proposed openstack/masakari-monitors master: Handle exception from guest agent correctly  https://review.opendev.org/c/openstack/masakari-monitors/+/90213923:09

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!