15:00:13 #startmeeting ironic 15:00:13 Meeting started Mon Jul 26 15:00:13 2021 UTC and is due to finish in 60 minutes. The chair is dtantsur. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:13 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:13 The meeting name has been set to 'ironic' 15:00:20 o/ 15:00:21 #chair TheJulia 15:00:21 Current chairs: TheJulia dtantsur 15:00:24 o/ 15:00:24 o/ 15:00:25 o/ 15:00:25 hello everyone! 15:00:32 welcome to the most ironic of all meetings 15:00:36 o/ 15:00:38 #link https://wiki.openstack.org/wiki/Meetings/Ironic is our agenda 15:00:39 o/ 15:00:44 wait!?! We added more ironic?!? 15:00:53 * TheJulia is confused with the levels of ironic 15:00:55 there is never enough, is there? 15:01:12 #topic Announcements / Reminder 15:01:13 \o 15:01:22 mmm, I wonder if #topic works any more :) 15:01:28 anyway 15:01:34 #info Spring 2 releases have been done! 15:01:44 thanks all for your contributions! 15:01:59 #info Please add ideas vote for the midcycle slot: https://etherpad.opendev.org/p/ironic-xena-midcycle 15:02:13 #link https://doodle.com/poll/mhh959u4s4rturxi doodle for midcycle 15:02:20 * dtantsur suspects some slots are already in the past 15:02:34 #info Please add your ideas for the PTG: https://etherpad.opendev.org/p/ironic-yoga-ptg 15:02:43 As a reminder, it will be in October, 18-22 15:03:11 anything else to announce or remind of? 15:03:55 #topic Review action items from previous meeting 15:04:00 Nothing from my end. 15:04:03 iurygregory has created a midcycle doodle, thank you! 15:04:28 #topic Review subteam status reports (capped at ten minutes) 15:04:41 #link https://etherpad.opendev.org/p/IronicWhiteBoard starting on line 66 15:05:00 the number of bugs has reduced. if somebody has been triaging/fixing them - thank you! 15:05:58 JayF, zer0c00l, how has the story if the anaconda CI finished? I don't remember the last week well :) 15:06:27 AFAIK, there is no planned time for us to spend on additional anaconda ci 15:07:00 TheJulia said she'd give the CentOS people a heads up about the artifacts we'd need to avoid doing image-building work in the gate 15:07:03 dtantsur: triaging/cleaning up as of recent 15:07:24 nice re cleaning up! 15:07:39 I'm not quite sure I remebering committing to that 15:07:46 JayF: not sure about the others, I feel pretty uneasy about having the whole deploy interface not really covered 15:07:47 s/remembering/remember/ 15:08:17 We likely need to have real discussion with opendev folks though, since it is a narly build in CI for us to build them every time 15:08:20 dtantsur: I don't like the situation either; but we just don't have the time, and building CI went up an order of magnitude in complexity when we realized that centos does not publish any compatible image 15:08:39 JayF: what exactly do you lack? 15:08:48 A second stage image 15:08:54 It's documented in the storyboard story linked in the subteam report 15:08:57 what is it technically? a root fs? 15:09:00 yeah 15:09:18 *but* that can be generated from centos native hosts if memory serves, and someone has recently posted some centos jobs to a number of projects.... 15:09:29 building it is documented in our docs, but the tools needed (yum/yum-utils) are not in the Ubuntu in our gate which is the first problem :( 15:09:41 JayF: this ugly stuff cannot help you? https://opendev.org/openstack/metalsmith/src/branch/master/playbooks/integration/centos-image.yaml#L85-L125 15:09:58 mm, yeah. we can have a bifrost job, it works on centos. 15:10:07 but note that centos8 doesn't have yum-utils either. 15:10:15 dtantsur: hm, I don't know. 15:10:24 It's all documented in that story, I haven't looked at it since filing that 15:10:33 JayF: that's what I did in metalsmith to get a kernel/ramdisk/root from a whole disk image 15:10:57 It looks closer than anything I've seen, but I just don't know for sure 15:11:36 Perhaps table this for now and schedule a specific time to get some contributors on a call to at least brainstorm and sort through it? 15:11:54 or maybe midcycle topic? 15:12:02 yeah, but let's not drop this on the floor. we have enough untested corners. 15:12:05 midcycle ++ 15:12:28 * TheJulia opens the etherpad to add it 15:12:36 dtantsur: that's why I documented the crap outta the problem and put it on the etherpad :) 15:14:02 okay, adding 15:14:46 arne_wiebalck: have you heard anything from kaifeng about the node history? 15:14:55 arne_wiebalck is on vacation 15:14:57 TheJulia: have you chatted with janders re cleaning? 15:14:59 ah 15:15:01 s/vacation/holiday/ 15:15:05 dtantsur: I have not 15:15:13 my last week was slammed 15:15:26 I see. I'll try to remember (maybe he'll see the ping) 15:15:43 hopefully 15:16:15 \o/ re RBAC done 15:16:32 yay! 15:16:33 would it be too much to ask for Bifrost support? ;) 15:16:54 dtantsur: uhhhh 15:17:04 the answer "yes" is perfectly acceptable :) 15:17:06 dtantsur: lets midcycle that because that would require some policy enforcement internals 15:17:20 as an MVP we can only do it when keystone is enabled 15:17:37 I guess I feel the need to have a clarifying doc written for rbac 15:17:52 oh, at least fix the 2 conflicting docs 15:17:55 since presently the policy enforcement code is bypassed in noauth model 15:18:03 there was a patch for it, seems abandoned though 15:18:05 or non-keystone auth models 15:18:11 orly? 15:19:18 https://review.opendev.org/789082 15:20:01 I guess that is a fair change to perform 15:20:41 \o/ ask refined 15:20:45 okay, is everyone done with the statuses? 15:20:50 we can continue the RBAC chat later 15:21:17 ++ 15:22:07 #topic Deciding on priorities for the coming week 15:22:25 I have 3 CI patches https://review.opendev.org/c/openstack/ironic-python-agent/+/801898 https://review.opendev.org/c/openstack/ironic-inspector/+/801873 https://review.opendev.org/c/openstack/ironic/+/801876 15:22:51 Riccardo Pittau proposed openstack/python-ironicclient master: Fix for missing fonts in PDF jobs https://review.opendev.org/c/openstack/python-ironicclient/+/802324 15:22:52 and https://review.opendev.org/c/openstack/ironic/+/800084 from cenne 15:23:03 any objections? 15:23:22 none 15:23:25 nope 15:23:58 We've got some backports on to ussuri that would be good, but they are presently failing CI, it appears due to random timeouts :\ 15:23:59 Ruby Loo proposed openstack/ironic master: Minor updates to anaconda doc https://review.opendev.org/c/openstack/ironic/+/802314 15:24:11 umm.. 15:24:15 * cenne raises hand 15:24:18 should we remove https://review.opendev.org/c/openstack/ironic-specs/+/764801 for now? it has been without updates for months 15:24:38 cenne: we're listening :) 15:24:39 sounds good 15:24:52 yeah, agree to remove 15:25:14 so, there's a few tests that are currently commented out that i was wondering if make sense to just remove 15:25:32 interesting, where is that? 15:25:44 and are there comments accompanying them? 15:26:32 ironic/tests/unit/conductor/test_manager.py Line 513, 604. 15:27:06 https://review.opendev.org/c/openstack/ironic/+/800084/11/ironic/tests/unit/conductor/test_manager.py 15:27:26 ah, you mean, in your patch, not in git master? 15:27:31 TheJulia: Sorry, I was going to ask today/this-week 15:27:41 Yes in the patch. 15:27:51 let's discuss it on the patch then? 15:27:58 okay. 15:28:00 ++ 15:28:19 anyone has anything else for the priorities? 15:29:14 not today 15:30:25 #topic OpenInfra Live 15:30:32 TheJulia: the mic is yours! 15:30:49 oh, so the tl;dr is some of us have started collaborating on a basic presentation on "what is ironic" 15:31:26 and the previously mentioned operator feedback loop session didn't really gain any traction as openinfra live. The consensus is we need the forums back. 15:32:11 That is it for OpenInfra Live. Any questions? Anyone want to collaborate on a high level what is ironic and maybe "why you should be using it!" 15:32:21 for which audience? 15:32:35 yay :) 15:32:38 for which part? 15:32:39 I'm not sure I understand what you propose if the proposal hasn't gained traction? 15:32:56 There were two proposals, one was for a feedback session and that didn't gain any traction 15:33:07 I actually wrote a few paragraphs of "why ironic" for my upcoming post, can share the draft with you 15:33:09 The ironic session for openinfra live is tentatively on August 12th 15:33:14 dtantsur: ++ 15:33:22 Aug 12th => no dtantsur 15:33:26 ack 15:33:27 I'll pm you once I render it 15:33:37 k 15:33:59 Safe to proceed to the next topic? 15:34:05 yep 15:34:06 #topic Cancelling review jams? 15:34:09 ~~When is it? (OpenInfra Live)~~. I am also working on a blog entry with similar title. 15:34:32 / August 12 ack. 15:34:45 So, JayF pointed out that I've been struggling to be present for Review Jams, and the tl;dr is I've had meetings take over my Tuesday calendar. 15:35:02 that's unfortunately also my case :/ 15:35:09 I mainly noticed nobody at all was attending them 15:35:10 except me. 15:35:19 We've also struggled with attendance after the last major release, which makes me think they are good for end of cycle get things sorted/merged activities 15:35:34 Our team now also has a partly conflicting meeting 15:36:04 JayF: Well, truth be told, It helps to have someone present who is kind of hitting the drum to drive people to the meeting and topics to get listed 15:36:27 I just can't hit the drum on the boat to waterskiiing speed on Tuesdays :) 15:36:44 I'll also generally note, as a reviewer, we haven't had many outstanding ironic/ipa/ironiclib patches recently 15:36:48 (think, someone drumming and a bunch of rowers to reach water skiing speeds) 15:36:51 they tend to get in pretty quick 15:37:12 So I guess, any objections to cancelling them at least until the midcycle? 15:37:26 If we identify priorities there then we should consider resumption and appropriate scheduling then? 15:37:42 * dtantsur agrees to both 15:37:59 ++ 15:38:01 Anyone objections? 15:38:14 none from me 15:38:18 * TheJulia believes the motion carries 15:38:41 So we're done with Discussion then 15:39:28 #topic Baremetal SIG 15:39:51 arne_wiebalck is out, and I don't believe there are any updates. Does anyone have anything sig related? 15:40:05 not me 15:40:33 #topic Who is going to run the next meeting? 15:40:35 Maybe a reminder, upcoming topics https://etherpad.opendev.org/p/bare-metal-sig <-- Around line 160 might interest people 15:40:44 a bit early, sorry :) 15:40:49 no worries 15:41:00 the question persists though: any volunteers? 15:41:12 I'll do it next week 15:41:12 I can run the next one 15:41:22 we have a conflict, the PTL will break the tie! 15:41:29 :D 15:41:31 lol 15:41:46 rpittau: if you really want to your welcome run the meeting :) 15:41:53 eheh ok :) 15:42:22 The PTL's judgement has been announced! 15:42:27 lol 15:42:33 #topic Open discussion 15:42:45 The floor is open, does anyone has anything? 15:42:48 I feel like my glorious purpose is to go reply to yet another email 15:42:51 I have one hopefully quick thing, the lower-constraints job is actually working as expected now https://review.opendev.org/c/openstack/ironic/+/801350 15:43:01 s/now/for now/ :D 15:43:13 lol 15:43:21 rpittau: you forgot the limit to master only 15:43:56 I didn't mention that in the commit message? 15:44:15 rpittau: I mean something like https://opendev.org/openstack/ironic-python-agent-builder/commit/963b23e0ae703f230815b474f1a2bf32cb4e807f 15:44:25 so that we don't have yet another thing to remember when creating branches 15:44:34 oh yeah, sure, I'll add that 15:44:49 I'm not 100% sure it works with templates.. but worth trying? 15:45:02 yep, let's give that a try 15:45:07 thanks! 15:45:15 np 15:46:55 anything else? 15:48:19 not from my point of view 15:48:59 thanks all! 15:49:12 #endmeeting