15:00:19 #startmeeting ironic 15:00:19 Meeting started Mon May 4 15:00:19 2020 UTC and is due to finish in 60 minutes. The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:20 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:20 o/ 15:00:22 The meeting name has been set to 'ironic' 15:00:24 Good morning everyone! 15:00:27 o/ 15:00:28 o/ 15:00:30 o/ 15:00:33 o/ 15:00:36 o/ 15:00:39 o/ 15:00:40 \o 15:00:44 o/ 15:00:45 o/ 15:01:02 o/ 15:01:10 I hope everyone had a wonderful weekend! 15:01:30 Our agenda can be found on the wiki as always. 15:01:32 #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting 15:01:47 #topic Announcements / Reminders 15:02:15 #info Ironic 15.0 was released this past week \o/. This was the version for our Ussuri release. 15:02:19 * dtantsur raises a glass for Ussuri 15:02:21 Thank you everyone who contributed and helped make that happen 15:02:36 \o/ 15:02:54 congrats and thanks!!! 15:02:54 Additionally, last week there was a note about the discussion to determine next steps for releasing ironic 15:03:14 dtantsur: I believe, if memory serves the agreement was for that to be a spec? 15:03:41 yes, and I'm still committed to writing it 15:03:50 k 15:04:07 got distracted by the fishy issues 15:04:10 #info Last week's discussion on next steps for ironic release management will move to a spec document. 15:04:13 Sorry about those fishy issues. 15:04:21 definitely not your fault :) 15:04:30 Anyway, does anyone have anything else they would like to announce or remind us of? 15:04:43 we should figure out the slots for the PTG 15:04:45 =) 15:05:08 Ahh yes, I did put that in the discussion area 15:05:14 iurygregory: You have poll and result data yes? 15:05:17 oh ok =) 15:05:22 TheJulia, yes 15:05:28 excellent 15:05:33 but we only have 12 participants 15:05:37 had* 15:05:42 Perhaps I should mention the fishy issues. 15:06:00 12 people is a nice small party 15:06:12 TheJulia: at least link to your ML post? 15:06:16 indeed! 15:06:40 For those interested, I emailed the mailing list on redfish driver issues, assumptions, and breaking to us changes. 15:06:43 #link http://lists.openstack.org/pipermail/openstack-discuss/2020-April/014543.html 15:07:26 I've proposed a workaround in two parts: https://review.opendev.org/725239 and https://storyboard.openstack.org/#!/story/2007610 15:07:27 patch 725239 - ironic - [PoC] redfish: handle hardware that is unable to s... - 3 patch sets 15:07:35 Thanks dtantsur 15:07:41 np, feedback welcome 15:07:45 Does anyone have anything else to announce or remind us of? 15:07:59 Oh! Victoria PTG! 15:08:02 #link https://etherpad.opendev.org/p/Ironic-VictoriaPTG-Planning 15:08:02 I don't =) 15:08:15 Will redfish be declared experimental? 15:08:21 Etherpad document for items/thoughts/ideas. Please try to spend a few minutes in the next day or two putting a couple thoughts 15:08:49 rpioso: *shrug* give the severity of the problem, I'd vote for that. But I won't be surprised to be in minority. 15:09:07 rpioso: I'm not sure, we likely need to weigh it against the push that exists out there for OEMs to disable raw setting capability on IPMI. Which is also... problematic and breaking in some ways 15:09:20 we cannot build a solid building on a shaking ground 15:09:26 even if shaking for a good reason 15:09:47 You know, we have earthquake code here :) 15:09:56 But I agree 15:10:06 Anyway, we should move on it seems 15:10:09 ++ 15:10:29 Looks like we had no action items last week 15:10:51 #topic Review subteam status reports 15:10:57 #link https://etherpad.opendev.org/p/IronicWhiteBoard 15:11:29 Lets keep in mind, new cycle, so we should likely update items and remove old/out of date data 15:11:55 we can probably remove the python 3 mirgation part 15:12:04 s/mirgation/migration 15:12:15 arne_wiebalck: I believe you tweeted some graphs and mention of performance impact for moving to conductor groups 15:12:24 TheJulia: yes 15:12:39 TheJulia: we will prepare a blog post 15:12:45 removing retirement 15:13:01 ++ 15:13:11 I'm having some trouble deploying with Bifrost using a VLAN: http://paste.openstack.org/show/793066/ 15:13:31 dking: we're in the middle of our weekly meeting, we'll have to circle back to that after our meeting 15:13:32 dking: we have a meeting now, could it maybe wait till it ends? 15:13:45 Oh, sure. Sorry about that. 15:13:51 Removing multitenancy 15:13:51 In about another hour? 15:13:56 dking: yes 15:14:45 Well, the deployment API somewhat falls under multi-tenancy.. but we can handle it separately 15:14:46 arne_wiebalck: would it make sense to try and have another doc review meeting on the baremetal whitepaper say... next week? 15:15:00 dtantsur: separate is better imho 15:15:04 TheJulia: yes, I will prepare a doodle 15:15:05 k 15:15:15 re whitepaper: I hope to see SuperCloud addition this week 15:15:27 we've been working on a draft with Jakob 15:15:39 <3 15:15:52 StackHPC also wanted to look into contributing. 15:16:29 Yes, Stig emailed me ?last? week, but it was a blur and email fell by the wayside for me last week 15:16:40 I do like the deploy steps note :) 15:16:46 victory in Victoria 15:16:58 I feel like this should be our motto this cycle? objections? 15:17:56 For software raid, do we consider that wrapped up? 15:18:07 no objections :) 15:18:29 modulo some cleanup we planned 15:19:14 so maybe we notes some cleanup but overall as done? 15:19:22 ++ 15:19:27 yeah 15:19:57 I think Derek has a v6 CI job, I need to review that this week 15:20:04 We should put that on the list of things to review this week! 15:20:36 +100 15:20:44 it actually looks good, I had a quick look, definitely need to review it this week 15:21:17 * TheJulia adds a todo note 15:21:26 Now... grenade 15:21:38 * rpittau ducks 15:21:53 iurygregory: any chance you can condense the notes, I'm just not sure what is important 15:22:08 TheJulia, sure 15:22:30 re the noted ci improvement and moving to dib based jobs, it looks like we're still in the process of reviewing those 15:23:50 seems so 15:24:03 and there are many more CI improvements to make IIRC 15:24:13 there are always 15:24:18 TheJulia: the ironic-inspector one is blocked by grenade at the moment 15:24:26 done 15:25:38 Are there any new subteam efforts to add to the list at this time? 15:25:57 at some point we need to start migrating bifrost to the new collections 15:26:15 maybe note it somewhere, so that we don't forget (like with python 3) 15:26:17 * iurygregory can work in that 15:26:22 iurygregory++ 15:26:34 talk to us on #openstack-ansible-sig if you need help 15:26:50 * iurygregory wow there is a channel for a sig 15:26:51 Have we successfuly mostly resolved the standalone job failures to everyone 15:26:56 * iurygregory :O 15:27:06 iurygregory: why not? :) 15:27:09 's present satisfaction? 15:27:23 let's put it this way: I don't think the standalone job fails more often than other jobs 15:27:24 we are just missing merge the job to cover the split of standalone =) 15:27:32 which is far from satisfaction, unfortunately 15:27:33 heh 15:27:59 I'm going to remove it from the known issues list since I think we sorted most of those race conditions we were hitting 15:28:08 the one huge one with netruon... yeah 15:28:22 neutron mmm, the events work? where have we left it? 15:28:58 dtantsur: they are posted to our API, but we don't do anything with them or even store them as a thing afaik 15:29:12 oh, so the events are coming already? was it implemented in neutron itself? 15:29:18 I believe so yes 15:29:31 can we put that on the priority list and find volunteers? 15:29:38 ++ 15:29:42 our side must be much less problematic 15:30:00 no, more, which is why forward progress stopped :( 15:30:28 added under replacing wsme 15:30:39 anyway, we can and should pick it up 15:30:51 * dtantsur looks around for volunteers 15:31:21 Well, I'm good to move on to priorities for the week! Everyone else? 15:31:30 ++ 15:31:30 ++ 15:32:11 Okay, onward then! 15:32:26 #topic Deciding on priorities for the coming week 15:32:28 Merged openstack/ironic master: Switch to unittest mock https://review.opendev.org/717979 15:32:31 I think one priority is PTG topics! 15:32:51 ++ 15:33:02 * dtantsur has proposed two topics already 15:33:07 #link https://etherpad.opendev.org/p/IronicWhiteBoard 15:33:16 Starting at line 108 15:33:20 * TheJulia cleans up stuff from last week 15:33:56 does anyone want to shepherd the network_data patches while Ilya is out? 15:34:26 (basically play the role of friendly neighborhood rebaser) 15:34:33 TheJulia, me 15:34:36 * kaifeng can take a look into the neutron events stuff 15:34:45 iurygregory++ kaifeng++ 15:34:49 many thanks! 15:34:54 I will take care of his patches to make sure they are updatodate 15:34:59 * dtantsur needs to dive into deploy steps again 15:35:03 * iurygregory enr space didn't work 15:36:20 so a fairly light list, is there anything in active review that is not on the list that people need eyes on? 15:36:26 I wonder when the neutron side has implemented the notification, my last memory on this was it was stuck for a while and we only have a minimum api entry. 15:36:26 dtantsur re neutron events: (I'm super delayed for connection issues) I can have a look at that, if noone else volunteered already 15:36:42 rpittau: kaifeng wanted, but I believe there is enough work for two there 15:37:09 checking the neutron side is action item #1 15:37:43 * rpittau reading irc messages after minutes :( 15:38:02 yeah, will do 15:39:19 can we move L363 into L276 so we will have handy links there? 15:39:19 TheJulia: do you want to add your V priorities spec? 15:39:35 Yeah, I think that would be good 15:39:46 kaifeng: +1 15:40:51 kaifeng, dtantsur: I had a look at that when refactoring the ironic notifiers in neutron, I'll be glad to help 15:40:59 dtantsur: added with additional context 15:41:27 rpittau would be good to have you onboard :) 15:41:48 Anything else to add to the list? Is everyone happy with the list? I guess we can add more items to it as they come up over the next week 15:42:16 looks okay 15:42:18 kaifeng: thanks, count me in :) 15:43:35 okay, then onward to discussion! 15:43:41 Merged openstack/tenks master: Drop CentOS 7 jobs https://review.opendev.org/725243 15:43:44 #topic Discussion 15:44:08 Two items, one is the PTG slots. We need to sign up for some 15:44:32 iurygregory: do you have the results available from the participation doodle? 15:44:38 re-doodle http://paste.openstack.org/show/793067/ 15:44:47 this was top3/4 in all days 15:44:54 from the 12 participants 15:45:04 mostly had 10 or 11 votes 15:45:16 https://doodle.com/poll/mbpr2x7z3t5hqec6 15:45:35 ok seems now we have 13 participants hehehe 15:45:54 there are several slots when everyone can attend 15:46:03 we need to make sure to include these 15:46:05 yep 15:46:17 I wonder if we can do we should try to do something like Thursday/Friday or Tuesday/Friday? 15:46:20 yep some with 12(13 now) 15:46:47 Tue doesn't work for iurygregory 15:46:54 makes sense to me, just wondering if we are going to have cross-project sessions 15:47:06 it seems we have 12/13 only wed-thu-fri 15:47:17 dtantsur, LOL I forgot Tue for some reason 15:47:22 LOL 15:47:26 could you re-vote then? 15:47:35 i.e. update your vote? 15:47:50 yes 15:48:10 I'd have more days rather than more hours per day 15:48:14 my guess is that these time slots may work well with other projects too. at least we aren't restricted by the number of rooms available :) 15:48:17 dtantsur: ++ 15:48:22 done 15:48:30 Lets do an hour on monday, and do a couple more small windows through the week? 15:48:39 for me any day is possible even in non working hours 15:48:45 i.e. for me 4 small slots on Tue, Wed, Thu, Fri is better than 2 big slots Thu+Fri 15:48:58 ++ for small slots 15:49:00 If we end up with say a total of 8 hours, I tink that would be fairly good plus give us flexibility to engage in other discussions 15:49:11 can we not do monday? And I agree with dtantsur 15:49:14 so people who need to participate in other projects won't have problems 15:49:22 I'd do Tue-Fri 2 hours each 15:49:33 yeah, I think I'll have a conflict on Thu with the Ansible SIG 15:49:38 but not sure 15:49:57 sshnaidm: when is the ansible collections slot at the PTG? 15:50:22 dtantsur: is that the automation sig ? 15:50:24 I don't see anything reserved by the edge wg 15:50:50 rpittau: no, do you have the ethercalc link handy? 15:50:57 https://ethercalc.openstack.org/126u8ek25noy ? 15:50:58 https://ethercalc.openstack.org/126u8ek25noy 15:51:05 that :) 15:51:07 dtantsur, 15:00 - 17:00 UTC in Thursday in Room14 Newton, and 1 hour for APAC/EU 7.00-8.00 UTC in Friday in Room14 Newton 15:51:10 https://etherpad.opendev.org/p/openstack-ansible-modules-ptg 15:51:24 thanks sshnaidm 15:51:44 oh ok it's missing from the ethercalc 15:52:16 it's OAM I guess 15:52:31 right... 15:53:36 So... we could do 14:00-16:00 wednesday/thursday/friday, say in icehouse? 15:53:49 things will be cold =) 15:53:55 TheJulia: and Tue as well? 15:53:57 Yes, but it is a nice reference 15:54:02 Tuesday, we could do later 15:54:07 UTC ? 15:54:16 or in antoher room 15:54:37 not too much later if possible 15:54:42 I guess the room doesn't really matter? 15:54:53 * dtantsur assumes UTC 15:54:55 no, but the same one would be nice 15:54:59 yes, all in UTC 15:55:03 15-17 on tue ? 15:55:06 so, looks like queen is unspoken for 15:55:10 queens 15:55:59 right 15:56:12 ++ same room. something in 14-17 UTC is good, overlaps with eg neutron and nova, in case we want to talk to them 15:56:16 stein empty as well 15:56:22 ++ 15:56:23 so 14-16 ? 15:56:33 wfm 15:56:36 wow, 4 mintues left 15:56:45 ++ 15:56:58 i think icehouse was the first ironic release or something? 15:57:22 yes, it was 15:58:01 (and if i recall, it was hot there!) 15:58:01 With regards to sig/wg overlap, I suspect we're going to need to create a master schedule of ironic topics 15:58:28 yeah 15:58:45 can everyone do outreach and begin to build a time schedule into our victoria ptg planning etherpad? 15:59:14 and by outreach, I mean if you are close to say the edge sig, try and sort out topics and when/etc 15:59:15 If you want I can send the email with the timeslots we will use 15:59:23 iurygregory: please 15:59:32 So, are we good to move on? 15:59:54 14 till 16 Queens correct? =) 15:59:55 seems so 16:00:01 iurygregory: yes 16:00:26 So the other topic is the proposed priorities for Victoria. I have an idea that perhaps we should re-think the cycle planning model in general 16:00:46 I would love discussion on that topic, but if it is relegated to the change set that works as well I guess 16:00:58 My hope was mroe that we better enable ourselves to kind of focus on lining things up 16:01:33 and mainly I mean by like We focus on x, y, z api changes in that order so we're able to line up work together. I'm just using API as an example because that is a case where it does help a lot 16:01:56 And while I toss that out there, is there anyone else with anything to raise this week before we try and end the meeting a little late? 16:02:01 this, unfortunately, will hit the contributor priorities issue (again) 16:02:52 I just wanted to say that stable/queens CI is broken at the moment 16:03:10 \o/ 16:03:39 on a positive note, we have a tinycore 11.x based tinyipa image built on focal :) 16:04:07 awesome! 16:04:08 nice 16:04:56 dtantsur: Well, I'm thinking if we try to rebalance or evaluate weekly, then contributor wants/needs could still move 16:05:10 Maybe it all comes down to being more proactive about stacking/ordering 16:05:17 Verification of a change to openstack/tenks failed: Py2 - Remove six, mock and future https://review.opendev.org/698922 16:05:23 instead of taking a... when the stars align merge approach 16:05:46 I don't know 16:06:08 Anyway, we're past time. Thanks everyone! 16:06:17 thank you! 16:06:31 email sent 16:06:34 iurygregory: thanks! 16:06:39 Have a wonderful week everyone! 16:06:42 #endmeeting