15:00:10 #startmeeting ironic 15:00:11 Meeting started Mon Sep 9 15:00:10 2019 UTC and is due to finish in 60 minutes. The chair is dtantsur. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:12 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:15 The meeting name has been set to 'ironic' 15:00:26 heya! who's here for the most ironic meeting in the openstack world? 15:00:27 o/ 15:00:30 o/ 15:00:34 o/ 15:00:39 o/ 15:00:52 o/ 15:01:00 o/ 15:01:04 #link https://wiki.openstack.org/wiki/Meetings/Ironic is our agenda (quite empty) 15:01:20 #topic Announcements / Reminder 15:01:51 #info Non-client libraries (metalsmith, ironic-lib and sushy) have been finalized for Train 15:02:06 all new development there will target U 15:02:12 (bug fixes are, of course, welcome) 15:02:26 #info Client libraries release deadline this week 15:02:39 o/ 15:02:43 actually, python-ironic-inspector-client has been released as well, and I'm about to process python-ironicclient 15:02:50 Ilya Etingof proposed openstack/ironic-prometheus-exporter master: Add metrics descriptions https://review.opendev.org/681007 15:02:50 last but not least: 15:03:01 o/ 15:03:11 #info Soft feature freeze begins on Thursday 15:03:43 I'd like us to slow down on new features this week, so that we can branch stable/train for all projects in one-two weeks 15:03:55 #link https://releases.openstack.org/train/index.html 15:04:01 #undo 15:04:02 Removing item from minutes: #link https://releases.openstack.org/train/index.html 15:04:05 #link https://releases.openstack.org/train/schedule.html 15:04:39 We must absolutely have stable/train by RC1 deadline, which is the week of Sep 23rd 15:05:08 better next week, so please plan accordingly 15:05:22 ack 15:05:33 if you're planning on feature work for train, let's raise it during the discussion section of this meeting 15:05:49 anything else to announce? any questions about the deadlines? 15:07:07 #topic Review subteam status reports (capped at ten minutes) 15:07:22 #link https://etherpad.openstack.org/p/IronicWhiteBoard around line 217 15:08:46 dtantsur: let's be explicit, and mention where we are punting to U 15:09:01 eg, deploy steps. ? 15:09:09 we could do it the other way around and mention what is NOT punted :) 15:09:19 dtantsur: true, either way. 15:09:41 dtantsur: historically, i think we've mentioned the stuff we've punted, but as long as it is clear to people :) 15:09:42 so yes, deploy steps are unlikely to make it, except for the vendor implementation patches (e.g. iDrac RAID) 15:10:04 idrac raid is WIP in the etherpad 15:10:29 i guess what i'd like to know is for deploy steps, there are 4 PRs; which do we think we can get into train? 15:11:10 most vendor PRs are in good shape, they're likely to make it 15:11:40 I'm working on the iDRAC RAID deploy step 15:11:45 well, it'd help if the status was updated; am doing that now... 15:11:49 punting other deploy steps work :( 15:11:55 sure 15:12:17 Richard G. Pioso proposed openstack/ironic master: Add idrac HW type support of Redfish interfaces https://review.opendev.org/681046 15:12:30 There's one ^^^ 15:12:41 so for deploy steps, mgoddard is working on idrac raid. what about deploy steps for ilo mgt? 15:13:03 shiv doesn't seem to be here. 15:13:03 rloo: stendulker has patches, I think they're in a good shape and close to approval 15:13:21 dtantsur: ok, i see you've looked at it. 15:13:26 yep 15:13:49 Indicators API remains unclear 15:14:00 this is a big chunk of work that has been largely unreviewed 15:15:15 opinions? etingof, will you have cycles to iterate on it this week? will anybody have time for reviewing it? 15:15:40 dtantsur, most certainly! 15:15:58 if it's a priority I can make some time to review them 15:16:25 etingof: how close do you think that work is? minor changes or .. ? 15:16:46 what worries me is that the last time I reviewed we ended up changing the API design :) 15:16:47 rloo, it works in devstack! 15:16:52 \o/ ship it 15:17:07 etingof: I see 3 patches mentioned in the etherpad, is that all? 15:17:10 and lol 15:17:55 rpittau, yes, other relevant ones have been merged 15:18:22 if etingof is avail to iterate and there are people willing to review, then i'm fine with it in train. we can see where we're at next Monday. 15:18:48 * etingof will do his best 15:18:52 do we have a *non Red Hat* person to review it? 15:19:31 wrt software raid, it is done but there is a comment at L324, 'Finished for Train, small fixes are still being posted'. If there are small fixes, could someone provide the links please? 15:19:32 * rpioso suggests expanding our cores' wardrobe 15:19:46 dtantsur: i can try later this week, but no promises 15:19:51 * rpittau still doesn't have a Red Hat :/ 15:19:52 thanks rloo 15:19:58 I'll add the sw raid links 15:20:15 * iurygregory is sad for rpittau 15:20:23 rpittau: don't worry, most of us don't have red hats :D 15:20:41 :D 15:21:15 * dtantsur has probably lost his already 15:21:22 anyway, are we ready to move on? 15:21:31 I updated the idrac vendor priority. It's the change above. We would like to see it land in Train. 15:21:38 k 15:22:13 #topic Deciding on priorities for the coming week 15:22:14 tssurya: when you are avail, would you please let us know if there are any more PRs related to power state callbacks to Nova. It is done except for a release note update: https://review.opendev.org/#/c/676400/, yes? 15:22:15 patch 676400 - ironic - Follow-up to power sync reno - 1 patch set 15:22:32 oh, not ready yet? 15:22:46 dtantsur: almost ready i think. just want to verify 15:22:55 so ipv6 is not going to get done, right? 15:23:00 rloo: yea that's correct 15:23:09 all done except thta follow-up 15:23:12 rloo: quite unlikely, yes. but since it's CI work, it's not affected by deadlines. 15:23:12 tssurya: thx for confirming! 15:23:12 that* 15:23:22 (unless it turns into a feature work, of course) 15:23:35 thanks to all of you for the review efforts and time 15:23:43 dtantsur: ok. and the doc generation stuff? is that affected by deadline? 15:24:24 L356 ^^ 15:24:50 rloo: nope. and it's fine to backport. 15:25:06 the docs job is basically done except a few patches in the review list 15:25:11 good to know dtantsur! 15:25:19 kaifeng: could you add them to the priority list? 15:25:35 etingof: same request for indicator patches ^^ 15:25:39 dtantsur: will do 15:25:45 dtantsur, ok 15:26:17 how about reviewing the latest ironic-prometheus-exporter additions? -- https://review.opendev.org/680731 15:26:17 patch 680731 - ironic-prometheus-exporter - Add Redfish metrics support - 3 patch sets 15:26:39 we probably have to, yes. if you plan on making it into train. 15:27:50 dtantsur, I'd like to 15:29:29 dtantsur: added at L168, molteniron and virtualpdf are not listed, they don't have doc job so I guess it's not required. 15:29:43 kaifeng: thanks! 15:30:10 etingof: when listing patches, could you add a short summary before each? 15:30:18 dtantsur, ack 15:31:04 we also have a couple of easy fixes for sushy-tools, those are good to merge because then the whole thing actually works 15:31:44 mgoddard: dtantsur: this client PR, is that something we want discussed/done in train? https://review.opendev.org/#/c/675940/ 15:31:44 I'd delay them, we already have a huge review backlog 15:31:45 patch 675940 - python-ironicclient - Fix AttributeError in negotiate_version - 1 patch set 15:32:15 rloo: ideally, although it can (and should) be backported 15:32:56 dtantsur: mgoddard: do you two know what the solution is? (I just skimmed the comments) 15:33:08 I don't 15:33:14 rloo: it would be nice, although as dtantsur says it's a bug so can be backported 15:33:29 I'd make it a priority for next week 15:33:40 and wrap up feature work this week 15:33:42 ok, i don't think it'll make it in this week then. the client deadline is this week, right? 15:33:59 yep 15:34:11 I actually wanted to request a release today 15:34:44 dtantsur: ok, let's try to fix that soon and backport it. i don't see anything other PRs for client that might need addressing right now, so a release today is fine. 15:35:01 dtantsur: we're bumping the major version or ? cuz we deleted ironic CLI, right? 15:35:15 right 15:35:45 dtantsur: yay. although i worry that we'll have broken people that didn't know/didn't get around to switching to openstackclient. 15:36:02 it would happen whenever we pull the trigger :) 15:36:20 dtantsur: yup. so even though i worry... pull the trigger anyway! 15:36:28 i'm good with the weekly priorities :) 15:36:36 now there are talks of migrating for python API to openstacksdk.. 15:36:39 anyway 15:36:50 any other comments on the priorities? anything to add or (better!) remove? 15:37:27 dtantsur: oh, that reminds me. the two PRs at L143 & 144 15:37:59 dtantsur: do we want to get them into train? if so we mgiht want to discuss L143, and someone needs to continue L144 15:38:16 I'd love to get 143 in, it's a constant source of complaints from the customers 15:38:31 dtantsur: yeah, me too... 15:39:08 wrt 144, there is a similar patch already gets in, proposed by mkrai 15:39:50 rloo: would be great if you could check my responses for L143 15:39:54 kaifeng: link handy? 15:40:10 dtantsur: ok, will discuss with you. prob tomorrow though. 15:40:49 dtantsur: https://review.opendev.org/#/c/677649/ it's doing the opposite 15:40:50 patch 677649 - ironic - Allow deleting unbound ports on active node (MERGED) - 3 patch sets 15:41:35 rloo: tl;dr I'm purposely disallowing "provide" action for nodes in maintenance 15:42:00 kaifeng: not exactly the opposite, although I am somewhat confused by these two patches.. 15:42:06 I have to check with a fresh head tomorrow 15:42:20 we can leave them on the list, so that we don't forget to follow-up/discuss, wdyt? 15:42:32 makes sense 15:42:49 i think mkrai's addresses a subset of the port deletion/with-or-without-vif problem. 15:43:00 but yeah, leave on the list. 15:43:13 okay. moving on? 15:43:20 mkrai_, I can take a look at the L144 patch tomorrow 15:43:31 thanks mkrai_ 15:43:38 #topic Open discussion 15:44:07 anything? I guess we've already discussed the feature freeze 15:44:15 * kaifeng was cheated by the double negative commit message 15:44:46 heh 15:45:07 double negatives are tricky; we should try to avoid them! 15:46:00 crickets? 15:46:09 apparently :) 15:46:17 it's still sort of summer, right? 15:46:42 +15 over here 15:46:48 sort of 15:46:49 crickets and dragonflies everywhere 15:46:55 thanks everyone, let's make a rocking train! :) 15:46:58 not in Brazil XD 15:46:58 #endmeeting