15:00:10 #startmeeting ironic 15:00:14 Meeting started Mon Apr 16 15:00:10 2018 UTC and is due to finish in 60 minutes. The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:16 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:18 The meeting name has been set to 'ironic' 15:00:18 o/ 15:00:18 o/ everone 15:00:29 o/ 15:00:30 o/ 15:00:30 o/ 15:00:32 o/ 15:00:33 o/ 15:00:37 Our agenda can be found on the wiki, and seems quite light today. 15:00:39 o/ 15:00:39 #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting 15:01:17 #topic Announcements/Reminders 15:01:18 o/ 15:01:35 \o 15:01:42 #info This week is the Rocky-1 Milestone 15:01:53 o/ 15:02:00 TheJulia: is there anything ironic needs to do wrt the milestone? 15:02:02 o/ 15:02:21 rloo: I do not believe so, we should only keep in mind that we need to be ready to release around milestone 3 15:02:26 or shortly after I should say. 15:02:47 TheJulia: right. 15:03:44 #info We have a board in Storyboard to aid in triaging stories. For those that do not know, we've been reviewing items across the community and adding a triaged tag if they have been reviewed. This allows us to make sure things are still valid, or if they are no longer valid, mark them as such. 15:03:56 link? :) 15:04:06 One moment! :) 15:04:38 It is slow to load, because it is running somewhat complex queries across ALL of the projects in ironic. Please keep that in mind. 15:04:41 #link https://storyboard.openstack.org/#!/board/67 15:04:55 nice, thanks 15:05:00 TheJulia: nice! 15:05:19 * rloo isn't crazy about 'panopticon though, seems somewhat negative' 15:05:19 It is also entirely automatic, so it will update as things change. No need to drag/drop. 15:05:37 rloo: I was thinking about architecture at the time... you can rename it if you want, you have that power :) 15:05:52 * jroll puts it in the 'dashboards' bit in the whiteboard 15:06:15 TheJulia: thx. Others are welcome too :) 15:06:18 Anyone have anything else to announce or reminds us of?? 15:06:36 there's elections for something coming up? 15:06:38 tc 15:06:39 Yes 15:06:54 and our CI is still down, right? 15:06:58 just a reminder that multinode grenade is currently broken on master, julia and I have been working on it 15:07:00 yes 15:07:02 :) 15:07:15 Yes, TC election starts next week 15:07:20 seems we probably have a fix, attempting to validate it 15:07:25 TheJulia: is there a deadline wrt getting forum topics? 15:08:03 #info Multinode Grenade CI testing is broken at the moment as Jim mentioned, "fix" being tested 15:08:08 rloo: That was Sunday :( 15:08:10 * rloo crosses fingers that fix works. 15:08:17 TheJulia: OH. ok, scratch that one off the list! 15:08:31 rloo: fwiw, the fix seems to work, but the job is timing out for (probably unrelated) reasons 15:08:39 waiting to see if it was a fluke or not 15:08:50 Anyway, we should move on if there is nothing else 15:08:57 grrrrr... could be another problem that snuck in... 15:09:10 + move on 15:09:16 #topic Review action items from previous meeting 15:09:36 #info One action item in preparation for the call we had later in the week. Completed. 15:09:49 #topic Review subteam status reports 15:09:56 #link https://etherpad.openstack.org/p/IronicWhiteBoard 15:10:57 So I realize our CI was broken for a good chunk of last week, and that prevented us from landing things in ironic, so the status does seem to be relatively unchanged. I went ahead and added some python-ironicclient and a housekeeping patch to the priority list for this week. 15:11:51 looks like decent progress on bios stuff 15:12:13 o/ 15:12:16 Yeah, it looks like we're just going to need to recheck some stuff once CI is happy again 15:13:11 It really is looking like it is in good shape and I think it would be good for us to try and review and land soon 15:13:32 Although I've not looked beyond the RPC patches yet, I need to do that. 15:14:03 wrt sighup support for ironic-inspector. i looked at the patch. it doesn' tuse oslo.service cuz it doesn't seem like flask can be used with it. Does anyone know? I suggested they send email to ask. 15:14:19 yeah, we can land the first few after CI is healthy 15:14:28 rloo: oh, this came up in keystone... lemme find a ML link 15:14:55 thread starts here: http://lists.openstack.org/pipermail/openstack-dev/2018-March/128471.html 15:15:02 there's some suggestions in replies 15:15:13 not sure if anyone has written shared code, though 15:15:26 jroll: ok, will add link to the patch, thx! 15:15:29 jroll: awesome, thanks 15:15:35 :) 15:15:45 TheJulia: wrt mox -- L369. have you done that? declared it done? 15:16:01 rloo: No, I've not, I've been literally swamped 15:16:22 TheJulia: it doesn't need to be you. any volunteers? 15:16:23 #action TheJulia to declare ironic done for mox 15:16:32 heh, that's another way to do it :) 15:17:00 :) 15:17:09 Anyone have any other questions or comments regarding status updates? 15:17:16 * jroll does not 15:17:25 TheJulia: wrt rescue, are there blockers on that? last week, there was mention of back/forth... 15:17:53 rloo: Great question, stendulker_ revised the patch that we're blocked on this morning after the back and forth last week. 15:18:09 I've not had a chance to look at it, but we really need to review and proceed forth in some direction with rescue this week. 15:18:43 TheJulia: so presumably he revised it on some agreed upon approach by the parties that were doing the back/forth? or are more eyes needed? 15:19:28 jroll, Was https://review.openstack.org/#/c/560884/ done to fix the multi-node grenade CI? 15:19:29 patch 560884 - networking-generic-switch - Sync hacking requirement with ironic (MERGED) 15:19:36 TheJulia: more eyes needed wrt direction? 15:19:50 He summarized the takeaways, he revised it, I think more eyes are needed once we have CI up and running to just go ahead and merge it if we're in consensus 15:20:05 But I've not looked yet this morning, so I can't say for sure 15:20:19 jlvillal: that was the hope, but alas it did not help 15:20:27 jroll, Thanks 15:20:30 TheJulia: ok, I'll try to look later this week. thx. 15:20:34 jlvillal: current thought: https://review.openstack.org/#/c/561358/ 15:20:34 patch 561358 - networking-generic-switch (stable/queens) - Ensure pep8 is present before trying to proceed 15:20:52 okay 15:21:05 heh 15:21:15 Anyway, seems like we're good to move along to priorities for this coming week? 15:21:21 ++ 15:21:23 ++ 15:21:36 #topic Priorities for the coming week 15:22:07 I've already updated the list in the etherpad of what I think would be good, it includes a few smaller patches that we can consider 15:22:55 I'm good with what is there, I'm not aware of anything that absolutely has to go on that list this week. 15:23:43 fine with it as well 15:23:55 +A'd that last one, but will leave it on the list in case it fails 15:24:13 jroll: awesome, thanks 15:24:16 Thx TheJulia, I'm fine with it (although rescue seems higher than client but i doubt it matters) 15:24:39 It really doesn't in the grand scheme of things 15:25:03 They are also fairly small patches for the client 15:25:39 Does anyone have anything they wish to raise before we move on? 15:26:02 Yes, about the XClarity driver 15:26:18 a story is created 15:26:36 Is it something that should be a project priority this week? 15:26:56 oh, perhaps not... 15:26:57 Otherwise it might be good to discuss during our Open Discussion 15:27:05 sorry 15:27:12 jiapei: no worries! 15:27:50 jiapei: you may want to take a look at https://etherpad.openstack.org/p/IronicWhiteBoard around line 148 15:28:15 Anyway, I guess I'll move on 15:28:29 #topic Bug Triaging for the coming week 15:30:10 I think this portion of the meeting can be skipped moving forward, at least until we've gone through our backlog of items and have a better idea of what is outstanding 15:31:05 + 15:31:07 + 15:31:08 + 15:31:09 I'm fine with that 15:31:11 :) 15:31:12 heh 15:31:19 Okay, moving on! 15:31:34 #topic Open Discussion 15:31:53 So what shall we discuss today? Taking over the world perhaps? :) 15:32:24 i wanted to thank mjturek for organizing the bug bash last week 15:32:37 thanks mjturek! 15:32:42 Thank you mjturek for organizing the bug bash last week :) 15:32:45 np rloo and TheJulia! 15:32:59 If we ever want to do another one I'd be happy to help. It was real fun 15:33:03 It was fun, and we should do it again... perhaps we can find a time that works every couple weeks? 15:33:15 +1 15:33:19 * rloo thinks once a month might be more realistic... 15:33:25 rloo: agreed 15:34:05 TheJulia: I'll send something out on the ML about it, maybe something like the first Thursday of the month or such 15:34:11 jiapei wanted to discuss xclarity 15:34:17 and yes, thanks mjturek :) 15:34:18 Yes 15:34:26 mjturek: Awesome, thanks! 15:34:27 np jroll :) 15:34:32 ty jroll 15:34:37 jiapei: You have the floor 15:35:00 Well, I've created a story 2001841 15:35:33 with 2 tasks, could you help review the patches :) 15:35:40 \o/ storyboard is tracking links again 15:35:49 well, gerrit really 15:35:58 #link https://storyboard.openstack.org/#!/story/2001841 15:36:26 oh whoops... that didn't work 15:36:35 nope, I need to do it :( 15:36:38 jiapei: that code patch needs some unit test updates and pep8 fixes, but yes, will try to take a look once they're updated 15:36:39 #link https://storyboard.openstack.org/#!/story/2001841 15:36:43 yay! (that the links are being added to storyboard) 15:37:42 jiapei: I was going to say what jroll said. Looks like your editor may be adding blank lines to the end of files being opened as well, which will cause pep8 errors. 15:38:05 Ah 15:38:17 It's blank line... 15:39:12 I tried to find the pep8 errors but didnt realize it's the blank that leads to 15:39:55 yeah, you can run the tests locally `tox -epep8`, `tox -epy35` and look at some of the result data to help you locally so your not having to wait on CI. 15:40:50 jiapei: there is some documentation here wrt running tests locally: https://docs.openstack.org/ironic/latest/contributor/dev-quickstart.html#running-unit-tests-locally 15:41:13 Running pep8 validation results in the various hacking rules getting loaded which govern things that the overall openstack community has agreed to ignore or be strict about. 15:41:26 I also have a question about the 3rd CI, do we have a doc about how to setup it? 15:42:26 Clark Boylan proposed openstack/ironic stable/queens: Preserve env when running vbmc https://review.openstack.org/559856 15:42:27 jiapei: There is the upstream documentation on how to connect to gerrit event streamer and there should be general docs on how to do that. Ironic specific you'd want to listen for change events on ironic that are appropriate for the xclarity driver to execute a test on 15:42:37 Clark Boylan proposed openstack/ironic master: Preserve env when running vbmc https://review.openstack.org/559811 15:43:07 jiapei: from there, it gets a little more complex depending on your test scenario, and typically everyone's requirements are a little different so we'll have to kind of work with you there 15:43:56 #link https://docs.openstack.org/infra/system-config/third_party.html 15:44:01 Yeah, I'll look through the links above 15:44:18 Seems a little complex 15:45:01 jiapei: Sadly it does tend to be a little complex, that last link also covers two different ways, which might seem a bit confusing 15:45:58 jiapei: Hope that at least helps. Feel free to ask questions, and welcome! 15:46:01 I have a thing, if jiapei is finished? 15:46:32 Will take silence as a yes 15:46:56 Always a good plan :_ 15:46:57 err :) 15:47:06 TheJulia: :) 15:47:15 We have a summer intern who has chosen a project called 'Save the inspector' 15:47:26 mgoddard: I like it! 15:47:54 It's perhaps a little optimistic, but will hopefully at least lead to some contributions 15:48:04 mgoddard: are they in need of a crash of sorts? 15:48:04 sweet! 15:48:06 I've worked with him before and he's very switched on 15:48:07 awesome 15:48:08 err, crash course 15:48:14 excellent! 15:48:25 he spent a year with us at cray so is familiar with openstack and ironic etc. 15:48:34 most excellent 15:49:00 so if anyone has any particular bugs or features that are urgent, let me know 15:49:18 did the HA work get finished? 15:49:23 when the time gets closer I'll start compiling a list of suggestions for him 15:49:38 rloo: essentially yes, but not entirely into TripleO AIUI 15:49:45 I think derekh was going to pick that up 15:50:04 is it enabled by default? 15:50:15 TheJulia: ah. the other thing that sticks in my mind might be too big a chunk, but we talked about refactoring/putting some inspector stuff into ironic 15:50:20 in OoO? 15:50:27 in inspector 15:50:56 mgoddard: There are a few things in storyboard for ironic-inspector, we had a few ideas about ways to improve interaction with the days of traits since properties -> trait based scheduling now 15:51:22 mgoddard: The HA model that is in place is slightly racey if I understand it correctly off hand, so in essence, yes 15:51:35 at least racey regarding the initial dhcp 15:51:38 which.. *shrug* 15:51:40 yeah, traits could be a useful thing to look at 15:51:54 there's also the config reload thing that needs doing 15:51:58 mgoddard: I don't think we have an explicit story for that, but it would go a long way 15:52:08 yes, config reload is also needed 15:52:48 Anyway, we have 8 minutes left of our scheduled time block, does anyone have anything else to discuss? 15:52:50 ok. All good ideas, I'll make a note. If anyone has any others, ping me. 15:52:54 Talking of crash courses for someone new to OpenStack and ironic, ... Any suggestions? 15:52:57 by 'config reload' you mean the sighup stuff? that is being worked on. 15:53:11 rloo: ah, ok 15:53:20 rpioso: I need to start getting things together for the Summit project onboarding 15:53:30 but I was planning on doing that next week 15:53:47 jroll: that was related to my question earlier about flask & oslo.service :) 15:54:01 rloo: oh right. thanks :) 15:54:09 mgoddard: dtantsur|pto will most likely have lots of ideas! 15:54:26 rloo: that's true :) 15:54:31 running ironic in production is the quickest way to learn :P 15:54:37 I've signed up to give a one hour introductory presentation to a few folks this Wednesday. 15:54:38 mgoddard: there's also he openstack priority wrt deploying with apache & wsgi in CI. 15:54:43 jroll: lol 15:55:30 one hour... that might just cover openstack in general :) 15:56:41 mgoddard: rloo: I think inspector going into wsgi would require for us to split the service, and I kind of doubt that will ever happen 15:56:53 then again, I might be surprised :) 15:57:30 Pointers to introductory content is probably all I'll have time to cover. 15:57:44 Any favorites? 15:58:04 rpioso: the bible https://docs.openstack.org/ironic/latest/ :D 15:58:28 lol 15:58:29 rloo: Amen :) 15:58:43 * jroll is hungry 15:58:43 lol 15:58:47 rpioso: perhaps maybe some presentations from summits might help, but they are going to be more than an hour 15:58:51 TheJulia: in the long term do you think inspector will take the same arch with ironic? 15:59:11 Anyway, seems like we're done meeting for today, we can discuss taking over the world in our normal channel discussion. 15:59:36 gn 15:59:47 thanks TheJulia 15:59:52 kaifeng: I don't know, that is going to be dependent upon the individual contributors to inspector. The original idea was that it was required to make it HA-ready, but in reality splitting it was not needed. 16:00:10 kaifeng: Any reason why you ask? 16:00:18 #endmeeting