17:00:06 <jroll> #startmeeting ironic
17:00:07 <openstack> Meeting started Mon Aug 22 17:00:06 2016 UTC and is due to finish in 60 minutes.  The chair is jroll. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:08 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:11 <openstack> The meeting name has been set to 'ironic'
17:00:12 <vdrok> o/
17:00:13 <dtantsur> o/
17:00:14 <mariojv> o/
17:00:14 <jroll> who's around?
17:00:16 <vsaienk01> \o
17:00:16 <thiagop> o/
17:00:17 <pas-ha> o/
17:00:18 <alineb> o/
17:00:20 <rpioso> o/
17:00:20 <cdearborn> \o
17:00:20 <jlvillal> o/
17:00:25 <jroll> agenda as usual is here
17:00:27 <jroll> #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting
17:00:27 <rloo> o/
17:00:33 <jroll> #topic announcements and reminders
17:00:44 <jroll> a few things
17:00:51 <jroll> we're closing in on the end of the cycle, schedule is here:
17:00:53 <jroll> #link http://releases.openstack.org/newton/schedule.html
17:01:03 <mat128> o/
17:01:05 <lucasagomes> o/
17:01:06 <jroll> #info non-client library final releases this week
17:01:12 <jroll> #info client library final releases next week
17:01:12 <devananda> o/
17:01:21 <thiagop> #info OneView CI is out due to another cloud breakage. Will be back yet this week.
17:01:39 <thiagop> Probably tomorrow, if everything goes as expected
17:01:41 <jroll> we don't typically follow feature freeze, but I'd like to stop accepting huge and/or risky features after september 2
17:01:57 <jroll> #info soft feature freeze (nothing huge and/or risky) september 2
17:02:06 <jroll> thiagop: thanks
17:02:11 <jroll> anything else here?
17:02:23 <krtaylor> o/
17:02:32 <vdrok> also please remember about string freeze
17:02:34 <jroll> oh, please chime in on this driver/CI email thread:
17:02:36 <jroll> #link http://lists.openstack.org/pipermail/openstack-dev/2016-August/101988.html
17:02:39 <jroll> vdrok: ++
17:02:43 <rloo> jroll: are we OK accepting any features/changes that have associated client changes, after the client library release next week?
17:02:44 <jlvillal> I'm assuming ironic-lib falls into the non-client library category
17:02:54 <devananda> jroll: any plans to do another release as the basis for newton-final?
17:02:55 <vdrok> jlvillal: correct
17:03:11 <lucasagomes> jroll, ++ on my list to reply to that
17:03:14 <jroll> rloo: yes, however that feature will not be in the "stable/newton" version of the client
17:03:32 <jroll> devananda: of course! definitely one the week of sept 26, maybe one before if we feel like it
17:03:37 <rloo> jroll: right. just wanted to make sure we're ok with it in ironic/newton, but not client/newton.
17:03:43 <TheJulia> o/
17:03:57 <jroll> rloo: yeah, that's okay, people can use a newer client if they need it
17:05:02 * jroll moves on
17:05:08 <jroll> #topic subteam status reports
17:05:14 <jroll> #link https://etherpad.openstack.org/p/IronicWhiteBoard
17:05:24 <jroll> starting at line 80
17:05:33 * jroll gives some time to review
17:05:53 <mjturek1> o/
17:05:57 <wajdi> o/
17:06:04 <rama_y> o/
17:07:07 <jroll> dtantsur: need help with that liberty IPA bug?
17:07:25 <rloo> dtantsur: that critical bug. I see that jlvillal proposed something but it is -1'd.
17:07:40 <dtantsur> jroll, yes please, I'm a bit out of time. I think we just have to make sure to use upper-constaints on liberty
17:07:42 <rloo> dtantsur, jlvillal: you got a handle on that? want to make sure someone is working on it.
17:07:53 <jroll> dtantsur: okay, I'll put it on my todo list
17:08:03 <jroll> unless jlvillal wants to pick it up
17:08:25 <jlvillal> jroll: If you want to grab it, that is okay. But I can put some time on it this week.
17:08:26 <jroll> oh, this is what he did
17:08:51 <jlvillal> jroll: The problem is that it got complicated with docker images and I haven't figured that part out.
17:09:04 <dtantsur> I'd say ignore docker images for now...
17:09:08 <rloo> wrt the portgroup patches. we're hoping to land those in newton, right? is it only in ironic? or are there nova/neutron-associated patches?
17:09:21 <jroll> jlvillal: I'd prefer to just land it and fix the dockerfile later
17:09:22 <jlvillal> dtantsur: That's what I wanted to do...
17:09:27 <jroll> but tonyb is -1 as well :/
17:09:45 <jroll> rloo: in ironic in newton, is the goal, and use them in nova in ocata
17:09:46 <dtantsur> I hope he proposed something instead :/
17:09:51 <jroll> he did
17:10:09 <jlvillal> I'll go see what I can do. Today or tomorrow.
17:10:11 <rloo> jroll: thx (wrt portgroups)
17:10:16 <dtantsur> thanks jlvillal
17:10:16 <jlvillal> But me down for that action item.
17:10:20 <jlvillal> s/But/Put/
17:10:24 <jroll> jlvillal: thanks
17:10:31 <mat128> jlvillal, jroll: I can handle the Dockerfile part
17:10:36 <jroll> \o/
17:10:45 <jroll> I agree it should be a separate patch, backported all the way
17:11:00 <mat128> ok, I'll do it in master then submit cherry picks in other branches
17:11:10 <jlvillal> mat128: Thanks!
17:11:11 <jroll> cool, thanks
17:11:31 <jlvillal> jroll: dtantsur: Does that mean want to land the non-docker part now? Or wait?
17:11:43 <jroll> jlvillal: now (well, maybe after the meeting
17:11:45 <jroll> )
17:11:48 <rloo> lucasagomes: ironic-lib freeze is this week. w/o the fix to the oslo specs_matcher, will the ironic-lib patches work?
17:11:50 <jlvillal> heh
17:12:18 <dtantsur> jlvillal, I'd love to unblock the stable gate, so yes
17:12:27 <lucasagomes> rloo, we could workaround that space problem by having an extra method that normalizes the string using urllib.parse.quote
17:12:27 <jlvillal> Okay
17:12:35 <lucasagomes> rloo, but it's really a workaround :-/
17:13:16 <rloo> lucasagomes: well, let me/us know what/if we should bother reviewing/getting those ironic-lib patches in this week.
17:13:17 <devananda> I did a bunch of doc work last week. there's not a spot on the 'pad to note it, but it will change the landing page, so I want to call attention to it here
17:13:28 <devananda> https://review.openstack.org/#/c/356788/
17:13:43 <rloo> devananda: but the landing page URL is the same, right?
17:13:43 <lucasagomes> rloo, yeah, I will do
17:13:49 <devananda> rloo: yes
17:13:59 <rloo> devananda: and the links from the old landing page will still work?
17:14:03 <devananda> rloo: yes :)
17:14:06 <devananda> rloo: see http://docs-draft.openstack.org/88/356788/3/check/gate-ironic-docs-ubuntu-xenial/ca94c68//doc/build/html/
17:14:09 <lucasagomes> rloo, thanks for volunteering to review it btw :-)
17:14:16 <rloo> lucasagomes: yw :)
17:14:38 <devananda> I didn't break any links - just made the frontpage layout more clear, with more words and sections
17:14:51 <rloo> devananda: ok, will comment in patch. (i think we had admin first cuz that stuff isn't in official openstack docs)
17:15:48 <rloo> nice work on inspector, dtantsur et al
17:15:56 <devananda> rloo: sure, please do. I also changed the order intentionally, because this is developer docs. and the TOC on the left makes it easy to find the admin guide now, and there's a note in the intro about admin docs.
17:16:45 <rloo> devananda: will do :)
17:16:48 <dtantsur> rloo, thanks :) most of the credit for inspector grenade work goes to milan and folks who did the ironic grenade work
17:17:28 <jroll> ++, thank you all for that
17:17:39 <jroll> and thank you deva for the awesome doc updates, I'll re-review that this week
17:17:44 <devananda> jroll: thanks
17:17:44 <vgadiraj> devananda: I can take a look at the docs patch as well
17:17:59 <devananda> I am hoping we can land the api-ref update soon as well
17:18:15 <devananda> it's been up for a while, and if anyone hsa more feedback, I'd really prefer to do thta in a follow up
17:18:53 <jroll> things I want to land this week: deva's doc update, api-ref update, anything in ironic-lib, osc things for the client
17:18:56 <devananda> our keystone and agent vendor work (both on the N-2 milestone must-have list) include an update to the api-ref as part of the work, which is not yet landed
17:19:04 <jroll> exactly :)
17:19:36 <jroll> anything else on this topic?
17:20:14 <jroll> I'm going to do a quick one not on the agenda
17:20:20 <jroll> #topic summit session allocations
17:20:32 <jroll> so thierry reached out to PTLs and asked how many slots everyone needs
17:20:37 <jroll> a couple changes this round
17:20:47 <jroll> workroom and fishbowls are happening Wednesday 4pm to Friday noon
17:20:59 <jroll> and the usual friday contributor meetups are only friday afternoon
17:21:14 <jroll> there are less slots than austin, and more teams
17:21:24 <jroll> so we should cut back from last time if possible
17:21:44 <jroll> last time we had 5 fishbowls, 5 workrooms, and half day contributor meetup
17:22:02 <jroll> I'd propose 4/3/half
17:22:15 <jroll> (assuming people can stick around for friday afternoon
17:22:17 <jroll> )
17:22:20 <TheJulia> That sounds reasonable
17:22:29 <jroll> agree/disagree/throw things at me now
17:22:29 <rloo> jroll: remind me again (cuz i always ask). fishbowl == big audience, workroom == design?
17:22:33 <devananda> in the past, folks often left on friday afternoon, so perhaps we should encourage folks to avoid that?
17:22:36 <jroll> rloo: correct
17:22:40 <devananda> rloo: yes
17:23:11 <clarkb> devananda: conference also starts on tuesday so you get a day back at the beginning of the week (may make friday more tolerable)
17:23:14 <rloo> i can't recall, but i thought the workrooms were more productive. I'll need to review what we did last summit.
17:23:19 <jroll> I could also deal with 3/4/half
17:23:21 <krtaylor> seems reasonable - we can shift QA/CI to a breakout in Fri contributor like we did in Austin
17:23:22 <devananda> jroll: I have often found the workrooms more productive than the fishbowls
17:23:28 <jroll> assuming we can keep non-contributors out of the workrooms
17:23:31 <devananda> jroll: 3/4/half would get my vote
17:23:32 <jroll> some of them got pretty full :)
17:23:34 <vdrok> is the contributor meetup going to be around 2pm-6pm?
17:23:40 <lucasagomes> rloo, I agree yeah
17:23:53 <jroll> vdrok: I don't have times yet - I'd assume 1pm-5pm or so?
17:24:01 <devananda> clarkb: noted
17:24:05 <jroll> in the afternoon, in any case
17:24:22 <vdrok> hm, I'll have to leave around 4pm :(
17:24:24 <rloo> jroll: fri AM is probably workroom/fishbowls?
17:24:38 <jroll> rloo: yes, "workroom and fishbowls are happening Wednesday 4pm to Friday noon"
17:25:02 <rloo> jroll: sorry, missed that. wow, 4pm wed...
17:25:03 <devananda> jroll: starting at 4pm wed? that's odd
17:25:22 <jroll> keep in mind cross-project is before that
17:25:24 <rloo> i'm guessing x-project will be before that on wed
17:25:34 <jroll> from the email:
17:25:36 <jroll> - Ops summit on Tuesday morning until 4pm
17:25:38 <jroll> - Cross-project workshops from Tuesday 4pm to Wednesday 4pm
17:25:40 <jroll> - Fishbowl/Workroom sessions from Wednesday 4pm to Friday noon
17:25:42 <jroll> - Contributors meetups on Friday afternoon
17:26:17 <jroll> also, as a reminder, ocata is only 5 months long, here's the proposed schedule: https://review.openstack.org/#/c/357214/
17:26:53 * dtantsur would prefer to think about topic first
17:27:20 <devananda> since Ocata is also going to be a shorter cycle, I think it will be fine to have fewer slots and focus on work we can accomplish in that time, rather than new design work
17:27:20 <rloo> i'm fine asking for 7 sessions, fishbowl or workroom. if we don't know the topics, maybe it is safer to ask for 4/3 as jroll proposed. we can design in fishbowl too.
17:27:21 <jroll> I'm meant to let thierry know by august 31, btw
17:28:08 <rloo> jroll: so we can decide aug 30?
17:28:28 <rloo> jroll: or err next Mon meeting?
17:28:48 <jroll> rloo: yeah, we can
17:28:58 <rloo> are we ready to ask for topics for summit?
17:29:02 <jroll> I guess everyone go think about it and we'll decide next meeting :)
17:29:09 <rloo> that can help inform our decision next week
17:29:10 <thiagop> makes sense
17:29:15 <devananda> jroll: how many total session slots are there in the timeframe of wed afternoon - friday noon?
17:29:18 <jroll> rloo: probably
17:29:21 <jroll> devananda: I do not know
17:29:57 <devananda> my recollection from past summits is that there were 8 per day
17:30:25 <jroll> so probably 14-15 total, then? (2-3 + 8 + 4)
17:30:27 <devananda> given this schedule, it sounds like 2 on Wed, 8 on Thu, 3 on Fri.
17:30:39 <devananda> yea ... something like that
17:31:10 <jroll> at any rate, let's make a decision next week
17:31:18 <rloo> so maybe 7 is too many, given the number of projects.
17:31:23 <jroll> rloo: could you kick a ML thread with an etherpad for topic suggestions?
17:31:29 <rloo> jroll: sure
17:31:33 <jroll> thanks
17:31:41 * rloo wants to practise kicking
17:32:01 <jroll> and everyone promise to noodle on this for a bit :)
17:32:09 <jroll> shall we move on?
17:32:16 <mat128> rloo: https://www.youtube.com/watch?v=jXhfikJzQ4w
17:32:39 <jroll> #topic open discussion
17:32:45 <rloo> mat128: :)
17:32:55 <jroll> I just have one thing here, I'll be out from (end of meeting) through wednesday (back thursday)
17:33:05 <jroll> dtantsur will be handling releases while I'm gone
17:33:11 <rloo> jroll: today's meeting?
17:33:13 <jroll> ping me on hangouts if you need something :)
17:33:15 <jroll> rloo: yep!
17:33:30 * jroll heading to mackinac island for a couple days
17:33:38 <rloo> jroll: enjoy!
17:33:45 <thiagop> As some of you may know, I've been trying to remove our tests from tempest.
17:33:51 <lucasagomes> jroll, enjoy the days off
17:33:59 <jroll> thanks :)
17:34:14 <devananda> I'm attending OSE and the ops midcycle this week, so will be only partly available.
17:34:15 <vsaienk01> I've prepared a spec https://review.openstack.org/343684 and PoC https://review.openstack.org/345963 for Neutron notifications
17:34:36 <vsaienk01> can we start reviewing them? :)
17:34:49 <thiagop> mrtreinish has asked me to change the approach from "this is not supported because ironic is available" to "ironic is not supported with this feature" or something
17:35:04 <jroll> vsaienk01: I'd prefer focusing on priorities for the next couple weeks :)
17:35:24 <thiagop> I had to move some sticks on devstack to make it happen, so I'd love if some of you take a look to check if https://review.openstack.org/#/c/358116/ makes sense
17:35:30 <devananda> vsaienk01: is this a necessary part of the newton planned integration work?
17:35:32 <thiagop> jlvillal: specially you^
17:35:38 <thiagop> :)
17:35:49 <thiagop> grenade may not be happy
17:36:00 <NobodyCam> anyone happen to know if Shivanand is planning on pushing up a new rev of 225115?
17:36:01 <vsaienk01> devananda: I don't think so, It is a separate task, with multitenancy it is specially actual
17:36:31 <devananda> thiagop: this moves to using tempest-lib and putting the tests in ironic, yes?
17:36:32 <jlvillal> thiagop: okay
17:36:35 * devananda actually looks at the patches now
17:36:48 <vsaienk01> since port binding with multitenancy may take a lot of time, and failure rate is much higher I think
17:37:07 <thiagop> devananda: in fact, I'm just removing the old BaremetalBasicOps from tempest
17:37:32 <devananda> thiagop: is there another / newer tempest test covers that functionality?
17:37:38 <thiagop> devananda: it's very out of date with what we have in the plugin
17:38:00 <jroll> devananda: there's no moving of tests, only removing all of the ironic-specific things from tempest's tree
17:38:01 <lucasagomes> NobodyCam, no idea, I think it would be ok to fix the current comments and add a patch-set ourselves
17:38:03 <thiagop> devananda: those are on the plugin for a while now
17:38:04 <lucasagomes> NobodyCam, it's nothing majo
17:38:06 <lucasagomes> major*
17:38:17 <jroll> devananda: deleting the things in the plugin, and deleting all the "if ironic is enabled" hacks
17:38:18 <lucasagomes> there's nothing major*
17:38:31 <devananda> thiagop: ah, I see
17:38:36 <NobodyCam> yea, I push up a fix so it can land if he's away
17:38:59 <thiagop> jroll: no test moving, just some configs (on 355591)
17:39:01 <NobodyCam> s/I/I'll/
17:39:09 <jroll> thiagop: yep :)
17:39:43 <thiagop> jroll: my doubts are around if it makes sense to disable the features I disabled in devstack if Ironic is present
17:40:14 <jroll> thiagop: okay, I'd normally take a look but am leaving after this meeting, hopefully some others can
17:40:19 <thiagop> if so, then I'll move to the next step
17:40:24 <jroll> JayF isn't here yet but will be later and is a good candidate
17:40:32 <thiagop> jroll: sure, have a good trip :)
17:40:40 <jroll> :)
17:41:42 <jroll> anyone have other things?
17:42:34 <jroll> alright, thanks for joining :)
17:42:35 <NobodyCam> :)
17:42:36 <jroll> #endmeeting