17:00:03 <NobodyCam> #startmeeting Ironic
17:00:03 <NobodyCam> #chair jroll
17:00:03 <NobodyCam> Welcome everyone to the Ironic meeting.
17:00:04 <openstack> Meeting started Mon Nov  9 17:00:03 2015 UTC and is due to finish in 60 minutes.  The chair is NobodyCam. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:05 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:07 <openstack> The meeting name has been set to 'ironic'
17:00:09 <openstack> Current chairs: NobodyCam jroll
17:00:12 <devananda> o/
17:00:15 <TheJulia> o/
17:00:16 <sambetts> o/ everyone
17:00:16 <lucasagomes> o/
17:00:20 <dtantsur> o/
17:00:22 <mgould> o/
17:00:24 <liliars> o/
17:00:25 <yuriyz> o/
17:00:26 <NobodyCam> morning all
17:00:27 <thiagop> o/
17:00:33 <thrash> o/
17:00:37 <vdrok> o/
17:00:43 <aarefiev22> hi
17:00:47 <NobodyCam> jroll said he may be a few minutes late
17:01:08 <cinerama> o/
17:01:13 <trown> o/
17:01:17 <NobodyCam> Of course the agenda can be found at:
17:01:17 <NobodyCam> #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting
17:01:24 <rloo> hi
17:01:30 <NobodyCam> #topic Greetings, roll-call and announcements
17:01:30 <NobodyCam> Roll-call: Who's here for the Ironic Meeting?
17:01:36 <NobodyCam> seems we just did that
17:01:38 <jlvillal> o/
17:01:55 * jlvillal waves from San Antonio
17:02:17 <lucasagomes> :-)
17:02:17 <NobodyCam> welcome all to the ironic meeting
17:02:34 <NobodyCam> hope everyone had safe travels back from tokyo
17:02:44 <wanyen> o/
17:02:46 <krtaylor> o/
17:02:54 <NobodyCam> Shall we jump in to the fire
17:03:03 <NobodyCam> #topic SubTeam: status report
17:03:03 <NobodyCam> Posted on Whiteboard
17:03:03 <NobodyCam> #link https://etherpad.openstack.org/p/IronicWhiteBoard
17:04:50 <NobodyCam> any one with status updates for today?
17:05:12 <rloo> NobodyCam: looks like folks are updating as we stare :)
17:05:15 <jlvillal> :)
17:05:28 <NobodyCam> yep
17:05:48 <NobodyCam> I also may be slow do to unshakeable jetjeg
17:05:55 <NobodyCam> lag even
17:05:58 <rloo> nice to see testing picking up steam!
17:06:10 <NobodyCam> yes!
17:06:10 <krtaylor> woot!
17:07:11 <NobodyCam> also I was going to share the proitories review link. but I seem to have miss placed it
17:07:57 <krtaylor> thanks for the CI spec reviews, I'll have a new revision today and incorporate everyones comments
17:08:14 <krtaylor> here is the priorities link: https://review.openstack.org/#/c/241223/
17:08:15 <rloo> NobodyCam: http://specs.openstack.org/openstack/ironic-specs/priorities/mitaka-priorities.html
17:08:32 <lucasagomes> #link http://specs.openstack.org/openstack/ironic-specs/priorities/mitaka-priorities.html
17:09:27 <NobodyCam> Ahh thank you rloo, lucasagomes and krtaylor
17:09:29 <NobodyCam> :)
17:09:40 <rloo> do we want subteam reports for each of those priorities?
17:10:09 * jlvillal thinks that would probably be a good idea for the future.
17:10:23 <devananda> rloo: as the cycle progresses, I think that would be good
17:10:38 <lucasagomes> sounds good indeed
17:10:55 <NobodyCam> I thibk it may be worth at least a line item on the whiteboard so folks can easly track the progress but not sure we need to spend a lot of time in the meetings unless there are things to discuss / go over
17:10:58 <rloo> I'll add subteams for those to the etherpad after this meeting
17:11:28 <NobodyCam> Thank you rloo
17:13:06 <NobodyCam> at thing (else) from any of the subteams
17:13:07 <rloo> dtantsur: what is 'RFE'?
17:13:25 <sambetts> rloo: Request for Enhancement
17:13:48 <rloo> sambetts: thx. I was thinking it was some Fulltime Engineer or something :)
17:14:21 <dtantsur> rloo, oh, you scared me at first :D actually this note was written by Sam
17:14:59 <rloo> I'm good to move onto the next meeting item :)
17:15:16 <NobodyCam> okay looks like we have a couple item inthe discussion section so
17:15:18 <jroll> hey, sorry I'm late
17:15:21 * jroll catches up
17:15:47 <NobodyCam> hey hey jroll just thru the subteam reports
17:16:05 * NobodyCam hands control pver to jroll
17:16:16 <jroll> so subteam reports are done?
17:16:32 <NobodyCam> yep / or being updated on the pad
17:16:39 <lucasagomes> yeah seems so, in the etherpad
17:16:46 <rloo> jroll: well, you might have something to add?
17:16:48 <jlvillal> jroll, Unless you want to walk through them.
17:17:11 <rloo> jroll: neutron/ironic work. patches up for review? :)
17:17:31 <jroll> yes, those have been up for review, a couple small issues
17:18:14 <yuriyz> i have some questions about in the gerrit, wait for answers for anyone
17:18:29 <jroll> yuriyz: on neutron?
17:18:51 <yuriyz> Ironic net isolation support
17:19:08 <jroll> ok, yeah we'll answer in the review
17:19:10 <yuriyz> https://review.openstack.org/#/c/139687/
17:19:22 <yuriyz> https://review.openstack.org/#/c/213262/
17:19:52 <jroll> alright, let's move on then if nobody has more subteam things
17:20:02 <jroll> #topic Do we need to have a midcycle?
17:20:26 <jroll> so, a few projects (glance/ceilometer/more?) are skipping the midcycle this cycle
17:20:32 <jroll> for a variety of reasons
17:20:35 <lucasagomes> that's a hard on, I like to think about midcycles as a place to hack on code
17:20:53 <lucasagomes> if we have something big to hack on, it would be good to have one yes
17:20:57 <jroll> and I wanted to ask folks if we think we should have one, based on the fact we have clear priorities and not everyone will make it anyway
17:21:01 <dtantsur> I'd like to see you all again asap, but probably midcycle is not really needed...
17:21:28 <krtaylor> neutron has skipped a midcycle also, was thinking it would be good to have at same time with them
17:21:32 <jroll> I think the midcycle is super valuable from a see-people-in-person perspective, otherwise I'm pretty indifferent
17:21:56 <jroll> please throw all of your thoughts at me - if we decide to have one, I'll plan it ASAP otherwise happy hacking :D
17:22:05 <rloo> much as I like you all, I'm fine seeing you every 6 months :)
17:22:10 <sambetts> I think small team virtual hack sessions would be a cool thing to have, using hangouts or something
17:22:11 <lucasagomes> lol
17:22:12 <jlvillal> I like having a mid-cycle. Hard to beat in-person interaction.
17:22:35 <devananda> ++ to seeing everyone and maintaining those connections four times a year instead of just twice a year, and ++ when we need to decide on plans or architecture, which I don't think we need this cycle as much as last cycle
17:22:39 <mgould> where would it be, physically?
17:22:54 <NobodyCam> i good chunch of what we have out standing is actually from last cycle, might be good to have a meetup to keep things on track
17:22:59 <jroll> if we do have one, it will likely be US based on the feedback I've seen
17:22:59 <dtantsur> wherever we find a venue, I guess
17:23:00 * mgould would like to meet you all, but hates travelling
17:23:04 <krtaylor> if not, maybe a couple vsprints  -> https://wiki.openstack.org/wiki/VirtualSprints
17:23:10 <lucasagomes> sambetts, ++ I would love to try some virtual hack session
17:23:14 <NobodyCam> and how may folks could / would attend?
17:23:25 <dtantsur> USA is unlikely for me for a usual reason
17:23:56 <vdrok> I'd like to meet everyone in person but not sure I'll attend
17:24:21 <dtantsur> oh yeah, meeting vdrok would be a serious reason :)
17:24:27 <jroll> I like the idea of virtual sprints as well
17:24:29 <sambetts> the US is not as hard for me as dtantsur, but its still a long way to travel and then deal with the after effects
17:24:51 <devananda> what about coordinated regional meetups?
17:25:11 <jroll> devananda: how did that work out when we did a split midcycle? :)
17:25:28 <devananda> it requires more planning than I have had bandwidth to do in the past, but other projects have had some success in the past with it as they grew
17:25:38 <dtantsur> I would try something like that
17:25:46 <dtantsur> the only problem is timezones
17:25:47 <devananda> jroll: I didn't intend to do a split midcycle last winter
17:25:55 <lucasagomes> yeah sounds like something worth trying indeed
17:26:26 <devananda> jroll: also they were not overlapping
17:26:37 <NobodyCam> having them at the same time instead of one after the other may be better
17:26:44 <NobodyCam> devananda: ya
17:27:07 <devananda> if we  located a meetup in western EU and eastern US on overlapping days, there would be a few hours each day for video conferencing
17:27:33 <jlvillal> That sounds interesting :)
17:27:38 <devananda> not saying I think we need that -- but it is an option we haven't tried before
17:27:46 <NobodyCam> that may work
17:27:53 <sambetts> I'm interested
17:28:48 <jroll> okay, would folks like me to continue investigating that?
17:28:50 <NobodyCam> are there locations that could work for that?
17:29:03 <lucasagomes> so we have some options: 1) normal midcycle, 2) virtual sprint, 3) splitted midcycle. Perhaps we should continue on the ML ?
17:29:12 <jroll> lucasagomes++
17:29:15 <rloo> 4) no midcycle
17:29:17 <jlvillal> +1 from me to continue investigating. As I like the idea of a mid-cycle
17:29:23 <lucasagomes> rloo, yeah, or no midcycle
17:29:34 <vdrok> when is it gonna be, approximately?
17:29:47 <dtantsur> January?
17:29:48 <vdrok> if it will be
17:29:52 <jroll> likely late january - early february, I think
17:30:00 <jroll> preferably on the earlier end
17:30:07 * dtantsur tries to remember when FOSDEM is
17:30:11 <lucasagomes> early february ++
17:30:16 <jroll> dtantsur: jan 29-30 iirc
17:30:23 <dtantsur> something like that
17:30:25 <jroll> lucasagomes: would you be able to lay out the options on the ML and we can go from there?
17:30:27 <dtantsur> we can have it right after
17:30:31 <lucasagomes> dtantsur, 30-31 jan
17:30:54 <mgould> are many people here likely to be at fosdem?
17:31:14 <devananda> fwiw, right after FOSDEM was the timing last winter
17:31:39 <sambetts> myself, lucasagomes and dtantsur have discussed meeting at FOSDEM
17:31:53 <lucasagomes> that's right! I will be there
17:32:05 <dtantsur> mgould, I'll try to
17:32:28 <NobodyCam> where is FOSDEM?
17:32:37 <jroll> brussels
17:32:42 <lucasagomes> Brussels, Belgium
17:32:42 <jroll> lucasagomes: would you be able to lay out the options on the ML and we can go from there?
17:32:42 <sambetts> #link https://fosdem.org/2016/
17:32:58 <lucasagomes> jroll, ack, will do
17:33:35 <jroll> thanks
17:33:36 * lucasagomes adds to the TODO list for tomorrow morning
17:33:46 <jroll> anyone else have anything on this topic?
17:34:22 <jroll> #topic release notes management
17:34:39 <jroll> a couple things here
17:35:05 <jroll> 1) we need to move to using a thing called reno for release notes, see this for more info on how that works:
17:35:06 <jroll> #link http://lists.openstack.org/pipermail/openstack-dev/2015-November/078301.html
17:35:34 <jroll> I'd love it if patches that deserve release notes come in with the release note in the same patch; we can also do it after the fact
17:35:40 <jlvillal> RElease NOtes I'm guessing
17:36:10 <jroll> I've started the necessary work here:
17:36:12 <jroll> #link https://review.openstack.org/#/q/status:open++branch:master+topic:add-reno+owner:%22Jim+Rollenhagen+%253Cjim%2540jimrollenhagen.com%253E%22,n,z
17:36:14 <jroll> jlvillal: yes
17:36:25 <rloo> some guidelines wrt what patches 'deserve' release notes, would be useful
17:36:50 <jroll> sure, I can try to come up with something - to start, think about what we've put for rel notes in the past
17:37:00 <jroll> mostly features or major bug fixes, upgrade notes, etc
17:37:19 <rloo> jroll: right. so not detailed stuff and not minor features
17:37:25 <jroll> yep
17:37:30 <jroll> so that's just an announcement, the bigger part I want folks to start thinking about is
17:37:41 <jroll> 2) we can move off of launchpad for feature tracking if we'd like
17:38:06 <jroll> it's not required, but also launchpad is the worst :)
17:38:08 <devananda> unfortunately, I need to step away for a bit (darn travel schedule) ... but I, for one, welcome our new ReNo overlords
17:38:22 <jroll> so, we can talk about that here or we can go think about it and come back to it later
17:38:24 <jroll> devananda: \o
17:38:25 <lucasagomes> devananda, see ya later!
17:38:30 <dtantsur> "User Jim Rollenhagen %3Cjim%40jimrollenhagen.com%3E not found"
17:38:33 <NobodyCam> safe travels devananda
17:38:37 <devananda> also huge ++ from me on moving away from LP for feature tracking
17:38:49 <devananda> but pls keep bugs there
17:38:50 <lucasagomes> jroll, do you have any suggestion of alternatives to launchpad?
17:39:25 <jroll> dtantsur: gerrit is the worst, idk what to say
17:39:34 <lucasagomes> jroll, also, bugs still needs to be tracked in launchpad right?
17:39:40 <jlvillal> dtantsur, Firefox?
17:39:41 <jroll> the main reno thing is here:
17:39:43 <jroll> #link https://review.openstack.org/242147
17:39:54 <NobodyCam> jroll: would moveing off LP for feature track make it difficult for other project contributors to work with us
17:39:54 <jroll> lucasagomes: bugs in launchpad, yes - I don't think that's painful today
17:39:57 <jlvillal> dtantsur, With Chrome it worked for me, but failed with Firefox :(
17:40:00 <jroll> feature tracking on launchpad is horrible
17:40:11 <jroll> NobodyCam: I'm not sure, that's something to consider
17:40:45 <jlvillal> Would reno take over for feature tracking also?  In addition to release notes.
17:40:52 <jlvillal> Or something else for feature tracking?
17:40:59 <jroll> reno is only release notes
17:41:01 <jroll> so
17:41:13 <jroll> some folks on the infra team are working on standing up phabricator, which is an option
17:41:14 <rloo> seems like every openstack project will be asking. to use LP or not, that is the question.
17:41:18 <NobodyCam> I do agree that feature tracking is horrible, but I am concerned that we'll make it harder for folks outside the ironic project to contribute
17:41:27 <jroll> and likely the best for consistency with other projects down the road
17:41:58 <rloo> it makes sense for (most of) openstack to use the same feature tracking whatever.
17:42:10 * jroll thinks this is probably best for the mailing list, but wanted to get folks thinking about it
17:42:28 <lucasagomes> rloo, ++
17:42:36 <NobodyCam> jroll: ++ yes would love to hear from other projects on this
17:42:47 <sambetts> #link http://phabricator.org/
17:42:56 <jroll> ok
17:43:00 <lucasagomes> yeah I would like to see it being cross project, many times we have features that are cross project
17:43:10 <lucasagomes> and having to work across different systems may be messy/painful
17:43:11 <jroll> yeah
17:43:17 <jroll> blueprints don't do cross project at all fwiw
17:43:42 <lucasagomes> right, but at least it's all in the same "thing" (launchbad)
17:43:50 <jroll> anyway, we won't make a decision here, but thanks in advance for putting some thought into it :)
17:44:07 <jroll> I'm going to go to open discussion, we can keep talking about this, or anything else on your mind
17:44:19 <jroll> #topic open discussion
17:44:22 <yuriyz> IMO there is a problem in Ironic with slow spec review process Do we 1) add more people to spec cores 2) cores should set more -2 (minimize specs in "hanging" state), more?
17:44:37 <yuriyz> my specs is an example
17:45:01 <jroll> probably both and more? :)
17:45:12 <jroll> spec review always slows down a bit around summits
17:45:25 <yuriyz> +1 for both
17:45:27 <rloo> i htink spec reviews are always slow
17:45:42 <NobodyCam> yuriyz: I can only speak for myself here: I know I have been slow reviewing... and plan on improving that volicity starting this week
17:46:15 <jroll> yeah, they are always a bit slow - I think we also need focus on which specs we review, sometimes we are very sporadic with them
17:46:19 <rloo> would it be useful to have spec sprints (or whatever they are called), once a week to review specs that are close or have contentious issues?
17:46:30 <jroll> rloo: I think it would
17:46:31 <yuriyz> -2 if spec is not suitable for Ironic/current cycle
17:46:42 <lucasagomes> like a spec jam
17:46:44 <yuriyz> should be set early
17:46:51 <rloo> i'm not sure we should -2 wrt 'current cycle'
17:46:57 <jroll> yuriyz: I agree with -2 for "not suitable for ironic" but not current cycle
17:47:13 <yuriyz> jroll ok
17:47:43 <jroll> the hard part is specs that might be good for ironic but need a complete rewrite
17:47:57 <jroll> because the details don't make sense
17:48:01 <jroll> do we -1 or -2 those?
17:48:03 <jroll> etc
17:48:16 <NobodyCam> if we are going to start to keep a up todate proitiries spec we could come up with rules that help keep specs aligned with those
17:48:26 <yuriyz> yes, -1 but not to do review line-by-line
17:48:37 <rloo> yeah. I might be a bit slow, I'll start to read a spec and then get totally confused cuz I don't know what the spec is trying to say.
17:48:49 <yuriyz> just mention mega issues
17:48:49 <jroll> NobodyCam: no, priorities are just our focus, other work is still welcome
17:48:50 <jlvillal> But specs might be useful for future cycle priorities. Most of the priorities for Mitaka had specs land in Liberty.
17:49:13 <thiagop> maybe we need some caution not to push away good ideas by using a harsh policy
17:49:14 <lucasagomes> jroll, I think we should -1 for that reason and ask the author to clarify
17:49:42 <yuriyz> thiagop +1
17:49:45 <jroll> lucasagomes: yeah, it's really case by case too
17:49:48 <wanyen> I like the idea of weekly spec review jam
17:50:00 <jroll> thiagop: I agree, I won't let that happen :)
17:50:17 <jroll> rloo: how do you feel about organizing review jams? :)
17:50:29 <NobodyCam> I also like the review jam idea as long we we set a time, and can keep it the same
17:50:30 <jlvillal> Wonder if weekly spec review is something for entire cycle or a specific time frame of the cycle?
17:50:49 <rloo> if you're asking me to organize it, I would prefer not. I'm not a proponent of review jams even though I suggested it.
17:50:58 <jroll> heh
17:51:01 <lucasagomes> lol
17:51:07 <rloo> jroll: oh well. i suppose i could do it if no one else wants to.
17:51:10 <jroll> rloo: do you think it would be useful, though?
17:51:25 <lucasagomes> also, prior to have a review jam would be good to have a list of specs that we are confident that are almost there
17:51:27 <lucasagomes> and could be merged
17:51:29 <rloo> jroll: it could be useful. i was also going to ask. how much effort do folks put in writing their specs?
17:51:29 <jroll> I mean, I can find somebody else to organize those too
17:51:39 <jroll> lucasagomes++
17:51:43 <rloo> lucasagomes: yes, i agree.
17:51:44 <lucasagomes> or something that needs discussion to unstuck
17:51:45 <NobodyCam> lucasagomes: ++ limited jam scope ++
17:51:45 <jroll> great question :)
17:51:56 <dtantsur> yeah, review jam would be useful if we at least don't read these specs in advance
17:52:40 <sambetts> at least don't read them? or have at least read them?
17:52:58 <jroll> so how's this - let's collect stuck/close specs in the meeting agenda. spend 5 minutes in each meeting taking a look at the list and (if needed) plan a jam for that week
17:52:59 <yuriyz> add actions?
17:53:01 * lucasagomes thinks it's "at least read them"
17:53:03 <NobodyCam> 7 minutes left btw
17:53:40 <NobodyCam> jroll: ++ I loke that to start
17:53:41 <lucasagomes> jroll, sounds like a good start
17:53:49 <NobodyCam> like even
17:54:01 <jroll> cool, I'll add the bullet point in the agenda
17:54:01 <vdrok> ++
17:54:11 <rloo> jroll: i'm not sure i like spending time in the meeting figuring out if we need another meeting, but let's try/see how it goes.
17:54:27 <dtantsur> yeah, sorry, I meant we should read them :)
17:54:38 <jroll> rloo: more like "what time works for folks?
17:54:40 <jroll> "
17:55:03 <jroll> rloo: otherwise it's going to involve catching people randomly in irc and mailing list emails etc
17:55:13 <rloo> jroll: maybe we need a specs subteam
17:55:33 <NobodyCam> we have spec's cores thats kinda a subteam already
17:55:35 <yuriyz> we already have spec cores
17:55:40 <yuriyz> ==
17:55:45 <rloo> jroll: like dmitry does for bugs, and jlvillal & mrda for nova bugs. someone to highlight specs that need more attention
17:56:02 <jroll> rloo: maybe
17:56:12 <rloo> just thinking out loud.
17:56:16 <jroll> yeah
17:56:23 <jroll> let's try this next week and go from there?
17:56:33 <NobodyCam> ++
17:56:33 <lucasagomes> yeah, specs are kinda hard to one person to be responsable for
17:56:42 <yuriyz> +1
17:56:46 <lucasagomes> they are long and require specific knowledge sometimes
17:56:54 <rloo> lucasagomes: not responsible for. just to have an idea of status.
17:57:05 <lucasagomes> so... I don't know, I think collectively it would work better
17:57:11 <lucasagomes> rloo, right
17:57:23 <jroll> ok, I've gotta run a couple minutes early, sorry. NobodyCam can you hit the endmeeting button when we're done? :)
17:57:29 <dtantsur> we can assign devananda while he's out :D
17:57:32 <NobodyCam> I can
17:57:39 <lucasagomes> dtantsur, and now jroll
17:57:41 <thiagop> I don't know if you guys knows about next-review
17:57:42 <jroll> thanks all for coming today
17:57:44 <NobodyCam> thaNK you jroll
17:57:44 <jroll> lol
17:57:53 <jlvillal> thiagop, What's that?
17:57:56 <lucasagomes> jroll, see ya
17:58:08 <lucasagomes> thiagop, I don't know it
17:58:14 <thiagop> a tool created by Dolph to help find things to review
17:58:19 <NobodyCam> thiagop: I have not heard of it
17:58:20 <thiagop> https://github.com/dolph/next-review
17:58:36 * dtantsur does not think he has problems finding things to review... he finds too many things to review
17:58:37 <NobodyCam> #link https://github.com/dolph/next-review
17:58:49 <rloo> dtantsur: ++
17:58:57 <lucasagomes> thiagop, oh I will take a look, thanks for it!
17:59:11 <NobodyCam> one minute
17:59:13 <thiagop> this follows a simple philosophy and, when you guys have made a push to review all older specs, will become a very helpful
17:59:25 <thiagop> it may become*
17:59:37 <lucasagomes> oh that's pretty cool! It's like my todo list but automated!
17:59:39 <sambetts> The etherpad has an updated ironic-inbox gerrit page which i find very useful to help manage my reviews
17:59:42 <rloo> i don't think the onus is on cores to review the specs. everyone should be reviewing.
17:59:43 <NobodyCam> thiagop: will take a look
17:59:43 <dtantsur> I'd prefer to add rules to gertty
18:00:07 <NobodyCam> thats time
18:00:15 <NobodyCam> Thank you all for attending
18:00:20 <lucasagomes> thanks
18:00:26 <dtantsur> thanks
18:00:29 <sambetts> thanks everyone o/
18:00:33 <yuriyz> thanks
18:00:34 <NobodyCam> #endmeeting