19:02:48 <devananda> #startmeeting Ironic
19:02:48 <openstack> Meeting started Mon Jun 16 19:02:48 2014 UTC and is due to finish in 60 minutes.  The chair is devananda. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:02:48 <linggao> \o
19:02:49 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:02:52 <openstack> The meeting name has been set to 'ironic'
19:02:54 <gmatefi> o/
19:02:57 <rloo_> o/
19:02:57 <devananda> hi all!
19:02:58 <mrda> \o
19:03:01 <ifarkas> o/
19:03:08 <GheRivero> o/
19:03:08 <NobodyCam> hi devananda
19:03:15 <romcheg> Good morning/evening everyone!
19:03:17 <devananda> #chair NobodyCam
19:03:18 <openstack> Current chairs: NobodyCam devananda
19:03:35 <devananda> as usual, our agenda can be found here
19:03:38 <devananda> #link https://wiki.openstack.org/wiki/Meetings/Ironic
19:03:53 <devananda> #topic announcements
19:04:13 <devananda> I think there's a few announcements to make today
19:04:24 <devananda> which I didn't add to the agenda -- sorry
19:04:37 <jroll> devananda: I have one as well when you're done
19:04:39 <devananda> most notably, we landed a lot of bug fixes last week -- thanks to everyone who helped with reviews !
19:04:54 <devananda> and the J1 milestone was tagged on thursday
19:05:25 <devananda> also, sdague did a lot of work fixing the gate issues, one of which was removign ironic from the integrated gate
19:05:28 <mrda> \o/
19:05:40 <JayF> /o\ \o/
19:05:46 <devananda> so we currently have our own gate queue, meaning we can land patches much faster than we could in the last few weeks
19:05:55 <NobodyCam> :)
19:05:56 <devananda> so that should help us get some *features* landed in J2 :-D
19:06:04 <lucasagomes> :)
19:06:13 <devananda> ok, two more announcements, which are on the wiki
19:06:25 <devananda> bug tags -- dtantsur had started work on that, and I added several more
19:06:29 <devananda> and updated the main wiki
19:06:34 <NobodyCam> Thank you sdague !
19:06:40 <devananda> #link https://wiki.openstack.org/wiki/Bug_Tags#Ironic
19:06:54 <devananda> so when filing or triaging bugs, please try to tag them when there's something appropriate
19:07:17 <NobodyCam> oh I too have a minor thing
19:07:29 <devananda> that'll help folks find bugs in areas they are familiar with / want to work on
19:07:32 <devananda> lastly
19:07:32 <comstud> o/ (late)
19:07:33 <devananda> the midcycle
19:07:44 <devananda> we're confirmed and all that to be colocated with nova
19:08:05 <devananda> details are here
19:08:06 <devananda> #link https://wiki.openstack.org/wiki/Sprints/BeavertonJunoSprint
19:08:24 <devananda> if you are going to attend, please RSVP here: https://www.eventbrite.com/e/openstack-ironic-juno-mid-cycle-developer-meetup-tickets-11886066545
19:08:41 * NobodyCam notes to do that toda
19:08:43 <devananda> so I can tell the venue how many are coming
19:08:45 <NobodyCam> today
19:08:59 <devananda> also, I'd love us to have an agenda
19:09:03 <romcheg> devananda: What about invitations for those who do not live in the US?
19:09:05 <devananda> anyone want to coordinate that?
19:09:34 <devananda> #info anyone needing a VISA invitation, please email me directly
19:09:46 <lucasagomes> #link https://etherpad.openstack.org/p/juno-ironic-sprint
19:09:52 <devananda> romcheg: ^ I'll see waht I can do, but can't promise anything ... also such things often take a long time
19:10:02 <lucasagomes> please put the topics to be discussed in the juno mid-cycle sprint on that etherpad ^
19:10:03 <devananda> longer than the ~6 weeks until the sprint
19:10:04 <romcheg> devananda: Thanks for the info!
19:10:15 <NobodyCam> over the weekend I put up https://review.openstack.org/#/c/100063 which Moves check-tripleo-ironic-undercloud-precise from the experimental queue
19:10:18 <NobodyCam> to the check queue for TripleO and Ironic.
19:10:37 <linggao> is the mid cycle paid by openstack fundation ? :-)
19:10:37 <devananda> NobodyCam: please hold off on review discussions until later in the meeting
19:10:41 <devananda> linggao: no
19:11:10 <linggao> (-:
19:11:29 <devananda> we can talk more about the midcycle a bit later too
19:11:39 <devananda> that's it for my announcements. jroll, you had something quick?
19:11:47 <jroll> for anybody that happens to be in SF for gigaom structure conference (or otherwise) the SF rackspace office is having a structure after party on thursday night. you're all invited - our team (mostly in this office) has really enjoyed working with y'all :)
19:12:10 <jroll> that's all :)
19:12:12 <devananda> jroll: awesome! thanks for sharing
19:12:33 <devananda> jroll: have a link to more info (like dates)?
19:12:53 <comstud> this thursday at 5pm
19:13:01 <devananda> oh, this thursday. heh :)
19:13:03 <devananda> ok, moving on
19:13:04 <jroll> I should have one by end of day today, but yes, thursday at 5
19:13:15 <devananda> #topic release cycle progress reports
19:13:37 <devananda> as i mentioned, j1 was tagged
19:13:58 <devananda> #link https://launchpad.net/ironic/juno/juno-1
19:14:05 <linggao> when will be j2?
19:14:06 <devananda> 62 bug fixes, 0 blueprints
19:14:24 <devananda> j2 is targeted for july 24
19:14:47 <devananda> just before the sprint
19:15:05 <devananda> we are tracking things that we target to J2 here
19:15:07 <devananda> #link https://launchpad.net/ironic/+milestone/juno-2
19:15:20 <devananda> at the moment, we have only had one spec approved
19:15:27 <devananda> lucas' instance-info spec
19:15:35 <lucasagomes> yes
19:15:46 <devananda> i think we need to spend some time this week hammering on specs
19:15:52 <devananda> which we think we can implement by j2
19:15:59 <devananda> so that they're approved ASAP
19:16:24 <linggao> so how to make sure one blueprint/feature get into j2?
19:16:26 <jroll> ^ we definitely need more cores looking at specs, I have two and have seen very little action in the reviews
19:17:02 <rloo_> devananda: there is a separate list of folks that can +2 specs, vs reviews, right?
19:17:11 <devananda> rloo_: correct
19:17:30 <jroll> oh? who is on that list?
19:17:47 <rloo_> i thought russell was on it but i could be wrong.
19:17:50 <dtantsur> me
19:17:54 <lucasagomes> me
19:18:15 <jroll> rloo_: russell_h ?
19:18:16 <devananda> the lists are separate to allow for different types of contribution
19:18:21 <rloo_> jroll: be nice to those guys ;)
19:18:33 <romcheg> rloo_: https://review.openstack.org/#/admin/groups/352,members
19:18:39 <devananda> IMHO, not everyone reviewing specs needs to be reviewing code, but they do need to understand the project's issues
19:18:46 <devananda> anyway
19:18:49 <linggao> so if a spec is landed does that mean it approved for Juno?
19:18:55 <devananda> linggao: no
19:19:05 <JayF> Do specs require 1x+2 or 2x+2?
19:19:08 <linggao> I have some specs there, but not sure when do I supposed to start write code.
19:19:17 <lucasagomes> JayF, 2x+2
19:19:18 <devananda> let me find the link ...
19:19:30 <devananda> there is a good description of the process we're following in the wiki
19:19:33 <devananda> also
19:19:40 <linggao> I saw some patches checked in before the specs got landed.
19:19:48 <devananda> i am not infinitely scalable :(
19:19:54 <jroll> devananda: here's that link https://wiki.openstack.org/wiki/Blueprints#Nova
19:20:03 <mrda> devananda: not true! :P
19:20:21 <lucasagomes> linggao, yeah, some patches came from old blueprints that were transformed into a spec later (like the management interface one)
19:20:30 <devananda> jroll: yep, thanks
19:20:47 <linggao> lucasagomes, also iLO.
19:20:58 <lucasagomes> yeah
19:21:03 <devananda> linggao: approving a spec means that the proposal is accepted, but the code/feature may still not be targeted
19:21:04 <rloo_> devananda: could we add 'ironic' to that Blueprints wiki?
19:21:10 <linggao> lucasagomes,  also the whole disk support.
19:21:10 <devananda> linggao: that is tracked on launchpad
19:21:22 <dtantsur> linggao, they weren't landed, were they?
19:21:25 <devananda> we still have *A LOT* of features proposed without specs
19:21:36 <devananda> #link https://blueprints.launchpad.net/ironic
19:22:00 <devananda> that's not a call to write specs for all of them
19:22:05 <linggao> dtantsur, no.  I just want to know when is a code time to *start* write code.
19:22:23 <dtantsur> oh, and I'll have a question what to do with bugs like https://bugs.launchpad.net/ironic/+bug/1327187
19:22:24 <uvirtbot> Launchpad bug 1327187 in ironic "node extra attribute as capability in host states  " [Undecided,Invalid]
19:22:32 <devananda> linggao: that depends. what's really a question for reviewers is when do we /approve/ the code
19:22:46 <devananda> linggao: which should be after teh spec is approved and the feature is targeted to a milestone
19:23:04 <rloo_> linggao: I think you're free to write code whenever, but you'd need to be aware that if the spec isn't approved or the direction/design is changed, your code will also have to correspond to those changes.
19:23:07 <dtantsur> I understand people start writing some code before BP is targeted, we just should not approve it
19:23:13 <lucasagomes> rloo_, +1
19:23:36 <rloo_> to be honest, given the number of reviews, I won't even look at a review associated with a blueprint unless the blueprint is approved.
19:23:39 <linggao> rloo, that's my worry
19:23:43 <devananda> core reviewers should be getting better at landing code that has a spec approved, and blueprint targetred to current milestone
19:23:46 <devananda> aside from bug fixes
19:23:59 <devananda> rloo_: thank you :)
19:24:02 <mrda> dtantsur: so leave big code changes as WIP until spec approved? Or flag in the review that the spec is awaiting review?
19:24:04 <devananda> rloo_: you're doing it right :)
19:24:25 <devananda> rloo_: feel free to -2 code changes that link to blueprints which are not yet approved
19:24:33 <rloo_> devananda: i guess the question is (and was when this new blueprint process came up), is if the specs process will delay things for the developers.
19:24:40 <devananda> rloo_: yes, it will
19:24:52 <dtantsur> mrda, I suggest cores just never +2 it (maybe just +1)
19:24:54 <linggao> just do not like to write throw-away code.
19:24:55 <devananda> rloo_: but it should be a huge help to the core review team
19:25:12 <rloo_> is there a need for a 'blueprint spec' review day?
19:25:34 <devananda> rloo_: we do need to review the specs
19:25:36 <romcheg> rloo_: It's hard to do
19:25:37 <lucasagomes> problem with a spec review day is that we can't really land following patches
19:25:39 <JayF> I would be very curious as to the review stats for specs vs. code. I just don't see much movement in review for specs.
19:25:47 <lucasagomes> I think we need to dedicate a bit more time per day to review it
19:25:51 <romcheg> rloo_: Specs require careful reading
19:25:53 <lucasagomes> (will do it)
19:26:00 <devananda> so
19:26:07 <devananda> all of this is really a call to action for folks to review specs
19:26:16 <dtantsur> got it :)
19:26:20 <devananda> I think we need to land some big features this cycle
19:26:20 <romcheg> +!
19:26:22 <romcheg> +1
19:26:39 <rloo_> and get ironic graduated?
19:26:48 <devananda> rloo_: right
19:26:58 <devananda> rloo_: but a few of the feature proposals ARE towards taht end
19:27:01 <devananda> like the async api
19:27:13 <devananda> anyway... i'm goign to side track myself soon.
19:27:43 <devananda> goign to move on lest we spend the whole meeting talkign about specs
19:28:10 <devananda> #info need more people to review specs, particularly for features that we want to land in j2 or j3
19:28:18 <devananda> #topic sub team status reports
19:28:30 <devananda> Shrews, adam_g: hi! any updaets on the tempest work?
19:28:52 <adam_g> the  new tempest patches are still in review
19:29:09 <adam_g> biggest news here has been the ironic gate jobs becoming non-voting
19:29:28 <adam_g> question.. is this what we want? or should we be actively trying to get that change reverted?
19:29:58 <rloo_> adam_g: what are adv vs disadvs?
19:30:06 <adam_g> #link http://lists.openstack.org/pipermail/openstack-dev/2014-June/037424.html
19:30:08 <devananda> adam_g: aiui, you're referring to gate-tempest-dsvm-virtual-ironic being non-voting on ironic itself
19:30:23 <adam_g> devananda, yes, and on DIB
19:30:33 <devananda> adam_g: i feel quite strongly that our tempest tests should be voting in our gate
19:30:52 <Shrews> devananda: turns out adding new ironic tests to tempest is complicated by the fact that only a single bm node is available for the tests. This means parallel testing will not work (that's not done anyway), and we must add code to make sure the bm node is available for the next ironic test. I've added code for the second part (going through testing now).
19:30:59 <lucasagomes> yeah, I rely on that test, because that's the only test that actually deploys a vm end to end
19:31:14 <adam_g> rloo_, AFAIU, it allows ironic to land patches quicker and if our job ends up with a rate of failure, it does not affect the main gate queue for other projects (which was the case during the gate fire of  the last two weeks)
19:31:25 <adam_g> okay
19:31:32 <devananda> AIUI, sdague made that change because a) that test was failing often due to network issues, and b) ironic was in the integrated gate
19:31:43 <devananda> both of which have been addressed, afaik
19:31:46 <adam_g> yeah
19:31:53 <rloo_> adam_g: so those are pros. cons?
19:32:02 <adam_g> rloo_, pros, i suppose. cons are we dont get that test coverage as part of the gate :)
19:32:14 <romcheg> I think we should try to revert that change
19:32:28 <adam_g> the job was actually made non-voting because some fixes had not synced to all devstack slaves. but we appear to be stable again http://no-carrier.net/~adam/openstack/ironic_gate_status.html
19:32:31 <romcheg> Since it won't affect the main gate anymore
19:32:34 <devananda> Shrews: ah, a cleanup between tempest tests? yep, i think we need that
19:32:52 <dtantsur> +1 for making it voting again
19:33:10 <devananda> adam_g: are there any cons for reverting that change?
19:33:14 <adam_g> okay. i have a patch that should address some of the other structual issues  identified with the ironic job. hoping we can get it voting again once that merges, tho it may take some time to prove stability once again
19:33:19 <rloo_> the test is being run, just not voting?
19:33:25 <devananda> rloo_: correct
19:33:28 <rloo_> is it passing > 90% of the time now?
19:33:33 <devananda> rloo_: it votes in our check queue and runs in gate, but doesn't vote in gate
19:33:38 <Shrews> devananda: more like a wait for the cleanup to complete
19:33:44 <adam_g> devananda, not that i can tell, its just a matter of convincing others that we are stable.
19:34:12 <devananda> adam_g: that link seems to only show stats for check queues
19:34:37 <adam_g> devananda, oh, dag. ill fix that after the meeting
19:34:57 <devananda> NobodyCam: any updates on the tripleo tests?
19:35:06 <NobodyCam> over the weekend I put up https://review.openstack.org/#/c/100063 which Moves check-tripleo-ironic-undercloud-precise from the experimental queue to the check queue for TripleO and Ironic.
19:35:10 <NobodyCam> :-p
19:35:36 <NobodyCam> it will be non voting until it proves it self as working
19:35:39 <NobodyCam> but it should
19:36:03 <devananda> sounds good
19:36:09 <devananda> thanks!
19:36:34 <NobodyCam> :)
19:36:37 <devananda> NobodyCam: i think that should remove check-tripleo-ironic-seed now, no?
19:37:01 <NobodyCam> its still a valid check no?
19:37:13 <devananda> NobodyCam: it's a subset of the undercloud test
19:37:31 <lucasagomes> booting the undercloud already include creating the seed vm and all?
19:37:45 <lucasagomes> if so yeah, sounds good to remove it
19:37:51 <devananda> NobodyCam: I'm not totally sure. please check with lifeless after the meeting on that, but I think that's what he wants
19:37:54 <NobodyCam> if undercloud proves it self as working then I will remove seed when switching it to voting
19:38:21 <devananda> #topic bugs
19:38:25 <NobodyCam> and check with lifeless too
19:38:25 <devananda> dtantsur: hi!
19:38:28 <dtantsur> hi!
19:38:41 <dtantsur> first some numbers about the situation: 2 New bugs; 118 Open bugs; 41 In-progress bugs; 1 Critical bug; 19 High importance bugs
19:38:42 <devananda> dtantsur: welcome back :)
19:38:48 <dtantsur> yeah, thanks :)
19:39:04 <dtantsur> I'm going to count differences every week, so that we feel, where we're going
19:39:24 <dtantsur> note that we have 41 in-progress bugs, isn't it too much?  <-- open question
19:39:48 <devananda> dtantsur: thanks! that should be very helpful
19:40:03 <devananda> how many of those 41 are actually in progress?
19:40:28 <dtantsur> that's the question :) I've cleaned bugs that are _obviously_ not in progress
19:40:47 <dtantsur> the remaining are up to their owners, I guess
19:41:01 <rloo_> in-progress == there's a review associated with it?
19:41:10 <dtantsur> rloo_, we're aiming for this
19:41:22 <devananda> hmm
19:41:25 <dtantsur> but actually people need some time before review. and this time can grow and grow...
19:41:40 <dtantsur> we're committed to no more than 7 days before patch arrives
19:41:47 <dtantsur> (that was 2 weeks ago)
19:41:59 <dtantsur> I'll be keeping eye on this
19:42:06 <mrda> sounds like we need some more reviewing done :)
19:42:20 <devananda> one thing I'd like to call out is setting proper importance on bugs
19:42:22 <devananda> #link https://wiki.openstack.org/wiki/Bugs#Importance
19:42:24 <rloo_> dtantsur: sorry, i think you mean that in progress == assigned?
19:42:40 <dtantsur> rloo_, to me it's the same
19:42:51 <devananda> I think folks (myself sometimes as well) are using priority more as an "how soon do we want this fixed"
19:42:54 <rloo_> dtantsur: I don't view it as the same, that's why I wanted clarification ;)
19:43:03 <devananda> rather than what it means to the rest of openstack -- importance or impact of this bug
19:43:28 <devananda> there are some low-impact bugs that we may want fixed soon
19:43:38 <devananda> the way to communicate that via LP is to set the priority "low" but target it to the next milestone
19:43:39 <dtantsur> for me "in progress" means we have a responsible one who is moving it. assigned means someone is doing something on it, not just sitting :)
19:43:41 <lifeless> devananda: NobodyCam: the seed job
19:44:00 <lifeless> yes, for now we can replace it with the undercloud one, which means that ironic in the *seed* is being exercised.
19:44:27 <lifeless> we actually want two other things - a) a seed-uc-overcloud job and b) to run tempest against each of the stages
19:44:27 <NobodyCam> lifeless: awesome ty...
19:44:39 <devananda> dtantsur: let's see if we can get the # of inprogress bugs to actually reflect # of code reviews fixing bugs
19:45:02 <dtantsur> devananda, ok :)
19:45:11 <dtantsur> I have 2 more things on agenda...
19:45:12 <devananda> fwiw, i think it's quite possible we have that many open reviews proposign bug fixes
19:45:37 <dtantsur> maybe, I didn't research this numbers quite well
19:46:13 <dtantsur> next, launchpad can show some guidelines to one reporting bug. I'm trying to write such guidelines and I need your help:  https://etherpad.openstack.org/p/ironic-launchpad-bug-guidelines
19:46:24 <dtantsur> at least I suggest asking people for logs
19:46:28 <dtantsur> they're often missing
19:46:37 <dtantsur> anyone can help me?
19:46:47 <romcheg> dtantsur: I will take a look
19:46:54 <dtantsur> thnx!
19:47:13 <dtantsur> And one more open question re bugs and specs: what to do with bugs like https://bugs.launchpad.net/ironic/+bug/1327187
19:47:17 <uvirtbot> Launchpad bug 1327187 in ironic "node extra attribute as capability in host states  " [Undecided,Invalid]
19:47:23 <dtantsur> that's actually a non-written spec
19:47:29 <jroll> dtantsur: IMO you did the right thing there
19:47:50 <devananda> dtantsur: reject the bug, have them file a spec, exactly as you did
19:47:51 <dtantsur> (I've put it into letter to devananda last week, but afaik the letter didn't get to you, right? :( )
19:47:52 <lucasagomes> +1 agreed, cause that bug fix might touch many parts of the code
19:48:06 <devananda> fwiw, I saw a patch recently that had both closes-bug and implements-bp in the commit message
19:48:07 <lucasagomes> driver code, api for better filtering the capabilities
19:48:20 <lucasagomes> so I'd say it should go to the spec process
19:48:25 <devananda> dtantsur: hm, i dont think i saw it :(
19:48:48 <dtantsur> devananda, well, nothing interesting except for this question :)
19:49:11 <dtantsur> I also suggest triagers to clearly evaluate, whether some bug requires writing a spec
19:49:39 <dtantsur> A bit complex case is bugs that are actually requests for specs :)
19:49:55 <dtantsur> like https://bugs.launchpad.net/ironic/+bug/1328939
19:49:56 <uvirtbot> Launchpad bug 1328939 in ironic "Setting instance default_ephemeral_device should be more intelligent" [Low,Triaged]
19:50:00 <jroll> (10 minute bell)
19:50:13 <NobodyCam> ty jroll :)
19:50:20 <dtantsur> oh, I take too much time :( Well, this is generally all for me
19:50:24 <devananda> dtantsur: thanks. a good point to bug triagers
19:50:30 <devananda> moving on
19:50:39 <devananda> #topic IPA subteam
19:50:46 <jroll> hi!
19:50:47 <devananda> jroll, russell_h: hi!
19:50:52 <jroll> I'll be quick
19:51:00 <jroll> I've updated our todo list: https://etherpad.openstack.org/p/ipa-todos
19:51:09 <jroll> there's some new specs linked in there
19:51:21 <jroll> I've outlined the current patch chain
19:51:28 <jroll> and what needs to be split from 84795
19:51:53 <devananda> jroll: fantastic
19:51:56 <jroll> please please please review our specs and the patches we're depending on :)
19:52:00 <dtantsur> jroll, how are chances to get everything done in Juno?
19:52:12 <jroll> dtantsur: that's my primary goal in life right now :)
19:52:25 <jroll> (that might be exaggerating a bit :P )
19:52:25 <dtantsur> jroll, great! Anything I can help in addition to reviews?
19:52:36 <jroll> nope! reviews are where we're most limited
19:52:37 <devananda> my rough thoughts as of right now ...
19:52:39 <dtantsur> we're very interested in auto-discovery here
19:52:48 <lucasagomes> for the instance_info stuff we are still blocked by a tempest test
19:52:54 <JayF> dtantsur: as are we :)
19:52:56 <lucasagomes> I see the patch as part of the IPA chain there
19:53:01 <devananda> getting more eyes on your specs will be really helpful over the next month
19:53:08 <devananda> and landing the common dependencies
19:53:10 <devananda> like instance info
19:53:11 <lucasagomes> #link https://review.openstack.org/#/c/95789/
19:53:18 <jroll> dtantsur: we should talk more about what that actually means, but our main goal right now is for feature parity with the pxe driver
19:53:59 <jroll> dtantsur: after that lands, we'll be working on better features :) (like the things listed to be split from 84795)
19:53:59 <devananda> I think we should be able to easily achieve that refactoring and landing common code before the midcycle
19:54:08 <jroll> ^
19:54:10 <jroll> I fully agree
19:54:12 <dtantsur> jroll, yeah, that's obvious :) I wonder if we can get a bit further during Juno and get nodes discovery
19:54:27 <devananda> human time would be the only reason we dont land the spec by then. from what i've sen it's coming along really well (need to review this week again)
19:54:54 <devananda> jroll: do you think you guys will have enough done by the midcycle that others can insatll and test locally?
19:55:04 <devananda> eg, some docs, devstack integration, etc
19:55:06 <jroll> dtantsur: I hope that everything in that etherpad will be done for juno, plus decom
19:55:20 <jroll> devananda: I hope so
19:55:37 <devananda> jroll: ok. i think that will have an impact on what we can do for IPA at the midcycle
19:55:37 <jroll> devananda: dwalleck_ has been working on devstack/tempest integration lately, that's a large part of that work
19:55:45 <jroll> devananda: and then docs
19:55:53 <jroll> I expect it to be doable
19:56:03 <devananda> if others can play with it easily (myself included) by then, I think we'll have a good shot at landing in j3
19:56:11 <dtantsur> jroll, btw are you planning getting rid of docker?
19:56:17 <jroll> devananda: indeed
19:56:22 <dtantsur> (was discussed some time ago, but I forgot)
19:56:25 <jroll> dtantsur: we no longer run in docker, it's only a build step
19:56:34 <lucasagomes> as soon as it lands, are we moving to have IPA as the default deploy mechanism?
19:56:37 <devananda> jroll: ^^ glad to hear that :)
19:56:40 <jroll> dtantsur: patches for alternate build systems (DIB) welcome :)
19:56:55 <JayF> dtantsur: I think that at Rackspace, our prefered way of running it will continue to be inside a container under CoreOS. I am willing to help and review patches to add DIB elements
19:56:55 <devananda> lucasagomes: let's talka bout that post-meeting
19:57:00 <lucasagomes> devananda, ack
19:57:07 <jroll> lucasagomes: my hope is juno or K1 :)
19:57:09 <devananda> 3 minute bell, moving on
19:57:11 <dtantsur> jroll, JayF, I see :)
19:57:11 <romcheg> 3 minutes
19:57:15 <devananda> #topic Oslo
19:57:17 <lucasagomes> jroll, sure
19:57:18 <devananda> GheRivero: any news here?
19:57:20 <GheRivero> oslo.db finally has a alpha release! The oslo.db review has been updated but is not functional yet (openstack pypi mirror not updated yet)
19:57:23 <GheRivero> But what it's really important to us are the alembic/opportunistic migrations tests.
19:57:26 <GheRivero> There are a couple of reviews (99965 and 93424) to address that.
19:57:29 <GheRivero> Once they land, there will be another oslo.db release (I hope), so we can fully start testing ironic within oslo.db (but there is no rush in that now)
19:57:32 <GheRivero> That's all
19:57:34 <GheRivero> :)
19:57:35 <devananda> thanks!!
19:57:46 <devananda> #topic nova db migration
19:57:52 <devananda> romcheg: hi!
19:57:54 <romcheg> Hi!
19:58:29 <devananda> romcheg: any news?
19:58:30 <romcheg> So I have installed nova and Ironic on my lab and tested what changed in both DBs
19:58:54 <romcheg> I managed to update the migration scripts and will post that tomorrow
19:59:06 <romcheg> So I have two questions and will try to be quick
19:59:20 <romcheg> Does this feature require a spec?
19:59:43 <devananda> it's covered by my spec in nova
19:59:44 <devananda> https://review.openstack.org/#/c/95025/
19:59:49 <romcheg> Ah, great
20:00:06 <romcheg> Is there anyone who wants to volunteer on making the research of how to test these migrations?
20:00:20 <romcheg> That is another big topic
20:00:29 <devananda> yea, will require grenade tests
20:00:35 <NobodyCam> beep (time bell)
20:00:49 <devananda> sdague had volunteered at the summit to give some guidance, burt we'll need someone to work on that
20:01:06 <devananda> #info need a volunteer to help with grenade testing of nova-bm -> ironic db migration code
20:01:13 <devananda> thanks all! we're out of time
20:01:15 <jroll> perhaps we should find a nova-baremetal user for that :P
20:01:19 <NobodyCam> thank you all
20:01:20 <devananda> #endmeeting