16:00:33 <TheJulia> #startmeeting ironic_bfv
16:00:38 <openstack> Meeting started Thu Jun 29 16:00:33 2017 UTC and is due to finish in 60 minutes.  The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:39 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:42 <openstack> The meeting name has been set to 'ironic_bfv'
16:00:46 <dtantsur> o/
16:01:09 <TheJulia> Who is here for a meeting about booting baremetal nodes from remote volumes?
16:01:20 <hshiina> o/
16:01:35 <TheJulia> Awesome!
16:01:55 <mjturek> o/
16:02:10 <phuongnh> o/
16:02:11 <TheJulia> Our agenda is on the wiki, as always.
16:02:13 <TheJulia> #link https://wiki.openstack.org/wiki/Meetings/Ironic-BFV#Agenda
16:02:26 <TheJulia> #topic Announcements/Reminders
16:03:45 <TheJulia> #info Feature Freeze and client library freeze if fast approaching. We should attempt to try and get the API, and client library patches reviewed and moving as soon as possible
16:03:48 <TheJulia> #link https://releases.openstack.org/pike/schedule.html
16:04:37 <TheJulia> #info Additionally, what is not covered on our status is three process/procedural patches that will be required before nova can begin reviewing the change in their repository. We should expect those patches to take about a week.
16:04:57 <TheJulia> Anyone have any other announcements or reminders to raise?
16:06:22 * TheJulia hears crickets and guesses we can move on to the next portion of the meeting
16:06:26 <dtantsur> the patches are merging \o/
16:07:15 <mjturek> yaaay
16:07:33 <hshiina> awesome
16:08:04 <TheJulia> That is true, we had two patches merge today. Some follow-ups here and there will be needed, but that is to be expected.
16:08:07 <TheJulia> Moving on
16:08:25 <TheJulia> #topic Current Status
16:08:53 <TheJulia> I reviewed our etherpad earlier today, and updated it with the merged patches, and a few notes.
16:08:57 <TheJulia> #link https://etherpad.openstack.org/p/Ironic-BFV
16:09:16 <TheJulia> hshiina: I saw the comment regarding the api patch, was that you?
16:09:35 <hshiina> TheJulia: yes
16:09:48 <TheJulia> Okay
16:10:35 <TheJulia> So for everyone else, I -1'ed the API patch yesterday because we didn't match the proposed spec in blocking changes when the power state was on for connectors/targets.
16:11:33 <TheJulia> hshiina: I guess implementers choice on if you want it to go in with the api patch, or go into the conductor as a follow-up patch.  I would think a follow-up patch would be easier on us to review at this point, but I don't know if other people have strong feelings either way.
16:12:14 <TheJulia> dtantsur: you might have the strongest feelings regarding that fix, so if you feel strongly, let us know :)
16:12:23 <hshiina> TheJulia: i agree with follow-up patch
16:12:33 <dtantsur> what I don't want to see is us first allowing the updates, then disallowing them in the next patch
16:12:54 <dtantsur> it's not a strong feeling, as the gap between the changes is going to be quite small
16:13:03 <TheJulia> I see and agree with that being the case, but we haven't landed the client library changes either way
16:13:30 <dtantsur> people may use API directly (though I'm pretty sure here they won't)
16:13:37 <TheJulia> True
16:14:00 <dtantsur> anyway, why make the fix go first? I anyway suspect it's going to land easier than the API..
16:14:06 <TheJulia> hshiina: it could be proposed prior?
16:14:29 <TheJulia> dtantsur: very true
16:14:39 <hshiina> TheJulia: yes, i will create a patch for conductor
16:15:19 <TheJulia> Okay, and we will just go from there I guess.
16:15:50 <TheJulia> hshiina: One other thing, have you heard anything from nova on reviewing the BFV patch once it is unblocked?
16:16:35 <TheJulia> For what its worth, I did reach out to Walter regarding his current review on the nova patch, to see if he had seen your reply.
16:17:06 <hshiina> i heard nothing
16:17:26 <TheJulia> Okay, thanks!
16:17:46 <TheJulia> Anyone have anything else for current status?
16:19:34 <TheJulia> I'm taking silence as a no, and that we can move on shortly
16:20:44 <TheJulia> Moving on!
16:20:59 <TheJulia> #topic Planning/Priorities
16:21:30 <TheJulia> We largely covered our resulting priorities in reviewing our current status.
16:21:55 <mjturek> +1
16:22:17 <TheJulia> But to summarize, we need to get the API patch landed, and then the python-ironicclient patches landed. From there, it is client release process, requirement, etc.
16:22:39 <TheJulia> And then nova is unblocked to land our BFV patches in their tree
16:23:16 <TheJulia> #info Priority: Getting the API reviewed and landed, getting the client library patches reviewed and on their way to being landed.
16:23:23 <TheJulia> Anyone have anything else?
16:25:33 <TheJulia> I guess not...
16:25:46 <TheJulia> Moving on if there are no objections
16:26:03 <TheJulia> #topic Discussion
16:26:17 <TheJulia> Looks like we have one topic raised for discussion, I don't know by who off-hand
16:26:31 <tiendc> Hi, I want to discuss about the tempest patch https://review.openstack.org/#/c/472740/
16:26:55 <TheJulia> tiendc: one moment
16:27:37 <TheJulia> Looks like someone wants to propose a dummy patch for testing, but I think that already exists minus gate changes...
16:28:02 <mjturek> TheJulia: I think this topic came up last week, sure it's not a stale topic?
16:28:31 <TheJulia> mjturek: I wonder if it was added after the meeting?  I thought I cleaned it out
16:28:39 <TheJulia> Anyway, we also discussed this some last week I remember.
16:28:45 <TheJulia> I guess we are good to talk about tempest
16:28:52 <mjturek> weeeird
16:28:54 <tiendc> It got -1 Jenkins, and as what I found, that is because of the ironiclient gets installed from pypi instead of git
16:28:55 <TheJulia> Hi tiendc!
16:29:05 <TheJulia> Ahh, fun.
16:29:27 <TheJulia> I think that might be intentional :\
16:29:31 <tiendc> Does anyone have idea how to get through this?
16:29:55 <mjturek> hmmm so do we need to merge ironicclient changes before this can pass?
16:30:16 <TheJulia> mjturek: I think that is essentially what has to happen since it uses pypi
16:30:26 <TheJulia> jlvillal: you around?
16:30:27 <mjturek> daaaang
16:30:38 <jlvillal> TheJulia: I am
16:30:45 <dtantsur> tiendc: only by temporary hacking your patch
16:30:52 <jlvillal> TheJulia: What's up?
16:31:01 <dtantsur> the intent is to use global-requirements for all libraries in the services CI
16:31:08 <TheJulia> wait....
16:31:17 <TheJulia> jlvillal: nevermind
16:31:26 * jlvillal wonders if this is the Best Friends Virtually meeting?
16:31:32 <dtantsur> so yeah, to properly test it, we need the ironicclient change merged AND released
16:31:57 <TheJulia> what dtantsur said, but also tempest jobs should not require client library changes
16:32:13 <TheJulia> because tempest is intended to be the independent verification of the api contract
16:32:18 <dtantsur> yeah, this is a good point. we don't use ironicclient in tempest. we may use it in devstack though..
16:32:33 <TheJulia> Very true
16:32:58 <tiendc> yeah, got it
16:33:40 <TheJulia> tiendc: Sorry for the headache, hopefully we can get it cleared up soon
16:33:47 <hshiina> i guess nova virt driver uses ironicclient in tempest
16:34:11 <dtantsur> correct, it depends on a new ironicclient release as well
16:34:13 <tiendc> TheJulia: no problem
16:34:17 <TheJulia> jlvillal: (joking) we can rename the meeting after the feature is fully landed and all ;)
16:34:40 <jlvillal> :)
16:35:22 <hshiina> i think we can merger this tempest test after ironicclient is merged
16:35:43 <TheJulia> hshiina: merged, released, global requirements updated, global requirements patches merged.
16:36:40 <TheJulia> It is a process, it is the same as to what I was trying to convey in the announcements since that does take some time.
16:37:11 <hshiina> TheJulia: got it, thanks
16:38:07 <tiendc> TheJulia: thanks
16:38:22 <TheJulia> I guess I'll move us to Open Discussion, and if there is nothing else we can all get about 20 minutes back.
16:38:25 <hshiina> rather than this grenade job, we would need to run this new tempest test with cinder as an experimental job
16:38:45 <TheJulia> #topic Open Discussion
16:40:37 <TheJulia> hshiina: for existing grenade... I have a strange feeling that might have to wait into this next cycle, the tempest test will be the real important item as long as it can determine if the storage interface is available or not.  So possibly experimental to start out
16:40:49 <TheJulia> I just worry about memory consumption as well
16:41:36 <dtantsur> grenade? Oo
16:41:54 <dtantsur> well, it makes sense, but we have so many features not covered in grenade..
16:41:59 <TheJulia> It is not a this cycle problem if we want to test on that.
16:42:51 <TheJulia> Grenade is largely testing that our upgrade processes work.  Tempest runs on each side, it would be good to have it eventually, perhaps. Or as an optional setting that could be triggered with an experimental job
16:43:11 <TheJulia> s/work/works/
16:43:26 * dtantsur wonders if tempest has a smoke test with cinder
16:43:38 <TheJulia> dtantsur: good question...
16:43:53 <TheJulia> I think its just basic ops....
16:44:11 <TheJulia> but one of us will need to investigate that
16:44:15 <dtantsur> it depends what is "basic" for nova :)
16:44:20 <TheJulia> true :)
16:45:07 <TheJulia> tiendc, did you raise the question in hope of getting it running with a grenade job?
16:45:20 <dtantsur> no, no cinder in the basic ops
16:45:31 <tiendc> TheJulia: yes
16:45:41 <TheJulia> dtantsur: thank you for looking it up
16:46:18 <TheJulia> tiendc: Okay, I feel like we've at leased raised the questions and answers that you needed
16:46:46 <tiendc> TheJulia: yeah, thanks
16:46:56 <TheJulia> Okay, anyone have anything else?
16:47:01 <dtantsur> then there a "minimum basic" test, but it's not a smoke test https://github.com/openstack/tempest/blob/master/tempest/scenario/test_minimum_basic.py
16:47:07 * dtantsur is confused
16:47:21 <TheJulia> dtantsur: I'll look too after I get some lunch :)
16:47:29 <dtantsur> cool :)
16:48:38 <TheJulia> Seems like we are done for today! Thank you everyone. We'll see you all here next week, and in #openstack-ironic in the mean time.
16:49:03 <dtantsur> thanks!
16:49:21 <mjturek> thanks all!
16:49:27 <phuongnh> thanks
16:49:31 <hshiina> thanks
16:49:33 <tiendc> thanks
16:49:38 <TheJulia> :)
16:49:40 <TheJulia> #endmeeting