09:03:25 <masahito> #startmeeting blazar
09:03:26 <openstack> Meeting started Tue Feb 20 09:03:25 2018 UTC and is due to finish in 60 minutes.  The chair is masahito. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:03:27 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:03:29 <openstack> The meeting name has been set to 'blazar'
09:03:45 <masahito> hi, blazar folks.
09:03:51 <masahito> #topic RollCall
09:03:56 <priteau> o/
09:04:18 <masahito> today's agenda is
09:04:26 <masahito> 1. PTG schedule
09:04:28 <masahito> 2. AOB
09:04:32 <masahito> anything else?
09:04:41 <masahito> less topics today
09:04:44 <masahito> priteau: hi
09:04:47 <priteau> Release?
09:04:57 <priteau> I mean preparation for the release
09:05:15 <priteau> What's still left to submit, merge, etc.
09:06:15 <masahito> #topic PTG schedule
09:06:54 <masahito> I created rough time schedule for the upcomming PTG.
09:06:57 <masahito> https://etherpad.openstack.org/p/blazar-ptg-rocky
09:07:10 <hiro-kobayashi> Hi priteau, masahito
09:07:23 <priteau> Hello!
09:07:39 <masahito> Monday morning is for review Queens tasks and afternoon is for Cross project and new features.
09:08:20 <masahito> Tuesday morning is for testing and afternoon is for feedback from SIG and summarizing the meeting.
09:09:28 <masahito> Monday morning is warm-up and less topics. We could start from 10am.
09:09:45 <masahito> hiro-kobayashi: hi
09:11:06 <masahito> any comment?
09:11:27 <priteau> It depends how much Queens work we manage to do this week. We could try to merge as many patches as possible before the meeting.
09:11:36 <hiro-kobayashi> thank you for arranging!
09:12:39 <priteau> I am arriving late on Sunday night in Dublin so I don't mind starting a bit later ;-)
09:12:59 <priteau> But we should keep enough time to make Queens a good release
09:14:05 <masahito> priteau: yes. I imagine the session is not for Queens itself, but for checking remaining tasks we couldn't and moving its to Rocky.
09:15:30 <priteau> What about we put an official start of 10 AM, but if we are already there before, we can start early?
09:16:28 <masahito> priteau: make sense.
09:17:16 <masahito> Then let's start at 10am on Monday.
09:18:14 <masahito> #topic Q release preparation
09:19:23 <masahito> There're some patches related to Q release in review board.
09:21:56 <masahito> 1. https://review.openstack.org/#/c/542815/
09:22:02 <masahito> 2. https://review.openstack.org/#/c/545773/
09:22:02 <priteau> I have backported some Python 3 patches for the client that were merged after branching
09:22:10 <masahito> 3. https://review.openstack.org/#/c/546056/
09:22:44 <masahito> Do you know other patches related to the release?
09:23:34 <masahito> If there, it's good time to merge or comment it.
09:23:53 <priteau> https://review.openstack.org/#/c/545802/
09:23:57 <priteau> https://review.openstack.org/#/c/545803/
09:24:23 <hiro-kobayashi> https://review.openstack.org/#/c/526272/
09:24:23 <masahito> I put W+1 to #2
09:24:40 <priteau> I also added some release notes, to master first
09:25:03 <priteau> I am not sure what should be the approach for release notes that were part of Queens but were not added before branching
09:25:08 <hiro-kobayashi> https://review.openstack.org/#/c/526235/
09:26:36 <masahito> priteau: patch for multi region support?
09:27:00 <masahito> s/patch/the releaenote/
09:27:04 <priteau> masahito: yes, for example https://review.openstack.org/#/c/546056/
09:27:30 <priteau> If we merge it in master, than cherry pick to queens, will it appear in Queens release notes and not in current series?
09:28:34 <masahito> maybe.
09:29:35 <priteau> Next release we should add these release notes early :)
09:29:53 <masahito> According to reno docs, it looks like some ignore-notes section is required.
09:29:54 <masahito> https://docs.openstack.org/reno/latest/user/usage.html#updating-stable-branch-release-notes
09:30:11 <masahito> priteau: yes.
09:32:02 <priteau> masahito: I will try it in my env
09:33:38 <masahito> priteau: thanks.
09:34:36 <priteau> I will also check history to see if we're missing any other release note
09:35:23 <hiro-kobayashi> thanks!
09:37:48 <masahito> thanks
09:37:53 <masahito> #topic AOB
09:38:39 <masahito> anything else to share/discuss/etc?
09:39:06 <priteau> Nothing on my side
09:39:30 <hiro-kobayashi> Nothing from my side, neither
09:39:38 <GeraldK> hi. sorry for being late. was stuck in another meeting.
09:39:50 <hiro-kobayashi> hi GeraldK!
09:39:54 <GeraldK> looks like you are at the end of the meeting already :)
09:40:39 <masahito> GeraldK: right :) Do you have anything to share/discuss?
09:40:55 <GeraldK> just a quick question:
09:41:27 <GeraldK> it seems Openstack Blazar uses PUT method to do the lease update.
09:41:47 <masahito> yes
09:42:08 <GeraldK> it seems there is also the alternative to use PATCH update method.
09:42:37 <GeraldK> has anyone already thought/discussed about changing it and what would be the implications?
09:43:51 <GeraldK> background: we are currently trying to introduce the host reservation in ETSI NFV IFA and are discussing on some options on how to best do the update of a host reservation.
09:45:06 <masahito> I didn't thought it. IIRC, openstack standard for update API is PUT method instead of PATCH method.
09:45:40 <priteau> Some APIs are using PATCH, e.g. Ironic
09:45:45 <masahito> I'm checking if there is docs about its definition.
09:45:55 <GeraldK> with the PUT, if I understand correctly, we have to send the full updated reservation data. with PATCH it would be possible to only send the updated parts.
09:47:04 <hiro-kobayashi> With the PUT, we can send just parameters to update.
09:47:18 <hiro-kobayashi> for current Blazar implementation
09:47:45 <GeraldK> we don't have to decide/agree on anything today. I just want to understand better the current situation in Blazar so that we can agree on the best way for the IFA specifications.
09:48:07 <priteau> I proposed that we should use PATCH to manage extra capabilities, because that allows you to do a <op> <arg> request
09:48:28 <priteau> GeraldK: This should be a discussion for a v3 API
09:48:33 <GeraldK> hiro-kobayashi: okay, that's good to know.
09:48:43 <hiro-kobayashi> https://developer.openstack.org/api-ref/reservation/v1/index.html#update-lease
09:48:44 <masahito> For current situation hiro-kobayashi's comment is right.
09:50:35 <masahito> v3 migration is one of topics in the PTG.
09:50:37 <GeraldK> hiro-kobayashi: it seem with the PATCH one would be able to even update parts of e.g. the reservation.resource_properties, whereas with the PUT you need to send the full reservation.resource_properties
09:51:32 <hiro-kobayashi> GeraldK: make sense
09:51:52 <GeraldK> thanks for your feedback. I will look more deeply into this. I agree with priteau to discuss about this option for the v3 API.
09:52:47 <GeraldK> my goal is to have maximal compatibility between Blazar and the standard and not introduce new gaps.
09:53:30 <masahito> GeraldK: What's the current situation in the standards?
09:53:49 <GeraldK> I will be in ETSI NFV meeting next week and will discuss there with some colleagues. I will keep you updated for your PTG discussions.
09:54:07 <GeraldK> masahito: we are now drafting that part of the standard
09:54:32 <masahito> GeraldK: got it. thanks.
09:54:51 <GeraldK> that's all from my side.
09:56:21 <masahito> last 4 mins.
09:57:16 <masahito> I announce the next meeting is skipped as we'll gather next week.
09:58:15 <GeraldK> okay. enjoy Dublin.
09:58:16 <masahito> looks no more topics. thanks all
09:58:26 <GeraldK> thanks and bye to all.
09:58:35 <masahito> see you in Dublin.
09:58:45 <masahito> bye.
09:58:52 <masahito> #endmeeting