09:01:05 #startmeeting blazar 09:01:06 Meeting started Tue Oct 24 09:01:05 2017 UTC and is due to finish in 60 minutes. The chair is masahito. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:01:07 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:01:09 The meeting name has been set to 'blazar' 09:01:41 #topic RollCall 09:01:54 hellp 09:02:01 o/ 09:02:05 o/ 09:02:08 o/ 09:02:51 bertys priteau hrito: hello 09:03:40 today's agenda is 09:03:44 1. Team Mascot 09:03:50 2. On Boarding 09:03:52 3. AOB 09:03:57 anything else? 09:04:42 #topic Team Mascot 09:04:54 It's just a fyi 09:05:19 I sent a request of Team Mascot to the Foundation. 09:05:38 thanks! 09:05:56 They said they're busy now because of preparation for upcoming Summit. 09:06:23 They will begin it after mid-Novenver. 09:06:40 that's from my side. 09:07:16 Thanks for requesting our mascot 09:07:17 anything else? 09:08:18 hiro-kobayashi: priteau: np 09:08:39 let's move on next 09:08:46 #topic OnBoarding 09:09:01 This topic is also fyi. 09:09:31 I'm still writing a slide deck for the session. 09:10:07 I'll share the deck once I have finished to write it. 09:10:19 I hope I'll share it before next meeting. 09:10:46 any comment? 09:11:05 masahito: nice, thanks 09:11:46 okay, jumping on next 09:11:52 bertys: np 09:11:58 #topic AOB 09:12:13 we have less topics today. 09:12:23 Does someone have something? 09:12:32 I have some updates 09:12:45 masahito: quick question, what is the spec freeze deadline for Blazar? 09:13:27 bertys: we're small team. so there is no restrict deadline. 09:13:44 bertys: maybe Q-3. 09:13:56 hiro-kobayashi: you have a floor. please go a head. 09:13:59 masahito: okay. thanks 09:14:00 Spec freeze is a project-specific event 09:14:13 I agree with masahito that we can wait longer 09:14:20 Nova has already frozen theirs 09:14:24 +1 09:14:30 Queens-3 is Jan 22 - Jan 26 09:14:43 Who plan to propose specs? 09:15:26 I'll do. but I've not written it well yet.... 09:16:18 I would like to discuss another topic 09:16:32 I've proposed 2 specs. And if nobody takes the multi-freepools blueprint, I'll take it. Then, I will propose 1 more spec. 09:17:10 priteau: okay, before that I'd like to hiro-kobayashi to share his update. 09:17:19 priteau: Is it okay? 09:17:20 But I'm not sure I can complete the multi-freepools in Queens... 09:17:21 sure 09:17:29 I wasn't sure if we were finished or not 09:17:47 thanks 09:17:54 My update is: 09:18:01 1. Updated Blazar wiki 09:18:15 2. Specs in review 09:18:44 about 1, I've updated the wiki https://wiki.openstack.org/wiki/Blazar 09:19:06 it's based on the README that was merged today: https://review.openstack.org/#/c/505110/ 09:19:54 Now there's a few information at wiki. I expect people to jump to the reathedocs. 09:20:15 s/reathedocs/readthedocs/ 09:20:34 Of course openstack.docs after we publish docs on it 09:21:24 hiro-kobayashi: thanks for updating it. 09:21:31 I expect blazar docs can be published after the patch for openstack-infra is merged: https://review.openstack.org/#/c/511176/ 09:22:03 that's all for 1 09:22:42 about 2, I've updated specs, so please review them. 09:23:45 And I'm working on implementation of the state-machine blueprint. So, I'll push code in this week. Please also check it. 09:23:50 Thanks hiro-kobayashi! 09:24:23 It helps understanding what the state-machine blueprint does. 09:24:34 That's all for my updates. 09:25:24 That's great works!! workstate-machine BP helps all to understand Blazar well. 09:26:01 hiro-kobayashi: you're working on lots of tasks now. If you need help, feel free to ask this team :-) 09:26:34 masahito: thanks :-) 09:26:51 I appreciate it if anyone can take multi-freepools blueprint 09:28:43 any commets for hiro's update? if nothing, let's move on priteau's topic. 09:28:48 I'll take it if no one take it after I complete state-machine, resource-monitoring, host-dashboard 09:29:39 Thanks hiro-kobayashi, you're doing a lot of work! 09:29:39 hiro-kobayashi: great! 09:30:36 my pleasure ;-) 09:31:30 The topic I wanted to discuss is the keystoneauth1 change in python-blazarclient 09:32:31 It broke the client API without warning. I would like to discuss what we should do about it. 09:32:56 priteau: right. I also noticed the problem. 09:33:25 On Chameleon we have some CI code using the client. We had to change it to use blazarclient 0.3.0 09:33:26 I noticed it break blazar-dashboard too. https://bugs.launchpad.net/blazar/+bug/1724757 09:33:27 Launchpad bug 1724757 in Blazar "blazar dashboard fails to use blazarclient" [High,Triaged] 09:34:37 I'm thinking if no one can fix it in one or two weeks. we can revert the patch once. 09:35:03 Oh, sorry. I should have care it. 09:35:38 Is there a way to provide compatibility with the old interface, or should we have a new client namespace 'blazarclient.v2.client.Client'? 09:36:14 I think we can provide backword compatibility. 09:37:13 priteau: just a question. you're using v2 client? 09:37:45 I don't think so. Do we have a v2 client? 09:38:47 hrito: What's your plan? 09:38:49 priteau: no. I imagined you have private v2 client which is related to v1. 09:39:06 no, we use python-blazarclient installed directly from Git 09:39:31 priteau: okay :-) 09:39:42 but I wasn't talking about REST API v2. Instead, asking if we need to create a v2 of the client (still using the REST API v1, but with keystoneauth1) 09:40:16 if hrito can provide backward compatibility, this is not neede 09:40:20 *needed* 09:41:12 hiro-kobayashi: Enabling to pass token to blazarclient and if token is passed to blazarclient, then create keytone token session. 09:42:45 Sounds good. Can you push patch for it? 09:42:54 hrito: From my understand, the client break happens because of the arguments change at blazarclient.v1.client.Client. 09:44:09 I understood hrito's plan is to revert those args. Is it correct? 09:44:12 masahito: that's right. So we need to enable to pass token again in blazarclient.v1.client.Client and change it as a keystone session 09:44:59 We may be able to support both, I think that's what hrito wants to do 09:45:09 Add back kwargs blazar_url and auth_token 09:45:13 The class receives blazar_url as a first argument and auth_token as a second argument. 09:45:19 If they are provided, use the old auth code 09:45:24 priteau: that's my plan 09:45:32 +1 09:45:34 masahito: they are kwargs 09:45:36 hrito: okay 09:46:00 I'll try to fix this bag 09:46:42 The client class supports same position arguments, blazar_url, auth_token, and kwards which includes session and other args, right? 09:47:12 masahito: yes 09:47:46 Anyway, thanks to hrito! I'll wait your patch :-) 09:48:54 masahito: okay 09:50:05 Thanks hrito 09:50:28 any comments or something to share? 09:51:32 just fyi. 09:51:47 next PTG will be held at Dublin. 09:51:53 https://www.openstack.org/ptg/ 09:52:12 It's Euro area. 09:52:43 I wanted to mention it too. I am hoping I will attend. 09:53:14 I hope Blazar team can physically gather there ;-) 09:53:50 Will you be able to attend? It's a long trip for you? 09:54:19 I plan to go there. 09:55:54 last 5 mins. 09:58:04 I plan to go, too 09:58:51 okay, I'll be around this channel by its endtime. If someone has something, feel free to mention it. 09:59:11 s/its endtime/ meeting end/ 09:59:34 Thanks everyone 09:59:42 Good bye 09:59:55 Thanks all! bye! 09:59:59 bye 10:00:05 thanks all. see you in next meeting. 10:00:31 #endmeeting