08:59:59 #startmeeting blazar 09:00:00 Meeting started Tue Nov 19 08:59:59 2019 UTC and is due to finish in 60 minutes. The chair is priteau. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:00:01 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:00:03 The meeting name has been set to 'blazar' 09:00:06 #topic Roll call 09:01:24 o/ 09:02:19 o/ 09:02:54 Hi bertys & tetsuro 09:03:52 Let's wait a few minutes in case Masahito joins 09:06:26 Agenda for today: 09:06:27 * Ussuri planning and priorities 09:06:30 * Stable backports 09:06:33 * AOB 09:06:43 #topic Ussuri planning and priorities 09:07:49 We've set priorities in the Etherpad document: 09:07:54 Network segment reservation 09:07:54 Calendar view 09:07:54 Usage enforcement 09:07:56 Quota 09:07:58 Preemtible 09:08:27 It would be good to set some timelines against it so they don't get delayed 09:08:56 For network segment reservation, we have reviews requesting changes on the spec. 09:09:14 I'll talk with Jake this week and see if he can get an updated spec out for review 09:09:55 Goal would be updated spec by end of November, hopefully approved by Christmas? 09:10:18 Then in the beginning of the year, or earlier if ready, we can focus on merging the code 09:11:50 I am adding some tentative schedule on the Etherpad 09:11:53 #link https://etherpad.openstack.org/p/blazar-ptg-ussuri 09:12:13 Do we want a spec for calendar view, or code would be enough? 09:13:19 For usage enforcement, quota, preemptible, we definitely want a spec. 09:13:25 Right 09:14:19 Do we have any controversial point on calendar view? 09:15:39 As I don't come up with the points, I'm okay that it is a specless blueprint 09:15:52 It's currently being used only with host reservation. It's not clear how it could be used to easily visualise instance reservation, since there are 3 resources to consider (vCPUs, memory, disk) 09:16:25 We could have an initial version that displays only host reservations. 09:17:47 Ah, I see. I haven't read any specs that are for GUI, but looks like it is better to have the spec if we have several implementation alternatives? 09:18:18 We could have a short spec that explains the limitations 09:19:28 I will talk with Jake about it 09:19:55 Does the schedule look appropriate to you? 09:21:27 Since spec can be a good guide line for reviewers of the function, it is worth doing. 09:21:52 +1 09:23:15 I've set assignes to each task, though I don't have any for the quota spec. Any volunteer? 09:26:20 ...I'm not sure I can complete it in U release, but I can start it if no one is gonna take it. 09:26:52 That's fine tetsuro, no pressure on this one. 09:27:27 Okay. Thanks. 09:28:53 Let's move to the next topic then 09:28:58 #topic Stable backports 09:29:56 I heard through Stig who was at the summit that staff from Nectar cloud (Australia) is evaluating Blazar and planning to run it in production soon 09:30:10 I got in touch with them, they're going to run Rocky 09:30:32 So we should consider actively maintaining this release with backports of bug fixes. 09:30:40 priteau: Heard this too in Shanghai. Good news. 09:30:53 great 09:31:12 I believe Chameleon is running Stein (and soon Train) so they wouldn't care much about the state of stable/rocky. 09:31:46 So if you submit or review a bug fix, please check if it is relevant for train, stein *and* rocky. 09:32:17 I know it's extra work but another new deployment would be a good boost for the project 09:34:31 I will take a pass at reviewing the state of these branches. 09:34:44 #action priteau review stable branches for bug fixes 09:35:53 Hopefully the gate still works... 09:37:34 #topic AOB 09:37:37 Anything else to share today? 09:39:10 Not from me 09:40:37 Nothing from me either 09:41:02 Thanks for joining today 09:41:06 Have a good week! 09:41:14 #endmeeting