15:00:40 #startmeeting horizon 15:00:41 Meeting started Wed Jul 22 15:00:40 2020 UTC and is due to finish in 60 minutes. The chair is vishalmanchanda. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:42 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:44 The meeting name has been set to 'horizon' 15:01:12 hi anyone around for horizon meeting. 15:01:51 o/ 15:02:05 rdopiera: hi. 15:02:47 e0ne: Ivan is going to join a little late. 15:03:16 let's wait for a few minutes if anyone joins. 15:03:45 hi 15:03:49 hi 15:03:59 #chair e0ne 15:04:00 Current chairs: e0ne vishalmanchanda 15:05:21 let's start 15:05:35 #topic Notices 15:05:52 we'll reach Victoria-2 milestone next week 15:06:07 #link https://releases.openstack.org/victoria/schedule.html 15:06:14 nothing special for horizon 15:06:25 just FYI 15:06:45 have we dropped deprecated features we need to drop in victoria? 15:07:01 amotoki: I didn't check it yet 15:07:12 but it should be safe until FF 15:07:12 I haven't checked it yet, but perhaps it is better to drop them if any before milestone-2 15:07:26 sure, earlier is better 15:08:45 #action e0ne to check for deprecated features to drop in Victoria 15:09:24 Community goals for V release: https://governance.openstack.org/tc/goals/selected/victoria/index.html 15:09:49 nothing special for Horizon team to implement 15:10:57 is there no things we need to work explicitly? 15:11:37 need to check our zuul config once more 15:12:02 I plan to check the alternate scss libraries, but that's more research than coding 15:12:03 yeah. let's check :) 15:12:34 amotoki: I checked it few weeks ago and didn't find any legacy jobs there 15:12:36 iirc there are no legacy zuul jobs for horizon, you should be fine 15:12:47 tosky: thanks! 15:12:52 I think we can test horizon on focal like others https://review.opendev.org/#/q/topic:migrate-to-focal 15:13:03 tosky: I appreciate your help with CI 15:13:06 e0ne: yes, I cleanedup all legacy jobs in the early stage of zuulv3 migration. 15:13:45 oops, we've got some bionic-based jobs https://github.com/openstack/horizon/search?q=bionic&unscoped_q=bionic 15:13:53 I'll propose a patch after the meeting 15:14:33 e0ne: +1. 15:16:21 are those nodeset different from the default ones? 15:16:25 nodesets* 15:16:34 well, just one, ubuntu-bionic 15:16:43 nodejs10 is available in focal, so it would be simpler. 15:17:01 tosky: they should not be. need to check if there are no issues with nodejs 15:17:07 the nodeset definition was introduced when we used trusty 15:17:30 it may be defined only for explicitly. I am not sure. 15:17:37 TBH, I didn't use nodejs from ubuntu packages 15:18:37 tosky: do you know where is the default node defined? 15:19:53 amotoki: iirc in openstack-zuul-jobs 15:20:12 https://opendev.org/openstack/openstack-zuul-jobs/src/branch/master/zuul.d/nodesets.yaml 15:20:55 nodejs jobs inherit jobs definitions from zuul-jobs (not from openstack-zuul-jobs) 15:20:56 https://opendev.org/openstack/horizon/src/branch/master/.zuul.d/nodejs-jobs.yaml#L3 15:21:01 but I guess it depends on what nodejs-run-lint use 15:21:31 so perhaps explicit nodeset might make sense 15:21:52 yeah, I guess so 15:22:04 it's a less common case than the usual devstack jobs 15:22:15 where the default can just be inherited 15:25:59 as far as I checked, other jobs inherit from openstack-zuul-jobs. nodejs jobs look the only exceptions. 15:26:14 *look like* 15:28:54 let's move on the next topic 15:29:20 next OpenStack Summit will be virtual again 15:29:23 #link https://www.openstack.org/summit 15:29:46 I'm sorry, Open Infrastructure Summit 15:29:48 Oct 19-23, 2020 15:30:18 it will be less complicated than virtual PTG 15:30:45 and it sounds absolutely reasonable with the current situation 15:31:16 and one more announcement from me 15:31:36 I created a poll for Victoria virtual mid-cycle poll as we discussed at the PTG 15:31:41 #link http://lists.openstack.org/pipermail/openstack-discuss/2020-July/016040.html 15:32:11 I don't expect a lot of discussion, so 1 hour should be enouph 15:32:27 otherwise, we can schedule one more session 15:32:50 e0ne: cool. already voted! 15:33:00 candidate dates are July 27-31, but you said please response by Aug 4.... 15:33:10 I tried to select timezone to be suitable both for US and Asia 15:33:58 amotoki: my bad. I meant Aug 5-7 15:34:04 will send an update 15:34:17 e0ne: thanks 15:34:31 it's a crazy day for me :( 15:34:45 the week of Aug 5 is a summer vacation week in my company :( 15:35:14 oh... we can schedule a week later I guess 15:35:49 I'm sorry, I have to run away in 10 minutes:( 15:36:05 If it is better to have it that week, I can coordinate my schedule but it is my schedule at the moment. 15:36:38 due to the current situation I have no specific plan during that off week 15:37:05 amotoki: please, drop a note to the thread what dates are good for you 15:37:30 e0ne: any week is okay except that. 15:38:03 e0ne: I just wonder scheduling it later leads to less days till FF 15:38:28 that's the issue 15:39:09 the week of Sep 07 - Sep 11 is Feature Freeze 15:40:28 vishalmanchanda: rdopiera: what about you? 15:41:03 I am okay to move my off week if the week of Aug 5 is fine to you all. 15:41:17 amotoki: I can be available on any day except 3rd August. 15:41:40 * e0ne is afk now 15:43:16 rdopiera: what about you? 15:43:50 I mean Aug 5 week works for you? 15:44:16 vishalmanchanda: it does 15:44:42 rdopiera: cool. 15:46:05 then we can plan mid-cycle in August 5 week except Monday. 15:47:04 e0ne: around? 15:47:39 okay, I will coordinate my schedule internally next week. (next two days are holidays in Japan) 15:48:46 amotoki: cool. 15:49:04 e0ne: I think Ivan is not around. 15:50:48 Guys please vote to Ivan's poll for mid-cycle timing link http://lists.openstack.org/pipermail/openstack-discuss/2020-July/016040.html 15:51:03 moving on to next topic. 15:51:17 #open-discussion 15:51:30 vishalmanchanda: you need '#topic' :p 15:51:38 oops. 15:52:00 #topic open-discussion 15:52:50 amotoki: I have fixed your comments on this patch https://review.opendev.org/#/c/708069/. 15:53:12 Please take a look once you have time. 15:53:20 rdopiera: you too. 15:54:27 do we have a consensus on whether we should add 'details' to all exceptions.handle() or automatically add details to error messages? 15:56:14 as long as in the horizon repo it looks better to add details automatically because you changes all occurrences of exceptions.handle() 15:57:12 adding details to all exceptions.handle looks fine to me. 15:57:16 I am not sure about horizon plugins, but if we plan to make similar changes for all plugins an explicit 'details' argument would not make sense. 15:57:39 rdopiera: Do you want to add something here? 15:59:04 amotoki: what is the issue in adding a additional parameter i.e 'details'. 15:59:46 nothing 16:00:00 vishalmanchanda: if all callers specify details=str(ex) it is just redundant. 16:00:16 just the fact that details don't always have to come from the exception 16:00:33 ok we can continue on #horizon channel. 16:00:38 #endmeeting