15:00:11 #startmeeting horizon 15:00:11 Meeting started Wed Oct 6 15:00:11 2021 UTC and is due to finish in 60 minutes. The chair is vishalmanchanda. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:11 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:11 The meeting name has been set to 'horizon' 15:00:15 o/ 15:00:21 o/ 15:00:27 hello everyone. 15:01:01 Let's wait for few minutes if more people join us. 15:02:04 o/ 15:02:18 e0ne: around? 15:02:24 hi 15:02:39 Let's start the meeting. 15:02:46 #topic Notices 15:03:02 Openstack Xena was officially released today. 15:03:12 http://lists.openstack.org/pipermail/openstack-discuss/2021-October/025214.html 15:03:50 Thanks to everyone for contributing in Xena release and now the Yoga release is open for development. 15:04:27 Also Please start adding your topic which you want to discuss during the PTG. 15:04:29 vishalmanchanda: thanks for wearing a PTL hat this release! 15:04:43 +1 15:04:48 e0ne: 🙂 15:05:12 etherpad link https://etherpad.opendev.org/p/horizon-yoga-ptg 15:06:10 I can see we have only a few topics for discussion as of now. 15:06:37 I will also add if I got any topic to discuss and assign time slots for each topic by Friday. 15:07:12 Feel free to reach me in case if you like to attend some topics discussion but the assigned time slot doesn't work for you. 15:07:48 Also looks like we have to cancel Monday, October 18 time slot for ptg because I have to attend TC+PTL interaction session at the same time. 15:08:42 Anyway I will share the same info. on open-discuss ML about the PTG time slots. 15:09:06 that's all announcement from my side for this week. 15:09:19 vishalmanchanda: I think it is better to add it to the schedule to join TC/PTL interaction :) 15:09:25 and as a friendly reminder let's register the PTG if you join https://www.openstack.org/ptg/ 15:09:56 amotoki: +1 15:09:57 amotoki: +1. 15:10:49 moving to next topic. 15:10:58 #topic Stable-branches 15:11:49 Please take a look at the patch which fixes the stable branches gate, actually it just drops lower-constraints job from stable branches. 15:11:57 https://review.opendev.org/q/topic:%2522drop-l-c%2522+status:open+project:openstack/horizon 15:12:26 Also, amotoki raise a concern that when we should drop lower-constraints job from stable branches. 15:12:27 vishalmanchanda: sure. rdopiera already reviewed them, thanks 15:13:05 When a stable branch is created or when a lower constraint job starts failing for that branch and we should document this point. 15:13:34 I am fine with both option but I would like to hear from the team. 15:14:36 IMHO I tend to prefer to dropping it when it starts to fail as it sometimes raises a good point to us. 15:14:56 (and if a fix is trivial it is worth fixing it) 15:15:27 either is reasonable in my opinion 15:15:41 keeping it if it works seems like a win 15:15:54 this is just my preference. I am okay with either too. 15:17:13 ok I will abandon my patch from stable/xena branch as lower-constraints job is not failing there. 15:18:02 Although I did not follow all discussions, there is a discussion on pinning base python dependencies like setuptools/pip in stable branches. It might help us. 15:18:52 that's all topic from my side for this week. 15:19:05 #topic open-discussion 15:19:58 does anyone have bugs to be raised here? 15:20:16 I would like to see more feedback on rdopiera patch https://review.opendev.org/c/openstack/horizon/+/810224 15:20:43 I already mention one of my concern there. 15:21:37 vishalmanchanda: what is your concern? 15:22:16 does it change the current behavior with the default policy rules? 15:22:21 I admit it's a very rare case they ran into. 15:22:26 amotoki: yes. 15:22:40 amotoki: it also hides 'port security' checkbox even with the current default policy. 15:22:59 vishalmanchanda: okay. I will check it this week 15:23:06 amotoki: thanks. 15:23:30 Does anyone have any other topic to discuss? 15:23:30 rdopiera: I don't think so. customizing policy rules is not rare 15:23:42 vishalmanchanda: does it? it should only hide it if you don't have the rights to change it 15:24:11 I would like to talk briefly about the system scope tokens 15:24:46 I looked briefly into what we will need, and I created bugs in keystoneclient and keystoneauth to add the APIs we will need 15:24:56 rdopiera: yes, I tried a couple of cases, I will add my comments in patch review. 15:25:09 https://bugs.launchpad.net/python-keystoneclient/+bug/1945649 15:25:15 https://bugs.launchpad.net/keystoneauth/+bug/1945657 15:25:42 I also made a patch for keystoneclient for this, but I will need help making the patch for keystoneauth 15:26:25 the patch is failing CI right now for unrelated reasons -- seems their gate is broken 15:26:57 vishalmanchanda: thanks, if it does, it's a bug in my patch 15:27:49 rdopiera patch link https://review.opendev.org/c/openstack/python-keystoneclient/+/811966 15:28:38 I would like to say sorry in this area in xena (as I felt a burnout a bit....) 15:28:49 rdopiera: is it documented in the keystone API reference? 15:29:05 amotoki: yes, the bugs have links to api docs 15:29:18 well, ok, one of them has a link 15:29:24 I need to add the link to the other 15:29:43 good point, thanks 15:29:43 ah... I just opened the first one :-( 15:30:10 it seems we will need an addiitonal parameter for Token creation 15:30:17 to make it system-scoped 15:31:07 from unscoped token to system-scoped? 15:31:12 yeah 15:31:33 that's the keystoneauth bug basically 15:31:44 I will add details 15:31:54 ah, I just tried keystone API directly... 15:32:43 we should probably be using the library, ultimately 15:32:59 yeah, agree 15:33:09 we can of course continue developent with direct calls while this is not merged 15:33:16 development* 15:33:48 amotoki: I've been pretty much psychologically blocked with this myself too. 15:34:32 amotoki: It's difficult. No worries. 15:35:03 rdopiera: hehe. I will be involved in the effort in Yoga too :-) 15:35:28 we do want to try to get this working until our downstream freeze on Nov 25th 15:35:29 We can check with keystone team if we can have priority for this patch. 15:35:41 but obviously it's a bit of wishful thinking at this point 15:36:09 rdopiera: if you work on it publicly, i will try to help it 15:36:26 amotoki: thanks a lot! I really appreciate it! 15:37:26 vishalmanchanda: I need to go to them and ask for help implementing it anyways. 15:37:54 rdopiera: +1. 15:38:14 perhaps we can call the keystone API directly first (if the libs do not work as expected) 15:38:25 anyway feedback to keystone would be useful 15:38:39 amotoki: sounds good 15:39:00 amotoki: do you have any code already, or was it all lost in that crash? 15:39:35 rdopiera: I did not try it after that.... but I can investigate it again 15:39:55 amotoki: I will try to put up a WIP patch this week 15:40:09 and we can discuss it on there then 15:40:15 rdopiera: great. i will follow it up 15:40:33 awesome. 15:41:27 Does anyone have any other topic to discuss? 15:41:50 none from me 15:43:06 nothing 15:43:17 ok Let's end this meeting. 15:43:28 Thanks everyone for joining! 15:43:50 #endmeeting