15:00:25 #startmeeting horizon 15:00:25 Meeting started Wed Dec 15 15:00:25 2021 UTC and is due to finish in 60 minutes. The chair is vishalmanchanda. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:25 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:25 The meeting name has been set to 'horizon' 15:00:52 hi anyone around for horizon meeting? 15:00:56 o/ 15:01:20 hi (but not feeling good :( ) 15:01:22 o/ 15:01:47 e0ne: ping 15:02:01 amotoki: hope you will get well soon! 15:02:04 hi 15:02:06 Let's start the meeting. 15:02:09 o/ 15:02:20 #topic Notices 15:02:36 few announcement from my side 15:02:58 As usual Yoga Schedule reminder https://releases.openstack.org/yoga/schedule.html 15:03:10 This week is R-15. 15:03:55 Next announcement is about python runtime testing for Yoga cycle. 15:04:12 Python runtime for Yoga can be found here https://governance.openstack.org/tc/reference/runtimes/yoga.html#python-runtimes-for-yoga 15:04:38 I have already pushed a patch to update the reference in horizon https://review.opendev.org/c/openstack/horizon/+/821800 15:04:47 Please take a look. 15:05:47 Next announcement is about adding review-priority check in horizon gerrit dashboard. 15:06:17 I was thinking if we can add review priority check for horizon in gerrit, so patches with high priority get some attention. 15:06:55 further, we monitor those in our weekly meeting if any priority patches need attention like this dashboard 15:07:04 https://review.opendev.org/q/label:Review-Priority%253D%252B2+status:open+(project:openstack/cinder) 15:07:05 +1. it's a good thing to have 15:07:21 vishalmanchanda: do you plan to update the horizon config in gerrit-dash-creator? 15:08:13 I am okay with either as I have the similar one in neutron and other projects as e0ne mentioned. 15:08:34 my question is how the flag is handled. 15:08:38 amotoki: no I will add in project config like this one https://review.opendev.org/c/openstack/project-config/+/739550 15:09:41 project-config defines the flag. it is a provider side. 15:10:06 ok If no objections here I'll add the review priority for the horizon in gerrit. 15:11:09 amotoki: you mean I have to update the horizon config in gerrit-dash-creator as well? 15:11:23 vishalmanchanda: it is up to you. it is optional. 15:11:28 amotoki: that I will check. 15:11:45 ok moving to next announcemnet 15:12:39 I am on vacation next week till Thursday(23rd Dec.), So I'll not be able to chair the next meeting(22nd December) 15:13:03 on 29th December week, I guess it is the holiday for all of us, So I was thinking if we can Cancel next two meetings. 15:13:41 +1 15:14:28 sounds good 15:14:58 ok, then our Next Weekly meeting will be on 5th January 2022. 15:16:17 I will be around for any discussion after 23rd December, so feel free to reach me in case of any help. 15:16:32 moving to next topic 15:16:37 #topic release priorties 15:16:59 No update from my side on this topic. 15:17:14 Patches are up for few priorities items, please review them asap. 15:17:21 Most of the patches already have one +2. 15:17:59 dropping the patches link here https://review.opendev.org/c/openstack/horizon/+/816368 15:18:20 https://review.opendev.org/c/openstack/horizon/+/817814 , https://review.opendev.org/c/openstack/horizon/+/816303 15:19:18 rdopiera: would you like to add anything regarding System Scope topic? 15:20:21 vishalmanchanda: I don't have anything to add at this moment. I have to talk to the developers from Nova and other projects to figure out what should exactly be visible and work in the system scope, and what shouldn't. 15:20:42 because currently half of the admin tab doesn't work 15:20:58 and I don't know if that is a bug, a work in progress, or expected 15:21:53 rdopiera: hmm we can discuss this with policy-popup team, I guess. 15:22:22 vishalmanchanda: do you have a link to that? 15:22:35 rdopiera: Now we have your initial patch which at least adds the menu. 15:23:12 I think I can check it during next week. at least I can check it from neutron perspective. 15:24:07 rdopiera: no I don't have a link now, but I will check and drop a link here for you. 15:24:24 vishalmanchanda: thanks a lot! 15:24:30 rdopiera: yw. 15:24:33 I missed it when it was announced 15:24:54 moving to next topic 15:25:17 #topic Bug deputy report 15:25:53 We have one new bug reported for this week https://bugs.launchpad.net/horizon/+bug/1954619 15:26:41 amotoki: I also added some comments for this bug, please take a look. 15:26:50 vishalmanchanda: where do we define the default device name in Angular impl? I am not sure. 15:28:44 amotoki: honestly I haven't check the code for angular impl yet but I have tested it in my devstack env. 15:29:18 vishalmanchanda: so, do you confirm that horizon sends 'vda' to nova API? 15:29:19 I see the same result with both implementations. 15:30:37 amotoki: that I will check again after the meeting. 15:30:52 yeah, let's double check anyway :) 15:31:03 ok. 15:31:11 moving to next topic 15:31:24 #topic On-Demand Agenda 15:32:05 staring with first topic i.e. Horizon pain-point-elimination discussion 15:33:30 So there is video call meeting regarding the pain point discussion on Wednesday last week 15:33:44 Summary can be found here http://lists.openstack.org/pipermail/openstack-discuss/2021-December/026169.html 15:34:12 I and amotoki also joined that meeting. 15:34:56 Pain Point for horizon can be found here https://etherpad.opendev.org/p/pain-point-elimination (at line 214) 15:35:39 So Can everyone please go through the list of pain point related to horizon. 15:36:08 We need to check if that's valid a issue and it's a new feature request or a bug which horizon team already know and add the link for that bug/feature. 15:36:59 You can use FEATURE REQUEST/DISCUSSION NEEDED/FIX NEEDED/INFO NEEDED tags. 15:37:22 If you find need more information about that issue/pain-point then use INFO NEEDED tag and ask for more information. 15:37:36 like amotoki did at line 232. 15:40:09 Right now I am not sure about the all issue/pain point but if you know anyone of those please add your comment and then we discuss it further and also if we not have any volunteer who can work on those issue that can be discussed later. 15:41:02 speaking of pain points, I want to get rid of all the django template and deprecation warnings we have accumulated 15:41:05 like for glance issue at line 218 I know there is already a similar bug in horizon 15:41:14 https://bugs.launchpad.net/horizon/+bug/1935001 15:41:27 the amount of them we have now is ridiculous and it's hard to find the error messages between them 15:43:03 rdopiera: make sense, I use development server method for development and see lot of error/deprecation messages. 15:44:10 amotoki: Did I missed anything from that meeting that you would like to add here? 15:45:20 vishalmanchanda: nothing special to add. some have been fixed and some are not. some are not clear. 15:46:27 I havent' checked one by one though 15:48:18 Ok Please everyone go through that list and add your thoughts about that issue asap. 15:48:33 I will also add my comment there. 15:49:25 Also, Feel free to discuss any pain point here. 15:50:05 ok next topic is a paste for a unit test https://paste.opendev.org/show/811691/ 15:50:16 I am not sure you have added that. 15:50:17 that's me ^ 15:50:24 ganso: ok go ahead. 15:50:43 I'm struggling with unit tests for that network page pagination. More specifically, what I highlighted in the pastebin 15:50:51 basically, I added 1 extra network for completeness of my tests 15:51:22 upon doing that, I broke a bunch of stuff in a weird way. I still need to go through most failures, but the 2 that strike me most are the 2 ones I included in the pastebin 15:51:39 so one of them has 2 extra quota calls when adding one network, that is strange 15:52:17 the other failure, is basically the opposite of what we would expect. It was expected to fail with the message "CIDRs allowed for user private ipv6 networks are fc00::/9." 15:52:37 but the result was 15:53:22 I'm inclined to thing those unit tests were artificially working, may be coded incorrectly, as I currently do not understand how adding one extra network is breaking them in that way 15:54:27 I'm a bit lost on this and spending a significant amount of time digging on those failures, so I'm also thinking about removing that extra network for my patch so it wouldn't break those unit tests 15:54:41 but it is somewhat a yellow flag 15:55:04 I just wanted to raise this concern 15:55:18 we do have a lot of tests that instead of find bugs just pin the implementation in place 15:55:41 they were converted from using a library that encouraged this kind of testing 15:55:50 rdopiera: like removing that specific extra network for those specific set of tests ? 15:55:58 ganso: honestly, I also struggle with these unit test, so what I do update these to get passed. 15:56:02 like checking that a certain function was called an exact number of times and so on 15:56:14 ganso: Althoght I am really for the late for the review, regarding the test failure, is the failure in the latest patch set of https://review.opendev.org/c/openstack/horizon/+/806454 the thing we should check? 15:56:51 amotoki: it is not included in that patchset (I think), I can push my latest so you can see those failures 15:57:04 ganso: okay 15:57:05 ganso: like If I see this error AssertionError: 7 != 9 I will update it to 9. 15:57:19 ganso: to get test passed. 15:57:36 vishalmanchanda: lol yea I thought about that but it bothers me a bit if I don't understand why 15:57:53 ganso: I checked the behavior of neutron this week and 'shared' attribute only affects networks exactly shared to a specific project even with the admin role, so the approach looks good in general. 15:57:54 vishalmanchanda: but the second one, I can't just update, the behavior is completely inverted 15:58:34 ganso: hmm for second need to check. 15:59:06 the TemplateResponse probably needs an explicit conversion of the object to a string 15:59:34 rdopiera: the response code would be different than 200 for the error message expected 16:00:04 ganso: ok, please push your updated patch to gerrit then we can check and help. 16:00:15 vishalmanchanda: sure, will do. Thanks! 16:00:26 ok we are out of time. 16:00:46 Thanks everyone for joining. 16:00:57 amotoki: TC, get well soon. 16:01:24 #endmeeting