15:00:17 #startmeeting horizon 15:00:17 Meeting started Wed Jun 28 15:00:17 2023 UTC and is due to finish in 60 minutes. The chair is vishalmanchanda. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:17 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:17 The meeting name has been set to 'horizon' 15:00:25 o/ 15:00:37 hello everyone 15:02:07 agenda of the meeting can be found here https://etherpad.opendev.org/p/horizon-release-priorities#L32 15:04:07 only two of us:( 15:04:19 anyway let's start the meeting 15:04:40 I have no announcements for this week. 15:04:58 so moving release priorities topic 15:05:12 #topic Release priorities 15:05:29 horizon-integration job failure. 15:06:16 As discussed in the last weekly meeting, I have added a integration job which uses chrome browser. 15:08:34 we can also see browser logs for "horizon-integration-tests-with-chrome" job. 15:10:20 unfortunately I am not succeed in getting the same issue in my local env. 15:11:03 I am trying to get the same error in my local env., so that I can fix those warning which we can see in ``browser.log``. 15:11:41 rdopiera: around? 15:12:21 vishalmanchanda: sorry I'm late 15:12:26 As you suggested to use latest version of selenium, I tried that in our CI job but all tests fail with that version. 15:12:57 see https://zuul.opendev.org/t/openstack/build/d8d6833078c94ceabcbab8d808272f81 15:13:05 vishalmanchanda: well, our code is not compatible with the last version of selenium, by "I will try to update selenium" I meant I will work on our code to make it work with it 15:13:31 we will want to upgrade it eventually anyways 15:13:31 I was investigating Volume tests failures, and I've got the same issues on my local env, but that was because I had old cirros image. After fixing it, I was not able to reproduce the issues :/ 15:14:41 yeah horizon code not compatible with selenium version 4.1.0 for now. 15:15:18 I'm working onit 15:16:00 cirros image again got change to 0.6.2 from 0.6.1 version, I have updated in our test patch 886867 15:16:35 I wonder if we could somehow get that image name automatically? 15:17:09 tmazur: you are running integration test with chrome or firefox in your local env? 15:17:18 Both 15:17:39 ok 15:18:18 did you done any changes in karma.conf file for running it with chrome browser? 15:18:22 My env is rather old, so I have old cirros image there, but I've just added a new one manually 15:18:27 the main problem with chrome right now is that because the instances page refreshes right after starting an instance, we never see the info message that confirms the launch 15:18:44 Karma.conf is for npm tests, and I was not touching it 15:18:47 karma.conf is for the unit tests 15:18:57 ohh 15:20:09 For running it with chrome browser I just manually start the tests like this 15:20:13 INTEGRATION_TESTS=1 SELENIUM_CHROME=1 pytest ./openstack_dashboard/test/integration_tests --ds=openstack_dashboard.test.settings -v -k test_volumes 15:21:02 btw, we should add a pytest.conf file with that --ds=openstack_dashboard.test.settings setting, so that we don't have to specify it each time 15:21:24 but that's after we fix everything, of course 15:22:26 good idea 15:23:20 rdopiera: for get that image name automatically, you mean image version doesn't depends on devstack config? 15:23:45 for horizon integration job? 15:24:11 vishalmanchanda: we could get it from that devstack config 15:24:30 hmm 15:24:45 or just run an openstack cli command before running the tests to check the image name and populate the test config 15:24:53 that's what we do downstream 15:25:18 well, not for image name, but for other things, like network name 15:26:08 ok 15:27:09 but again, let's fix everything first, then we can talk about improvements 15:27:32 yeah. 15:27:55 I have tested some things with pytest, and I have a lot of ideas for improvements to the tests 15:28:35 tmazur: one last question you are using chromium or google-chrome for running the job? 15:29:35 I have Chrome 15:29:36 also on which Os your devstack env. running if you can tell. 15:30:01 The one with volumes is 18.04 15:30:58 ohh on older env. , integartion jobs are running for me but I am facing issue with ubuntu jammy 15:31:32 is it too... jammy? 15:31:39 yeah, I am planning to deploy a new one. But again, the issue with wrong image looks the same 15:31:43 now I will try again with older env. where it is running fine and try to reproduce the issue. 15:32:01 with jammy I face many issues. 15:32:22 even with firefox I am unable to run horizon-integration job. 15:32:33 some firefox profile issue. 15:32:39 Do we run our tests on jammy? 15:33:20 firefox really became the new msie 15:33:41 for our CI we are now using debian-bullseye 15:34:06 vishalmanchanda, thanks 15:34:29 because again some snap package issue with ubuntu jammy for horizon integration job 15:35:25 I am fine with skipping 4-5 tests to at least make the CI green for now. 15:35:51 because CI is failing for almost 3 weeks for now. 15:42:33 Do we have any other topic to discuss? 15:44:17 #topic open-discussion 15:46:05 If no other topic to discuss, let's end this meeting and try to fix our CI job asap. 15:46:21 Thanks everyone for your contributions. 15:46:27 See you next week. 15:46:39 o/ 15:47:00 JFYI I am on PTO from 30th June - 4th July. 15:47:16 #endmeeting