15:00:10 <vishalmanchanda> #startmeeting horizon
15:00:10 <opendevmeet> Meeting started Wed May 24 15:00:10 2023 UTC and is due to finish in 60 minutes.  The chair is vishalmanchanda. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:10 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:10 <opendevmeet> The meeting name has been set to 'horizon'
15:00:16 <rdopiera> o/
15:00:30 <vishalmanchanda> Hello everyone
15:01:25 <tmazur> o/
15:02:21 <amotoki> hi
15:02:24 <vishalmanchanda> let's start the meeting
15:02:36 <vishalmanchanda> agenda of meeting can be found here https://etherpad.opendev.org/p/horizon-release-priorities#L39
15:03:34 <vishalmanchanda> I have no announcements for this week, so moving to release priorities topic.
15:03:46 <vishalmanchanda> #topic Release priorities
15:04:11 <vishalmanchanda> Update-jQuery
15:04:45 <vishalmanchanda> I have pushed a patch to bump XStatic-JQuery-Migrate version from 1.2.1.1->3.3.2.1 and horizon-integration job also pass.
15:05:02 <vishalmanchanda> Please take a look https://review.opendev.org/c/openstack/horizon/+/883660
15:05:53 <vishalmanchanda> The next step would be updating the XStatic-jQuery version from 1.12.4.1->3.5.1.1 in horizon CI jobs.
15:07:23 <vishalmanchanda> I hope it doesn't broke the CI jobs.
15:07:40 <vishalmanchanda> specifically horizon-integration job.
15:08:06 <vishalmanchanda> moving to next thing
15:08:40 <amotoki> the failure rate of the integration job is already super high, so I don't think we can evaluate whether the version bump affects it :-(
15:08:40 <vishalmanchanda> Setuptools issue https://bugs.launchpad.net/horizon/+bug/2007574
15:09:45 <vishalmanchanda> amotoki: yeah, but I am testing it manually for all the panels but few things I may miss in my testing.
15:10:32 <amotoki> vishalmanchanda: the failure rate of the integration job is another topic. perhaps we need to discuss it separately
15:11:42 <vishalmanchanda> amotoki: yes, I was saying it didn't breaks due to changes I am doing for jQuery update.
15:11:51 <vishalmanchanda> For setuptool, patch to use latest version of XStatic =1.0.3 is merged in both openstack/requirements and openstack/horizon.
15:12:17 <vishalmanchanda> Now it is using XStatic =1.0.3, see https://zuul.opendev.org/t/openstack/build/730cdebb1a864d6289d0afca99d6e39e/log/kolla/build/horizon.log#145
15:12:35 <vishalmanchanda> but still getting the same error https://zuul.opendev.org/t/openstack/build/730cdebb1a864d6289d0afca99d6e39e/log/kolla/build/horizon.log#1883
15:13:31 <vishalmanchanda> rdopiera: did you get the time to look into it?
15:14:07 <rdopiera> vishalmanchanda: no
15:14:27 <rdopiera> vishalmanchanda: I suppose that means we have to make that change to all xstatic packages...
15:14:45 <rdopiera> but before we do that, I need to test it
15:14:52 <rdopiera> on some fake packages
15:16:15 <vishalmanchanda> rdopiera: yes, we start with XStatic-Moment-Timezone pypi
15:18:40 <vishalmanchanda> that's all update from my side on this topic
15:19:16 <vishalmanchanda> moving to next topic
15:19:25 <vishalmanchanda> #topic open-discussion
15:20:15 <vishalmanchanda> I have pushed a patch to rename x/xstatic-angular-fileupload->openstack/xstatic-angular-fileupload but getting some errors.
15:20:25 <vishalmanchanda> https://review.opendev.org/c/openstack/horizon/+/883995
15:21:28 <vishalmanchanda> Does anyone have any idea how to fix it?
15:21:56 <amotoki> we hit the same issue on xstatic-fond-awesome too
15:22:00 <vishalmanchanda> I think this series of patch needs to be merged till stable/ussuri branch to fix it https://review.opendev.org/q/I9c75b084705dfdc6df52aca1659f89444c8a1728
15:22:08 <vishalmanchanda> amotoki: yes.
15:22:08 <amotoki> I think it is better to ask it to the opendev infra team
15:22:34 <vishalmanchanda> amotoki: ok, I will check with them
15:22:41 <amotoki> stable/ussuri syntax error is just what we see
15:22:55 <amotoki> it is just because ussuri branch zuul config is checked first
15:23:08 <vishalmanchanda> ohh.
15:23:11 <amotoki> I think we will hit the same error one branch by one
15:24:12 <amotoki> but it is just my guess, so let's discuss it with the TACT SIG member
15:24:32 <vishalmanchanda> yeah, we have to rename x/xstatic-angular-fileupload and remove xstatic-font-awesome in a single patch in stable branches, but anyway I will check with them
15:26:22 <vishalmanchanda> We have 1 new bug reported this week https://bugs.launchpad.net/horizon/+bug/2020630
15:26:52 <vishalmanchanda> I haven't looked at the bug yet, If anyone get some time please take a look.
15:27:45 <tmazur> I will take a look. Might be connected to the angular upgrade.
15:28:09 <vishalmanchanda> tmazur: thanks.
15:29:19 <vishalmanchanda> Does anyone setup keystone Federation env. on devstack and provide me some references?
15:29:25 <amotoki> as I mentioned above, the failure rate of the integration job is super high https://grafana.opendev.org/d/260356fb6c/horizon-failure-rate?orgId=1&from=1684335968345&to=1684940768345&viewPanel=4
15:29:39 <amotoki> does it happen for purely random test?
15:29:49 <amotoki> or does it happen in a specific set of tests?
15:30:33 <amotoki> if it fails in a specific set of tests, over 90% failure might be a good indicator to skip them
15:32:48 <vishalmanchanda> I have noticed pagination and filteration related integration tests fails too much.
15:33:41 <vishalmanchanda> we can also rewrite those random failing tests rather then skipping it.
15:34:12 <amotoki> theose tests often fail for long.... but the failure rate these weeks are too high. I wonder something has changed
15:34:42 <amotoki> at least "recheck random failure" mesage is confusing
15:35:30 <vishalmanchanda> hmm, we can add more detail info. while rechecking for a patch.
15:35:44 <amotoki> vishalmanchanda: do you suggest to continue recheck recheck recheck.... until we succeed to rewrite them?
15:36:56 <vishalmanchanda> amotoki: you want to skip those failed tests?
15:37:09 <tmazur> rdopiera and I were also thinking about rewriting integration tests as scenario-based. This will help to decrease the amount of tests running
15:37:31 <amotoki> vishalmanchanda: IMHO we can skip them once and explore how to make them stable
15:38:21 <amotoki> we are wasting CI resources and power. we are not energy friendly :p
15:38:47 <tmazur> Basically we test create something, and then we test update the same thing, and for this we run create again. So we can skip the create test and include it into the scenario
15:38:53 <vishalmanchanda> ok, I am fine with that but just wondering about side-effect of that.
15:40:04 <vishalmanchanda> tmazur: I like the idea.
15:40:25 <amotoki> we are doing almost the same thing as "bare recheck" and other tests verifies the basic operation, so I think we can skip them
15:40:43 <vishalmanchanda> amotoki: that's true.
15:45:49 <vishalmanchanda> what we can do now, make a list of these random failing integration tests and then decide to skip that test or not.
15:47:05 <vishalmanchanda> where we can save these type of data about the failed integration test which everyone of us can see and update?
15:47:10 <vishalmanchanda> etherpad?
15:47:29 <vishalmanchanda> or somewhere else in tabular format.
15:47:59 <amotoki> I think we can use etherpad and collect failing tests for example last week
15:48:34 <tmazur> +1 for etherpad
15:48:47 <vishalmanchanda> ok
15:48:58 <amotoki> launchpad bug can be used as an entry point too (as it allows us to link a bug from a gerrit change)
15:49:20 <amotoki> but we can collect stats on failing tests in an etherpad
15:49:55 <vishalmanchanda> I will create a new etherpad and add some info. their.
15:50:24 <vishalmanchanda> Feel free to update and add some more info. there.
15:50:37 <amotoki> thanks
15:52:05 <vishalmanchanda> https://etherpad.opendev.org/p/random-failed-horizon-integration-tests
15:52:36 <vishalmanchanda> Does anyone have any other topic to discuss?
15:56:12 <amotoki> none from me
15:56:21 <tmazur> nothing
15:58:03 <vishalmanchanda> Does anyone of you ever setup Keystone Federation env. on devstack and provide me some references?
16:00:46 <vishalmanchanda> ok let's end this meeting.
16:00:50 <amotoki> I have no experience on it. devstack itself does not support it and perhaps you need extra config after running devstack
16:01:03 <tmazur> Never did that
16:01:08 <vishalmanchanda> ok
16:01:17 <vishalmanchanda> thanks, everyone for joining.
16:01:26 <vishalmanchanda> #endmeeting