amotoki | vishalmanchanda: hi, do you have any information on my question in https://review.opendev.org/c/openstack/horizon/+/791532 ? you are the original author of the volume backup panel, so you may know something. | 09:59 |
---|---|---|
vishalmanchanda | amotoki: looking. | 10:02 |
vishalmanchanda | amotoki: whoami-rajat replied to your queries, please take a look. | 10:24 |
vishalmanchanda | whoami-rajat: thanks:) | 10:24 |
whoami-rajat | np | 10:30 |
amotoki | vishalmanchanda: whoami-rajat: thanks for the info. | 10:31 |
amotoki | I wonder horizon needs to provide a configuration option to mirror cinder filter configuration. | 10:32 |
amotoki | otherwise, users cannot filter resources if such field is not supported. for example, what happens 'size' is disabled? | 10:32 |
amotoki | vishalmanchanda: any thought on this? | 10:32 |
whoami-rajat | amotoki, can we query list-filters first from horizon to know the enabled filters and only use those? | 10:44 |
vishalmanchanda | amotoki: is it even possible to disable 'size' field for backup? | 10:45 |
amotoki | whoami-rajat: yes, it can. perhaps we need to add a common logic to handle cinder generic filter support. the discussion started from backup filter support. | 10:46 |
whoami-rajat | yes, that sounds good, the common logic can be used for all GET resources APIs | 10:48 |
amotoki | vishalmanchanda: do you talk about horizon side? if so, a subclass of horizon.tables.actions.FilterAction can support it. the subclass for cinder can adjust filter_choices dynamically based on the result of list-filters call. | 10:51 |
vishalmanchanda | amotoki: ok, I will take a look | 10:54 |
vishalmanchanda | #startmeeting horizon | 15:00 |
opendevmeet | Meeting started Wed May 25 15:00:13 2022 UTC and is due to finish in 60 minutes. The chair is vishalmanchanda. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:00 |
opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:00 |
opendevmeet | The meeting name has been set to 'horizon' | 15:00 |
tmazur | o/ | 15:00 |
vishalmanchanda | hello everyone | 15:01 |
rdopiera | o/ | 15:01 |
vishalmanchanda | ok let's start the meeting | 15:02 |
vishalmanchanda | agenda of meeting can be found here https://etherpad.opendev.org/p/horizon-release-priorities#L39 | 15:02 |
vishalmanchanda | #topic Notices | 15:02 |
vishalmanchanda | A casual reminder about schedule. We passed Zed-1 milestone last week. | 15:03 |
vishalmanchanda | This week is R-19 week. | 15:03 |
vishalmanchanda | For more info about schedule please refer https://releases.openstack.org/zed/schedule.html | 15:03 |
vishalmanchanda | I have no other announcements to make, if anyone have any announcement to make. please go ahead | 15:04 |
vishalmanchanda | moving to next topic | 15:05 |
vishalmanchanda | #topic Release priorities | 15:05 |
vishalmanchanda | few updates from my side on this topic. | 15:06 |
vishalmanchanda | tmazur: I am waiting for your review on deprecation patch https://review.opendev.org/c/openstack/horizon/+/838333 | 15:07 |
vishalmanchanda | Could take a look asap, so can merge this asap. | 15:07 |
tmazur | I'll review it asap | 15:07 |
vishalmanchanda | tmazur: thanks. | 15:07 |
vishalmanchanda | about migration to nodejs16 https://review.opendev.org/c/openstack/horizon/+/831929 is merged last week | 15:08 |
vishalmanchanda | Now waiting for plugins to use new nodejs template | 15:08 |
vishalmanchanda | Gate is passing for all the plugins but unfortunately, few plugins maintainers/core team is not active in horizon plugins. | 15:10 |
vishalmanchanda | So if you guys can take a look, would be great. | 15:10 |
vishalmanchanda | https://review.opendev.org/q/topic:migrate-to-nodejs16 | 15:10 |
vishalmanchanda | that's all update from my side on release-priorities topic. | 15:11 |
vishalmanchanda | If anyone would like to share any updates on this topic, please go ahead. | 15:12 |
tmazur | vishalmanchanda: I have some news about the migration to angular 1.8.2 | 15:13 |
tmazur | It's done, and I tested it on my env. The patch is coming today I hope, just need some grooming. | 15:15 |
vishalmanchanda | tmazur: awesome🙂) | 15:15 |
tmazur | But we really need to test it as much as we can, since the changes are critical | 15:15 |
vishalmanchanda | tmazur: +1. | 15:16 |
tmazur | Also, I had to skip a couple tests on promise rejections, since the rejection process itself has been changed after 1.5.8 | 15:16 |
tmazur | So a rejection is like a resolved promise now, which makes those tests useless, and they need to be rewritten completely | 15:17 |
tmazur | But that's out of the current scope for sure | 15:18 |
vishalmanchanda | ok, will wait for your patch then let's discuss it in detail. | 15:19 |
vishalmanchanda | moving to next topic | 15:20 |
vishalmanchanda | #topic Bug deputy report | 15:20 |
vishalmanchanda | we have 3 new bugs reported in last two weeks. | 15:21 |
vishalmanchanda | for the bug https://bugs.launchpad.net/horizon/+bug/1974184 I have already added my thoughts in bug summary. | 15:22 |
amotoki | as I commented, when disabling the create button, the usage does not exceed the quota. it just reaches the quota. I think this is the point of the bug. | 15:25 |
vishalmanchanda | I am also fine with using 'Quota reached' or 'Quota exceeded' | 15:25 |
vishalmanchanda | because neutron also uses 'Quota exceeded' | 15:25 |
amotoki | in case fo API response, it means the usage would exceed the quota if a requested resource is crated. | 15:26 |
vishalmanchanda | We can use 'Quota reached' as it looks more correct. | 15:27 |
vishalmanchanda | moving to next bug https://bugs.launchpad.net/horizon/+bug/1974470 | 15:29 |
vishalmanchanda | 2nd bug looks valid to me | 15:31 |
rdopiera | we were looking already into using the new apis | 15:33 |
rdopiera | but this is more an RFE than a bug | 15:34 |
vishalmanchanda | I am still investigating the 2nd bug, will take a look how to use placement api | 15:34 |
vishalmanchanda | rdopiera: yeah. | 15:34 |
vishalmanchanda | rdopiera: sorry, what you mean, are we already using placement apis? | 15:36 |
rdopiera | vishalmanchanda: tmazur and me were already looking into adding the placement apis to horizon, but we don't have anything working yet | 15:37 |
vishalmanchanda | rdopiera: ok. | 15:38 |
rdopiera | and I think that getting the libraries updates is higher priority right now | 15:38 |
vishalmanchanda | yeah, anyway I will take a look at this(use placement-api in horizon) but it will take time. | 15:39 |
vishalmanchanda | moving to last bug https://bugs.launchpad.net/horizon/+bug/1971592 | 15:40 |
vishalmanchanda | for 3rd bug, we already know this a known issue. | 15:40 |
vishalmanchanda | I have also replied to the author on open-discuss ML. | 15:41 |
vishalmanchanda | but I guess we have to add it in horizon documentation as a known issue. | 15:42 |
vishalmanchanda | Otherwise people are going to ask it again and again. | 15:43 |
vishalmanchanda | I can push a patch to add this in horizon documentation. | 15:44 |
vishalmanchanda | but I forgot, will do that asap. | 15:44 |
vishalmanchanda | moving to next topic | 15:46 |
vishalmanchanda | #topic On-Demand Agenda | 15:46 |
vishalmanchanda | Restructuring of integration tests | 15:46 |
vishalmanchanda | rdopiera: Floor is yours. | 15:47 |
rdopiera | right, so I was thinking a little bit more about the integration tests, because they give us a lot of headache recently | 15:48 |
vishalmanchanda | 100% | 15:48 |
rdopiera | and I noticed that we have a lot of duplicate tests in there, for example | 15:48 |
rdopiera | we have a test that checks if you can log in into horizon | 15:49 |
rdopiera | but then every other test also logs in into horizon | 15:49 |
rdopiera | and the login test doesn't even run first | 15:49 |
rdopiera | which makes it completely useless | 15:49 |
rdopiera | the thing is, we wrote those tests like one would write unit tests -- each testing one specific thing | 15:50 |
rdopiera | when for integration tests, we should instead write them from user stories | 15:50 |
rdopiera | so we would, for example, have one "story" about how flavors work, and all flavor operations would be tested in it | 15:51 |
rdopiera | of course it would be a lot of work to rewrite those tests we have right now | 15:51 |
rdopiera | so as a stop-gap solution, I propose that we for now simply skip the tests that add no coverage -- like the login test -- specifying which other test also tests the same thing | 15:51 |
rdopiera | that should reduce the time of running the tests | 15:52 |
vishalmanchanda | Is there any specific reason or history, we login again for each test? | 15:52 |
rdopiera | you want to be able to run each test independently | 15:53 |
rdopiera | this brings me to a second point | 15:53 |
rdopiera | we need to make use of the setup and teardown functions for the tests -- especially for deleting any leftover objects when a test has failed | 15:54 |
amotoki | +1. perhaps addCleanup would be better rather than teardown though | 15:55 |
rdopiera | right, pytest has some cool mechanisms for this | 15:56 |
rdopiera | that could even let us avoid having to log in every time | 15:56 |
vishalmanchanda | rdopiera: I am agreed with both of your points. | 15:57 |
vishalmanchanda | As it is really irritating to do recheck on patches. | 15:57 |
rdopiera | the rechecks are mostly due to no cleanup | 15:57 |
rdopiera | then it's stuck broken until the vm is reprovisioned | 15:58 |
vishalmanchanda | No patch is merged in horizon last week due to these random integration tests failure. | 15:58 |
vishalmanchanda | rdopiera: ok. | 15:58 |
vishalmanchanda | rdopiera: are you already working on that patch to fix it. | 15:59 |
rdopiera | vishalmanchanda: I wanted to make sure there is an agreement to that first | 15:59 |
rdopiera | I will start on it next week | 16:00 |
vishalmanchanda | I have no issue with it. | 16:00 |
vishalmanchanda | amotoki: tmazur : please let us know, if you see any issues here? | 16:00 |
vishalmanchanda | we are out of time. | 16:00 |
amotoki | I am fine with it | 16:01 |
rdopiera | I suppose the cleanup part is the most urgent now | 16:01 |
vishalmanchanda | let me just end this meeting and continue the discussion here. | 16:01 |
vishalmanchanda | #endmeeting | 16:01 |
opendevmeet | Meeting ended Wed May 25 16:01:34 2022 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:01 |
opendevmeet | Minutes: https://meetings.opendev.org/meetings/horizon/2022/horizon.2022-05-25-15.00.html | 16:01 |
opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/horizon/2022/horizon.2022-05-25-15.00.txt | 16:01 |
opendevmeet | Log: https://meetings.opendev.org/meetings/horizon/2022/horizon.2022-05-25-15.00.log.html | 16:01 |
vishalmanchanda | rdopiera: yes. | 16:01 |
tmazur | +1 for the cleanup first | 16:02 |
amotoki | vishalmanchanda: note that we have a meeting in our own irc channel, so one hour is not so strict | 16:02 |
vishalmanchanda | amotoki: ok. | 16:02 |
vishalmanchanda | Does anyone have any other topic to discuss? | 16:03 |
vishalmanchanda | looks nothing more to discuss, thanks everyone for joining. | 16:05 |
vishalmanchanda | See you next week. | 16:05 |
amotoki | rdopiera: how does the fact that we don't clean up resources properly affect failures in the integration tests? | 16:05 |
amotoki | does it cause pagination test faiures? | 16:05 |
rdopiera | amotoki: pagination tests fail | 16:05 |
rdopiera | yup | 16:05 |
rdopiera | mostly | 16:05 |
rdopiera | sometimes it's some other test, when it selects the wrong object | 16:06 |
rdopiera | another option would be to do a cleanup at the start of the test | 16:06 |
rdopiera | but I think proper teardown is the right way to do it | 16:07 |
rdopiera | we could do both | 16:07 |
amotoki | agree. proper teardown is the right way | 16:07 |
rdopiera | if we do both, this will also clean the current problem | 16:07 |
amotoki | rdopiera: thanks for detail investigation | 16:11 |
rdopiera | I think that in fact I should add the cleanup at the beginning of the test first, merge that, let it clean all the problems, and then in another patch replace that with proper teardown | 16:11 |
vishalmanchanda | rdopiera: +1. | 16:13 |
amotoki | we don't run the integration tests in parallel, so cleanup at the beginning of each test should work | 16:13 |
opendevreview | Merged openstack/horizon stable/wallaby: Add horizontal scrollbar to role dropdown https://review.opendev.org/c/openstack/horizon/+/810197 | 16:34 |
opendevreview | Merged openstack/horizon master: Bump tox minversion to 3.18.0 https://review.opendev.org/c/openstack/horizon/+/842864 | 16:34 |
opendevreview | Merged openstack/horizon master: Pass real client IP to keystoneauth original_ip https://review.opendev.org/c/openstack/horizon/+/838859 | 16:49 |
opendevreview | Tatiana Ovchinnikova proposed openstack/xstatic-angular master: Update XStatic-Angular to 1.8.2 https://review.opendev.org/c/openstack/xstatic-angular/+/843342 | 19:35 |
opendevreview | Tatiana Ovchinnikova proposed openstack/horizon master: Migrate to AngularJS v1.8.2 https://review.opendev.org/c/openstack/horizon/+/843346 | 20:36 |
*** tosky_ is now known as tosky | 21:49 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!