[09:15:21] <hashar>	 kostajh: good morning :]
[09:16:05] <hashar>	 I should mute those  INFO:backend.PhpWebServer log message I don't think they have much purposes
[09:16:12] <hashar>	 at least if they are a 2xx or 3xx
[09:16:36] <hashar>	 so for that build
[09:16:44] <hashar>	 3 minutes just to rsync the cache :-\
[09:17:44] <kostajh>	 hashar: Bonjour
[09:17:52] <hashar>	 the patch was for EventLogging and we ran every single selenium tests from AbuseFilter, Cite, Echo, FileImporter, GrowthExperiments, MobileFrontend, ProofreadPage, VisualEditor, Wikibase and MinervaNeue
[09:17:58] <kostajh>	 I’m afk but back later on today
[09:18:08] <hashar>	 and that approaches to blindly run every single tests  doesn't work
[09:18:22] <hashar>	 kostajh: no worries, I will continue on crafting the apache images for php73+
[09:18:23] <hashar>	 :]
[09:18:27] <kostajh>	 I have patches to split the selenium jobs into two groups and also to parallel install npm which I think will help
[09:18:36] <hashar>	 yeah that might
[09:18:58] <hashar>	 but we really should tackle the functional side of the problem: aka which tests do we actually want to run ;)
[09:19:24] <hashar>	 oh also 00:18:08.708 npm WARN old lockfile HttpErrorGeneral: 503 Service Unavailable - GET https://registry.npmjs.org/grapheme-splitter - Service Unavailable
[09:19:39] <hashar>	 with a fairly large delay
[09:19:51] <kostajh>	 well we have the principle of running all tests in case a patch breaks something, right? Eg if Minerva breaks something in growthexperiments we want to know about it
[09:20:00] <hashar>	 like 6 minutes timeout after a 503 from npmjs
[09:20:12] <hashar>	 which explains part of the unusually slow build
[09:20:18] <hashar>	 yeah
[09:20:32] <hashar>	 then for a patch to EventLogging, I am not sure we need to run every single tests of Wikibase
[09:20:52] <hashar>	 then determinating which test to run for another extension is a bit complicated ]
[09:21:04] <hashar>	 there is the same issue for the PHPUnit integration tests :\
[09:21:17] <hashar>	 anyway, I will be around this afternoon
[09:21:21] <hashar>	 going to refactor the docker images
[11:39:02] <kostajh>	 hashar: i'm back, if you want to chat more
[11:42:52] <hashar>	 yeah we can 
[11:43:00] <hashar>	 let me copy paste files here and there first
[11:52:51] <hashar>	 kostajh: I am back and joined https://meet.google.com/vjr-ahpb-ygz :]
[11:53:27] <kostajh>	 hashar: ok, let me get some tea, I'll be there in a few minutes.
[11:53:34] <hashar>	 same :D
[11:59:36] <wikibugs>	 (03CR) 10Kosta Harlan: Add phpcs-and-lint stage (031 comment) [integration/quibble] - 10https://gerrit.wikimedia.org/r/747617 (https://phabricator.wikimedia.org/T297561) (owner: 10Kosta Harlan)
[15:39:31] <hashar>	 kostajh: it works! :D
[15:39:43] <hashar>	 I will just get rid of the quibble-buster-php72-apache ;)
[15:52:25] <hashar>	 docker run --entrypoint=quibble-with-apache   --rm -it -v ~/projects:/srv/git:ro docker-registry.wikimedia.org/releng/quibble-buster-php72:1.2.0-s5 --run=qunit
[16:39:27] <hashar>	 [0-2] RUNNING in chrome - /tests/selenium/specs/user.js
[16:39:27] <hashar>	 [0-2] PASSED in chrome - /tests/selenium/specs/user.js
[16:40:22] <hashar>	 success
[16:40:46] <hashar>	 https://gerrit.wikimedia.org/r/748104 is the change