[08:25:08] holy x [08:25:50] I am starting to wonder whether using memcached cause those race conditions / failing tests [08:27:09] MediaWikiIntegrationTestCase gave us again `LogicException: Trying to delete mock tables, but table prefix 'parsertest_' does not indicate a mock database (expected 'unittest_')` [08:29:36] which is T292239 [08:29:51] the tests got flagged with `@group Broken` [08:29:53] will try again [08:30:37] (03CR) 10Hashar: [C: 03+2] "Trying again after the flappy test got properly flagged with "@group Broken" https://gerrit.wikimedia.org/r/c/mediawiki/core/+/771467" [integration/quibble] - 10https://gerrit.wikimedia.org/r/770518 (https://phabricator.wikimedia.org/T300340) (owner: 10Kosta Harlan) [08:44:08] (03Abandoned) 10Hashar: Jenkins job validation (DO NOT SUBMIT) [integration/quibble] - 10https://gerrit.wikimedia.org/r/663557 (owner: 10Hashar) [08:56:37] (03Merged) 10jenkins-bot: LocalSettings: Move Memcached block and define MemCachedServers [integration/quibble] - 10https://gerrit.wikimedia.org/r/770518 (https://phabricator.wikimedia.org/T300340) (owner: 10Kosta Harlan) [09:33:45] HURRAH [11:15:43] hehe [11:16:12] hashar: no I don't think memcached had anything to do with those failures, because it was happening consistently in core CI builds [13:17:31] kostajh: yeah that one is/was an issue in mediawiki indeed [16:35:22] (03PS1) 10Hashar: release: Quibble 1.4.4 [integration/quibble] - 10https://gerrit.wikimedia.org/r/771636 (https://phabricator.wikimedia.org/T300340) [16:35:58] (03PS1) 10Hashar: release: Start 1.4.5 cycle [integration/quibble] - 10https://gerrit.wikimedia.org/r/771637 [16:37:18] (03CR) 10Kosta Harlan: [C: 03+2] release: Quibble 1.4.4 [integration/quibble] - 10https://gerrit.wikimedia.org/r/771636 (https://phabricator.wikimedia.org/T300340) (owner: 10Hashar) [16:37:30] (03CR) 10Kosta Harlan: [C: 03+2] release: Start 1.4.5 cycle [integration/quibble] - 10https://gerrit.wikimedia.org/r/771637 (owner: 10Hashar) [17:08:08] (03Merged) 10jenkins-bot: release: Quibble 1.4.4 [integration/quibble] - 10https://gerrit.wikimedia.org/r/771636 (https://phabricator.wikimedia.org/T300340) (owner: 10Hashar) [17:09:04] (03Merged) 10jenkins-bot: release: Start 1.4.5 cycle [integration/quibble] - 10https://gerrit.wikimedia.org/r/771637 (owner: 10Hashar) [19:20:28] hashar: maybe we should disable the parallel npm install flag until we can make the output easier to read [19:20:45] yup [19:20:51] looks like I forgot about that task [19:21:03] Also I’m not sure it had a big performance benefit; we could revisit when we parallelize the selenium tests more generally [19:21:20] it also apparently randomly deadlock https://phabricator.wikimedia.org/T303270 [19:21:33] Right [19:21:42] Ok, well it was worth a try :) [19:22:11] I am build the quibble 1.4.4 images [20:26:58] kostajh: I will do the Jenkins job update tomorrow