[00:00:15] (03Merged) 10jenkins-bot: layout: Switch from mediawiki-phpunit-coverage-patch-docker to mwcore-phpunit-coverage-patch [integration/config] - 10https://gerrit.wikimedia.org/r/518125 (owner: 10Jforrester) [00:13:14] (03CR) 10Jforrester: [C: 03+2] jjb: Rename mediawiki-{core-code|phpunit}-coverage-(patch-)?docker for consistency, pt. 2 [integration/config] - 10https://gerrit.wikimedia.org/r/518126 (owner: 10Jforrester) [00:13:19] (03CR) 10Jforrester: [C: 03+2] "Deployed." [integration/config] - 10https://gerrit.wikimedia.org/r/518126 (owner: 10Jforrester) [00:15:31] (03Merged) 10jenkins-bot: jjb: Rename mediawiki-{core-code|phpunit}-coverage-(patch-)?docker for consistency, pt. 2 [integration/config] - 10https://gerrit.wikimedia.org/r/518126 (owner: 10Jforrester) [00:18:48] 10Continuous-Integration-Infrastructure, 10serviceops: Provide php72 on contint1001 rather than php56 - https://phabricator.wikimedia.org/T226224 (10Jdforrester-WMF) [00:19:15] 10Continuous-Integration-Infrastructure, 10serviceops: Provide php72 on contint1001 rather than php56 - https://phabricator.wikimedia.org/T226224 (10Jdforrester-WMF) [00:19:20] (03CR) 10Jforrester: [C: 04-1] "Blocked by T226224." [integration/config] - 10https://gerrit.wikimedia.org/r/516570 (owner: 10Jforrester) [00:19:22] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team, 10Patch-For-Review: Drop php56 testing support - https://phabricator.wikimedia.org/T224906 (10Jdforrester-WMF) [00:42:25] (03PS2) 10Jforrester: layout: Replace negative branch matches of quibble jobs with positive ones [integration/config] - 10https://gerrit.wikimedia.org/r/518088 [00:42:27] (03PS1) 10Jforrester: tests: Explicitly test REL1_3[0-4] [integration/config] - 10https://gerrit.wikimedia.org/r/518179 [00:43:59] (03CR) 10jerkins-bot: [V: 04-1] layout: Replace negative branch matches of quibble jobs with positive ones [integration/config] - 10https://gerrit.wikimedia.org/r/518088 (owner: 10Jforrester) [00:44:01] (03CR) 10jerkins-bot: [V: 04-1] tests: Explicitly test REL1_3[0-4] [integration/config] - 10https://gerrit.wikimedia.org/r/518179 (owner: 10Jforrester) [00:47:01] (03PS2) 10Jforrester: tests: Explicitly test REL1_3[0-4] [integration/config] - 10https://gerrit.wikimedia.org/r/518179 [00:48:56] (03PS3) 10Jforrester: layout: Replace negative branch matches of quibble jobs with positive ones [integration/config] - 10https://gerrit.wikimedia.org/r/518088 [00:50:31] (03CR) 10jerkins-bot: [V: 04-1] layout: Replace negative branch matches of quibble jobs with positive ones [integration/config] - 10https://gerrit.wikimedia.org/r/518088 (owner: 10Jforrester) [01:08:44] (03PS4) 10Jforrester: layout: Replace negative branch matches of quibble jobs with positive ones [integration/config] - 10https://gerrit.wikimedia.org/r/518088 [01:10:24] (03CR) 10jerkins-bot: [V: 04-1] layout: Replace negative branch matches of quibble jobs with positive ones [integration/config] - 10https://gerrit.wikimedia.org/r/518088 (owner: 10Jforrester) [01:30:33] 10Deployments, 10Release-Engineering-Team, 10HHVM, 10MW-1.34-notes (1.34.0-wmf.3; 2019-04-30), 10Upstream: hhvm aborts on extensions/CirrusSearch/tests/phan/stubs/intl.php blocking deployment sync - https://phabricator.wikimedia.org/T221814 (10MaxSem) 05Open→03Resolved a:03MaxSem Since the immediat... [01:33:17] PROBLEM - Free space - all mounts on deployment-fluorine02 is CRITICAL: CRITICAL: deployment-prep.deployment-fluorine02.diskspace._srv.byte_percentfree (<33.33%) [02:06:30] (03PS1) 10Kosta Harlan: Provide option to skip installation of MediaWiki [integration/quibble] - 10https://gerrit.wikimedia.org/r/518180 [05:46:23] (03PS1) 10Legoktm: Add CI config for mediawiki/libs/{IPUtils,Services} [integration/config] - 10https://gerrit.wikimedia.org/r/518195 [05:49:01] (03PS1) 10Legoktm: Get rid of special coverage job for RemexHtml [integration/config] - 10https://gerrit.wikimedia.org/r/518196 [05:49:13] (03CR) 10Legoktm: [C: 03+2] Add CI config for mediawiki/libs/{IPUtils,Services} [integration/config] - 10https://gerrit.wikimedia.org/r/518195 (owner: 10Legoktm) [05:49:18] (03CR) 10Legoktm: [C: 03+2] Get rid of special coverage job for RemexHtml [integration/config] - 10https://gerrit.wikimedia.org/r/518196 (owner: 10Legoktm) [05:50:45] (03Merged) 10jenkins-bot: Add CI config for mediawiki/libs/{IPUtils,Services} [integration/config] - 10https://gerrit.wikimedia.org/r/518195 (owner: 10Legoktm) [05:51:33] (03Merged) 10jenkins-bot: Get rid of special coverage job for RemexHtml [integration/config] - 10https://gerrit.wikimedia.org/r/518196 (owner: 10Legoktm) [05:51:59] !log deployed https://gerrit.wikimedia.org/r/518195 https://gerrit.wikimedia.org/r/518196 [05:52:01] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [06:04:14] !log deleted remexhtml-phpunit-coverage-docker-publish jenkins job [06:04:16] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [06:30:43] (03PS1) 10Legoktm: Publish doxygen for RunningStat library [integration/config] - 10https://gerrit.wikimedia.org/r/518204 (https://phabricator.wikimedia.org/T185724) [06:30:52] (03CR) 10Legoktm: [C: 03+2] Publish doxygen for RunningStat library [integration/config] - 10https://gerrit.wikimedia.org/r/518204 (https://phabricator.wikimedia.org/T185724) (owner: 10Legoktm) [06:32:24] (03Merged) 10jenkins-bot: Publish doxygen for RunningStat library [integration/config] - 10https://gerrit.wikimedia.org/r/518204 (https://phabricator.wikimedia.org/T185724) (owner: 10Legoktm) [06:34:06] (03PS1) 10Legoktm: Publish coverage for base-convert [integration/config] - 10https://gerrit.wikimedia.org/r/518205 [06:34:08] (03PS1) 10Legoktm: Publish doxygen for Assert library [integration/config] - 10https://gerrit.wikimedia.org/r/518206 [06:34:25] !log deployed https://gerrit.wikimedia.org/r/518204 [06:34:26] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [06:35:36] (03CR) 10Legoktm: [C: 03+2] Publish coverage for base-convert [integration/config] - 10https://gerrit.wikimedia.org/r/518205 (owner: 10Legoktm) [06:35:41] (03CR) 10Legoktm: [C: 03+2] Publish doxygen for Assert library [integration/config] - 10https://gerrit.wikimedia.org/r/518206 (owner: 10Legoktm) [06:37:08] (03Merged) 10jenkins-bot: Publish coverage for base-convert [integration/config] - 10https://gerrit.wikimedia.org/r/518205 (owner: 10Legoktm) [06:37:14] (03Merged) 10jenkins-bot: Publish doxygen for Assert library [integration/config] - 10https://gerrit.wikimedia.org/r/518206 (owner: 10Legoktm) [06:37:20] (03PS1) 10Legoktm: doc: Add Assert and RunningStat PHP libraries [integration/docroot] - 10https://gerrit.wikimedia.org/r/518207 (https://phabricator.wikimedia.org/T185724) [06:37:44] !log deployed https://gerrit.wikimedia.org/r/518205 https://gerrit.wikimedia.org/r/518206 [06:37:45] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [06:39:30] (03CR) 10Legoktm: [C: 03+2] doc: Add Assert and RunningStat PHP libraries [integration/docroot] - 10https://gerrit.wikimedia.org/r/518207 (https://phabricator.wikimedia.org/T185724) (owner: 10Legoktm) [06:39:59] (03Merged) 10jenkins-bot: doc: Add Assert and RunningStat PHP libraries [integration/docroot] - 10https://gerrit.wikimedia.org/r/518207 (https://phabricator.wikimedia.org/T185724) (owner: 10Legoktm) [06:40:07] (03CR) 10jenkins-bot: doc: Add Assert and RunningStat PHP libraries [integration/docroot] - 10https://gerrit.wikimedia.org/r/518207 (https://phabricator.wikimedia.org/T185724) (owner: 10Legoktm) [06:41:19] 10Continuous-Integration-Config, 10Librarization, 10Performance-Team, 10RunningStat, and 2 others: Publish Doxygen for RunningStat library - https://phabricator.wikimedia.org/T185724 (10Legoktm) 05Open→03Resolved a:03Legoktm [06:44:02] 10Continuous-Integration-Infrastructure, 10serviceops: Provide php72 on contint1001 rather than php56 - https://phabricator.wikimedia.org/T226224 (10hashar) `contint1001.wikimedia.org` is running Jessie and we haven't backported PHP 7.2 to it. php is probably only used for the small website https://integration... [06:58:04] 10Continuous-Integration-Infrastructure, 10serviceops: Provide php72 on contint1001 rather than php56 - https://phabricator.wikimedia.org/T226224 (10Joe) Yes, the burden of maintaing a backport just for contint is not justifiable IMHO. I'm declining the ticket, but if this is blocking something important, plea... [06:58:18] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team, 10Patch-For-Review: Drop php56 testing support - https://phabricator.wikimedia.org/T224906 (10Joe) [06:58:21] 10Continuous-Integration-Infrastructure, 10serviceops: Provide php72 on contint1001 rather than php56 - https://phabricator.wikimedia.org/T226224 (10Joe) 05Open→03Declined [07:08:19] RECOVERY - Free space - all mounts on deployment-fluorine02 is OK: OK: All targets OK [07:23:16] (03PS1) 10Reedy: Add more libs [integration/docroot] - 10https://gerrit.wikimedia.org/r/518211 [07:24:08] (03Abandoned) 10Reedy: Add more libs [integration/docroot] - 10https://gerrit.wikimedia.org/r/518211 (owner: 10Reedy) [07:34:07] 10Continuous-Integration-Config, 10Parsoid-PHP: Parsoid/PHP post-merge doc publishing task needs updating to use PHP 7.2 - https://phabricator.wikimedia.org/T219319 (10hashar) PHPUnit failed somehow: ` 00:02:09.049 Generating code coverage report in HTML format ... done 00:02:13.226 Build step 'Execute shell'... [07:40:44] 10Continuous-Integration-Config, 10Parsoid-PHP: Parsoid/PHP post-merge doc publishing task needs updating to use PHP 7.2 - https://phabricator.wikimedia.org/T219319 (10hashar) Fun things: https://github.com/sebastianbergmann/phpunit/issues/1760 "Do not error out on invalid @covers tag" which apparently is i... [07:56:11] PROBLEM - Host integration-slave-docker-1034 is DOWN: CRITICAL - Host Unreachable (172.16.7.168) [07:56:27] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (Kanban): Rebuild integration-slave-docker-* instances to use less RAM, new name and Stretch - https://phabricator.wikimedia.org/T226233 (10hashar) [08:14:25] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (Kanban): Rebuild integration-slave-docker-* instances to use less RAM, new name and Stretch - https://phabricator.wikimedia.org/T226233 (10hashar) [08:16:48] 10Continuous-Integration-Infrastructure, 10Wikimedia-Incident: Add two more zuul-merger process - https://phabricator.wikimedia.org/T222645 (10hashar) p:05Triage→03Low [08:25:14] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team: mediawiki-config (and others?) should ride gate-and-submit-swat not gate-and-submit - https://phabricator.wikimedia.org/T225252 (10hashar) TLDR: we do not need to move `operations/mediawiki-config` to `gate-and-submit-swat`. The repository c... [08:29:32] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (Kanban): Rebuild integration-slave-docker-* instances to use less RAM, new name and Stretch - https://phabricator.wikimedia.org/T226233 (10hashar) `counterexample E: Version '18.06.2~ce~3-0~debian' for 'docker-ce' was not f... [08:32:34] 10Release-Engineering-Team-TODO, 10Scap, 10serviceops, 10User-jijiki: Deploy scap 3.10.0-1 - https://phabricator.wikimedia.org/T224915 (10jijiki) p:05Triage→03Normal [08:34:28] 10Gerrit, 10Release-Engineering-Team-TODO, 10Operations, 10serviceops: Gerrit Hardware Upgrade - https://phabricator.wikimedia.org/T222391 (10jijiki) p:05Triage→03Low [08:36:09] 10Phabricator-Sprint-Extension: Call to undefined method SprintProjectProfilePanelEngine::buildNavigation() when accessing Burndown since 2019.16 - https://phabricator.wikimedia.org/T222586 (10Shed4) @Aklapper Also suffering from the problem here. Is there any workaround or maybe some patch to test? [08:37:42] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Operations: Upload docker-ce 18.06.3 upstream package for Stretch - https://phabricator.wikimedia.org/T226236 (10hashar) [08:42:00] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Operations: Upload docker-ce 18.06.3 upstream package for Stretch - https://phabricator.wikimedia.org/T226236 (10hashar) And maybe we could use some `reprepro` configuration to ease further upgrades? [08:48:01] 10Scap, 10Operations, 10serviceops-radar, 10User-jijiki: Introduce state to Scap - https://phabricator.wikimedia.org/T209881 (10Joe) [08:49:18] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team (Kanban), 10Patch-For-Review: Rebuild integration-slave-docker-* instances to use less RAM, new name and Stretch - https://phabricator.wikimedia.org/T226233 (10hashar) I have also removed the `zuul` package from the Docker... [08:58:01] 10Phabricator-Sprint-Extension: Call to undefined method SprintProjectProfilePanelEngine::buildNavigation() when accessing Burndown since 2019.16 - https://phabricator.wikimedia.org/T222586 (10Aklapper) @Shed4: The workaround would be someone to take a look at the previous comment and fix the code. As https://ph... [09:00:12] 10Gerrit, 10Release-Engineering-Team, 10serviceops-radar: Gerrit http threads stuck behind sendemail thread - https://phabricator.wikimedia.org/T224448 (10akosiaris) [09:00:55] PROBLEM - Puppet errors on integration-slave-jessie-1002 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [3.0] [09:08:09] 10Release-Engineering-Team, 10serviceops, 10Patch-For-Review: Rebuild integration/config images based on jessie - https://phabricator.wikimedia.org/T219748 (10Joe) 05Open→03Stalled p:05Triage→03Normal The situation is, release engineering doesn't feel confident doing general upgrades like this, and a... [09:09:22] 10Release-Engineering-Team, 10Scap, 10serviceops: Define a mediawiki "version" - https://phabricator.wikimedia.org/T218412 (10Joe) [09:18:43] 10Gerrit, 10Release-Engineering-Team: Create mirror of Gerrit repositories for consumption by various tools - https://phabricator.wikimedia.org/T226240 (10hashar) [09:19:21] 10Scap, 10serviceops: Scap2 to use etcd for target servers - https://phabricator.wikimedia.org/T218328 (10jijiki) a:05jijiki→03None [09:22:14] PROBLEM - Puppet errors on integration-slave-jessie-1004 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [3.0] [09:23:53] PROBLEM - Puppet errors on integration-slave-jessie-1001 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [3.0] [09:26:34] 10Gerrit, 10Release-Engineering-Team: Create mirror of Gerrit repositories for consumption by various tools - https://phabricator.wikimedia.org/T226240 (10hashar) Candidates to be switched to a new Git mirror would be: * codesearch https://codesearch.wmflabs.org/ #vps-project-codesearch . Quoting Legoktm //A... [09:27:04] 10Gerrit, 10Release-Engineering-Team, 10VPS-project-codesearch, 10Patch-For-Review: Gerrit thread use GC thrashing - https://phabricator.wikimedia.org/T221026 (10hashar) I have missed the follow up about creating a mirror for our git repositories. I have filled it as T226240 [09:28:37] 10Gerrit, 10Release-Engineering-Team: Create mirror of Gerrit repositories for consumption by various tools - https://phabricator.wikimedia.org/T226240 (10Lucas_Werkmeister_WMDE) > The CI slaves do hammer Gerrit :-/ Couldn’t they use a mirror as well? If they fetch a specific ref (`refs/changes/ef/abcdef/xy`)... [09:59:52] 10Release-Engineering-Team-TODO, 10serviceops: Our docker base images lack tags - https://phabricator.wikimedia.org/T218342 (10Joe) p:05Triage→03Normal [10:07:30] 10Continuous-Integration-Config, 10Release-Engineering-Team, 10Wikimedia-Site-requests, 10serviceops-radar: Consider creating a puppet-compiler equivalent for mediawiki-config.git - https://phabricator.wikimedia.org/T220775 (10Joe) [10:12:21] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10User-Michael, 10phan: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10Michael) [10:12:30] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10User-Michael, 10phan: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10Michael) p:05Triage→03High [10:13:29] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10User-Michael, 10phan: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10Michael) [10:14:27] 10Release-Engineering-Team, 10Librarization, 10serviceops-radar: Travis tests for mediawiki-libs-etcd broken - https://phabricator.wikimedia.org/T215864 (10Joe) [10:23:15] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10User-Michael, 10phan: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10Daimona) AIUI, only the last mentioned patch is culprit. The dependency on Wikibase should only be used to tell what MW extensio... [10:43:17] 10Gerrit, 10Release-Engineering-Team: Create mirror of Gerrit repositories for consumption by various tools - https://phabricator.wikimedia.org/T226240 (10hashar) The delay would be just a few seconds for sure, the devil is that the CI jobs fetch branches (`refs/heads/*`) which thus change constantly. At the t... [10:46:40] 10Phabricator: Set up Phabricator to accept OAuth requests from discuss-space.wmflabs.org - https://phabricator.wikimedia.org/T226246 (10Qgil) [10:47:22] 10Phabricator: Set up Phabricator to accept OAuth requests from discuss-space.wmflabs.org - https://phabricator.wikimedia.org/T226246 (10Qgil) [10:52:20] Hi, it has been a while that I don t ask anything about Phabricator on IRC, and now I am not sure whether this is the right channel... However, I see andre__ and twentyafterfour are here anyway, so let me ask. [10:53:12] If you could have a look at https://phabricator.wikimedia.org/T226246, I would be very grateful. :) [10:53:44] It shouldn't be complicated for you (?) and it would remove one big blocker for Community Relations. [10:54:01] Please. [10:54:21] 10Phabricator, 10Release-Engineering-Team: Set up Phabricator to accept OAuth requests from discuss-space.wmflabs.org - https://phabricator.wikimedia.org/T226246 (10Reedy) [10:54:44] 10Phabricator, 10Release-Engineering-Team: Set up Phabricator to accept OAuth requests from discuss-space.wmflabs.org - https://phabricator.wikimedia.org/T226246 (10Reedy) p:05Triage→03High [10:55:48] I guess I do not understand the urgency [10:57:42] We want to present a proposal next Tuesday. [10:58:05] And we want to enable Phabricator login and test it. [10:58:17] To avoid local accounts from the start. [10:59:26] 10Phabricator, 10Release-Engineering-Team: Set up Phabricator to accept OAuth requests from discuss-space.wmflabs.org - https://phabricator.wikimedia.org/T226246 (10Aklapper) @Qgil: Please try accessing https://phabricator.wikimedia.org/oauthserver/client/view/5/ for the token / secret. [10:59:35] qgil: ^ here you go [11:01:06] Thank you! Checking and testing.... [11:06:19] 10Phabricator (Upstream), 10Upstream: Cannot close "Custom Policy" dialog window when viewing the policy of an OAuth application - https://phabricator.wikimedia.org/T226249 (10Mainframe98) [11:07:45] 10Phabricator, 10Release-Engineering-Team: Set up Phabricator to accept OAuth requests from discuss-space.wmflabs.org - https://phabricator.wikimedia.org/T226246 (10Qgil) Thank you @Aklapper ! Hm. I think everything is correct in Discourse's side (I am also comparing with discourse-mediawiki from the admin in... [11:14:19] 10Phabricator, 10Release-Engineering-Team: Set up Phabricator to accept OAuth requests from discuss-space.wmflabs.org - https://phabricator.wikimedia.org/T226246 (10Qgil) 05Open→03Resolved a:03Qgil FIXED! (there was an extra space at the end of the URL, apparently........) Thank you so much. [11:15:42] ... and Resolved! Thank you so much. [11:16:07] \o/ [11:16:29] Fun fact: The URL didn't work initially because an extra space at the end, that Phabricator apparently took literally as part of the URL instead of ignoring it. Picky... [11:39:58] hashar: FYI there are a few "easy" patches lined up for quibble, ahead of the browser tests patch (which I've adjusted as we discussed) [11:45:38] awight: hello, ah yeah [11:45:48] I think one of them removed some unlreated test [11:46:02] That should be cleaned up in the latest PS [11:46:11] Good catch! [11:51:26] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10noarave) @Daimona unfortunately requiring phan config 0.6.1 in composer.json did not seem to have an effect. [11:51:30] I should grab a coffee and merge all those pending patches yeah [11:51:31] :\ [11:51:52] no rush! I'm just writing to let you know I responded to the CR [11:52:08] cool ;) [11:58:07] (03PS1) 10Hashar: docker: port zuul-cloner to Stretch [integration/config] - 10https://gerrit.wikimedia.org/r/518241 (https://phabricator.wikimedia.org/T224908) [11:58:37] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team, 10Patch-For-Review: Drop jessie testing support - https://phabricator.wikimedia.org/T224908 (10hashar) [12:03:09] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10hashar) That did cause the container to trigger the installation of phan 1.3.4 instead of the outdated 0.8.0 fallback.... [12:13:05] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10Daimona) @noarave It actually did. It's just that phan completed its analysis and found several errors. Of note, you sho... [12:14:21] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10alaa_wmde) Another possible source? WikibaseLexeme `.phan/config.php` looks different from Wikibase one. For example, o... [12:20:16] anyone else seeing failed CI builds with Composer\Downloader\TransportException ? [12:20:40] started relatively recently I think, not more than an hour ago I’d say [12:25:12] seeing this even in core builds – I’ll create a #shared-build-failure task [12:28:13] https://phabricator.wikimedia.org/T226253 [12:30:30] 10Gerrit, 10Release-Engineering-Team: Create mirror of Gerrit repositories for consumption by various tools - https://phabricator.wikimedia.org/T226240 (10mmodell) @hashar: we already have a read-only replica of most repositories on phabricator. Is that not satisfactory? It's kept up to date, with a lag of ju... [12:37:58] 10Phabricator, 10Release-Engineering-Team (Kanban), 10Operations, 10Traffic: Set up a subdomain for Phame to enable caching - https://phabricator.wikimedia.org/T226044 (10mmodell) This seems like a good idea. The upstream documentation has a warning that there are some issues with an external blog / dedic... [12:41:56] 10Scap: Track scap syncs that are part of a given SWAT window - https://phabricator.wikimedia.org/T193311 (10mmodell) [12:41:58] 10Release-Engineering-Team (Kanban), 10Scap, 10User-MModell: Document scap swat command - https://phabricator.wikimedia.org/T196411 (10mmodell) 05Stalled→03Declined I no longer believe that `scap swat` is the right solution to the problem that it originally attempted to address. In the near future we wi... [12:47:03] 10Diffusion, 10Release-Engineering-Team (Kanban), 10Operations, 10Packaging, and 2 others: Cannot connect to vcs@git-ssh.wikimedia.org (since move from phab1001 to phab1003) - https://phabricator.wikimedia.org/T224677 (10mmodell) @LucasWerkmeister that could work, though if the fix is as simple as it appea... [12:48:44] Lucas_WMDE: I guess packagist.org has some issue :- [12:48:47] :-\ [12:49:03] ^ I was just typing the exact same thing, so +1 [12:49:35] we should really set up our own mirrors [12:52:32] hashar: ok :/ [12:52:32] Lucas_WMDE: hashar: I tweeted them: https://twitter.com/20after4/status/1142052311923142657 [12:52:32] heh maybe someone over at packagist will see that [12:52:32] thanks :) [12:56:51] (03CR) 10Hashar: [C: 03+2] Clean up tests [integration/quibble] - 10https://gerrit.wikimedia.org/r/516980 (owner: 10Awight) [12:57:32] (03Merged) 10jenkins-bot: Clean up tests [integration/quibble] - 10https://gerrit.wikimedia.org/r/516980 (owner: 10Awight) [12:58:01] (03CR) 10jenkins-bot: Clean up tests [integration/quibble] - 10https://gerrit.wikimedia.org/r/516980 (owner: 10Awight) [12:58:39] 10Diffusion, 10Release-Engineering-Team (Kanban), 10Operations, 10Packaging, and 2 others: Cannot connect to vcs@git-ssh.wikimedia.org (since move from phab1001 to phab1003) - https://phabricator.wikimedia.org/T224677 (10ArielGlenn) If we build it for reals, I'd ask @MoritzMuehlenhoff about all that. If we... [13:00:21] Interesting, it looks like tox-docker defaults to never running tests tagged with `integration`. [13:15:31] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10alaa_wmde) >>! In T226244#5273849, @gerritbot wrote: > Change 518256 had a related patch set uploaded (by Daimona Eaytoy... [13:18:30] PROBLEM - Long lived cherry-picks on puppetmaster on deployment-puppetmaster03 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [13:23:31] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10Daimona) >>! In T226244#5273853, @alaa_wmde wrote: >>>! In T226244#5273849, @gerritbot wrote: >> Change 518256 had a rel... [13:30:09] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10alaa_wmde) @Daimona understood. I also think the clean-up work is necessary. I'm not such a fan of statements of the sor... [13:34:38] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10Daimona) >>! In T226244#5273866, @alaa_wmde wrote: > @Daimona understood. I also think the clean-up work is necessary. I... [13:36:56] (03CR) 10Hashar: [C: 03+2] "One less crazy test :-] The removal of the test could have been made in a different change but given this is the last of the serie... tha" [integration/quibble] - 10https://gerrit.wikimedia.org/r/513518 (https://phabricator.wikimedia.org/T223752) (owner: 10Awight) [13:37:14] awight: one more approved :] [13:37:44] (03Merged) 10jenkins-bot: Browser tests as command [integration/quibble] - 10https://gerrit.wikimedia.org/r/513518 (https://phabricator.wikimedia.org/T223752) (owner: 10Awight) [13:38:13] (03CR) 10jenkins-bot: Browser tests as command [integration/quibble] - 10https://gerrit.wikimedia.org/r/513518 (https://phabricator.wikimedia.org/T223752) (owner: 10Awight) [13:42:49] Woohoo, thanks for the push! [13:43:36] DEBUG:quibble.cmd:PHPUnit None suite (without database) [13:43:36] :D [13:43:49] harrhar [13:44:15] Is that with change 511749? [13:44:39] I can debug... [13:44:43] yeah with dry run but that comes from other patches [13:44:54] that is the __str__ or __repr__ from some commands which are hmm clumsy :D [13:45:04] DEBUG:quibble.cmd:Browser tests in /home/hashar/workspace/src: ['qunit', 'selenium (maybe)'] using DISPLAY=None [13:45:08] display=None :D [13:45:13] it is not important for this change [13:45:14] yeah I shouldn't have used a magic method, IMO [13:45:24] PROBLEM - Disk space on contint1001 is CRITICAL: DISK CRITICAL - free space: / 2514 MB (5% inode=45%): https://wikitech.wikimedia.org/wiki/Monitoring/Disk_space [13:45:42] bah [13:46:06] I'll write a fixup patch separate from the rest of the pile... [13:46:56] RECOVERY - Disk space on contint1001 is OK: DISK OK https://wikitech.wikimedia.org/wiki/Monitoring/Disk_space [13:50:27] (03CR) 10Hashar: [C: 03+2] Separate planning and execution phases (031 comment) [integration/quibble] - 10https://gerrit.wikimedia.org/r/511749 (https://phabricator.wikimedia.org/T223752) (owner: 10Awight) [13:50:54] (03CR) 10Hashar: "recheck" [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [13:51:05] (03CR) 10jerkins-bot: [V: 04-1] Move parallel_run to util.py [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [13:51:07] (03Merged) 10jenkins-bot: Separate planning and execution phases [integration/quibble] - 10https://gerrit.wikimedia.org/r/511749 (https://phabricator.wikimedia.org/T223752) (owner: 10Awight) [13:51:36] (03CR) 10jenkins-bot: Separate planning and execution phases [integration/quibble] - 10https://gerrit.wikimedia.org/r/511749 (https://phabricator.wikimedia.org/T223752) (owner: 10Awight) [13:52:55] awight: and that https://gerrit.wikimedia.org/r/#/c/integration/quibble/+/515090/ needs a rebase :-\ [13:54:23] (03PS6) 10Hashar: Rename native composer command [integration/quibble] - 10https://gerrit.wikimedia.org/r/515009 (owner: 10Awight) [13:54:55] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10Patch-For-Review, and 2 others: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10alaa_wmde) > Ah, I haven't been clear. Extensions *should* override the default config, e.g. to define additional depend... [13:55:14] (03PS2) 10Kosta Harlan: Provide option to skip installation of MediaWiki [integration/quibble] - 10https://gerrit.wikimedia.org/r/518180 [13:55:28] (03CR) 10Hashar: [C: 03+2] Rename native composer command [integration/quibble] - 10https://gerrit.wikimedia.org/r/515009 (owner: 10Awight) [13:55:30] (03PS1) 10Awight: Clean up string rendering of commands [integration/quibble] - 10https://gerrit.wikimedia.org/r/518269 [13:56:07] (03Merged) 10jenkins-bot: Rename native composer command [integration/quibble] - 10https://gerrit.wikimedia.org/r/515009 (owner: 10Awight) [13:56:09] (03CR) 10jerkins-bot: [V: 04-1] Clean up string rendering of commands [integration/quibble] - 10https://gerrit.wikimedia.org/r/518269 (owner: 10Awight) [13:56:37] (03CR) 10jenkins-bot: Rename native composer command [integration/quibble] - 10https://gerrit.wikimedia.org/r/515009 (owner: 10Awight) [13:56:52] (03PS4) 10Awight: Move parallel_run to util.py [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 [13:57:00] (03CR) 10Awight: "PS4: manual rebase" [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [13:58:26] (03PS2) 10Awight: Clean up string rendering of commands [integration/quibble] - 10https://gerrit.wikimedia.org/r/518269 [13:58:48] (03PS1) 10Kosta Harlan: Codehealth: Always append code coverage properties [integration/config] - 10https://gerrit.wikimedia.org/r/518271 [13:59:48] (03PS1) 10Kosta Harlan: Codehealth: Bump jobs to 0.4.10 [integration/config] - 10https://gerrit.wikimedia.org/r/518272 [14:01:06] awight: sorry I have cherry picked one of the changes out of the chain :-\ [14:01:19] No worries! [14:02:35] The only stressful part of this, from my perspective, is that we're messing with a tool that our colleagues depend on 1000x/day. I don't know how you've managed it all these years... [14:02:57] File "/home/hashar/projects/integration/quibble/quibble/commands.py", line 44, in __str__ [14:02:57] self.kwargs) [14:02:57] AttributeError: 'ZuulCloneCommand' object has no attribute 'kwargs' [14:02:58] :D [14:03:00] with quibble -n [14:03:01] :) [14:03:04] ooh [14:03:12] oh [14:03:13] well [14:03:14] hmm [14:03:18] I don't know either [14:03:28] then that Quibble is fairly recent [14:03:49] and it is really not that complicated. Well now with all you refactoring it is a bit more complicated and simpler [14:03:51] but in the end [14:03:58] quibble is just the icing on a seven-tier cake :-) [14:04:06] it is really just a lot of magic boilerplate for something that could be a oneliner [14:04:19] (git clone mediawiki/core && composer install && install.php && phpunit && whatever) [14:04:26] yeah [14:04:45] I had two goals when I created it: [14:05:01] 1) make it possible to reproduce a CI build locally [14:05:17] 2) offload logic from the jenkins jobs / shell scripts we had [14:05:30] so yeah that more or less work [14:05:49] but I am absolutely terrible at communicating or marketing a project. So not much happened :] [14:06:25] Probably just because CI is a scary rabbit hole and devs are unsure if they will emerge uneaten. [14:06:48] hashar: I assume the error at 10:04:28 is an problem with packagist.org and not anything on our end, but wanted to check if you've seen it before: https://integration.wikimedia.org/ci/job/quibble-vendor-mysql-hhvm-docker/55002/consoleFull [14:06:52] (03CR) 10Hashar: "Not from this patch but related:" [integration/quibble] - 10https://gerrit.wikimedia.org/r/518269 (owner: 10Awight) [14:07:29] kostajh: most probably :-( https://twitter.com/20after4/status/1142052311923142657 and https://phabricator.wikimedia.org/T226253 [14:08:12] hashar: thx :( [14:08:42] awight: I dont bite :] [14:09:23] no but subtle errors introduced by people unfamiliar with CI's idiosyncracies (/me points at self) do bite [14:09:53] that is how you learn to bake bread anyway [14:09:57] gotta process with mistakes [14:10:03] and I do a ton of them as well :\ [14:11:08] Am I just being lazy about this? [14:11:10] > DEBUG:quibble.cmd:Zuul clone with parameters {"zuul_project": null, "branch": null, "projects": ["mediawiki/core", "mediawiki/skins/Vector", "mediawiki/vendor"], "zuul_branch": null, "cache_dir": "ref", "workspace": "/home/adamw/wikimedia/ci-workspace/src", "zuul_url": null, "project_branch": [], "zuul_ref": null, "zuul_newrev": null, "workers": 4} [14:12:36] hmm [14:13:21] I say, it might be good enough for now [14:13:29] then [14:13:43] yeah I don't know really [14:13:54] at least if there are no errors. That is an improvement! [14:16:53] (03PS3) 10Awight: Clean up string rendering of commands [integration/quibble] - 10https://gerrit.wikimedia.org/r/518269 [14:17:18] I pruned the empties, at least: DEBUG:quibble.cmd:Zuul clone with parameters {"cache_dir": "ref", "workers": 4, "workspace": "/home/adamw/wikimedia/ci-workspace/src", "projects": ["mediawiki/core", "mediawiki/skins/Vector", "mediawiki/vendor"]} [14:17:40] (03CR) 10Kosta Harlan: (wip) Add option for using Apache as server (031 comment) [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) (owner: 10Kosta Harlan) [14:18:56] (03PS3) 10Awight: Inline remaining cmd.py steps [integration/quibble] - 10https://gerrit.wikimedia.org/r/515182 [14:23:31] (03CR) 10Hashar: "That is a crazy option to have since most commands do require a MediaWiki installation? But I guess it could be useful for some corner cas" [integration/quibble] - 10https://gerrit.wikimedia.org/r/518180 (owner: 10Kosta Harlan) [14:23:52] kostajh: what was your use case to skip mediawiki installation in Quibble? (re: https://gerrit.wikimedia.org/r/#/c/integration/quibble/+/518180/ ) [14:24:52] hashar: the codehealth pipeline for MW core is ~10 minutes. I think we could eliminate 4 minutes since we don't need npm/composer dependencies or a working mediawiki install in order to statically analyze the code [14:25:24] kostajh: then you probably don't need Quibble either? [14:26:02] hashar: eventually I would like to generate the coverage report for JS and PHP, and there is also an argument to keep the setup for core & extensions as similar as possible [14:26:19] (03PS7) 10Awight: Run node browser tests in each repo [integration/quibble] - 10https://gerrit.wikimedia.org/r/510709 (https://phabricator.wikimedia.org/T199116) [14:26:19] but you're right, probably, we don't need quibble. It just simplifies things to inherit from the quibble template, use the same image, etc [14:27:09] 10Release-Engineering-Team (Kanban), 10Security-Team, 10phan-taint-check-plugin: Phan-taint-check-plugin not available for PHP > 7.0 - https://phabricator.wikimedia.org/T207344 (10sbassett) p:05Triage→03Normal @Daimona @Jdforrester-WMF - apologies for the disappearing act from the #Security-Team on this.... [14:27:44] hashar: plus, I don't know, suppose you want to to run `quibble --skip-install --skip-deps`, do some intermediary step after all the code is checked out, then run `quibble --skip-zuul`? [14:28:08] (03CR) 10Hashar: [C: 03+2] "\o/" (031 comment) [integration/quibble] - 10https://gerrit.wikimedia.org/r/518269 (owner: 10Awight) [14:28:47] (03Merged) 10jenkins-bot: Clean up string rendering of commands [integration/quibble] - 10https://gerrit.wikimedia.org/r/518269 (owner: 10Awight) [14:28:55] kostajh: yeah that is becoming too meta for me :] Anyway the code is ifne and there might be some use case for it so yeah +1 [14:28:56] and [14:29:13] I guess one would thus be able to use its own little LocalSettings.php this way [14:29:28] yeah that too [14:29:35] btw I'm going to try to rework the apache patch today [14:29:47] (03CR) 10Hashar: [C: 03+2] "Another use case I guess is that if one has a LocalSettings.php , there is no need to install MediaWiki :-]" [integration/quibble] - 10https://gerrit.wikimedia.org/r/518180 (owner: 10Kosta Harlan) [14:30:13] kostajh: I must say I haven't looked at it :-\ [14:30:35] (03Merged) 10jenkins-bot: Provide option to skip installation of MediaWiki [integration/quibble] - 10https://gerrit.wikimedia.org/r/518180 (owner: 10Kosta Harlan) [14:30:52] About localsettings... I've been daydreaming about having a convention for tests to load configuration snippets into a localsettings.d directory, which are then cleared out between test suites. [14:31:07] (03CR) 10jenkins-bot: Clean up string rendering of commands [integration/quibble] - 10https://gerrit.wikimedia.org/r/518269 (owner: 10Awight) [14:31:30] (03CR) 10jenkins-bot: Provide option to skip installation of MediaWiki [integration/quibble] - 10https://gerrit.wikimedia.org/r/518180 (owner: 10Kosta Harlan) [14:35:26] (03PS4) 10Hashar: Inline remaining cmd.py steps [integration/quibble] - 10https://gerrit.wikimedia.org/r/515182 (owner: 10Awight) [14:36:07] (03CR) 10Hashar: [C: 03+2] "Fixed a trivial conflict" [integration/quibble] - 10https://gerrit.wikimedia.org/r/515182 (owner: 10Awight) [14:36:45] hashar: you left enough of a review on it for me to rework it anyway :) [14:36:47] (03Merged) 10jenkins-bot: Inline remaining cmd.py steps [integration/quibble] - 10https://gerrit.wikimedia.org/r/515182 (owner: 10Awight) [14:37:15] now we're down to the crazy stuff :-) Please feel free to push back on either of my remaining patches, of course. [14:37:17] (03CR) 10jenkins-bot: Inline remaining cmd.py steps [integration/quibble] - 10https://gerrit.wikimedia.org/r/515182 (owner: 10Awight) [14:37:36] who knows :-\ [14:40:42] and https://gerrit.wikimedia.org/r/#/c/integration/quibble/+/515090/4/quibble/commands.py [14:40:42] hm [14:40:54] pretty sure flake8 has a rule to have each import on their own line [14:41:04] or I am just confused [14:41:07] which is a sign of fatighe [14:41:16] See you next week <3 [14:43:32] (03PS5) 10Hashar: Move parallel_run to util.py [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [14:43:48] (03PS6) 10Hashar: Move parallel_run to util.py [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [14:44:58] (03CR) 10Hashar: "PS5 move the Apache license header as well." [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [14:45:10] AZOE [14:45:23] awight: kostajh and I can't do a release of Quibble right now [14:45:33] because the CI containers are build using the latest tag that is available [14:45:34] so hmm [14:45:45] if someone build a container it will end up with the wrong quibble [14:45:46] :\ [14:45:50] so I guess it will be for monday [14:46:02] (03CR) 10Hashar: [C: 03+2] Move parallel_run to util.py [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [14:46:06] oh no! I mean, this is Friday anyway, but "latest" does seem like a twitchy target [14:46:42] (03Merged) 10jenkins-bot: Move parallel_run to util.py [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [14:47:12] (03CR) 10jenkins-bot: Move parallel_run to util.py [integration/quibble] - 10https://gerrit.wikimedia.org/r/515090 (owner: 10Awight) [14:50:20] o/ [15:00:58] ah [15:01:01] flake8 not passing [15:01:06] it is broken and never report :-] [15:01:22] bisected to 609369af484e8af23e1a76377f4045066cc4b09f "Test setup, make lint pass" by thcipriani ;] [15:01:35] or maybe not [15:09:43] off [15:09:48] be back for the meeting [15:09:49] ;) [15:51:39] (03PS2) 10Kosta Harlan: Add option for using Apache as server [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) [15:51:45] (03CR) 10jerkins-bot: [V: 04-1] Add option for using Apache as server [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) (owner: 10Kosta Harlan) [15:53:02] (03PS3) 10Kosta Harlan: Add option for using Apache as server [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) [16:00:55] ^ I'm confused about what's supposed to happen with ports, do we use 9412 or 4881, or something else? [16:18:59] (03CR) 10Kosta Harlan: "> Note the various config files can be put in their own directory and defined as data. Pip would then install them in a special directory" (032 comments) [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) (owner: 10Kosta Harlan) [16:19:01] 10Gerrit, 10Operations, 10Traffic: When downloading from git using HTTPS: HTTP 500 / GnuTLS recv error (-110) - https://phabricator.wikimedia.org/T225347 (10sbassett) Hello @Ciencia_Al_Poder - Apologies for the delay on a response to this issue. Due to an ongoing security incident [0], certain IP ranges co... [16:19:33] kostajh: we've been using 9412 since ~ 2014 for the in-jenkins MW localhost install. Something I came up with, pretty much random number. [16:19:48] it's not required by anything, just a random number I used that wasn't clashing with anything else [16:19:58] this is the first time I see 4881 though, where is that used? [16:20:20] (at the time it was for Apache, hashar since re-used it for php -S) [16:20:57] If it only ever runs inside the container with nothing else web-y, I suppose 8000 or 8080 or some other default would be fine too [16:21:20] I believe I originally tried to spell "QUnit" with the port number, as it was only used for that at the time. [16:21:20] it's the default port for DevWebServer [16:21:47] Yeah, antoine coded that in last year when converting from the persistent puppetised Apache-based CI slaves to Quibble. [16:22:34] I suppose the outer code will need to know the port used for env vars and request urls. Could be kept consistent indeed to make that work as an optional flag. I don't know if we need to support both side by side though. [16:22:51] So if you swap to Apache, can change the input port as well. Presumavly its a variable somewhere? [16:23:01] if it doesnt work out for some reason I assume we'd have to revert anyway. [16:23:36] (03PS4) 10Kosta Harlan: Add option for using Apache as server [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) [16:23:47] or if we need to support both, but want to avoid changing apache's number, could also make it return it with a getter instead of inject, and then Apache's implemetnation will return a differnet number instead. [16:24:18] I'm not sure we can easily override the apache port at runtime [16:24:38] we'd need to chown /etc/apache2 to nobody:nogroup which I could experiment with but probably not today :) [16:25:20] alternative [16:25:29] we finally start the project to get quibble to run on kubernetes [16:25:38] and use a helm chart that has mysql + apache + quibble container [16:25:45] but hmm. We are far from there I guess :-] [16:25:47] +1, but how far away is that? [16:26:12] there is preliminary work on the local dev tooling which will help that [16:26:26] but the list of prerequisite is rather long I guess [16:27:16] IMO the apache patch for quibble is not that far away, and if in production it cuts down the selenium test run times then it would be a nice holdover until the future kubernetes world arrives [16:27:30] +1 [16:27:58] kostajh: maybe you can reach out to mutante, he has lot of experience with apache and might have some idle cycle to help polish the apache config [16:29:02] mutante (and Krinkle and whoever else), contributions to the patch would be very welcome! [16:30:08] (03PS8) 10Awight: Run node browser tests in each repo [integration/quibble] - 10https://gerrit.wikimedia.org/r/510709 (https://phabricator.wikimedia.org/T199116) [16:30:10] (03PS3) 10Awight: Experiment with frozen classes [integration/quibble] - 10https://gerrit.wikimedia.org/r/515889 [16:30:16] kostajh: not sure I understand. Why would it need to do it as nobody at run-time? [16:30:25] The apache conf is provisioned in the Dockerfile right? [16:30:52] I mean, aside from 8080 or whatever being fine, seems like it needs a conf file either way and 9412 seems straight forward lookinga t the diff you did that already :) [16:31:53] I thought the idea was that you could configure the port when you run `quibble --run selenium`. If we don't need to support that use case, then that's much simpler [16:32:46] (03CR) 10Jforrester: "Nice!" [integration/config] - 10https://gerrit.wikimedia.org/r/518196 (owner: 10Legoktm) [16:32:59] seems to be passed in as a constant HTTP_PORT [16:33:06] but with an unused default that is different [16:33:07] no idea [16:33:23] hashar might know. but not exposed as far as I can see, so that should be fine to drop I think? [16:33:46] 9412 yeah I borrowed it from our old ci infra [16:33:53] it is passed to php -S [16:34:05] the port is arbitrary really [16:35:05] and it is passed to karma/qunit and wdio/selenium so they know where to point at [16:35:13] and indeed it is a constant yeah [16:35:30] kostajh: so no need for the http port to be configurable. Hardcoded 9412 sounds good [16:35:32] so, yes let's stick with that then :) [16:36:17] cant we pass values to Apache? [16:39:20] hashar: in which context? When we build the image, when quibble starts Apache? [16:39:49] when Quibble starts Apache [16:39:54] but I guess we can just hardcode all values [16:40:18] Yeah I think we should hardcode [16:40:35] I don't think we need to pass any parameters [16:40:48] it is just that by hardcoding settings, we have them defined both in Apache AND in Quibble code [16:40:49] Because quibble runs as nobody, we’d need to also chown etc/apache2 if we wanted to modify config at runtime [16:41:03] which might be prone to error. But that surely can be tested via an integration test anyway ;] [16:41:14] yeah [16:41:24] just hardcode! [16:41:24] Right [16:41:27] Lol [16:41:31] and later on we can figure out a better solution [16:41:49] I would guess the priority is to figure out whether Apache actually gives us a performance boost compared to php -S [16:41:51] ad [16:41:56] whether that will work on the ci infra [16:42:05] for perf boost, you did the benchmark, it seems faster [16:42:14] on the ci infra, I don't see why it would not work :] [16:42:33] Yeah testing it is tricky [16:42:56] (03PS1) 10Jforrester: jjb: [docker-zuul-cloner] Point at docker image 0.2.0 based on stretch [integration/config] - 10https://gerrit.wikimedia.org/r/518297 [16:42:58] I need to run a test build locally to make sure the npm problems are sorted [16:43:23] Assuming it works, and is merged, we could have an experimental job that uses it [16:43:31] And then do some timing comparisons [16:44:20] (03CR) 10Jforrester: [C: 03+1] "Shall we proceed?" [integration/config] - 10https://gerrit.wikimedia.org/r/518241 (https://phabricator.wikimedia.org/T224908) (owner: 10Hashar) [16:47:38] (03PS2) 10Jforrester: Codehealth: Bump jobs to 0.4.10 [integration/config] - 10https://gerrit.wikimedia.org/r/518272 (owner: 10Kosta Harlan) [16:47:46] kostajh: and I will cut a new Quibble release on monday and deploy it to the infra [16:47:50] eg git tag [16:47:56] (03PS3) 10Jforrester: jjb: [codehealth] Bump jobs to 0.4.10 docker image [integration/config] - 10https://gerrit.wikimedia.org/r/518272 (owner: 10Kosta Harlan) [16:47:56] then rebuild all the containers and switch the jobs [16:48:48] (03CR) 10Jforrester: [C: 03+2] Codehealth: Always append code coverage properties [integration/config] - 10https://gerrit.wikimedia.org/r/518271 (owner: 10Kosta Harlan) [16:50:23] (03Merged) 10jenkins-bot: Codehealth: Always append code coverage properties [integration/config] - 10https://gerrit.wikimedia.org/r/518271 (owner: 10Kosta Harlan) [16:50:25] hashar: sounds good. I’ll do some more work on it today then [16:54:31] kostajh: then the apache would be in the next release anyway :D [16:54:34] it is too scary [16:54:40] Psh, scary. [16:54:43] but I don't mind cutting a new release just for the apache support [16:54:45] it is worth it [16:55:12] !log Docker: Publishing java8-sonar-scanner 0.4.10. [16:55:14] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [16:57:28] (03CR) 10Jforrester: [C: 03+2] jjb: [codehealth] Bump jobs to 0.4.10 docker image [integration/config] - 10https://gerrit.wikimedia.org/r/518272 (owner: 10Kosta Harlan) [16:57:38] 10Continuous-Integration-Config: composer test fails with Composer\Downloader\TransportException - https://phabricator.wikimedia.org/T226272 (10Urbanecm) [16:58:43] !log Updated all four codehealth jjb jobs to use java8-sonar-scanner 0.4.10 [16:58:44] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [16:59:37] (03Merged) 10jenkins-bot: jjb: [codehealth] Bump jobs to 0.4.10 docker image [integration/config] - 10https://gerrit.wikimedia.org/r/518272 (owner: 10Kosta Harlan) [16:59:51] 10Continuous-Integration-Config: composer test fails with Composer\Downloader\TransportException - https://phabricator.wikimedia.org/T226272 (10Urbanecm) p:05Triage→03Unbreak! Blocks (at least some) builds => qualifies as UBN in my opinion, at least before an initial remedy is in place. [17:03:22] 10Continuous-Integration-Config: composer test fails with Composer\Downloader\TransportException - https://phabricator.wikimedia.org/T226272 (10kostajh) [17:04:05] thanks kostajh [17:04:40] Urbanecm: np, thanks for filing the task anyway. My default reaction was to come ask here :P [17:05:49] hashar: about to use `--skip-install` fwiw :) I'm running `quibble --run selenium` to get a baseline, then I'll run `quibble --run selenium --skip-zuul --skip-deps --skip-install --webserver=apache2` [17:10:14] hashar: hmm, `quibble --run selenium` didn't install composer/npm dependencies. maybe an issue with one of the recently merged patches. [17:11:03] ahhhh [17:11:05] hashar: https://phabricator.wikimedia.org/P8640 [17:11:13] dependencies for core or the extension? [17:11:28] WARNING:quibble.cmd:ZUUL_PROJECT not set. Assuming mediawiki/core [17:11:29] :) [17:11:39] hashar: maybe I don't have the very latest source for quibble though, I think one patch was merged since I branched [17:11:58] yeah AttributeError: 'ZuulCloneCommand' object has no attribute 'kwargs' [17:12:05] that got fixed by awight and merged [17:12:46] so update I guess? [17:12:51] oh [17:12:54] or rebase your patch yeah [17:13:31] 10Gerrit, 10Operations, 10Traffic: When downloading from git using HTTPS: HTTP 500 / GnuTLS recv error (-110) - https://phabricator.wikimedia.org/T225347 (10Ciencia_Al_Poder) Ok, thanks for the update. I only need a read-only access to the repos, so I'd have to live with the github mirror... [17:14:30] weirdly, still can't find npm in the image I've built [17:14:49] :-\\\ [17:14:53] despite this change https://gerrit.wikimedia.org/r/c/integration/quibble/+/516729/4/Dockerfile#17 [17:14:56] James_F: any ideas? [17:15:07] wrong path maybe? [17:15:19] OH [17:16:46] Do I have to manually copy over npm from that image? [17:16:55] as was done with https://gerrit.wikimedia.org/r/c/integration/quibble/+/516729/4/Dockerfile#b71 ? [17:17:09] It doesn't inherit? [17:17:35] (03CR) 10Hashar: Add option for using Apache as server (033 comments) [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) (owner: 10Kosta Harlan) [17:17:44] kostajh: https://gerrit.wikimedia.org/r/#/c/integration/quibble/+/516729/4/Dockerfile [17:17:49] FROM foo [17:17:51] FROM bar [17:18:04] that DOES NOT result in a container that would be a merge of the foo and bar FS [17:18:10] but I want magic multiple inheritance :P [17:18:12] it just result in an image that is the same as bar [17:18:17] well [17:18:19] ha [17:18:22] hmm [17:18:23] ok [17:18:29] so it is past 7pm on a friday [17:18:30] :-) [17:18:32] alright, I'll try it [17:18:33] later! [17:18:44] I dont think I am still subject to the friendly working space policy [17:18:48] so here is my bold statement: [17:18:50] BECAUSE DOCKER SUCKS [17:18:51] ;] [17:18:55] hashar: Want to move ahead with https://gerrit.wikimedia.org/r/c/integration/config/+/518241 ? [17:19:11] so yeah hmm the thing that they introduced is you can do multiple FROM statement [17:19:23] which makes docker build to build several images independently [17:19:29] and you can then COPY from one to another [17:19:38] Also, can we just delete quibble-jessie? It's unused now, right? [17:19:41] anyway [17:19:42] FROM npm [17:19:55] FROM quibble [17:20:10] COPY --from=npm /usr/local/lib/node_modules /usr/local/lib/node_modules [17:20:13] something like that [17:20:23] more details at https://docs.docker.com/develop/develop-images/multistage-build/ [17:20:28] comes from Docker 17.05 [17:20:34] but yeah [17:20:46] it is a trap :-\ I have fall into it once [17:21:44] James_F: yeah apparently releng/quibble-jessie is unused [17:22:22] and earlier this afternoon, I have ported releng/zuul-cloner to Stretch but still have to test it [17:22:43] It seems to work for me? [17:23:08] (03CR) 10Hashar: "It builds, but I haven't tested it locally yet. Should have been more clear sorry." [integration/config] - 10https://gerrit.wikimedia.org/r/518241 (https://phabricator.wikimedia.org/T224908) (owner: 10Hashar) [17:23:11] might be [17:23:14] :] [17:23:27] at least it builds properly [17:23:31] and runs the entry point [17:23:45] so I guess yeah, one can just build and deploy it and see what happens (tm) [17:24:17] I am off [17:24:18] hashar: reproducible builds 😂 [17:24:22] * James_F grins. [17:24:24] kostajh: James_F happy hacking!!! :] [17:24:25] enjoy your weekend [17:24:29] danke [17:29:54] (03PS1) 10Jforrester: dockerfiles: Drop quibble-jessie, unused [integration/config] - 10https://gerrit.wikimedia.org/r/518302 (https://phabricator.wikimedia.org/T224908) [17:31:44] (03PS1) 10Jforrester: dockerfiles: Drop hhvm-jessie, unused [integration/config] - 10https://gerrit.wikimedia.org/r/518304 (https://phabricator.wikimedia.org/T224908) [17:32:27] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team, 10Patch-For-Review: Drop jessie testing support - https://phabricator.wikimedia.org/T224908 (10Jdforrester-WMF) [17:34:18] (03CR) 10Jforrester: [C: 03+2] dockerfiles: Drop hhvm-jessie, unused [integration/config] - 10https://gerrit.wikimedia.org/r/518304 (https://phabricator.wikimedia.org/T224908) (owner: 10Jforrester) [17:34:23] (03CR) 10Jforrester: [C: 03+2] dockerfiles: Drop quibble-jessie, unused [integration/config] - 10https://gerrit.wikimedia.org/r/518302 (https://phabricator.wikimedia.org/T224908) (owner: 10Jforrester) [17:35:53] (03Merged) 10jenkins-bot: dockerfiles: Drop quibble-jessie, unused [integration/config] - 10https://gerrit.wikimedia.org/r/518302 (https://phabricator.wikimedia.org/T224908) (owner: 10Jforrester) [17:35:56] (03Merged) 10jenkins-bot: dockerfiles: Drop hhvm-jessie, unused [integration/config] - 10https://gerrit.wikimedia.org/r/518304 (https://phabricator.wikimedia.org/T224908) (owner: 10Jforrester) [17:36:21] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Release-Engineering-Team, 10Patch-For-Review: Drop jessie testing support - https://phabricator.wikimedia.org/T224908 (10Jdforrester-WMF) [17:42:27] 10Release-Engineering-Team (Kanban), 10Security-Team, 10phan-taint-check-plugin: Phan-taint-check-plugin not available for PHP > 7.0 - https://phabricator.wikimedia.org/T207344 (10Jdforrester-WMF) Thank you! I'm happy to help in whatever way I can. [17:45:17] hmm, running `quibble --run selenium` shows attempts to access 127.0.0.1:59250 and other very high port ranges [17:48:13] (03PS5) 10Kosta Harlan: Add option for using Apache as server [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) [17:57:17] 10Continuous-Integration-Config: Move all CI generic tasks from PHP70 to PHP72 - https://phabricator.wikimedia.org/T225457 (10Jdforrester-WMF) [17:57:19] 10Continuous-Integration-Config, 10Parsoid-PHP: Parsoid/PHP post-merge doc publishing task needs updating to use PHP 7.2 - https://phabricator.wikimedia.org/T219319 (10Jdforrester-WMF) 05Open→03Resolved a:03Jdforrester-WMF OK, the task as filed is indeed fixed. Will put in a follow-up. [18:13:51] 10Continuous-Integration-Config, 10Wikidata, 10Wikidata-Campsite, 10User-Michael, 10phan: Phan fails with php72 on WikibaseLexeme - https://phabricator.wikimedia.org/T226244 (10Jdforrester-WMF) 05Open→03Resolved a:03alaa_wmde Thank you for fixing this. [18:28:29] 10Continuous-Integration-Infrastructure, 10serviceops: Provide php72 on contint1001 rather than php56 - https://phabricator.wikimedia.org/T226224 (10Jdforrester-WMF) OK, will switch out with T224591. [18:28:39] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team, 10Patch-For-Review: Drop php56 testing support - https://phabricator.wikimedia.org/T224906 (10Jdforrester-WMF) [18:28:42] 10Continuous-Integration-Infrastructure, 10serviceops: Provide php72 on contint1001 rather than php56 - https://phabricator.wikimedia.org/T226224 (10Jdforrester-WMF) [18:28:48] 10Continuous-Integration-Infrastructure (phase-out-jessie), 10Operations: Migrate contint* hosts to Stretch/Buster - https://phabricator.wikimedia.org/T224591 (10Jdforrester-WMF) [18:28:51] 10Continuous-Integration-Infrastructure, 10Release-Engineering-Team, 10Patch-For-Review: Drop php56 testing support - https://phabricator.wikimedia.org/T224906 (10Jdforrester-WMF) [18:35:28] 10Release-Engineering-Team, 10Wikimedia-production-error (Shared Build Failure): Composer\Downloader\TransportException in CI builds for multiple repositories - https://phabricator.wikimedia.org/T226253 (10Jdforrester-WMF) [19:35:15] (03PS6) 10Kosta Harlan: Add option for using Apache as server [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) [20:01:59] (03CR) 10Jforrester: [C: 03+1] Run node browser tests in each repo [integration/quibble] - 10https://gerrit.wikimedia.org/r/510709 (https://phabricator.wikimedia.org/T199116) (owner: 10Awight) [20:22:07] 10Continuous-Integration-Infrastructure, 10Quibble, 10Patch-For-Review: Consider httpd for quibble instead of php built-in server - https://phabricator.wikimedia.org/T225218 (10kostajh) Some benchmarks with quibble + [this patch](https://gerrit.wikimedia.org/r/c/integration/quibble/+/516729) + docker on mac... [20:35:15] 10Gerrit, 10Operations, 10Traffic: When downloading from git using HTTPS: HTTP 500 / GnuTLS recv error (-110) - https://phabricator.wikimedia.org/T225347 (10sbassett) 05Open→03Resolved [20:48:44] !log Fix Jenkins Console Section pattern for "setup quibble mw-install" to not eat all of output from phan (differentiate from phpunit, composer-test and phan). regexcomplexity++ [20:48:46] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [20:50:12] 10Release-Engineering-Team, 10Scap, 10serviceops, 10PHP 7.2 support, and 2 others: Enhance MediaWiki deployments for support of php7.x - https://phabricator.wikimedia.org/T224857 (10thcipriani) >>! In T224857#5268558, @Joe wrote: > To summarize the situation a bit, we have 2 problems right now that need to... [20:52:00] !log Add "post-build scripts" Jenkins Console section to collapse most unrelated output and cascading failures from artefact readers etc. which are confusing when looking for "the" error [20:52:01] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [21:25:36] PROBLEM - Work requests waiting in Zuul Gearman server on contint1001 is CRITICAL: CRITICAL: 50.00% of data above the critical threshold [140.0] https://grafana.wikimedia.org/dashboard/db/zuul-gearman?panelId=10&fullscreen&orgId=1 [21:28:48] To ssh://gerrit.wikimedia.org:29418/labs/libraryupgrader [21:28:48] ! [remote rejected] HEAD -> refs/for/master (internal server error: Error inserting change/patchset) [21:28:50] that's a new one [21:29:01] pushed again and then it worked [21:29:19] legoktm: gerrit is trying to tell you to take a kit-kat pause :) [21:35:46] RECOVERY - Work requests waiting in Zuul Gearman server on contint1001 is OK: OK: Less than 30.00% above the threshold [90.0] https://grafana.wikimedia.org/dashboard/db/zuul-gearman?panelId=10&fullscreen&orgId=1 [21:36:05] (03CR) 10Awight: "Nice initiative, thanks! Minor notes inline." (039 comments) [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) (owner: 10Kosta Harlan) [21:37:41] legoktm: I've had that before. Too much load on gerrit. [21:40:30] Heh [21:41:51] legoktm: according to Gerrit you hit a lock [21:42:08] *failure [21:48:17] (03PS1) 10Brennen Bearnes: config: Add basic validation steps [integration/pipelinelib] - 10https://gerrit.wikimedia.org/r/518333 (https://phabricator.wikimedia.org/T225335) [21:48:56] (03CR) 10jerkins-bot: [V: 04-1] config: Add basic validation steps [integration/pipelinelib] - 10https://gerrit.wikimedia.org/r/518333 (https://phabricator.wikimedia.org/T225335) (owner: 10Brennen Bearnes) [21:51:11] still very WIP https://libraryupgrader2.wmflabs.org/vulns/npm [21:52:40] 10Release-Engineering-Team, 10MediaWiki-extensions-General: Fix flapping tests in gated extensions as of 2019-06-06 - https://phabricator.wikimedia.org/T225162 (10hashar) [21:53:08] "ugh, npm." <-- loltrue [21:53:22] ain't that covered by the weekly github security alerts? [21:55:07] don't worry, I'lll leave the "ugh" line for every language :p [21:55:22] and I don't want to rely on a proprietary solution to manage the security of our dependencies [21:55:32] Except Ruby, where it's a more gestalt, primal scream. [21:57:41] although it looks to me security issues should be kept hidden until resolved [21:58:20] (03CR) 10Awight: Add option for using Apache as server (031 comment) [integration/quibble] - 10https://gerrit.wikimedia.org/r/516729 (https://phabricator.wikimedia.org/T225218) (owner: 10Kosta Harlan) [22:14:42] Ugh Ruby [22:15:13] 10Continuous-Integration-Config, 10MediaWiki-Core-Testing, 10MediaWiki-ResourceLoader, 10Performance-Team: Run ForeignResourceManager verification on MediaWiki core commits - https://phabricator.wikimedia.org/T203694 (10Jdforrester-WMF) [22:16:13] * James_F grins. [22:22:24] legoktm: It'd be nice if you distinguished between "risks" from devDependencies and real risks from dependencies. [22:23:10] If you're going to all the effort of building a custom dashboard anyway, we might as well get something neat. :-) [22:26:14] James_F: if you use Fresh-node (click click, recommended everywhere!) indeed. If you instead run npm on your work station directly, devDeps means access to Gerrit, Phab, wikimedia.org docs/mail, office wiki, and potentially scp/ssh from random prod hosts. [22:27:17] James_F: sure, wanna file a bug so I dont forget? [22:28:27] legoktm: I feel like "vulnerabilities in npm dependencies" is almost a pleonasm [22:28:33] Hehehe [22:28:38] Patches welcome [22:29:34] just joking :) [22:30:00] most vulnerabilties reported are abuse vectors, which indeed doesnt' apply to devDeps [22:30:14] but actual malicious/compromised packages, does. [22:30:26] I would hope npm pulls those rather than warning, though. [22:33:50] Krinkle: Who runs other people's code directly on their machines and not in a docker image?! ;-) [22:33:56] legoktm: Will do. Against LibUp? [22:35:08] (Filed as T226288.) [22:35:08] T226288: LibUp vulnerability dashboard should distinguish dev-only/devDependency issues from "real" ones. - https://phabricator.wikimedia.org/T226288 [22:35:58] Yep [22:36:00] Same repo [22:36:07] Slowly rewriting the old broken code [22:38:45] Cool. [23:53:08] legoktm: I like the dashboard a lot (sorry for starting with the one nitpick that didn't matter). Also nice how it includes the summaries from upstream advisories right there in context and de-dup by repo. [23:53:13] Very actionable :)