[04:08:01] Yippee, build fixed! [04:08:02] Project selenium-MultimediaViewer ยป safari,beta,OS X 10.9,BrowserTests build #368: 09FIXED in 12 min: https://integration.wikimedia.org/ci/job/selenium-MultimediaViewer/BROWSER=safari,MEDIAWIKI_ENVIRONMENT=beta,PLATFORM=OS%20X%2010.9,label=BrowserTests/368/ [08:21:16] (03CR) 10Hashar: "> We be nicer to use the same jobs in test and in gate for clarity. To remove any doubt about them going out of sync. Will that be possibl" [integration/config] - 10https://gerrit.wikimedia.org/r/313028 (https://phabricator.wikimedia.org/T107529) (owner: 10Hashar) [08:34:33] (03PS10) 10Hashar: Decouple repos from mediawiki gate queue [integration/config] - 10https://gerrit.wikimedia.org/r/313028 (https://phabricator.wikimedia.org/T107529) [08:34:39] (03PS11) 10Hashar: Decouple repos from mediawiki gate queue [integration/config] - 10https://gerrit.wikimedia.org/r/313028 (https://phabricator.wikimedia.org/T107529) [08:38:13] hashar: o/ [08:38:41] elukey: I am about to brea^H^H^H^Hdeploy a change for CI :D [08:39:16] ah okok just wanted to ask your help for the usual jobrunner/timeout thing, whenever you have time let me know :) [08:39:57] elukey: sure shout the idea here, I will catch up :-} [08:40:08] I am reviewing a diff of two 33MBytes files [08:41:54] yesterday I tried to modify /srv/mediawiki/wmf-config/jobqueue.php manually on jobrunner02 to apply the persistent conn setting [08:42:13] but nothing really changed.. am I doing something wrong? (disabled on tin in its scap config) [08:42:47] elukey: have you touched InitialiseSettings.php ? [08:43:04] the timestamp of that file is used to burst the config cache [08:43:17] suuuuure [08:43:21] * elukey ignorant [08:43:27] if ( @filemtime( $filename ) >= filemtime( "$wmfConfigDir/InitialiseSettings.php" ) ) { [08:43:35] // do stuff for real [08:43:39] else load from a .cache file [08:43:50] I thought that it was enough to change the file [08:47:30] PROBLEM - Puppet errors on deployment-redis02 is CRITICAL: CRITICAL: 55.56% of data above the critical threshold [0.0] [08:48:31] !log temporary disable puppet on deployment-tin to remove jobrunner02 from scap dsh; manually enable persistent connection between it and rdb redis hosts [08:49:23] elukey: maybe we can give it a shot on some production canaries on monday morning [08:49:32] note I am only around on monday and tuesday next week [08:50:09] sure! [08:50:49] hashar: so I modified jobqueue.php and edited/saved /srv/mediawiki/wmf-config/InitialiseSettings.php [08:50:56] enough right? [08:53:50] hopefully yeah [08:54:04] then you can confirm the settings via mwscript eval.php --wiki=enwiki [08:54:08] and vardump the variable [08:56:01] where do I find mwscript? [08:59:23] ah something like terbium for deployment prep right [08:59:23] ? [08:59:44] like tin :) [09:03:53] elukey: deployment-tin [09:04:05] then you have done the change on deployment-jobrunner02 [09:04:14] though that host should have mwscript as well (or maybe not) [09:04:24] anyway [09:04:34] you should see less TIME-WAIT [09:05:22] yeah this is the issue, I don't :( [09:05:45] sudo -u www-data php /srv/mediawiki/multiversion/MWScript.php eval.php --wiki=enwiki [09:05:49] on jobrunner02 [09:05:54] that should simulate mwscript [09:06:23] or just scap it fleet wide from tin ? [09:06:56] ah yes another option [09:07:04] local hack, scap deploy [09:07:25] sorry for the stupid questions but what should I do to "vardump the variable"? [09:07:28] never ever done it [09:07:33] elukey: and there is also /srv/mediawiki/wmf-config/jobqueue-labs.php [09:07:56] yeah but IIRC we figured out that it was not picked up [09:07:58] no? [09:08:03] ohh [09:08:07] I have no idea really :( [09:08:16] :D [09:08:27] try to change that one as well [09:08:40] *trying [09:09:58] (03PS12) 10Hashar: Decouple repos from mediawiki gate queue [integration/config] - 10https://gerrit.wikimedia.org/r/313028 (https://phabricator.wikimedia.org/T107529) [09:22:28] RECOVERY - Puppet errors on deployment-redis02 is OK: OK: Less than 1.00% above the threshold [0.0] [09:25:37] (03CR) 10Hashar: [C: 032] "I have switched the jobs in check/test pipelines to use mwgate- prefix. Fixed a few duplicate entry in the process." [integration/config] - 10https://gerrit.wikimedia.org/r/313028 (https://phabricator.wikimedia.org/T107529) (owner: 10Hashar) [09:26:51] (03CR) 10jerkins-bot: [V: 04-1] Decouple repos from mediawiki gate queue [integration/config] - 10https://gerrit.wikimedia.org/r/313028 (https://phabricator.wikimedia.org/T107529) (owner: 10Hashar) [09:26:57] OH NO [09:28:23] (03CR) 10Hashar: [C: 032] Decouple repos from mediawiki gate queue [integration/config] - 10https://gerrit.wikimedia.org/r/313028 (https://phabricator.wikimedia.org/T107529) (owner: 10Hashar) [09:29:55] (03Merged) 10jenkins-bot: Decouple repos from mediawiki gate queue [integration/config] - 10https://gerrit.wikimedia.org/r/313028 (https://phabricator.wikimedia.org/T107529) (owner: 10Hashar) [09:30:52] !log hack reverted on tin and scap pull performed on jobrunner02 [09:30:56] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [09:31:01] not really lucky this morning [09:32:41] !log Zuul: deploying "Decouple repos from mediawiki gate queue" 7a79f752363a / T107529 [09:32:45] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [09:32:46] T107529: free some repositories from their unintended chain to mediawiki/core - https://phabricator.wikimedia.org/T107529 [09:53:33] (03Abandoned) 10Hashar: Remove mw-checks-test from all skins/* [integration/config] - 10https://gerrit.wikimedia.org/r/323248 (owner: 10Paladox) [09:53:38] (03Abandoned) 10Hashar: Update skin-tests template [integration/config] - 10https://gerrit.wikimedia.org/r/323245 (owner: 10Paladox) [11:06:45] PROBLEM - Puppet errors on deployment-etcd-01 is CRITICAL: CRITICAL: 100.00% of data above the critical threshold [0.0] [12:10:32] 10Continuous-Integration-Config: Jenkins: Fail on BOM in submitted files - https://phabricator.wikimedia.org/T40233#3201185 (10hashar) 05Open>03declined [12:16:11] 10Continuous-Integration-Config, 07Browser-Tests, 07Ruby: Cucumber linter should run for all repositories that contain Cucumber code - https://phabricator.wikimedia.org/T58251#3201194 (10hashar) 05Open>03declined This is 2+years old task. The main use case for cucumber is for the selenium tests and we a... [12:21:41] (03PS1) 10Hashar: [AhoCorasick] publish doxygen doc [integration/config] - 10https://gerrit.wikimedia.org/r/349420 (https://phabricator.wikimedia.org/T104048) [12:23:13] (03CR) 10Hashar: [C: 032] [AhoCorasick] publish doxygen doc [integration/config] - 10https://gerrit.wikimedia.org/r/349420 (https://phabricator.wikimedia.org/T104048) (owner: 10Hashar) [12:24:56] (03Merged) 10jenkins-bot: [AhoCorasick] publish doxygen doc [integration/config] - 10https://gerrit.wikimedia.org/r/349420 (https://phabricator.wikimedia.org/T104048) (owner: 10Hashar) [12:25:17] !log T104048 zuul enqueue --trigger gerrit --pipeline postmerge --project AhoCorasick --change 345433,1 [12:25:21] Logged the message at https://wikitech.wikimedia.org/wiki/Release_Engineering/SAL [12:25:21] T104048: Publish AhoCorasick documentation on docs.wikimedia.org - https://phabricator.wikimedia.org/T104048 [12:27:21] (03PS1) 10Hashar: AhoCorasick documentation [integration/docroot] - 10https://gerrit.wikimedia.org/r/349421 (https://phabricator.wikimedia.org/T104048) [12:28:02] 10Continuous-Integration-Config, 10AhoCorasick, 13Patch-For-Review: Publish AhoCorasick documentation on docs.wikimedia.org - https://phabricator.wikimedia.org/T104048#3201223 (10hashar) 05Open>03Resolved a:03hashar https://doc.wikimedia.org/AhoCorasick/master/ And it is on https://doc.wikimedia.org/ [12:28:21] (03CR) 10Hashar: [C: 032] AhoCorasick documentation [integration/docroot] - 10https://gerrit.wikimedia.org/r/349421 (https://phabricator.wikimedia.org/T104048) (owner: 10Hashar) [12:28:52] (03Merged) 10jenkins-bot: AhoCorasick documentation [integration/docroot] - 10https://gerrit.wikimedia.org/r/349421 (https://phabricator.wikimedia.org/T104048) (owner: 10Hashar) [12:28:58] (03CR) 10jenkins-bot: AhoCorasick documentation [integration/docroot] - 10https://gerrit.wikimedia.org/r/349421 (https://phabricator.wikimedia.org/T104048) (owner: 10Hashar) [12:34:09] 10Continuous-Integration-Config, 13Patch-For-Review: Add support for skin and extension dependencies in new skin unit test - https://phabricator.wikimedia.org/T151593#3201239 (10hashar) [12:34:11] 10Continuous-Integration-Config, 13Patch-For-Review: Set up phpunit structure tests for MediaWiki skin repositories - https://phabricator.wikimedia.org/T68926#3201240 (10hashar) [12:34:14] 10Continuous-Integration-Config, 07Easy, 13Patch-For-Review: Fix structure tests for all skins - https://phabricator.wikimedia.org/T113860#3201237 (10hashar) 05Open>03Resolved That is mostly fixed. We have the structure tests running on most of the skins we care about. [12:56:51] 10Continuous-Integration-Config, 07Documentation: Jenkins: Set up generated php documentation for MediaWiki extensions - https://phabricator.wikimedia.org/T27978#315375 (10hashar) [12:59:01] 10Continuous-Integration-Infrastructure: Let Jenkins-mwext-sync clean up own open unmergable patch sets - https://phabricator.wikimedia.org/T99552#3201312 (10hashar) 05Open>03Resolved a:03hashar jenkins-mwext-sync is no more in use. That was a workaround for a bug in Gerrit. [14:19:58] (03PS1) 10Hashar: Clean out jsonlint/jshint [integration/config] - 10https://gerrit.wikimedia.org/r/349439 (https://phabricator.wikimedia.org/T119973) [15:39:42] 10Continuous-Integration-Config: Set up phpunit structure tests for MediaWiki skin repositories - https://phabricator.wikimedia.org/T68926#3201655 (10Volker_E) [16:06:42] RECOVERY - Puppet errors on deployment-etcd-01 is OK: OK: Less than 1.00% above the threshold [0.0] [16:54:04] PROBLEM - Puppet errors on deployment-puppetmaster02 is CRITICAL: CRITICAL: 33.33% of data above the critical threshold [0.0] [16:57:12] <_joe_> that's me ^^ [16:57:20] <_joe_> I'll fix it befor going away [17:15:23] 10Deployment-Systems, 10Scap (Scap3-MediaWiki-MVP): Install conftool on deployment masters - https://phabricator.wikimedia.org/T163565#3201910 (10mmodell) [17:15:36] 10Deployment-Systems, 10Scap (Scap3-MediaWiki-MVP): Install conftool on deployment masters - https://phabricator.wikimedia.org/T163565#3201922 (10mmodell) [17:16:10] 10Scap (Scap3-MediaWiki-MVP), 10scap2, 06Operations: Depool proxies temporarily while scap is ongoing to avoid taxing those nodes - https://phabricator.wikimedia.org/T125629#3201925 (10mmodell) [17:16:11] 10Deployment-Systems, 10Scap (Scap3-MediaWiki-MVP): Install conftool on deployment masters - https://phabricator.wikimedia.org/T163565#3201910 (10mmodell) [17:16:15] 10Scap (Scap3-MediaWiki-MVP), 03releng-201617-q4, 10scap2, 06Operations, and 2 others: Make scap able to depool/repool servers via the conftool API - https://phabricator.wikimedia.org/T104352#3201926 (10mmodell) [17:16:25] 10Scap (Scap3-MediaWiki-MVP), 10scap2, 06Operations: Depool proxies temporarily while scap is ongoing to avoid taxing those nodes - https://phabricator.wikimedia.org/T125629#1993054 (10mmodell) [17:16:27] 10Deployment-Systems, 10Scap (Scap3-MediaWiki-MVP): Install conftool on deployment masters - https://phabricator.wikimedia.org/T163565#3201910 (10mmodell) [17:16:30] 10Deployment-Systems, 10Scap (Scap3-MediaWiki-MVP): Install conftool on deployment masters - https://phabricator.wikimedia.org/T163565#3201910 (10mmodell) [17:16:35] 10Scap (Scap3-MediaWiki-MVP), 03releng-201617-q4, 10scap2, 06Operations, and 2 others: Make scap able to depool/repool servers via the conftool API - https://phabricator.wikimedia.org/T104352#1414314 (10mmodell) [17:16:44] 10Scap (Scap3-MediaWiki-MVP), 10scap2, 06Operations: Depool proxies temporarily while scap is ongoing to avoid taxing those nodes - https://phabricator.wikimedia.org/T125629#1993054 (10mmodell) [17:16:46] 10Deployment-Systems, 10Scap (Scap3-MediaWiki-MVP): Install conftool on deployment masters - https://phabricator.wikimedia.org/T163565#3201910 (10mmodell) [17:16:53] 10Scap (Scap3-MediaWiki-MVP), 03releng-201617-q4, 10scap2, 06Operations, and 2 others: Make scap able to depool/repool servers via the conftool API - https://phabricator.wikimedia.org/T104352#1414314 (10mmodell) [17:20:12] 10Deployment-Systems, 10Scap (Scap3-MediaWiki-MVP), 06Operations-Software-Development: Install conftool on deployment masters - https://phabricator.wikimedia.org/T163565#3201910 (10mmodell) [17:29:04] RECOVERY - Puppet errors on deployment-puppetmaster02 is OK: OK: Less than 1.00% above the threshold [0.0] [19:02:04] hasharAway: Argh. You've made patches adding jshint to dozens of repos even though that's deprecated and shouldn't be added anywhere. :-( [19:18:39] ah [19:18:45] James_F: yeah I replied on https://gerrit.wikimedia.org/r/#/c/349459/1 [19:19:01] the aim is to drop the old Jenkins job that runs some old version of jshint [19:19:06] clean up CI of that job [19:19:10] 10Continuous-Integration-Config, 05MW-1.27-release (WMF-deploy-2016-03-08_(1.27.0-wmf.16)), 13Patch-For-Review: Convert all repos to use npm Jenkins job with jsonlint and eslint - https://phabricator.wikimedia.org/T119973#3202308 (10Jdforrester-WMF) [19:19:30] seems to me migrating from jshint on CI node to grunt-contrib-jshint in the repos is straightforward [19:19:36] 10Continuous-Integration-Config, 05MW-1.27-release (WMF-deploy-2016-03-08_(1.27.0-wmf.16)), 13Patch-For-Review: Convert all repos to use npm Jenkins job with jsonlint and eslint - https://phabricator.wikimedia.org/T119973#1841667 (10Jdforrester-WMF) >>! In T119973#3201889, @hashar wrote: > I am sending 190 p... [19:19:40] while doing jshint > eslint is a bit more cumbersome [19:21:38] Eh. It's pretty trivial. [19:21:53] not really [19:22:03] I spent my whole afternoon on those patches and making sure they all pass [19:22:07] If they fail just add the an override to the rules. [19:22:14] (which includes the CI config change to be honest) [19:22:33] Umherirrender and other have been doing great work to fix up bad CI. [19:23:34] Also if you're touching package.json you really should be updating the grunt version to 1.0 and dropping grunt-cli and all the other things we do whenever we touch these things. [19:24:15] yeah that is all a pain in the a**** [19:24:25] That's what scripts are for. :-) [19:24:32] ideally we would just add a package wikimedia-style-nitpicker [19:24:39] and that would set up everything :D [19:24:57] That's my plan once we've finished the eslint switch. [19:26:14] James_F: and my aim is to drop the jsonlint/jshint jobs : https://gerrit.wikimedia.org/r/#/c/349439/ [19:26:27] then I guess most extensions will have the exact same set of jobs [19:26:41] and I can work on unifying and running a single job per ext that will do everything [19:26:51] Yeah. [19:27:47] an idea I had but I am not smart enough to implement: collect the package.json / composer.json of all repos and branches [19:27:56] push them in a database/elasticsearch [19:28:09] and build some style dashboard out of it [19:28:15] We sort-of have this from third parties. [19:28:27] https://david-dm.org/wikimedia/oojs-ui?type=dev [19:28:41] but but! [19:28:58] And legoktm built https://www.mediawiki.org/wiki/User:Legoktm/ci [19:29:08] But a full dashboard would be lovely. [19:29:24] pretty sure Kunal system can be extended [19:29:50] one of the pain point is having to git clone every repos :( [19:30:04] Well. [19:30:06] but I think Gerrit now supporst git archive, so we can just query it for the composer.json/package.json [19:30:12] We could just .. yeah. [19:30:39] then send a few thousands query over an ssh tunnel and the data gathering should be quite fast [19:30:51] hasharAway: Eurgh, OK, I'll merge your commits to move everything over to jshint but ... let's kill it soon. [19:31:08] yeah I am well aware of the eslint task [19:31:22] been too lazy to figure out the jshint -> eslint transition/rules tweak for each of the repos :( [19:31:31] commits are cheap_ [19:32:45] James_F: let me check, maybe I have added jshint on repos already moved to eslint [19:33:19] hasharAway: I'm manually reading all of them before merging, obviously. [19:35:45] I caught two already on eslint (Score and TemplateSandbox). Though the patchs havent been sent to gerrit [19:37:16] James_F: dont CR+2 them too fast though. That pills up in CI :] [19:38:41] hasharAway too late https://integration.wikimedia.org/zuul/ [19:41:32] hasharAway: OK, will give it a bit before the next lot. [20:10:29] hasharAway it seems that this https://gerrit.wikimedia.org/r/#/c/348258/ test fails with a missing file (extension). [20:10:44] But it should work with https://phabricator.wikimedia.org/diffusion/CICF/browse/master/zuul/parameter_functions.py;2fe240f69d8c6fe5e9c7cc43cb755d58b8ee5506$182 [20:10:48] paladox: I replied there [20:11:02] oh [20:11:04] i see [20:11:07] thanks [20:11:09] maybe Umherirrender will fix it up :] [20:11:37] How do extensions load other extensions using extension.json? [20:11:39] I forgot [20:13:02] i think so [20:13:11] paladox: iirc via some php scripts in integration/jenkins.git mediawiki/conf.d/ [20:13:25] oh [20:13:36] oh you mean for jenkins... that i am unsure [20:13:36] there is a snippet injected in LocalSettings.php that tries to include Foo.php / else wfLoadExtension [20:13:39] or something like that [20:13:41] hasharAway: It'd be nice to have the dashboard so we know what repos to do. Right now I use GitHub, search for "jshint" in JSON files on Wikimedia repos, and pick a handful that catch my eye, but... [20:14:23] Oh, so just use wfLoadExtension? [20:14:39] paladox: yeah look at the scripts there :] [20:14:45] ok :) [20:14:46] thanks [20:14:55] James_F: probably Kunal tool can be extended [20:15:13] hasharAway: Maybe. He's been busy with other things though. [20:15:32] yup and apparently the tool is no more running since November :/ [20:18:01] hasharAway https://gerrit.wikimedia.org/r/#/c/349548/ :) [20:49:30] twentyafterfour not sure if this https://secure.phabricator.com/rP5c1e4488dedafda08684b33a8a4786cf687d2811 security issue affects us. [20:59:35] @seen luke081515 [20:59:51] greg-g: heh, I'm here [20:59:57] oh! [21:00:12] luckily I still get pings for my nick :D [21:04:47] paladox: we're not cc twentyafterfour [21:05:04] ok [21:13:35] 06Release-Engineering-Team, 15User-greg: Investigate using Phlogiston for releng - https://phabricator.wikimedia.org/T119270#3202802 (10greg) 05stalled>03declined Maybe revisit later. [22:00:27] (03PS1) 10Mholloway: Android: Set -no-boot-anim command line flag for emulator [integration/config] - 10https://gerrit.wikimedia.org/r/349630 [22:01:03] (03PS2) 10Mholloway: Android: Set -no-boot-anim command line flag for emulator [integration/config] - 10https://gerrit.wikimedia.org/r/349630 [22:02:03] (03PS3) 10Mholloway: Android: Set -no-boot-anim command line flag for emulator [integration/config] - 10https://gerrit.wikimedia.org/r/349630 [22:27:05] I wonder is the wikibugs bot being hit by the rate limit in phabricator. [22:42:11] PROBLEM - Work requests waiting in Zuul Gearman server https://grafana.wikimedia.org/dashboard/db/zuul-gearman on contint1001 is CRITICAL: CRITICAL: 50.00% of data above the critical threshold [140.0] [22:46:11] PROBLEM - Work requests waiting in Zuul Gearman server https://grafana.wikimedia.org/dashboard/db/zuul-gearman on contint1001 is CRITICAL: CRITICAL: 78.57% of data above the critical threshold [140.0] [22:48:36] that is fine, btw, before anyone asks :) ^ [22:59:14] greg-g: Is that "James should slow down"? ;-) [22:59:29] James_F: heh, whatever, it's Friday afternoon ;) [22:59:46] I stopped 20 minutes ago, in my defence. ;-) [23:36:59] paladox: it's possible [23:47:11] RECOVERY - Work requests waiting in Zuul Gearman server https://grafana.wikimedia.org/dashboard/db/zuul-gearman on contint1001 is OK: OK: Less than 30.00% above the threshold [90.0]