[00:48:09] 3Wikimedia / 3Continuous integration: Jenkins: Zuul queue stuck indefinitely if dependent pipeline change has merge conflict? - 10https://bugzilla.wikimedia.org/72113 (10Krinkle) 3NEW p:3Unprio s:3critic a:3None Created attachment 16783 --> https://bugzilla.wikimedia.org/attachment.cgi?id=16783&act... [00:48:38] 3Wikimedia / 3Continuous integration: Jenkins: Zuul queue stuck indefinitely if dependent pipeline change has merge conflict? - 10https://bugzilla.wikimedia.org/72113 (10Krinkle) 5NEW>3ASSI p:5Unprio>3Highes a:3Antoine "hashar" Musso (WMF) [00:55:04] (03PS1) 10Krinkle: Remove broken symlinks to _shared (no longer exists) [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166928 [00:55:19] (03CR) 10Krinkle: [C: 032] Remove broken symlinks to _shared (no longer exists) [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166928 (owner: 10Krinkle) [00:55:21] (03Merged) 10jenkins-bot: Remove broken symlinks to _shared (no longer exists) [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166928 (owner: 10Krinkle) [00:55:31] (03CR) 10Krinkle: "Fixed broken symlinks in 6c003c9b8d8853c87a27746da9b0b109a9afa399." [integration/jenkins] - 10https://gerrit.wikimedia.org/r/155727 (https://bugzilla.wikimedia.org/51717) (owner: 10Hashar) [01:22:20] (03PS1) 10Krinkle: zuul: Simplify graphite query [integration/docroot] - 10https://gerrit.wikimedia.org/r/166931 [01:22:22] (03PS1) 10Krinkle: zuul: Change stack graph from orange/blue to blue/green [integration/docroot] - 10https://gerrit.wikimedia.org/r/166932 [01:22:28] (03CR) 10Krinkle: [C: 032] zuul: Simplify graphite query [integration/docroot] - 10https://gerrit.wikimedia.org/r/166931 (owner: 10Krinkle) [01:22:30] (03Merged) 10jenkins-bot: zuul: Simplify graphite query [integration/docroot] - 10https://gerrit.wikimedia.org/r/166931 (owner: 10Krinkle) [01:22:34] (03CR) 10Krinkle: [C: 032] zuul: Change stack graph from orange/blue to blue/green [integration/docroot] - 10https://gerrit.wikimedia.org/r/166932 (owner: 10Krinkle) [01:22:36] (03Merged) 10jenkins-bot: zuul: Change stack graph from orange/blue to blue/green [integration/docroot] - 10https://gerrit.wikimedia.org/r/166932 (owner: 10Krinkle) [02:36:00] (03PS1) 10Krinkle: tools: Remove broken symlink for local grunt-lib-phantomjs module [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166937 [02:38:06] (03CR) 10Krinkle: [C: 032] tools: Remove broken symlink for local grunt-lib-phantomjs module [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166937 (owner: 10Krinkle) [02:38:08] (03Merged) 10jenkins-bot: tools: Remove broken symlink for local grunt-lib-phantomjs module [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166937 (owner: 10Krinkle) [03:25:56] 3Wikimedia / 3Continuous integration: CI browser test dashboard takes 100 seconds to appear on first load - 10https://bugzilla.wikimedia.org/70671#c2 (10spage) I don't expect miracles :) but it's getting worse. The last few times I've loaded the dashboard it times out altogether and I have to reload to get... [03:52:16] (03PS3) 10Krinkle: tools: Update dev packages [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166900 [04:06:58] Project browsertests-Echo-en.wikipedia.beta.wmflabs.org-linux-firefox-sauce build #123: FAILURE in 9 min 41 sec: https://integration.wikimedia.org/ci/job/browsertests-Echo-en.wikipedia.beta.wmflabs.org-linux-firefox-sauce/123/ [04:17:42] Yippee, build fixed! [04:17:43] Project browsertests-Flow-en.wikipedia.beta.wmflabs.org-linux-chrome-monobook-sauce build #56: FIXED in 33 min: https://integration.wikimedia.org/ci/job/browsertests-Flow-en.wikipedia.beta.wmflabs.org-linux-chrome-monobook-sauce/56/ [04:45:32] Project browsertests-CirrusSearch-test2.wikipedia.org-linux-firefox-sauce build #203: FAILURE in 1 min 29 sec: https://integration.wikimedia.org/ci/job/browsertests-CirrusSearch-test2.wikipedia.org-linux-firefox-sauce/203/ [04:50:23] Yippee, build fixed! [04:50:23] Project browsertests-Echo-en.wikipedia.beta.wmflabs.org-linux-chrome-sauce build #109: FIXED in 9 min 20 sec: https://integration.wikimedia.org/ci/job/browsertests-Echo-en.wikipedia.beta.wmflabs.org-linux-chrome-sauce/109/ [05:13:59] (03CR) 10Krinkle: [C: 04-1] "Doesn't work. phantomjs throws binPath is null." [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166900 (owner: 10Krinkle) [06:13:49] !log updated salt minion on remaining beta instances to 2014.1.11 from salt ppa [06:14:06] hm still no bot eh? [06:29:32] Project browsertests-MultimediaViewer-en.wikipedia.beta.wmflabs.org-windows_7-internet_explorer-9-sauce build #69: FAILURE in 4 min 31 sec: https://integration.wikimedia.org/ci/job/browsertests-MultimediaViewer-en.wikipedia.beta.wmflabs.org-windows_7-internet_explorer-9-sauce/69/ [07:50:37] Project browsertests-UniversalLanguageSelector-commons.wikimedia.beta.wmflabs.org-linux-firefox-sauce build #223: FAILURE in 23 min: https://integration.wikimedia.org/ci/job/browsertests-UniversalLanguageSelector-commons.wikimedia.beta.wmflabs.org-linux-firefox-sauce/223/ [07:52:07] (03CR) 10Hashar: [C: 032] Update to phpunit 3.7.37 [integration/phpunit] - 10https://gerrit.wikimedia.org/r/164683 (owner: 10BryanDavis) [08:00:06] (03Merged) 10jenkins-bot: Update to phpunit 3.7.37 [integration/phpunit] - 10https://gerrit.wikimedia.org/r/164683 (owner: 10BryanDavis) [08:15:52] 3Wikimedia / 3Continuous integration: [upstream] Zuul: status.json API should be able to filter by change - 10https://bugzilla.wikimedia.org/47609#c4 (10Antoine "hashar" Musso (WMF)) Upstream has landed a change that would let one query a specific change https://review.openstack.org/#/c/124390/ > The suppor... [09:28:53] (03CR) 10Tobias Gritschacher: [C: 031] Fix broken jobs when BROWSER_TIMEOUT is not set (031 comment) [integration/config] - 10https://gerrit.wikimedia.org/r/166547 (owner: 10Zfilipin) [10:11:09] !log Updated our Jenkins Job Builder forked repository ( ee80dbc..7ad4386 ). No job configuration impact. [11:12:56] (03PS1) 10Zfilipin: Extract browsertests related things from macro.yaml [integration/config] - 10https://gerrit.wikimedia.org/r/166977 [11:35:17] (03PS1) 10Zfilipin: WIP Merge browsertest-website and browsertest-website-version builders [integration/config] - 10https://gerrit.wikimedia.org/r/166981 [12:20:54] hashar: around? [12:21:21] looks like there is a problem in mediawiki/core [12:21:27] Could not locate Gemfile [12:21:30] https://integration.wikimedia.org/ci/job/mediawiki-core-bundle-rubocop/8/console [12:21:51] looks like our bundle job does not cd into the correct folder (usually tests/browser) [12:26:17] hashar: the same here https://integration.wikimedia.org/ci/job/mwext-ContentTranslation-bundle-rubocop/7/console [12:27:27] but it works fine here [12:27:28] https://integration.wikimedia.org/ci/job/mwext-Flow-bundle-rubocop/47/consoleFull [12:36:21] (03PS2) 10Zfilipin: Extract browsertests related things from macro.yaml [integration/config] - 10https://gerrit.wikimedia.org/r/166977 [12:38:16] (03PS2) 10Zfilipin: Merge browsertest-website and browsertest-website-version builders [integration/config] - 10https://gerrit.wikimedia.org/r/166981 [12:38:54] Tobi_WMDE_SW: the above commit is ready [12:38:59] we even did it right :) [12:39:20] nice amount of code removed, behaves the same [12:39:50] more than 100 lines of code less [12:39:59] well done, sir :) [13:06:52] 3Wikimedia Labs / 3deployment-prep (beta): login busted - 10https://bugzilla.wikimedia.org/71532#c2 (10Andre Klapper) I guess this can be closed now as RESOLVED WORKSFORME / FIXED? [13:09:07] 3Wikimedia / 3Continuous integration: Zuul prepareRef does not handle failure to connect to Gearman - 10https://bugzilla.wikimedia.org/72113#c1 (10Antoine "hashar" Musso (WMF)) 5ASSI>3NEW p:5Highes>3Normal s:5critic>3normal Looking at debug.log.2014-10-15 23:21:17 - receives a code review +2 ev... [13:28:08] 3Wikimedia / 3Continuous integration: [upstream] Zuul prepareRef does not handle failure to connect to Gearman - 10https://bugzilla.wikimedia.org/72113#c2 (10Antoine "hashar" Musso (WMF)) Turns out I already filled a bug upstream in August. The case was the connection to Gearman timing out while preparing a... [13:37:58] (03CR) 10Addshore: [C: 04-1] "As before:" [integration/config] - 10https://gerrit.wikimedia.org/r/166024 (owner: 10Hashar) [13:38:00] zeljkof: sorry [13:38:10] zeljkof: was debugging a Zuul bug that occurred over night [13:38:49] which might be an easy fix :D https://review.openstack.org/#/c/128921/1/zuul/scheduler.py,unified [13:38:51] so rubocop [13:39:26] zeljkof: the bundle macro used by the template executes bundle install and bundle exec from the root directory [13:39:35] so one would have to add a /Gemfile containing rubocop [13:39:38] OR [13:39:57] we can craft a different job template [13:40:21] that would install rubocop and then use the rubocop macro (to avoid copy pasting it) [13:47:53] zeljkof: I'll miss today's meeting, but if you want to look over some refactoring: https://gerrit.wikimedia.org/r/#/c/166767/ also 166780 and 166874 they all should be straightforward [13:48:17] hashar: hm [13:48:21] will have to think about it [13:48:37] maybe the job could copy gemfile to root from tests/browser [13:48:46] or it should cd into tests/browser [13:49:30] chrismcmahon: sorry, I do not think I will have the time to take a look at that today [13:50:02] c'est la vie :-) [13:56:40] zeljkof: while we figure out something, can we stop triggering the rubocop jobs on every patchset? [13:56:44] cause that confuses developers [13:57:12] hashar: please no :) [13:57:16] if possible [13:57:20] why no? [13:57:26] I promise to figure out something today, or tomorrow [13:57:29] it serves no purposes as is beside reporting FAILURE (non voting) [13:57:31] if not, then disable it [13:57:41] I have proposed a change to move them to the experimental pipeline [13:57:48] if you think it should be disabled, go ahead [13:57:52] so they would only be triggered on demand by comment 'check experimental' [13:58:01] but I think I will be able to fix it this week [13:58:23] so lets move it to experimental, polish it up / fix it up this week by using 'check experimental' [13:58:25] I would prefer if it was enabled all the time, just did not have the time to work on it today [13:58:35] and once they are ready, reenable them on patch proposals? [13:58:36] hashar: sure, go ahead [13:58:46] you can iterate by commenting on a change :D [13:58:59] or at worth, trigger it manually from Jenkins web interface hehe [14:00:51] zeljkof: just +2 https://gerrit.wikimedia.org/r/#/c/166805/ :-D [14:00:52] hehe [14:01:11] that comments out all the rubocop jobs [14:01:22] and add them to experimental: [14:03:02] (03CR) 10Zfilipin: [C: 032] zuul: move rubocop jobs to experimental [integration/config] - 10https://gerrit.wikimedia.org/r/166805 (https://bugzilla.wikimedia.org/69245) (owner: 10Hashar) [14:03:17] hashar: done [14:03:47] for rubocop, if there is no /Gemfile we could craft our own that adds the rubocop gem so bundle install actually install it [14:04:19] hashar: yes, we actually need only Gemfile with one line, containing rubocop [14:04:26] I guess we don't want to move /tests/browser/Gemfile at the root of the repository do we ? [14:04:39] hashar: that is an option too [14:04:51] that would be ideal [14:04:58] but we would need to test if it breaks something else [14:05:05] zeljkof: generally, in a changeset, shouldn't the change-id be in the last line of the commit message? [14:05:30] just saw that in your changesets you've put something after it [14:05:32] Tobi_WMDE_SW: not sure, git review adds it automatically, I rarely change it by hand [14:05:44] no, probably git-review did something strange [14:05:57] zeljkof: ok. was just wondering because I was assuming gerrit not to work properly otherwise [14:06:03] I never change that line [14:06:03] (or I made a mistake) [14:06:07] feel free to fix, it was not on purpose [14:19:02] (03PS3) 10Tobias Gritschacher: Merge browsertest-website and browsertest-website-version builders [integration/config] - 10https://gerrit.wikimedia.org/r/166981 (owner: 10Zfilipin) [14:19:18] (03PS3) 10Tobias Gritschacher: Extract browsertests related things from macro.yaml [integration/config] - 10https://gerrit.wikimedia.org/r/166977 (owner: 10Zfilipin) [14:21:08] 3Wikimedia Labs / 3deployment-prep (beta): login busted - 10https://bugzilla.wikimedia.org/71532 (10Chris McMahon) 5NEW>3RESO/FIX [14:31:24] Project browsertests-Wikidata-SmokeTests-linux-firefox-sauce build #13: FAILURE in 14 min: https://integration.wikimedia.org/ci/job/browsertests-Wikidata-SmokeTests-linux-firefox-sauce/13/ [14:32:13] zeljkof: super lame Gemfile forger http://paste.openstack.org/show/121480/ :D [14:32:48] but overall, it is probably nicer to move the Gemfile at the root of the repositories [14:33:14] hashar: not sure where the rest of our scripts expect the gemfile to be [14:33:22] we can adjust [14:33:29] but maybe we just should tweak them [14:33:51] in that case, rubocop dotfile should also be in repo root [14:34:05] indeed [14:40:45] hashar: sorry, I am distracted, have to prepare monthly paperwork for my accountant today :( [14:40:50] booooring [14:41:03] flipping papers [14:42:03] damn [14:42:07] I forgot about it [14:50:40] Project browsertests-Wikidata-SmokeTests-linux-firefox-sauce build #14: ABORTED in 5 min 22 sec: https://integration.wikimedia.org/ci/job/browsertests-Wikidata-SmokeTests-linux-firefox-sauce/14/ [17:38:11] (03PS4) 10Krinkle: tools: Update dev packages [integration/jenkins] - 10https://gerrit.wikimedia.org/r/166900 [17:40:11] hi ryasmeen I'm going to merge this pretty soon so that the test will pass, but I'm not really sure that the auto-populate feature in the Webcite citation is really behaving as designed, it might be worth looking into: https://gerrit.wikimedia.org/r/#/c/167023/ [17:54:20] Yippee, build fixed! [17:54:20] Project browsertests-MultimediaViewer-en.wikipedia.beta.wmflabs.org-windows_7-internet_explorer-9-sauce build #70: FIXED in 28 min: https://integration.wikimedia.org/ci/job/browsertests-MultimediaViewer-en.wikipedia.beta.wmflabs.org-windows_7-internet_explorer-9-sauce/70/ [17:54:53] 3Wikimedia Labs / 3deployment-prep (beta): In VE, while adding any image using Firefox getting the error "Image corrupt or truncated" in Betalabs - 10https://bugzilla.wikimedia.org/72115 (10Rummana Yasmeen) [18:15:11] hi marxarelli do you have a few minutes? I'd like to do some guerilla merging in the browser test repo for Flow and I could use another set of eyes [18:15:45] chrismcmahon: sure, just wrapping up some mwv troubleshooting [18:15:57] > guerilla merging [18:16:03] That sounds fun [18:16:39] marxarelli: awesome. you already gave +1 to https://gerrit.wikimedia.org/r/#/c/166767/ and I have two more behind it [18:17:45] marktraceur: it's pretty exciting. I have the Flow tests refactored enough to actually make small granular commits now, but they languish in gerrit [18:18:54] Ooh. [18:19:37] :-) [18:20:30] chrismcmahon: cool, let me fire up my flow role [18:21:37] marxarelli: the other two are https://gerrit.wikimedia.org/r/#/c/166780/ and https://gerrit.wikimedia.org/r/#/c/166874/ [18:37:19] chrismcmahon: ok, git-update finally finished [18:40:57] chrismcmahon: any clue as to why jenkins-bot would -1 that first one? [18:41:37] chrismcmahon: looks like mwext-Flow-testextension is the culprit, but i don't know what that's checking for [18:42:21] marxarelli: I think some other tests failed, but I don't know if it's a Jenkins artifact or a real failure. Regardless, it's not the browser test code [18:42:47] chrismcmahon: no, the browser code is good to go (just ran it locally and it's green) [18:43:37] I took a brief look at the failure yesterday, but not sure what it was trying to do [18:44:37] chrismcmahon: no matter. +2' [18:44:43] yay [18:44:59] I'm getting the hang of RSpec 3.0 now :-) [18:46:16] chrismcmahon: cool! [19:04:05] (03PS4) 10Hashar: Extract browsertests related things from macro.yaml [integration/config] - 10https://gerrit.wikimedia.org/r/166977 (owner: 10Zfilipin) [19:24:53] marxarelli: I fixed the indent and merged it (after a nonsense rebase), the tests pass on beta labs, the failure seems to a known issue with performance when creating a new Flow topic https://bugzilla.wikimedia.org/show_bug.cgi?id=71476 [19:25:38] chrismcmahon: ok. it could have also been an issue with my vm. i'm seeing some strange behavior with hhvm and mysql connections [19:26:03] e.g. every request seems to open a new mysql connection [19:27:01] * marxarelli finds https://bugzilla.wikimedia.org/show_bug.cgi?id=71370 [19:29:18] huh [19:34:26] Project browsertests-Flow-en.wikipedia.beta.wmflabs.org-linux-chrome-monobook-sauce build #57: FAILURE in 24 min: https://integration.wikimedia.org/ci/job/browsertests-Flow-en.wikipedia.beta.wmflabs.org-linux-chrome-monobook-sauce/57/ [19:54:00] Yippee, build fixed! [19:54:01] Project browsertests-Flow-en.wikipedia.beta.wmflabs.org-linux-firefox-monobook-sauce build #69: FIXED in 38 min: https://integration.wikimedia.org/ci/job/browsertests-Flow-en.wikipedia.beta.wmflabs.org-linux-firefox-monobook-sauce/69/ [19:57:15] (03CR) 10Hashar: [C: 032] "Indeed noop :-)" [integration/config] - 10https://gerrit.wikimedia.org/r/166977 (owner: 10Zfilipin) [19:57:29] (03PS4) 10Hashar: Merge browsertest-website and browsertest-website-version builders [integration/config] - 10https://gerrit.wikimedia.org/r/166981 (owner: 10Zfilipin) [19:59:13] (03CR) 10Hashar: [C: 031] "Seems fine to me :-) Great to see less code duplication." [integration/config] - 10https://gerrit.wikimedia.org/r/166981 (owner: 10Zfilipin) [20:01:32] (03Merged) 10jenkins-bot: Extract browsertests related things from macro.yaml [integration/config] - 10https://gerrit.wikimedia.org/r/166977 (owner: 10Zfilipin) [20:12:34] Yippee, build fixed! [20:12:34] Project browsertests-CirrusSearch-test2.wikipedia.org-linux-firefox-sauce build #204: FIXED in 2 min 19 sec: https://integration.wikimedia.org/ci/job/browsertests-CirrusSearch-test2.wikipedia.org-linux-firefox-sauce/204/ [20:24:46] fixed is good [20:37:27] 3Wikimedia / 3Quality Assurance: create custom RSpec matchers for when_present, when_not_present, etc - 10https://bugzilla.wikimedia.org/72149 (10Chris McMahon) 3NEW p:3Unprio s:3normal a:3None see the discussion at https://lists.wikimedia.org/pipermail/qa/2014-October/002021.html this would allow... [21:27:53] 3Wikimedia Labs / 3deployment-prep (beta): In VE, while adding any image using Firefox getting the error "Image corrupt or truncated" in Betalabs - 10https://bugzilla.wikimedia.org/72115#c2 (10Greg Grossmeier) Created attachment 16790 --> https://bugzilla.wikimedia.org/attachment.cgi?id=16790&action=edit C... [21:29:20] (03PS1) 10John F. Lewis: Add myself to trusted users list [integration/config] - 10https://gerrit.wikimedia.org/r/167107 [21:32:16] (03CR) 10Dzahn: [C: 031] "yes, i did suggest that, so we can get jenkins to verify. examples of merged patches in ops/puppet: https://gerrit.wikimedia.org/r/#/q/ow" [integration/config] - 10https://gerrit.wikimedia.org/r/167107 (owner: 10John F. Lewis) [21:32:18] (03CR) 10Hoo man: [C: 031] Add myself to trusted users list [integration/config] - 10https://gerrit.wikimedia.org/r/167107 (owner: 10John F. Lewis) [21:34:00] (03CR) 10Dzahn: "ironically adding yourself to this list means the change itself doesn't get +2 from jenkins either" [integration/config] - 10https://gerrit.wikimedia.org/r/167107 (owner: 10John F. Lewis) [21:41:54] marxarelli|nom: one more if you'd like to look it over: https://gerrit.wikimedia.org/r/#/c/167104 [21:44:38] Yippee, build fixed! [21:44:39] Project browsertests-Flow-en.wikipedia.beta.wmflabs.org-windows_8-internet_explorer-sauce build #217: FIXED in 37 min: https://integration.wikimedia.org/ci/job/browsertests-Flow-en.wikipedia.beta.wmflabs.org-windows_8-internet_explorer-sauce/217/ [21:46:44] chrismcmahon: will do [21:48:58] marxarelli: on that one I'm not sure the "confirm" step really needs to be a block but I don't think it does any harm, and it makes for better readability [21:49:38] chrismcmahon: i think it makes sense [21:50:06] otherwise pretty straightforward I think [21:50:08] chrismcmahon: and, in general, the block syntax seems to improve readability with expect() [21:50:45] I am almost done with these, three more before I tackle the couple of really confusing things, time enough for that next week. [21:52:01] chrismcmahon: the tests are green but i see a "'#text_field' is deprecated" message [21:52:29] chrismcmahon: "Please, use '#textarea' method instead" [21:52:38] marxarelli: yeah, that is know, I filed an issue with the page_object maintainer. it's an old artifact of how Selenium treats tags [21:52:44] known, even [21:52:54] chrismcmahon: ah, i see [21:54:29] marxarelli: second comment here is the relevant one:https://github.com/cheezy/page-object/issues/248 has to do with an tag with no type associated [22:08:37] 3Wikimedia Labs / 3deployment-prep (beta): In VE, while adding any image using Firefox getting the error "Image corrupt or truncated" in Betalabs - 10https://bugzilla.wikimedia.org/72115#c3 (10Rummana Yasmeen) Created attachment 16792 --> https://bugzilla.wikimedia.org/attachment.cgi?id=16792&action=edit S... [22:54:32] Project browsertests-MultimediaViewer-en.wikipedia.beta.wmflabs.org-linux-chrome-sauce build #229: FAILURE in 17 min: https://integration.wikimedia.org/ci/job/browsertests-MultimediaViewer-en.wikipedia.beta.wmflabs.org-linux-chrome-sauce/229/