[00:00:51] (03PS1) 10Krinkle: (DRAFT) Hotfix: grunt-lib-phantomjs Date mockery protection [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125039 [00:02:41] (03PS1) 10Krinkle: Add bin/phantomjs symlink to avoid having to figure out the deep path [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125040 [00:03:14] (03PS2) 10Krinkle: Add bin/phantomjs symlink to deep node_modules path [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125040 [00:03:23] (03CR) 10Krinkle: [C: 032] Add bin/phantomjs symlink to deep node_modules path [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125040 (owner: 10Krinkle) [00:03:26] (03Merged) 10jenkins-bot: Add bin/phantomjs symlink to deep node_modules path [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125040 (owner: 10Krinkle) [00:58:28] (03PS1) 10Krinkle: Update tools/package.json to newer upstream versions [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125052 [01:01:30] (03PS2) 10Krinkle: Update tools/package.json to newer upstream versions [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125052 [01:01:36] (03CR) 10Krinkle: [C: 032] Update tools/package.json to newer upstream versions [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125052 (owner: 10Krinkle) [01:01:55] (03Merged) 10jenkins-bot: Update tools/package.json to newer upstream versions [integration/jenkins] - 10https://gerrit.wikimedia.org/r/125052 (owner: 10Krinkle) [01:14:58] (03PS1) 10Ori.livneh: *.hhvm.beta.wmflabs.org -> *.wikipedia.beta-hhvm.wmflabs.org [operations/apache-config] (betacluster) - 10https://gerrit.wikimedia.org/r/125058 [01:15:36] (03CR) 10Ori.livneh: [C: 032] *.hhvm.beta.wmflabs.org -> *.wikipedia.beta-hhvm.wmflabs.org [operations/apache-config] (betacluster) - 10https://gerrit.wikimedia.org/r/125058 (owner: 10Ori.livneh) [01:15:38] (03Merged) 10jenkins-bot: *.hhvm.beta.wmflabs.org -> *.wikipedia.beta-hhvm.wmflabs.org [operations/apache-config] (betacluster) - 10https://gerrit.wikimedia.org/r/125058 (owner: 10Ori.livneh) [07:26:23] hello [07:30:40] (03PS1) 10Hashar: Mark ContributionReporting has obsolete [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/125082 [07:30:49] (03CR) 10Hashar: [C: 032] Mark ContributionReporting has obsolete [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/125082 (owner: 10Hashar) [07:35:42] (03PS1) 10Hashar: Jobs for mw/ext/TextExtracts [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125085 [07:35:54] (03CR) 10Hashar: [C: 032] Jobs for mw/ext/TextExtracts [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125085 (owner: 10Hashar) [07:36:46] (03PS2) 10Hashar: Mark ContributionReporting has obsolete [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/125082 [07:36:55] (03CR) 10Hashar: [C: 032] Mark ContributionReporting has obsolete [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/125082 (owner: 10Hashar) [07:37:19] (03Merged) 10jenkins-bot: Jobs for mw/ext/TextExtracts [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125085 (owner: 10Hashar) [07:37:53] (03PS1) 10Hashar: Trigger jobs for mw/ext/TextExtracts [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/125087 [07:38:07] (03CR) 10Hashar: [C: 032] Trigger jobs for mw/ext/TextExtracts [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/125087 (owner: 10Hashar) [07:38:38] (03Merged) 10jenkins-bot: Mark ContributionReporting has obsolete [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/125082 (owner: 10Hashar) [07:38:44] (03Merged) 10jenkins-bot: Trigger jobs for mw/ext/TextExtracts [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/125087 (owner: 10Hashar) [07:42:03] hi hashar [07:42:52] hashar: I filed https://bugzilla.wikimedia.org/show_bug.cgi?id=63745 yesterday [07:45:37] hold on busy taking traces :-D [07:47:29] hashar: :) [07:52:26] bahhh [07:55:44] i must have deployed that manually on production [07:55:50] the dir has proper ownerships [08:06:05] Invalid parameter user at /etc/puppet/modules/contint/manifests/maven_webproxy.pp:21 [08:06:11] I sometime hate puppet [08:25:00] hashar: heh [08:25:06] being worked on :-] [08:25:08] puppet is running [08:25:09] hashar: it broke after a puppet run I guess [08:25:10] hashar: ok :) [08:25:14] * yuvipanda is intermittently available [08:25:30] I will retrigger the job to see if it is fixed and close bug if so [08:26:31] yuvipanda: should be good rebuilding at https://integration.wikimedia.org/ci/job/apps-android-wikipedia-maven-checkstyle/53/console [08:26:43] hashar: woot! :) [08:26:44] quick afk brb [08:37:32] re [08:50:43] hashar: so if I upload a new patchset it should re trigger I guess? [08:56:52] yuvipanda: yeah or you can look at the job failing and click the [Rebuild] link in the left menu [08:57:03] hashar: yeah, I just did a trivial rebase and pushed them all up :) [08:57:04] that would show a form with all parameters filled, then just click [rebuild] [08:57:07] saves you a new patchset [08:57:14] or you can do a trivial rebase / commit summary edit [08:58:21] hashar: are these checks run parallelly or serially? [08:58:27] just looking for ways to speed it up :) [08:58:28] serially [08:58:31] going to adjust that [08:58:45] ah ok :D [08:59:16] and i found out that the maven template has a typo :)D [08:59:31] hashar: hah! [08:59:32] - tirggers: [08:59:32] + triggers: [08:59:32] - zuul [08:59:39] oh wow. and it still worked [08:59:52] I've the checkstyle errors down to under 200. [08:59:57] yeah because Zuul does not uses the parameters [08:59:59] should get it to zero soon [09:00:02] it is only used when triggering build manually [09:00:06] not sure why it worked though [09:01:53] (03PS1) 10Hashar: Typo for Zuul triggers in '{name}-maven-checkstyle' [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125106 [09:01:56] (03PS1) 10Hashar: Make '{name}-maven-checkstyle' concurrent [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125107 [09:02:10] yuvipanda: ^^^:-] [09:02:28] woo! [09:02:47] bah apps-android-wikipedia-tox-flake8 is blocked :( [09:02:50] https://integration.wikimedia.org/zuul/ [09:03:22] hashar: oh? is that why jenkins-bot isn't voting? [09:03:29] yeah [09:03:33] the jobs are not being triggered [09:03:48] that has happened from time to time since ~ 3 weeks [09:03:48] hmm [09:03:51] no clue what is happening [09:03:57] what's the fix? [09:04:12] need to diagnosse it [09:04:23] :( [09:31:24] (03CR) 10Addshore: [C: 031] ":D" [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125106 (owner: 10Hashar) [09:31:36] (03CR) 10Addshore: [C: 031] ":D" [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125107 (owner: 10Hashar) [09:50:02] I am off for lunch, will be back later [09:50:17] (03CR) 10Hashar: [C: 032] "Deployed" [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125106 (owner: 10Hashar) [09:50:27] (03CR) 10Hashar: [C: 032] "Deployed" [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125107 (owner: 10Hashar) [09:51:09] (03Merged) 10jenkins-bot: Typo for Zuul triggers in '{name}-maven-checkstyle' [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125106 (owner: 10Hashar) [09:51:12] (03Merged) 10jenkins-bot: Make '{name}-maven-checkstyle' concurrent [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/125107 (owner: 10Hashar) [11:04:36] Zeljkof:Hi! I had installed SulimeText ..was wondering is it an IDE or more of a text editor [11:04:46] Jagori: sorry, in a meeting with Tobi_WMDE [11:05:07] Zeljkof:Ok sorry [11:07:44] Jagori: it is pretty much a text editor [11:08:13] Yuvipanda:Thanks..yes:) [11:08:51] Yuvipanda:Since I had been using RubyMine for a while. are there any open source IDE too [11:09:15] Jagori: I usually just use Vim :D [11:09:22] :) Ohh.. [11:09:57] Yuvipanda:thanks..no issues..then I think Imay use Sublimetext only..notepad ++ is also a good editr [11:10:19] thanks :) [11:14:01] Jagori: :) [11:14:05] Jagori: good luck! [11:23:37] (03PS1) 10Zfilipin: If environment variable HEADLESS is set to true, run a local browser [selenium] - 10https://gerrit.wikimedia.org/r/125169 [11:48:53] zeljkof: just saw that if I turn a pageObject into a module I cannot do things like on(SpecialModifyEntityPage)... anymore. it would give me "undefined method `new' for SpecialModifyEntityPage:Module (NoMethodError)" [11:49:03] any idea? [11:49:13] Tobi_WMDE: hm [11:49:21] can you push it to gerrit? [11:49:24] so I could take a look [11:49:32] can do [11:49:34] one moment [11:50:41] (03PS1) 10Zfilipin: Fixed link to Wikibase Jenkins in readme file [selenium] - 10https://gerrit.wikimedia.org/r/125171 [11:54:39] zeljkof: https://gerrit.wikimedia.org/r/#/c/125172 [11:55:23] (03CR) 10Tobias Gritschacher: [C: 031] Fixed link to Wikibase Jenkins in readme file [selenium] - 10https://gerrit.wikimedia.org/r/125171 (owner: 10Zfilipin) [11:56:13] Tobi_WMDE: looking [11:56:23] zeljkof: any objections in giving me +2 rights in mediawiki/selenium repository [11:56:38] Tobi_WMDE: I thought you already have them :) [11:56:53] I am not the one that gives the permission [11:57:02] hashar should be able to do it [11:57:18] probably somebody else too, but not sure who [11:57:30] can you create a bug in bugzilla and assign it to hashar? [11:57:38] (for +2 on the gem) [11:58:48] can do [12:00:05] zeljkof: what's the right component for that bug? [12:00:11] :) [12:00:15] probably wikimedia-qa [12:00:25] ok [12:00:34] it is important that it is in bugzilla, hashar uses it as a todo list :) [12:00:41] he can change the component if needed [12:00:56] feel free to add me to cc, I will give my +1 on the bug [12:02:35] zeljkof: you were added automatically [12:02:58] Tobi_WMDE: yes, now I remember :) [12:06:42] Tobi_WMDE: I see the problem [12:06:50] in the commit [12:07:09] zeljkof: yes? [12:07:21] yes, when you want to do on(SpecialModifyEntityPage), that calls method #new on SpecialModifyEntityPage [12:07:34] and classes have that method defined by default, modules do not [12:07:56] zeljkof: so I just add it? [12:08:00] what should it do? [12:08:03] nothing? [12:08:21] I am not sure would adding #new method work, who knows what page-object gem expects it to do :) [12:08:34] is there a reason you have converted all classes to modules? [12:08:58] can you just copy/paste the code from the module to the class? and delete the module? [12:09:04] is the module included in several classes? [12:10:00] zeljkof: yes they will be included in several classes [12:10:36] the reason why I converted them is to no longer use inheritance but inclusion [12:10:48] if you need the module and the class with the same name [12:10:50] but I can live with having them as classes [12:11:05] you can create SpecialModifyEntityPage and SpecialModifyEntityModule [12:11:11] pull all the code into the module [12:11:18] zeljkof: yeah, was thinking of this [12:11:20] and include the module in all the classes that need it [12:11:38] not sure if this is nicer than just having classes [12:11:44] SpecialModifyEntityPage can be almost empty, except for including page object and the module [12:11:56] Tobi_WMDE: depends :) [12:12:13] if you want to ride with the pack, we usually use composition, not inheritance [12:12:25] but you are free to ride alone :) [12:13:23] hehe [12:13:40] I'll think over it. currently trying to get the headless thing working [12:16:45] (03CR) 10KartikMistry: [C: 032] "LGTM" [selenium] - 10https://gerrit.wikimedia.org/r/125171 (owner: 10Zfilipin) [12:16:47] (03Merged) 10jenkins-bot: Fixed link to Wikibase Jenkins in readme file [selenium] - 10https://gerrit.wikimedia.org/r/125171 (owner: 10Zfilipin) [12:38:19] zeljkof: after running bundle update (and updating to cucumber 1.3.14 I no longer can run specific tests by specifying the feature file - I'm getting "No such file or directory - features/special_set_label.feature. Please create a features/special_set_label.feature directory to get started." [12:38:39] do you know something about that? [12:38:45] no :) [12:38:56] did you read release notes? [12:39:02] maybe they have changed something [12:39:10] zeljkof: sorry [12:39:13] was my mistakle [12:39:17] arr [12:39:21] too confused today [12:39:32] was unrelated to the update [12:39:58] :) [12:54:07] hashar: where is node: hasAndroidSdk defined? [12:54:20] ohh [12:54:26] I am not sure Android SDK is working anymore [12:54:35] yeah, I think the builds have been broken for a while [12:54:40] http://integration.wikimedia.org/ci/label/hasAndroidSdk/ [12:54:40] is there any way I can help set one up on labs? [12:54:50] that is only for the job https://integration.wikimedia.org/ci/job/apps-android-commons-build/ [12:54:56] and only the gallium slave has it [12:55:17] gotta find a way to install the android SDK properly [12:56:20] hashar: hmm, would that have to be puppetized? [12:56:35] ideally [12:56:36] or packaged [12:56:57] last time I checked (several months ago) there was no Debian package though :( [12:57:11] hashar: yeah, there are some ubuntu PPAs [12:57:14] hashar: but a debian package I think is unlikely [12:57:18] hashar: a shell script is more doable [12:57:19] oh [12:57:20] https://packages.debian.org/source/sid/androidsdk-tools [12:57:43] waat [12:57:54] oh wow [12:57:56] and Ubuntu Trusty has http://packages.ubuntu.com/trusty/android [12:58:10] zeljkof: I managed to fix https://gerrit.wikimedia.org/r/#/c/125172/ [12:58:15] * hashar assigns bug to yuvipanda :-] [12:58:35] yuvipanda: you could try out the Ubuntu Trusty android package. Maybe it can be installed on Precise [12:58:38] hashar: it doesn't actually have support for sdkmanager :( [12:58:44] else we will have to backport it (that might be possibly) [12:58:46] * yuvipanda checks trusty [12:59:16] hashar: I can play around with it on a labs instance. Should I do that on mobile project or contint? [12:59:41] probably easier on mobile [12:59:48] Tobi_WMDE: looks good [12:59:56] try to figure out whether the package can be installed and if you can use what it provides to build your apps [13:00:13] + the commands involved (probably just need maven package or something and it would find the androidSDK provided by the package) [13:00:21] hashar: could you please take a look at https://bugzilla.wikimedia.org/show_bug.cgi?id=63761 [13:00:28] if that works fine (which I doubt) we can ask ops to publish the androidsdk package on apt.wikimedia.org [13:00:44] then get it installed on the labs Jenkins slaves and have the maven job to run there [13:01:34] zeljkof: oh [13:01:55] zeljkof: aren't you a project owner for mediawiki/selenium ? :-] [13:02:07] * hashar points at https://gerrit.wikimedia.org/r/#/admin/projects/mediawiki/selenium,access [13:02:38] hashar: ah, right. I'll create something on mobile and try it out [13:03:07] hashar: I wonder if I should just setup a cron on toollabs. Should be rather trivial to do there, since there's no packaging requirement [13:03:17] zeljkof: https://bugzilla.wikimedia.org/show_bug.cgi?id=63761#c2 [13:03:17] and I can have it do it once a day or so which is 'good enough' [13:03:37] zeljkof: just add tobias to the mediawiki-selenium group and he will be a owner and probably get +2 / merge rights as a result [13:04:06] yuvipanda: if the android package from trusty works we can get it installed on labs slaves [13:04:17] hmm [13:04:23] yuvipanda: then you can use jenkins which would provide you with IRC notifications/ Zuul triggering / dashboards / check style whatever :] [13:04:28] :D [13:04:32] at the expense of a bit of effort :] [13:04:44] but probably less than setting up a cron based continuous integration system hehe [13:05:16] hashar: created an instance under mobile to test it out. let's see how it goes :) [13:05:18] yuvipanda: can you bug fill it so we can track the progress ? [13:05:23] hashar: sure. moment [13:06:49] hashar: https://bugzilla.wikimedia.org/show_bug.cgi?id=63765 [13:08:42] hashar: brb, sunset time [13:10:48] * hashar today I learned indian people stop working during sunset [13:30:05] hashar: when I open the page you have linked to in the bug, "members" text field is disabled :( [13:30:11] I can not enter new people there [13:30:27] bah [13:30:28] link? [13:30:43] (cleared backlog and closed bug sorry) [13:30:52] hashar: https://bugzilla.wikimedia.org/show_bug.cgi?id=63761 [13:31:12] https://gerrit.wikimedia.org/r/#/admin/groups/600,members [13:31:12] hmm [13:31:18] maybe you dont have the right to edit that group [13:31:25] could be [13:31:29] belong to "Project and Group Creators" [13:31:31] sorry [13:32:30] added Tobias there [13:32:37] hashar: thanks [13:32:38] he should get +2 and basically any rights on that repository [13:32:41] sorry :-( [13:32:43] great [13:32:52] I am able to add new people or not? [13:32:52] I thought members of a group could add members hehe [13:32:56] na you can't [13:32:58] I see [13:33:05] ok, we do not have to do that often [13:33:07] need to be a member of "Project and Group Creators" [13:33:14] i.e. more or less a Gerrit bureaucrat or something [13:33:15] it only happened a couple of times [13:33:27] will be happy to address such requests :] [13:33:39] is there a reason kartik does not automatically have +2 on mediawiki/selenium repo? [13:33:41] either via bug if you CC me or by IRC ping hehe [13:33:48] since he is wmf [13:34:01] he must be using an email which is not @wikimedia.org [13:34:07] hence we have to whitelist him in zuul [13:34:12] I see [13:34:29] ^kartik\.mistry@gmail\.com$ [13:34:34] apparently he is whitelisted [13:34:51] hashar: he is a member of the group now [13:34:58] nevermind, just a thought [13:34:58] \O/ [13:35:05] our tooling is strange sometimes [13:35:07] you can close https://bugzilla.wikimedia.org/show_bug.cgi?id=63765 [13:35:13] hashar: will do [14:03:46] who is the right person to add to a mediawiki ruby api change? [14:04:14] zeljkof: http://wdjenkins.wmflabs.org/ci/job/wikibase-browsertests-smoke/14/console [14:04:33] thats waht I'm getting when trying to run a test in HEADLESS firefox on our jebnkins [14:04:38] manybubbles: me :) [14:04:46] k. cool. already done then:) [14:04:51] it seems the headless firefox cannot reach beta [14:05:02] manybubbles: Juliusz too [14:05:27] Tobi_WMDE: ruby-2.1.1 is not installed. [14:05:42] zeljkof: that's unrelated I guess [14:05:48] we're using 1.9.3 [14:06:01] zeljkof: do you know Juliusz's gerrit account name? [14:06:15] manybubbles: something jgonera I think [14:06:28] thanks! done [14:06:29] Tobi_WMDE: are you sure you have the correct gems installed/used? [14:06:54] zeljkof: I think so.. you see the list of gems in the console output [14:08:10] zeljkof: the request times out [14:08:32] if there were a problem with ruby or the gems it would not go so far [14:09:12] I think I remember something similar when I tried phantomjs back then [14:09:27] Tobi_WMDE: can you try with phantomjs? [14:09:37] just to make sure it is not related to headless gem [14:09:57] also, why don't you change ruby version in Gemfile to the one you use? [14:10:05] just to make sure that is not the problem [14:10:29] zeljkof: should do that [14:10:30] :) [14:10:37] I think you've put it there [14:13:11] Tobi_WMDE: I thought you will use the new and shiny version, not the old and rusty one :) [14:13:30] zeljkof: yeah.. guess you're right [14:13:32] :) [14:13:49] I'm using the old and rusty one on my machine that's why I installed it on jenkins too [14:14:17] Tobi_WMDE: I have tested it recently, ruby 2.0 and cucumber now work on windows [14:14:18] zeljkof: this is with phantomjs: http://wdjenkins.wmflabs.org/ci/job/wikibase-browsertests-smoke/15/console [14:14:20] try it [14:14:21] same thing [14:14:35] you can upgrade to the newest and greatest ruby now [14:14:35] zeljkof: you think it's related? [14:14:43] no, just random thought [14:16:58] zeljkof: and that's the same test when using sauce: http://wdjenkins.wmflabs.org/ci/job/wikibase-browsertests-smoke/16/console [14:17:01] it's passing [14:17:54] hashar: Tobi_WMDE is having problems with jenkins ^ [14:18:20] can their jenkins access the intertubes? [14:18:36] ha [14:19:03] Tobi_WMDE: maybe sauce works because their browsers are on the internet, and your jenkins is maybe behind a firewall or something [14:19:33] zeljkof: surprisingly our jenkins can access beta when we're doing api calls to wikidata.beta with ruby (to e.g. create items) [14:19:46] ruby-2.1.1 is not installed. [14:19:46] To install do: 'rvm install ruby-2.1.1' [14:19:53] zeljkof: it's just the browser that cannot access it [14:20:09] hashar: I think this is not the problem [14:20:14] what is the browser trying to reach ? [14:20:21] is that the beta cluster wikis? [14:20:50] hashar: http://wikidata.beta.wmflabs.org/wiki/Special:SetLabel [14:20:57] yes [14:21:04] ah [14:21:24] so here is the deal. Your labs instance wdjenkins.wmflabs.org query the DNS for wikidata.beta.wmflabs.org [14:21:31] that gives it a public IP address [14:21:41] but that public IP is NATed to an instance IP [14:22:02] and when you try to reach that public IP from your wdjenkins instanceIP, that is unsupported by the network stack [14:22:09] it basically has no clue what to do with it [14:22:27] hashar: but when ruby makes calls to http://wikidata.beta.wmflabs.org/w/api.php it works fine [14:22:39] the evil hack of doom is to add some iptables rules on the wdjenkins to rewrite the destination IP from a public IP to a private IP. [14:22:41] taht's what we're doing since a long time [14:22:52] hashar: we have that hack already in the iptables [14:22:52] ohh [14:23:04] another possibility is to use a web proxy [14:23:18] it's just that it seems the headless firefox and phantomjs cannot reach beta [14:23:21] but I am not sure whether we have any web proxy reachable from labs [14:23:29] :-/ [14:24:26] hashar: I'm trying something [14:25:03] on wdjenkins can you try: curl http://wikidata.beta.wmflabs.org/w/api.php [14:25:07] just to confirm [14:25:17] as for phantomjs maybe it is set to use a web proxy that has the same issue? :-( [14:25:18] hashar: curl works fine [14:25:20] I tried that [14:25:21] good [14:25:26] so the iptables works \O/ [14:25:30] yes [14:26:40] hashar: ok, I just tried in the console [14:27:00] opening google.com with phantomjs works [14:27:06] opening http://wikidata.beta.wmflabs.org/wiki/Special:SetLabel times out [14:28:08] would need to figure out what phantomjs is doing exactly [14:28:27] hashar: same with firefox [14:29:18] hashar: arn't you using phantomjs also on wmf jenkins? [14:29:38] to run the qunit tests but they are running against a local virtualhost [14:29:46] hashar: ok [14:29:54] and I did work a bit on having browser tests but they hit the local virtual host as well :-( [14:30:00] maybe tcpdump would help ? :D [14:30:47] hashar: hmm [14:31:45] or find out whether a http proxy is set with the snippet https://github.com/ariya/phantomjs/blob/master/examples/printenv.js [14:31:58] though unlikely [14:33:54] doesn't seem like [14:37:10] * zeljkof is out to lunch [14:39:50] https://www.irccloud.com/pastebin/NBsXpCDb [14:39:55] hashar: ^^ [14:40:06] that's what phantomjs is doing [14:41:26] so at least packets are being exchanges with the text cache which serves wikidata.beta.wmflabs.org [14:41:32] that is deployment-cache-text02.eqiad.wmflabs [14:41:47] maybe the logs on beta cluster got more informating, digging in [14:43:01] yeah I see them just fine [14:43:10] hashar: means? [14:43:12] X-FORWARDED-FOR: 10.68.16.79, 127.0.0.1 [14:43:35] maybe it has trouble to login? [14:43:47] hashar: login where? [14:43:56] it should not be necessary to login [14:45:09] it shoud just open http://wikidata.beta.wmflabs.org [14:45:16] no need to login for that [14:45:38] I got some GET /wiki/Special:SetLabel [14:45:46] ok [14:46:12] what happened to them then? [14:47:28] Tobi_WMDE: an example http://paste.openstack.org/show/75481/ [14:48:03] and there is one after doing Start request GET /wiki/Special:SetLabel?setlang=en [14:49:13] hashar: yeah, guess the one with the setlang is from the actual browsertest [14:49:30] while the other one is from my manual try with phantomjs in the console [14:49:55] but it seems ok. so why is the request then timing out? [14:50:10] and from the apache log [14:50:10] http://paste.openstack.org/show/75482/ [14:50:27] do you have access on the beta cluster ? [14:50:35] might be user for you to tail the log while the job is running :] [14:51:30] the apache log is probably not relevant.. that could also be from one of the jobs that run on aucelabs.. they're fine [14:52:03] I can just try again with phantomjs [14:52:04] !log deployment-prep Adding Tobias Gritschacher to the project so he can look at udp2log / apache logs whenever needed :-] [14:52:38] and then we should have a look at the logs [14:52:48] to be sure to look at the right time [14:53:11] hashar: ok, request is running [14:53:18] zeljkof: I am running a little behind right now, figuring out if a Mobile test failure is a bug or not... [14:53:47] can you look at the logs again? [14:54:36] hashar: request still pending.. think it times out after 1 minute or so [14:55:17] -:( [14:55:39] hm.. timeout [14:56:01] is there anything in the logs now regarding /wiki/Special:_SetLabel [14:56:04] is there anything in the logs now regarding /wiki/SpecialSetLabel [14:56:10] ? [14:56:22] within the last 2 minutes? [14:56:49] [10/Apr/2014:14:37:32 +0000] "GET /wiki/Special:SetLabel HTTP/1.1" 200 5181 xfp: - [14:56:50] [10/Apr/2014:14:44:26 +0000] "GET /wiki/Special:SetLabel HTTP/1.1" 200 5181 xfp: - [14:57:06] I gave you access to the project. You can probably login on deployment-bastion.eqiad.wmflabs now [14:57:23] the set label seems to be working on apache side [14:58:02] hashar: the entries are to old.. they should be around minute 50 or so [14:58:25] 14:44 is like over 190 minutes ago [14:58:31] I mean 10 [14:59:47] ah time is UTC [14:59:55] yeah yeah 10 [15:00:27] hashar: me trying was at 55 or something [15:00:40] and right now again [15:00:50]