[06:19:24] floody floody [06:20:41] heh [06:24:18] not fun [06:26:46] legokm: ping [06:26:56] hi jeremyb [06:27:04] your connection's broke [06:27:07] pls fix ;) [06:27:12] yeah i just fixed it hopefully [06:27:14] sorry about that :( [06:27:24] legoktm: give Nikerabbit a cookie [06:27:35] !cookie Nikerabbit [06:27:35] To enable cookies for popular browsers, see [06:27:41] sigh. [06:27:42] haha [06:27:46] lol [06:27:54] * legoktm gives Nikerabbit a cookie :) [06:34:58] !b 38638 | Nikerabbit [06:34:58] Nikerabbit : https://bugzilla.wikimedia.org/show_bug.cgi?id=38638 [06:35:11] Bug 38638 - System messages needing rewording or documentation [06:35:28] oki [06:35:46] I was also thinking of creating a tracking bug for system messages to be split/errors to be added [06:35:55] There are a few dozens such requests [06:36:16] But I guess the odds of someone working on them systematically are even lower than for 38638 [06:36:24] (despite the cute number) [06:48:44] oh https://gerrit.wikimedia.org/r/#/c/34495/ [18:07:22] Reedy: with Chad out sick, do you feel comfortable leading the deployment session with csteipp and anomie? [18:08:15] We're a pretty rough crowd :) But if we need to reschedule, it's not a problem for me. [18:10:07] session == talk/demo ? or session == an actual deploy to prod ? [18:12:48] jeremyb- Sort of both. Showing us new people how it's done with a Google Hangout screenshare and narration or something along those lines. [18:13:00] k [18:18:34] Err [18:29:52] Reedy: was that "err" a "no"? [18:30:36] What am I doing? [18:31:40] so.... after you're done with 1.21wmf5, we'd like to deploy TemplateSandbox to test2 wiki. Rather than just do it, we need to figure out how to teach csteipp and anomie how to do it [18:32:07] fyi....I may drop offline a sec and get distracted, since I'm going to be in a meeting [18:56:09] So are we doing the Google Hangout for the deployment? Or not? [18:57:15] I'm still waiting for the localisation cache to push [18:57:19] so I can do it again [18:57:21] and push it again [19:02:10] Hangouts not working for me... are you guys on? [19:02:38] csteipp- "It's taking too long to connect you to this hangout"? Same here. [19:03:55] Reedy: did you do anything on test2 that would break UploadWizard? UW is spinning for me there [19:04:26] It's on 1.21wmf5 [19:04:42] If it's just spinning check your JS console [19:05:46] WFM [19:12:00] where is this Google Hangout for the deployment? not mentioned in http://www.mediawiki.org/wiki/MediaWiki_1.21/Roadmap#Schedule_for_the_deployments [19:12:19] spagewmf- It's in the Engineering calendar for today's deployment [19:14:42] "Pinging RobLa for hangout URL" ... [19:15:39] I didn't realise it was a free-for-all... [19:16:39] Reedy: wasn't really meant to be, but didn't preclude that. I figured Chad would be here, so it'd be ok, but since he's out, we can keep it small [19:16:52] lol [19:16:58] I don't mind [19:17:12] zeljkof and I are both unable to complete a file upload with http://test2.wikipedia.org/wiki/Special:UploadWizard mine is hanging on "Getting file information and previews..." in both FF and Chrome [19:18:02] is trunk UW broken then? [19:18:03] chrismcmahon: anything in your javascript console? [19:18:51] looking... [19:19:29] oh hey looky "Uncaught TypeError: Cannot read property 'latitude' of undefined" [19:19:42] (sorry, multitasking in a skype meeting too) [19:20:33] * robla looks up release notes [19:22:17] based on the error, this commit looks suspicous: https://gerrit.wikimedia.org/r/#/c/34968/ [19:22:51] https://gerrit.wikimedia.org/r/#/c/35143/ is supposed to have fixed it [19:23:36] Reedy: did that get merged pre-branch? [19:23:49] Updated Nov 26, 2012 6:13 PM [19:23:52] I guess that's a no [19:24:14] ah, I should have a free terminal now [19:25:16] by the way, chromedriver (part of selenium project) really does not like when file upload elements have negative margins [19:26:08] so our "upload file" test works fine in firefox, but fails in chrome [19:27:27] Should be fixed now [19:29:29] Reedy: see PM [19:29:36] ah, nm [19:30:02] Reedy: file upload via upload wizard should work now on test2? [19:31:16] yup [19:31:26] might need to wait for js cache to expire [19:33:51] Reedy: any way to force, clear cache? [19:34:43] Reedy: I mean, is there any way to force js cache to expire? would clearing browser cache do it? [19:34:56] or is it something server side? [19:38:08] Reedy: never mind, file upload works no (cc chrismcmahon) [19:38:42] great [19:39:42] chrismcmahon: let me check if the chrome hack for UW test works now :) [19:40:51] chrismcmahon: works! :) [19:42:24] awesome :) [19:42:42] chrismcmahon: submitting to gerrit in a minute [19:42:49] * robla patiently waits for Reedy [19:51:47] anomie: while we're waiting: here's a couple of docs you should be familiar with if you haven't seen them already: http://wikitech.wikimedia.org/view/How_to_deploy_code [19:51:59] http://wikitech.wikimedia.org/view/Wikimedia_binaries [19:52:42] also: http://wikitech.wikimedia.org/view/Heterogeneous_deployment [20:09:22] * jeremyb wonders about joining the hangout too? i can't tell if it's already running or not [20:10:09] * jeremyb waves robla [20:11:01] jeremyb: see PM [20:33:07] Hello. [20:34:29] Hi [20:36:23] Krenair: do you think the release notes rebase headaches are an important factor in people not updating the release notes much? [20:36:45] important = worth mentioning in http://thread.gmane.org/gmane.science.linguistics.wikipedia.technical/65080 or http://lists.wikimedia.org/pipermail/wikitech-l/2012-November/064630.html [20:36:58] Dereckson: did you notify all those wikis with super-bureaucrats in the end? [20:37:13] yes, excepted se.wikimedia [20:38:06] Can we see the pages you posted to? [20:38:13] and why didn't you tell sewikimedia? [20:39:58] I can tell them [23:23:02] RoanKattouw: any idea how to debug a script from bits becoming unresponsive? [23:26:31] pgehres: Hmm, could you give me more info about what's going on? [23:26:55] sadly, Firefox and Chrome don't give much [23:27:01] the issue is consistent on http://meta.wikimedia.org/w/index.php?title=Special:NoticeTemplate&offset=&limit=100 [23:27:29] FF at least blames: http://bits.wikimedia.org/meta.wikimedia.org/load.php?debug=false&lang=en&modules=jquery%2Cmediawiki%2CSpinner%7Cjquery.triggerQueueCallback%2CloadingSpinner%2CmwEmbedUtil%7Cmw.MwEmbedSupport&only=scripts&skin=vector&version=20121112T164603Z:97 [23:29:07] Wow [23:29:14] Looks like it's an infinite loop that's also leaking memory [23:29:18] indeed [23:32:41] That line number is somewhere in jQuery [23:34:54] A helper function for jQuery.clone() [23:35:02] ooh, fun times [23:43:43] pgehres: Alright, so now that my machine is recovering from near-swapdeath ... [23:44:08] It's pretty hard to figure out what exactly is causing this [23:44:20] yeah, awight just had to restart X [23:44:31] Chrome doesn't give me any info, Firefox just gives me a line number but no backtrace [23:44:33] now you see why I asked you [23:44:40] cause I am lost here [23:45:05] Chrome lets me pause execution at arbitrary moments but that's also not useful, because I just end up in some innocent-looking code that changes labels [23:45:26] well, that code is new-ish [23:46:08] is it possible that my label changing code is causing this? [23:48:09] Yes [23:48:10] I got a backtrace just now that confirms that [23:48:10] When I paused, var label (inside the each) contained 89