[02:27:38] !log LocalisationUpdate completed (1.20wmf4) at Wed Jun 13 02:27:37 UTC 2012 [02:27:46] Logged the message, Master [02:52:38] !log LocalisationUpdate completed (1.20wmf5) at Wed Jun 13 02:52:38 UTC 2012 [02:52:42] Logged the message, Master [04:26:31] liangent: is that still broken for you? [04:26:40] 12 17:55:10 < liangent> `host upload.wikimedia.org` works, `curl upload.wikimedia.org` returns "curl: (6) Couldn't resolve host 'upload.wikimedia.org'" [04:28:40] Ryan_Lane1: see [[labsconsole:Help talk:Access]]. i'm sure I triple checked the behavior first hand at the time. I don't remember exactly if I was logged in or not though. ;( [05:37:41] jeremyb: yes and finally I put it in hosts to work around it [07:33:22] hi [07:43:34] hello and gone, back in a while with more memory [07:43:55] hmm [07:44:03] looks like all engineers are upgrading their desktop RAM [07:44:11] I've had it in my queue for a month [07:44:12] er laptop RAM [07:44:15] hehe [07:44:22] finally got to it today (buying it) [07:44:32] I don't even know how much RAM I have [07:44:48] 4GB !!! [07:45:04] well what set it off today was waiting for swap while I was trying to test dumps at th same time I wanted to look at a message about an external store [07:45:15] kinda the last straw, having to sit and wait.... [07:45:32] ok talk to you in a bit (going to also get milk and stuff while I wait for poor lappy to cool off a bit) [07:45:46] have a good shopping! [07:45:59] will probably have a nap this morning anyway [08:04:50] ok I was impatient... woo hoo, back with 8gb finally! [08:04:57] and nw off for milk and other items [08:28:39] !log hashar synchronized wmf-config/InitialiseSettings.php '(bug 37482) Adding Proofread Page ext. namespaces on nl.wikisource' [08:28:43] Logged the message, Master [08:33:57] !log hashar synchronized wmf-config/InitialiseSettings.php '(bug 37482) Adding Proofread Page ext. namespaces on nl.wikisource' [08:33:59] the more I use our cluster [08:34:02] Logged the message, Master [08:34:06] the more I find it f**** up [08:47:17] !log hashar synchronized wmf-config/InitialiseSettings.php 'bug 37006 - fawiki: add Book namespace + aliases' [08:47:21] Logged the message, Master [09:05:12] !log hashar synchronized wmf-config/mobile.php [09:05:17] Logged the message, Master [09:05:29] !log hashar synchronized wmf-config/mobile-wmflabs.php [09:05:34] Logged the message, Master [09:05:48] !log hashar synchronized wmf-config/throttle.php [09:05:52] Logged the message, Master [09:06:19] ohh [09:06:20] !log hashar synchronized wmf-config/InitialiseSettings-wmflabs.php [09:06:20] sync-dir [09:06:24] Logged the message, Master [09:07:52] !log hashar synchronized wmf-config [09:07:57] Logged the message, Master [09:08:32] !log finished deploying my wmflabs related change. mediawiki-config is now at commit c0baf3e [09:08:37] Logged the message, Master [09:08:53] seems to work so far [09:24:12] !log hashar synchronized wmf-config/InitialiseSettings.php '(bug 37482) Adding Proofread Page ext. namespaces on nl.wikisource' [09:24:17] Logged the message, Master [14:57:47] @jeremyb [14:58:18] yellow [14:58:35] so, besides the radio default what else do you want? [14:58:58] also, the word is bug 37544 can be done @ TWN [15:00:21] reza1615 [15:00:32] hi [15:00:38] 13 14:58:35 < jeremyb> so, besides the radio default what else do you want? [15:00:41] 13 14:58:58 < jeremyb> also, the word is bug 37544 can be done @ TWN [15:00:59] I want to change [15:01:05] the defualt to [15:01:11] namse book [15:01:14] 102 [15:01:24] ویکی‌پدیا:کتاب‌ها [15:01:34] changed to کتاب: [15:01:51] i can't do unicode on IRC [15:02:08] or at least the 2nd line you wrote was fine but first was garbage [15:02:11] book; [15:02:15] book: [15:02:29] erm?! there's a semicolon? [15:02:35] no [15:02:41] wait [15:03:27] instead of [15:03:28] http://fa.wikipedia.org/w/index.php?title=%D9%88%DB%8C%DA%98%D9%87:%D9%86%D9%85%D8%A7%DB%8C%D9%87_%D9%BE%DB%8C%D8%B4%D9%88%D9%86%D8%AF%DB%8C&prefix=%D9%88%DB%8C%DA%A9%DB%8C%E2%80%8C%D9%BE%D8%AF%DB%8C%D8%A7:%DA%A9%D8%AA%D8%A7%D8%A8%E2%80%8C%D9%87%D8%A7/ [15:03:58] please set it to کتاب: [15:08:27] reza1615: i still don't understand ;( [15:08:40] reza1615: what langs do you speak best? [15:08:41] now the second [15:08:54] radio lable is set to wikipedia namespace [15:09:06] I want to set it too book name sapce 102 [15:09:48] farsi and english [15:15:18] reza1615: change this message: https://en.wikipedia.org/w/index.php?title=Special%3AAllMessages&prefix=Coll-community_book_prefix&filter=all&lang=en&limit=50 [15:15:37] https://fa.wikipedia.org/w/index.php?title=%D9%88%DB%8C%DA%98%D9%87:%D8%AA%D9%85%D8%A7%D9%85_%D9%BE%DB%8C%D8%BA%D8%A7%D9%85%E2%80%8C%D9%87%D8%A7&prefix=Coll-community_book_prefix&filter=all&&limit=50 [15:15:54] to "Book:" [15:16:01] or, rather [15:17:34] "کتاب:" [15:17:37] http://translatewiki.net/wiki/MediaWiki:Coll-community_book_prefix/fa [15:17:41] no [15:17:45] not at translatewiki [15:17:54] on wiki? [15:18:08] fawiki [15:20:29] ok I have admin access with my bot [15:20:44] so i asked other admin to do this would please check [15:21:33] https://bugzilla.wikimedia.org/show_bug.cgi?id=30743 [15:21:37] ok. what else? radio button? and? [15:22:06] I got a question: How do I export a Wikimedia Installation as fast as possible= [15:22:21] to HTML / Plain-Text (Mobile) [15:22:46] FlySoftiii: you mean mediawiki? [15:23:04] Doh, of course I mean mediawiki. [15:23:11] so #mediawiki :) [15:23:46] @jeremyb:https://bugzilla.wikimedia.org/show_bug.cgi?id=30743 [15:23:52] Okay... Why diddn't I notice that...? [15:23:56] Sorry! [15:24:36] reza1615: in the future http / https needs a space or < or something before it (not a colon and another word) [15:24:47] or just !b 30743 | jeremyb [15:24:50] !b 30743 | jeremyb [15:24:50] jeremyb: https://bugzilla.wikimedia.org/show_bug.cgi?id=30743 [16:47:51] Hi [16:48:55] On http://bindery.wikimedia.org/, are the mwlibrl languages files up to date (relevant file should be mwlib/rl/locale/fa/LC_MESSAGES/mwlib.rl.po)? [16:49:23] We got a translation issue on https://bugzilla.wikimedia.org/show_bug.cgi?id=37544 but this is already translated upstream. [16:55:39] Dereckson: it's not managed by us [16:55:55] As per the bug, it's pdf[1-3].wikimedia.org now [16:56:04] Got to wait until pediapress deal with it [16:57:09] which is usually, whenver they feel like it [16:57:28] it's already 5 months it's translated in their git repo [16:58:14] lol [16:58:29] Best you can do is to poke Tomasz and ask him to poke Pediapress [16:59:34] Reedy: power restored, or have you just wired your computer up to a bicycle generator? [17:00:15] It's restored, and been stable for today [17:00:20] Okay Reedy, thank you. [17:03:54] robla: Has Chris' baby arrived yet? Just wondering if I'm likely to get an email reply [17:05:44] Reedy: I haven't heard anything, which probably means he has [17:06:22] I wasn't sure, as it could easily be either way, no news -> no email or no news -> very busy [17:07:23] I suspect as he didn't reply yesterday, that's the case. I just need to kick Roan to publish that draft and get the tarballs finished (I noticed reviewing the patches that a couple of cherry pick revs weren't in the branches) [17:08:41] he's been pretty good at keeping me up to date on "not yet", so I'm guessing they're all at the hospital [17:10:14] Reedy: when on Thursday do you want to deploy? Looks like 17:00-20:00 UTC are taken, so maybe earlier? [17:10:27] (Ed Ext, that is) [17:21:49] * robla makes something up [17:36:36] !log preilly synchronized php-1.20wmf4/extensions/ZeroRatedMobileAccess 'changes for zero needed for carrier testing' [17:36:41] Logged the message, Master [17:37:12] !log preilly synchronized php-1.20wmf5/extensions/ZeroRatedMobileAccess 'changes for zero needed for carrier testing' [17:37:16] Logged the message, Master [18:06:46] Reedy: one change to backport before we get too far: https://gerrit.wikimedia.org/r/#/c/11083/ [18:07:00] (brb) [18:09:30] robla: not too bad, just need to recreate the tarball [18:11:43] Just need to get chad to delete the 1.18.4 tag for me [18:12:28] you can't just force it? [18:13:49] Reedy: we've the the non-WP deployment to start now [18:14:00] yup [18:14:10] jeremyb: no [18:14:10] ! [remote rejected] 1.18.4 -> 1.18.4 (can not update the reference as a fast forward) [18:14:10] error: failed to push some refs to 'ssh://gerrit.wikimedia.org:29418/mediawiki/core.git' [18:14:31] !log aaron synchronized php-1.20wmf5/extensions/FlaggedRevs 'deployed 537bb248bb93948844f195014227512f169a439b' [18:14:34] robla: for 1.17 and 1.18 it's ./make-release 1.17.5 1.17.4 && ./make-release 1.19.1 1.19.0 and then wait. 1.18 needs more poking, so it can wait ;) [18:14:36] Logged the message, Master [18:14:48] so we can do the deployment like now [18:14:52] cool [18:14:57] * Reedy opens a shell to fenari [18:16:11] (btw, the backport of gerrit 11083 is from master->1.20wmf5, in case that wasn't clear) [18:16:23] oh [18:16:30] right [18:18:39] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: special.dblist wikis to 1.20wmf5 [18:18:44] Logged the message, Master [18:22:04] Vector sync incoming [18:22:28] !log reedy synchronized php-1.20wmf5/extensions/Vector/ [18:22:33] Logged the message, Master [18:23:16] commons and the rest of special wikis are now on 1.120wmf5 [18:27:01] no obvious breakage [18:27:26] I say onward [18:28:58] 1 swift warning appearing, but that's nothing new [18:30:47] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikiversity to 1.20wmf5 [18:30:51] Logged the message, Master [18:31:58] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wiktionary to 1.20wmf5 [18:32:03] Logged the message, Master [18:32:48] that one looks fine. Next? [18:33:23] I'd recommend just doing the whole rest of the pile at once, and backing out if necessary [18:33:43] i'll do that next then [18:33:56] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikiquote to 1.20wmf5 [18:34:01] Logged the message, Master [18:35:17] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: Everything non wikipedia to 1.20wmf5 [18:35:22] Logged the message, Master [18:35:46] No other new errors either [18:36:46] That's odd, and seemingly unrelated [18:36:55] Where've all the padlocks from https links on mediawikwiki? [18:37:01] https://www.mediawiki.org/wiki/MediaWiki_1.20/wmf5 [18:39:20] Krinkle: ^ [18:39:23] Any ideas? [18:39:30] I'm sure they were there earlier.. [18:39:39] Reedy: I made them go away [18:39:49] Considered internal link now, just like [[bugzilla:]] [18:39:59] oh [18:40:15] they are meant for external sites, I think we all know that going to gerrit is trusted :P [18:40:57] Fatal error: Call to a member function getThreadObj() on a non-object in /usr/local/apache/common-local/php-1.20wmf5/extensions/LiquidThreads/pages/ThreadHistoricalRevisionView.php on line 111 [18:40:59] That looks new [18:41:37] Good luck with that ;P [18:41:46] Logged for werdna to ignore ;) [18:41:58] I think translatewiki's reporter has a string search for LiquidThreads even [18:42:05] (to surpress) [18:42:07] or had [18:42:13] $this->mDisplayRevision = ThreadRevision::loadFromId( $oldid ); [18:42:13] $this->thread = $this->mDisplayRevision->getThreadObj(); [18:42:14] I wonder [18:42:26] yup, lol [18:42:27] if ( !$row ) return null; [18:42:51] Getters/Setters ftw :) [18:43:06] robla: nothing to even care about in the error logs [18:43:36] w00t [18:45:30] * robla goes to grab some lunch downstairs [19:27:23] !log aaron synchronized php-1.20wmf4/includes/WikiPage.php 'temporary logging code.' [19:27:28] Logged the message, Master [19:37:42] !log aaron synchronized php-1.20wmf4/includes/WikiPage.php [19:37:47] Logged the message, Master [19:43:54] !log aaron synchronized wmf-config/CommonSettings.php 'Added debug log.' [19:43:59] Logged the message, Master [19:57:30] !log aaron synchronized php-1.20wmf4/includes/WikiPage.php [19:57:34] Logged the message, Master [20:31:36] Reedy: https://en.wikipedia.org/w/index.php?title=Special%3ALog&type=merge I think that's http://www.mediawiki.org/wiki/Help:Merge_history - but was there a discussion for enabling that? [20:32:28] I've no idea [20:32:53] No one has the right on enwiki [20:33:07] https://en.wikipedia.org/wiki/Special:ListGroupRights [20:33:32] yeah, it's on meta too and mediawiki [20:33:55] so I was wondering if you knew anything about it as release manager - maybe it came through in 1.20wmf4? [20:34:45] Again, no one on mediawiki.org has the right [20:34:48] it's not in any groups [20:35:06] and meta [20:35:14] So it's permissions have been changed [20:35:18] yeah I know that, but then why do we have it if we're not using it? [20:35:19] or it's just not giving an error message [20:35:22] I've no idea [20:35:34] yeah it must have been hidden before or something [20:36:08] <^demon> Empty logs sure are annoying. So many logs to...not look at. [20:54:07] ^demon: i'm sure we could make you some [21:04:07] can someone tell me what happened here? http://nl.wikipedia.org/w/index.php?title=Engels_voetbalelftal&diff=31454266&oldid=31454262 [21:04:15] no change made... [21:04:55] Romaine: https://bugzilla.wikimedia.org/show_bug.cgi?id=37225 ;) [21:07:38] thanks hoo [21:08:57] You're welcome [21:20:49] !log aaron synchronized php-1.20wmf5/includes/WikiPage.php 'deployed 82742bccf3b5f2da0d5df05630eb31978afbbce1' [21:20:54] Logged the message, Master [21:21:30] binasher: so will we be decommissioning a crapload of mc servers? [21:22:07] mc will no longer be running on any of the servers it currently running on [21:22:18] the servers themselves won't be decom'd though since they're all apaches or job runners [21:22:30] heh, right :p [21:25:12] binasher: how many of these new boxes are we getting? [21:25:23] 16 [21:25:34] lucky ones [21:25:41] AaronSchulz: moar memory! [21:25:46] (to mw php processes) [21:26:41] binasher: can we through the whole parser cache at it? [21:27:44] it should be enough to let us dump the mysql parsercache [21:28:08] db40 is currently using 1.6TB which is more than we'll have, but it hasn't been purged for a while [21:29:00] but the row caching will also demand some (probably comparatively small) space [21:29:34] AaronSchulz: throw [21:29:41] mysql parsercache ?? what are you talking about? [21:30:57] Platonides: see $wgObjectCaches['mysql-multiwrite'] in CommonSettings.php [21:32:13] !log lcarr synchronized wmf-config/mc.php 'replacing broken srv203 with working srv250' [21:32:18] Logged the message, Master [21:32:31] oh, we were using a SqlBagOStuff in wmf now? [21:32:58] he have been for months, just for parser cache [21:33:28] it got the hit rate up, though it's also fairly terrible [21:34:07] I thought we had very little parsercache misses [21:34:21] they were happening more due to editing, not to mc expiration [21:34:34] I remember some numbers about it in the past, though [21:34:37] maybe when it changed [21:34:48] no, it went up from like 40 to 80% or so, though it it's not quite as high as it peaked for some reason [21:35:41] !log reedy synchronized wmf-config/InitialiseSettings.php 'Bringing in numerous changes merged via gerrit' [21:35:45] Logged the message, Master [21:35:47] Platonides: http://www.mediawiki.org/wiki/Disk-backed_object_cache [21:36:44] I remember that Tim studied several alternatives months ago [21:37:01] but didn't remember an outcome of using mysql parsercache table [21:37:58] yep, i remembered jsut the ehcache [21:39:34] binasher: how long do think we can go without sharding the s1 master? :) [21:40:07] til 2011 :) [21:40:11] meh [21:41:00] we're going to have ssd based enwiki db's within a few months under the assumption that it won't happen any time soon [21:41:33] but that we'd still like to do things like multiple content-type support within the same page name space [21:41:42] I'm hoping the SELECT load while drop in wmf6 [21:44:37] AaronSchulz: wmf6 has the page latest via slaves change? [21:45:20] if it gets merged, it didn't make it to wmf5 [21:49:16] !log reedy synchronized wmf-config/ 'More changes from gerrit' [21:49:21] Logged the message, Master [21:56:27] binasher: I'm afraid to ask how much those drives would cost [21:56:52] all of the ssds? [21:57:01] ALL THE SSDS [21:57:13] orly [21:58:07] AaronSchulz: i think the servers with them will only come out to a few $k more than normal [21:58:37] I tried to make a case to Danese that we needed these as developer machines [21:58:42] She wasn't having any of it :( [22:01:36] AaronSchulz: Leslie merged the sync script backlogs :D [22:02:29] and here I said she was "no fun" [22:03:15] you said that? [22:03:22] :'-( [22:05:03] LeslieCarr: I WILL NEVER DOUBT YOU AGAIN [22:20:51] binasher: are you looking at https://bugzilla.wikimedia.org/show_bug.cgi?id=34997 ? [22:21:34] yes [22:22:12] API.upload p90 is uninteresting, but p99 is [22:22:25] the stash operations, even at their slowest, don't seem to be the problem [22:26:28] * AaronSchulz suspects a JS UI issue [22:31:39] possible [22:31:55] might be a network issue [22:34:02] yeah [23:06:04] gn8 folks [23:24:56] will I collide with anyone if I deploy a fix now? [23:25:20] MaxSem: is it scheduled? [23:25:31] is there anything on the calendar or on the wiki? [23:26:37] no, I've been asked to deploy a fix quickly [23:27:12] by no you mean you checked all those things and none of them have a deployment set, or no you didn't check? [23:28:33] no, I didn't check - looks to be clear now [23:28:57] Wednesday 18:00-20:00 UTC (11am-1pm PDT): MediaWiki general deployment window [Sam / Aaron / RobLa] was the last one for today [23:29:14] _-_ [23:29:46] that's my "You know what my answer to that's going to be, and I'm sure you can tell I'm disappointed" look ;) [23:29:47] yeah, we were done ages ago [23:30:07] MaxSem: *always* check the calendar and wiki :) [23:30:34] MaxSem: anyway, it's clear and no one is saying otherwise, so go for it [23:30:39] yeah, I'm a noob:) [23:30:52] Then you can deploy and go to lunch [23:31:05] though it's best to always ask if it's not your window;) [23:37:10] !log maxsem synchronized php-1.20wmf4/extensions/MobileFrontend/ 'Deploying MF fix' [23:37:14] Logged the message, Master [23:39:18] !log maxsem synchronized php-1.20wmf5/extensions/MobileFrontend/ 'Deploying MF fix' [23:39:23] Logged the message, Master [23:39:51] i'm quite happy with asking -- more notifications are always good [23:40:08] * jeremyb wonders what's up with git-deploy