[01:22:36] Katie Horn 0wns [02:28:42] * jeremyb wonders if there was spme trigger for spage to say that. i think no one was up speaking any more at that point, right? [02:29:57] !log LocalisationUpdate completed (1.20wmf11) at Sat Sep 15 02:29:57 UTC 2012 [02:30:10] Logged the message, Master [02:36:34] @add #wikimedia-otrs [02:45:37] (" ) [05:41:17] whym: sorry for not noticing the CLDR plural news [05:41:29] I made you lose a lot of time I think [05:42:27] but reading the whole https://www.mediawiki.org/wiki/MediaWiki_1.20/wmf11#Core_changes to find such stuff is a bit too much for me [05:52:28] Nemo_bis: sorry for not noticing here too, *I* should have double checked the page! [05:54:36] whym: looking for "plural" on the page would have revealed the commit, but for some reason I assumed the new parser wasn't going to change the actual rules... [05:54:52] and niklas has made an update https://bugzilla.wikimedia.org/show_bug.cgi?id=40251 [05:54:55] yep [05:55:05] I also notified mediawiki-i18n now [05:56:00] of course this all staff meetings in SF are slowing everything down... [06:05:43] Nemo_bis: I remember there is a saying something like "never do a major release before holidays" [06:09:03] hello [06:09:16] This is Butch Bustria from Wikimedia Philippines [06:09:37] We are currently having an outreach project right now on wiki editing [06:09:46] oh, again? [06:09:50] this is really bad timing [06:10:09] you need to give 5-10 days advance warning [06:10:10] Can we request to lift IP address 202.92.128.224 [06:10:18] and 202.92.130.3 [06:10:20] WMPH-Exec8: file a bug [06:10:26] i understood [06:10:35] its spontaneous [06:10:40] i don't think you understand [06:10:44] this is *really* bad timing [06:11:07] in fact i think you're not likely to get it fixed in the next 12 hrs [06:11:15] do you need it after 12 hrs from now? [06:11:25] anyway, file a bug [06:11:50] and tell us the bug # of course [06:12:56] seems to be 14:12 there now [06:13:03] is this a multi-day event? [06:14:36] ok, i'm off. back in 7-8 hrs [06:25:18] Bug 40263 - Request to lift IP address for mass creation - Philippines SFD 2012 [06:25:26] from now until 15:00 UTC [06:26:22] We appreciate your kind action. [06:26:27] WMPH-Exec8: you have no sysop at the event? [06:27:22] no sysop [06:28:57] WMPH-Exec8: en or tl? [06:29:31] both. sorry. wikipedia.org only [06:32:29] WMPH-Exec8: what's your username? [06:32:56] Exec8 [06:33:27] ok [06:33:31] oh, only 24 persons [06:33:35] https://www.mediawiki.org/wiki/Help:Mass_account_creation is for you [06:33:36] yes [06:33:54] you don't even need noratelimit I think [06:34:33] Nemo_bis: i don't follow? [06:35:16] Nemo_bis: tlwiki has no ability for crats to grant accountcreator and there are no current accouncreators [06:35:43] ok [06:35:44] Nemo_bis: easiest is to get an enwiki crat to grant it (since they're covering enwiki anyway) and then use that account [06:35:56] hm [06:36:23] en.wiki has a rule requiring identification for account creators [06:36:30] that must be new [06:36:45] I think they're using an OTRS queue or something [06:36:46] Nemo_bis: frankly i have no problem with just turning down this request outright. too little notice [06:36:58] do we need any intervention from any tl admins? [06:37:02] maybe a temp access won't need it [06:37:05] WMPH-Exec8: no [06:37:06] no, there's a toolserver app. ACC. at least if it's still in use [06:37:15] WMPH-Exec8: please write on the bug which wikis are affected [06:37:22] jeremyb: yes but where does data go? [06:37:33] Nemo_bis: toolserver of course [06:37:40] * Nemo_bis shrugs [06:37:49] is there any not too bureaucratic bureaucrat around? :) [06:37:56] WMPH-Exec8: you need a tl or en admin to show up at the event in person [06:38:08] * jeremyb runs away [06:39:00] Bluemask , jojit_fb and seav of TL are here [06:39:35] they are TL admins [06:39:43] wow [06:39:48] wowwww [06:39:51] easy tehn [06:39:56] just follow https://www.mediawiki.org/wiki/Help:Mass_account_creation [06:40:12] 15 06:26:27 < jeremyb> WMPH-Exec8: you have no sysop at the event? [06:40:12] 15 06:27:22 < WMPH-Exec8> no sysop [06:40:16] been to that site [06:40:26] i'm going to stab someone... [06:40:28] at our last WMIT event I had a tablet logged in with a sysop account borrowed from another person and went around creating accounts :p [06:40:43] seav is a bureaucrat [06:41:41] admin is enough [06:42:27] What's up? [06:43:32] aww superpowers coming [06:45:30] Reedy: account creation at a event with no crat [06:46:55] * p858snake|l bribes Reedy with stropwaffles for https://bugzilla.wikimedia.org/40263 [06:51:26] p858snake|l: they have a crat and 2 sysops! [06:51:30] and 2 IPs even [06:51:41] as we say, "grasso che cola" [06:59:53] does this make any sense? https://gerrit.wikimedia.org/r/gitweb?p=operations/mediawiki-config.git;a=blob;f=wmf-config/InitialiseSettings.php;hb=7f319eb18c36ee09effaa75058515d03c9c2cb3d#l6370 [07:03:52] we are in a computer laboratory and may prefer do the account creation themselves [07:04:35] they can [07:04:57] you only have to login in advance (and check they don't make disasters), did you check the help page? [08:42:50] does someone know when the new planet will be enabled? [08:43:25] haven't heard anything about it lately [08:43:28] the current planet is lacking several posts, like Wiki Loves Monuments in French, because feeds have been added only to the new planet config [08:43:31] https://meta.wikimedia.org/wiki/Planet_Wikimedia#Requests_for_inclusion [08:43:43] Should be out soon we're rolling out a new version of planet at the moment and it will be up when that's done (probably later today) Jalexander (talk) 19:30, 22 August 2012 (UTC) [08:43:51] the new system (already live here ) which should happen in the next day or two. Jamesofur (talk) 23:39, 26 August 2012 (UTC) [08:45:26] well slap some folks around :D [08:45:52] just bite a few thumbs and chuck a couple of gloves onto the ground? [08:46:47] Nemo_bis: or blog about it on a blog that is still on the current planet [08:47:14] or beg thedj to help [08:47:21] https://meta.wikimedia.org/w/index.php?title=Planet_Wikimedia&diff=4130472&oldid=4114918 [08:47:41] it's like four lines of copy and paste, come on :p [08:48:19] i'm not sure how the current config is running... [08:48:38] thedj: it's just a file on svn [08:48:38] it's still on svn ? [08:48:48] I think so [08:49:01] lemme switch computers [08:49:13] like https://www.mediawiki.org/wiki/Special:Code/MediaWiki/115645 [08:49:33] I suppose (and hope) they didn't do double work, made the new puppet config only for the new syste, [08:51:38] too bad: http://svn.wikimedia.org/viewvc/wikimedia/trunk/tools/planet [08:51:41] no longer exists [08:52:53] !log someone should deploy Planets update, or add blogs to the old config because otherwise WLM has passed before the blogs are added to the stream. [08:53:04] Logged the message, Master [08:53:44] aww [08:53:51] thanks for trying, thedj [08:54:36] that's not what I should use that for, but it's not like filing a bugzilla ticket ever results into someone noticing it. [08:54:55] and i'm not around 24/7 to poke with sticks either :D [08:55:14] indeed [08:56:02] so now where does the old config live, one wonders? [08:56:35] possibly only on disk atm... [09:40:26] Nemo_bis: actually it still exists. it's in mediawiki, not wikimedia. But it doesn't seem like i'm still allowed to commit there. [09:41:26] svn: E175013: Access to '/svnroot/mediawiki/!svn/act/f45d5d6d-8546-4181-ab80-929f3dbd6667' forbidden [09:45:11] thedj: sigh [09:47:50] posted on the ml. [09:48:03] gtg now [09:48:14] Just added throttle.php to noc conf ;) [09:48:43] Right... [09:56:38] !log reedy synchronized wmf-config/throttle.php 'Bug 40263 - Request to lift IP address for mass creation - Philippines SFD 2012' [09:56:47] Logged the message, Master [10:02:31] Reedy: are you also one of the "someone" who could update the old planet's config per DJ's request on wikitech? :D [10:18:16] Nemo_bis: need root for that one, I think [10:18:45] You can likely get mutante to do it on Monday [10:20:46] Reedy: what? you don't have root access on that machine? :-O /me surprised [10:21:00] I has no root anywhere [10:21:07] oh [10:21:13] * Nemo_bis facepalms [10:21:45] singer? [10:23:10] loooks so [10:24:42] yeah, can't even log into it seemingly [11:54:11] !log andrew synchronized wmf-config/throttle.php 'Fix undeclared variable error' [11:54:20] Logged the message, Master [13:28:27] Squid down, thanks. [13:28:36] Hi, I´m getting a Wikimedia foundation error here [13:28:51] oh... [13:29:14] wikimedia error... but I guess I'm not the only one [13:29:21] yeah, down here. [13:29:42] +1 [13:29:43] facepalms, its a Saturday zzz [13:30:55] haha, the french team arrived [13:31:04] Yes :) [13:31:19] Hi everybody [13:31:46] Hi Superjuju10 ! [13:31:50] x) [13:31:53] Superjuju10: Welcome here :D [13:31:57] Hi Jules78120 :D [13:32:01] Woozz: Hi too ! [13:32:20] Hi Woozz Superjuju10 and Jules78120 :) Where do you come from :-P [13:32:47] elfabixx: Tours, in France :) [13:33:24] xquirky servers :-/ [13:33:27] elfabixx: I think I know you [13:33:57] Jules78120: you sure ? [13:34:02] Jules78120: Superjuju10 elfabixx : it's working again [13:34:03] I live in Prugny in France ;) [13:34:20] Woozz: ya, i noticed it [13:34:41] on and off and on and off ... [13:34:49] elfabixx: Hum... Not sure but... [13:34:58] Superjuju10: Oh ! [13:35:02] flood [13:35:10] Superjuju10: We live in the same country ! [13:35:23] Yep [13:35:25] :O [13:35:28] Ow ! that so strange ! [13:35:34] And you, Woozz, where do you live ? [13:35:44] Superjuju10: Do you work in french wiki ? [13:35:47] -_- [13:36:01] Request: GET http://en.wikipedia.org/wiki/Weighing_scale, from 10.64.0.131 via cp1014.eqiad.wmnet (squid/2.7.STABLE9) to () [13:36:03] Error: ERR_CANNOT_FORWARD, errno (11) Resource temporarily unavailable at Sat, 15 Sep 2012 13:34:30 GMT [13:36:21] Jules78120: Yes I'm [13:36:21] fabriceferrer: Hello :) [13:36:31] Sj [13:36:31] Hello fabriceferrer the troll ;) [13:36:43] ... [13:37:19] Superjuju10: What's your real name ? Jules ? [13:37:29] fyi, we're looking to see why memcache has gone crazy and trying to fix the problem [13:37:30] No, my name is Julien ;) [13:37:46] Superjuju10: Okay ! [13:38:05] Superjuju10: Do you want to be my friend ? [13:38:17] Well.... perhaps should we stop, uh ? :) [13:38:21] thx LeslieCarr, appreciated [13:38:30] +1 [13:38:30] Jules78120: Yes :) [13:38:42] Thank you Superjuju10 :-* [13:38:50] You'r Welcom :-* [13:39:09] elfabixx: you're too serious ^^ [13:40:20] what's wrong with you guys ?? x) [13:41:01] elfabixx: do you know Superjuju10 ? He's a nice guy ! [13:42:37] nothing is wrong, Woozz, the french like to practice the art of proper conversation under all circumstances :) [13:44:03] * Superjuju10 confirm [13:44:06] Jan_eissfeldt: we're... crazy ! ;) [13:44:15] hey, is it up already? [13:44:20] Very versy crazy ;) [13:44:23] *very [13:44:54] Jan_eissfeldt: especially me and Superjuju10 [13:45:34] ;) [13:48:53] We are alone Jules78120, we are freedom ! :) [13:49:22] Superjuju10: :D [13:50:24] /wc [13:51:35] bye bye all ! [13:51:47] and happy birthday Superjuju10 ! [13:56:36] hi [13:56:54] Juandev: hi [13:58:28] Some idea, Exactly whats happend? [13:59:26] wilfredor: tropezaron con un cable y bueno, el resultado es el que ves :P [13:59:56] mafk: I see [14:11:52] RD: thanks for the reminder, i should have breakfast [14:12:03] aka unhatched chickens [14:12:07] w/ spinach [14:12:12] Sounds good [14:12:13] er [14:12:20] not the last bit [14:12:23] hah [14:12:30] I'll go without the spinach, more chickens. [14:14:52] thanks LeslieCarr, a lot better [14:16:56] sDrewth: it's still mostly dead I think? [14:17:04] moin odder [14:17:10] moin jeremyb [14:18:36] Hydriz: see what you did? [14:18:59] what? [14:19:12] you broke it again because you said it was fixed to early [14:19:38] it loads zzz [15:07:35] werdna: hey [15:07:48] liangent: just finished writing you a mail [15:07:55] we've been having some downtime [15:08:03] and that request that AdvancedSiteNotices makes is uncached [15:08:54] I remember we tested it and made sure it's cached during development [15:08:59] but I can't remember details [15:09:04] because main dev is not me [15:09:13] Well, the response has Cache-Control: maxage=0 [15:09:22] Or maybe private [15:09:26] I don't remember which [15:09:32] but in either case it is not cached by squids [15:15:47] werdna: I guess it became uncached because of https://bugzilla.wikimedia.org/show_bug.cgi?id=36142 [15:16:06] mark: ^^ [15:16:30] looking [15:17:01] hmmph [15:17:10] doesn't look related [15:17:18] liangent: but it's not diff pages in question [15:18:13] werdna: all un-auto-purgable pages are excluded from caching in the fix [15:18:26] mm I can see that now [15:19:24] werdna: so is there a param for index.php to make it set smaxage? [15:19:31] like the one for api.php [15:19:31] I don't know [15:19:37] where is the /cn.js call coming from? [15:22:01] werdna: anyway I can turn it to use api and set smaxage I think [15:22:15] liangent: can we wait for now? [15:22:26] just give us time to exclude it as the source of the site issues [15:23:01] werdna: ok [15:23:07] thanks [15:25:25] werdna: what do you suggest for the (s)maxage value I should use? [15:25:45] not sure… I think mark / LeslieCarr would have a better idea [15:27:06] and we will of course say: as long as you can make it ;) [15:27:34] hah [15:27:45] took me a few secs to parse that one [15:28:25] is 1hr ok? [15:31:23] liangent: i think outage first, your notice later [15:32:40] liangent: what's the notice about anyway? [15:33:00] jeremyb: replacement of MediaWiki:Sitenotice [15:33:28] huh? you have a notice about a notice? [15:33:44] jeremyb: I don't understand your question [15:33:57] iirc the first time it came out because MediaWiki:Sitenotice doesn't handle zh-hans/zh-hant well [15:34:26] what notice is supposed to show up at the top of the page right now? [15:34:36] is it actually a big deal that it's not shown? [15:35:20] jeremyb: not really I think [15:35:28] but I want to fix it [15:35:39] liangent: so what is it supposed to show now? [15:35:44] liangent: what's the message? [15:35:55] with that script disabled, nothing to show then [15:36:09] ... [15:36:24] i can just repeat myself i guess [15:36:34] and when it's not disabled? [15:38:16] jeremyb: current article writing activity, and some "ads" target specific people [15:38:20] eg local meetups [15:38:34] ok [15:40:14] werdna: so I fixed the script to use api.php with smaxage, and I'm going to remove your "return" so I can test it myself? [15:40:21] anyway it's still non-default in gadgets-definition [15:40:28] Can you do it tomorrow? [15:46:34] werdna: then I added these http://zh.wikipedia.org/w/index.php?title=MediaWiki:Gadget-AdvancedSiteNotices.js&diff=22808369&oldid=22808177 to allow debugging with local hack [15:48:55] i don't see how that could possibly help [15:50:02] jeremyb: in firebug I set a breakpoint at 'if ( _return )' and edit _return to false then continue [15:50:18] uhuh [15:50:33] can't you just have it check a GET param? [15:51:31] jeremyb: it needs more code and when debugging I have to use firebug [15:52:08] ok, if it makes you happy [15:52:10] ;) [15:55:18] how about not even tomorrow but next week when people are working ;) [15:55:34] i'm gonna get breakfast now [15:56:14] liangent: ^^. so monday or tuesday [15:56:59] eh every time I see your code deployment late by one day [15:57:16] because you're utc-?? and I'm utc+8 [15:57:26] so should be Tuesday or Wednesday for me then? [15:58:00] btw it's now using this url [15:58:00] http://zh.wikipedia.org/w/api.php?page=Template%3AAdvancedSiteNotices%2Fajax&variant=zh&action=parse&format=json&maxage=3600&smaxage=3600 [15:58:21] but it keeps returning MISS from X-Cache-Lookup and X-Cache [15:58:26] so what's wrong? [16:00:30] i got a hit [16:01:04] and age=38 [16:07:33] jeremyb: ok if you can confirm this works, I'll leave it as its current state and wait for Monday [16:07:47] liangent: well what are you getting? [16:08:58] jeremyb: X-Cache: MISS from cp1004.eqiad.wmnet, MISS from cp1018.eqiad.wmnet [16:08:58] X-Cache-Lookup: MISS from cp1004.eqiad.wmnet:3128, MISS from cp1018.eqiad.wmnet:80 [16:09:29] and how are you checking? [16:47:42] jeremyb: I load that url in browser and check firebug console [16:48:01] how are you loading it in browser? [16:49:47] by clicking reload or locating cursor in location bar then enter [16:49:49] neither works [16:50:41] well it's surely just cached in your browser's local cache and not making a network request at all [16:50:51] clear your cache and then location bar + enter [16:50:54] don't hit refresh [16:53:08] jeremyb: still miss [16:53:28] and I'm sure it's not cached locally because everytime it gives a new "date" header [16:55:04] liangent: errrr, i want to watch you do this... [16:55:09] liangent: can you tcpdump it? [16:55:16] liangent: are you logged in??? [16:55:42] liangent: you have to do it when not logged in. clear your wikipedia.org cookies and try again [16:56:21] I pressed "log out" but didn't clear cookies [16:56:29] clear em [16:58:12] jeremyb: ok finally HIT now [16:58:16] which cookie is annoying? [16:59:05] idk [17:27:49] hey [17:38:40] * Hazard-SJ knew something was wrong [17:42:43] hello ? [17:42:45] anyone ? [17:42:53] it's saturday... [17:43:02] I'm struggling to find anyone active [17:43:23] hi [17:43:57] hey paravoid [17:44:03] paravoid: you familiar with mediawiki extensions ? [17:44:09] jeremyb: the process is usually: Check if guillom is around, and if not, ping someone else on the team :) [17:44:24] average_drifter: we're in an emergency situation right now ... [17:44:29] LeslieCarr: oh ? why ? [17:44:49] because the site is falling over [17:44:53] average_drifter: try #mediawiki [17:44:58] average_drifter: see topic. ;) [17:45:02] guillom: heh [17:45:16] LeslieCarr: can I know more ? [17:45:26] not from LeslieCarr [17:45:30] alright [17:45:31] other people can tell you about it [17:46:19] average_drifter: there was an outage this morning about 4ish hrs ago [17:46:38] average_drifter: everything recovered and looked good. and now maybe it's back [17:46:49] average_drifter: or at least there's some current major problem [17:46:57] average_drifter: take a look at status.wikimedia.org [17:48:36] jeremyb: hum, interesting that the outage of the servers is so severe, that it even affects the nagios load ... [17:48:56] T3rminat0r: nah, maybe that was already yellow to begin with [17:49:10] T3rminat0r: but nagios does sometimes have problems [17:50:01] seeing the responsetimes rising again in the nagios graph on status. ... I guess that is because it has to do lots and lots of repeat-checks due to downed servers [17:52:39] jeremyb: can I ask you about the clicktracking extension ? [17:52:54] average_drifter: nope. you should ask on monday [17:53:08] ok [17:57:10] * Hazard-SJ is annoyed that vandals are doing there things and he can barely do anything [17:58:10] My cookie keeps getting eaten after the long wait to load the pages :'( [17:58:31] Hazard-SJ: yea, as the topic states, there are server issues... [17:59:28] T3rminat0r: I was aware of it, but the sad thing is that vandals are still at work. [17:59:52] ofcourse ... understand your problem ... but well, it will get fixed eventually ... :S [18:07:11] jeremyb: do you think the problems might be caused by the mediawiki update to 1.20wmf11 ? [18:07:45] i don't personally. if you made me guess [18:07:51] but i don't really know [18:08:21] ok [18:08:36] yea, I was interested in your opinion :) [18:10:49] Just FYI, I am getting this error message when I try to open a specific article. [18:11:13] don't do that [18:11:28] I can open other most articles without any problem. [18:11:30] T_suzu: please don't use /notice [18:11:32] T_suzu: 1) we know there's an outage, see the /topic [18:11:55] got it [18:12:08] T_suzu: 2) don't paste more than 1-2 lines at a time. use a pastebin or figure out a way to shorten/abbreviate it [18:12:28] T_suzu: 3) as you just heard from DeltaQuad, no /notice [18:12:35] danke ;) [18:13:11] jeremyb: will be more cautious next time. [18:13:34] T_suzu: the main problem is the /notice but also the pasting [18:13:50] T_suzu: telling us something we already know is not so horrible [18:14:10] makes some users machines go beep [18:17:27] My apologies if anyone had to climb out of bed [18:18:00] * jeremyb runs off to the WTM event [18:38:11] tstarling cleared profiling data [18:44:45] Wow, the sites are slow this eve. :-( [18:45:01] intentionally so [18:45:06] even [18:45:38] what's gone wrong? [18:45:52] no idea, but things are messed up [18:46:07] not intentionally certainly [18:46:35] !log tstarling synchronized php-1.20wmf11/includes/objectcache/MemcachedClient.php 'memcached profiling' [18:46:39] didn't you reduce maximal requests per client? [18:46:40] tstarling cleared profiling data [18:46:44] Logged the message, Master [18:47:06] i think you've misread [18:47:25] let me consult the docs [18:47:27] i've only heard [18:49:47] giftpflanze: http://httpd.apache.org/docs/2.2/mod/mpm_common.html#maxrequestsperchild [18:49:56] that's what was actually changed from 300 to 4 [18:50:14] ah, ok [18:50:22] ok, then the missinformation is on my bad, sorry [18:51:30] T3rminat0r: no, wasn't you. i think it was an unnamed someone else ;) [18:51:32] that's scary technical jargon … activities in a child lifetime, after which they die? :-/ [18:51:55]