[00:22:19] kaldari: do you want to comment on bug 45066 too? (now that you have on 42894) it's about twice the threshold you stated. idk what other criteria you might have. (or do we even have criteria written down somewhere) [00:30:18] hello, anyeone here? [00:30:23] jeremyb_: there are no criteria written down, so I don't really have any leg to stand on, but 25,000 is definitely too few articles. [00:30:34] o sry -.- [00:30:47] kaldari: right, but 45066? [00:31:17] that one's a harder case. Many of the other wikis that have disabled anon creation also have about 200,000 or 300,000 articles [00:31:33] turkish, indonesian [00:31:49] I think tamil is quite small though if I remember [00:33:50] I had a suggestion, which is prefered a short description or the full text? [00:36:14] the idea is this: We could have a user level between Banned and New users. Users demoted to this level would not be able to edit articles or talk pages but they would be allowed to post pending changes. [00:36:48] This would produce millions of aditional contributions. [00:37:57] gde33|2: why would you want non-new users to not be able to edit articles? [00:38:15] normally they would be banned [00:38:37] that seems a bit extreme [00:38:46] banning people? [00:38:47] gde33|2: which wiki? [00:38:49] yeah :) [00:38:58] editors get banned the whole time? [00:39:09] we just banned one with a million+ contributions [00:39:37] ah, I see your use case now [00:39:52] I'm not sure who would be the ideal test candidates [00:39:58] users that are sometimes useful and sometimes not [00:40:06] I thought conduct bans at first but those appear only to apply to new users [00:40:08] ottava [00:40:09] :-P [00:40:51] kaldari: most (all?) haven't had anon edits disabled recently though, right? i wonder if some of those currently disabled would make it through review these days. in any case it would be really nice to have real criteria. (even if more qualitative than quantitative) [00:41:30] jeremyb_: I agree, we need an RfC on meta [00:41:57] if only I have time to do such a thing :P [00:42:04] and definitely, the default should be to reject disabling anon edits. burden to prove (whatever we decide they need to prove) should be on the requestor [00:42:07] hah [00:43:12] pending changes also dramatically extends the review process [00:43:26] from 5 seconds to whole minutes, even hours? [00:44:16] gde33|2: it sounds like a good idea, but I don't think it would work in practice... [00:44:43] what cenario do you imagine where it wouldn't work? [00:44:49] if the semi-banned user was contributing content (reviewed or not), other users would need to be able to interact with them [00:45:05] why do you think they would need to interact with them? [00:45:24] the way I see it the user who approved the contribution is responsible for it. [00:45:24] to discuss/debate the content they want to add [00:46:34] they spoiled that for themselves or lost those privliages for some reason [00:46:48] 99% of wikipedia is uncontroversial [00:47:13] !log synchronized payments cluster to 12bc36c2385d8e56792c [00:47:15] talk is not even that productive on the 1%? [00:47:20] Logged the message, Master [00:48:58] gde33|2: you might be right, but my instincts say it would be trading one set of problems for another set. If someone isn't able to behave and we have to lose there edits, it seems like a small price to pay for a non-toxic community [00:49:23] after restriction you are no longer participating in the community [00:49:46] the community is the means to write the encyclopedia? [00:52:09] One user thought it was a bad idea because reviewers could be biased. I think we can ban overly biased reviewers just like overly biased editors. [00:52:23] To deny the content suggestion describes a whole different level of bias [00:52:33] one that doesn't involve reading the content suggestions [00:52:47] yeah, it's a sticky wicket as they say elsewhere [00:52:57] perhaps you should do an RfC on it [00:53:23] asking arround the idea seems much to complicated [00:55:13] thats why I thought I'd go to the nerd channel... ehh I mean tech [00:56:31] I mean, do you want these millions of contributions or not? [00:57:18] maybe :) [00:57:19] depends on the price [00:57:24] free [00:57:37] Should we read them first or should we reject them without reading? [00:57:40] contributions are always free (unless omse company is paying to get a better aticle written on them) [00:57:46] free is sometimes expensive if you know what I mean [00:57:56] there is a price to pay in the review process [00:58:04] I have a cleaver idea for that also [00:58:37] the moment you change a section that has a pending change you are promted to merge it into that section. [00:58:52] I had a very similar idea actually :) [00:59:36] it gives us the most knowledgable editor with the content fresh in his mind [01:01:18] An important thought imho is that people come here with the intention to help, then they try their very best and then we ban them [01:01:37] yep [01:01:48] the sensible thing to expect from a normal person is to go write essays about this rediculous wikipedia thing [01:02:28] time that would be better spend working on the wiki :D [01:02:58] gotta run [01:03:08] ok thanks, cu [01:03:13] good luck with that rediculous Wikipedia thing :) [01:03:20] yee haa [01:07:51] the wall of text is here http://en.wikipedia.org/wiki/User_talk:Jimbo_Wales#Contributors_banned_over_conduct [02:00:56] ok, LevelUp office hour starting now in #wikimedia-dev [02:16:09] oh, that's now... [05:05:27] https://en.wikinews.org/wiki/Wikinews:DynamicPageList#category "There is a maximum number of category parameters allowed by the extension, configurable at installation time (see top of the extension file)." how do I see that value please? [05:09:42] gry: Directly, you can't [05:09:44] Indirectly.. [05:09:55] can someone here simply read that config? [05:10:01] i don't see any sign that it's been customized [05:10:10] what's the default value? :) [05:10:14] Maybe check noc? [05:10:22] not sure what noc is [05:10:37] noc.wm.org [05:10:37] noc.wikimedia.org :) [05:10:44] /conf [05:11:07] now that is a new thing to me :) [05:11:09] $wgDLPmaxCategories = 6; // Maximum number of categories to look for [05:11:35] And still is [05:11:36] reedy@fenari:~$ mwscript eval.php enwikinews [05:11:36] > echo $wgDLPmaxCategories; [05:11:36] 6 [05:12:43] ok, thank you [05:13:09] Reedy: is it really DLP not DPL?? [05:13:20] it's intersection! [05:18:00] > [05:18:02] 60 # Configuration variables. Warning: These use DLP instead of DPL [05:18:02] 61 # for historical reasons (pretend Dynamic list of pages) [05:18:04] > [05:18:10] Heh. [05:18:14] ewwwwwwwwwww [05:18:32] That's not the gross part. [05:18:43] The gross part is that https://www.mediawiki.org/wiki/Extension:DynamicPageList_(Wikimedia) doesn't seem to mention them at all... [05:23:03] https://www.mediawiki.org/wiki/Extension:DynamicPageList_%28wikimedia%29 [05:23:25] err, not that either [05:23:29] https://www.mediawiki.org/wiki/Extension:DynamicPageList_%28Wikimedia%29 [06:03:18] * Jasper_Deng_busy pokes Betacommand [06:29:53] jeremyb_: where does the WMF use VirtualBox and/or VMWare? [06:30:27] errr? [06:30:36] * Jasper_Deng_busy knows that Labs uses it [06:30:52] (at least one of them) [06:31:07] no, certainly not [06:31:10] neither of them [06:31:48] Jasper_Deng_busy: we use KVM in Labs [06:32:41] Vagrant uses Virtualbox, by default. We have a way to create MediaWiki development virtual machines using vagrant [06:33:07] it's a good approach for having a pre-configured environment on your own laptop [06:35:53] * Jasper_Deng_busy likes and uses VB.... VMWare, not so much [06:36:02] (in the spirit of FOSS) [06:37:14] Ryan_Lane: PayPal switching to OpenStack is pretty cool [06:37:19] indeed [06:37:25] 10k nodes, apparently [06:40:41] IBM made a big announcement about OpenStack recently too [06:40:56] about switching all of their cloud services/products to it [06:41:12] we were using VMware till you turned it off recently ;) [06:41:54] that's pretty cool, re: IBM [06:42:00] Reedy: ;) [06:42:01] Reedy: SLEEEEEEP!!! [06:42:04] we were indeed [06:43:08] if only it was so easy ori-l... [06:43:37] i know it's E3's deployment window tomorrow, but still.. [06:43:43] I've been in bed 4 hours now. [06:44:50] considering I had been sleeping more from being really tired (including today), nfi why I'm wide awake [06:45:15] roll back to wmf11 [06:45:20] see if it fixes it [07:45:20] petan: we just spent a lot of time and effort removing sub-sub domains and now you want to add more of them? [07:45:34] "Oops! A server error occured. It's been logged." on etherpad.wikimedia.org [07:46:12] Nemo_bis: and then what? [07:46:15] you tried again? [07:46:39] and the same [07:46:48] oh, wow [07:46:55] petan: no. please. no [07:46:59] * jeremyb_ begs [07:48:19] jeremy.b.bots.labs.wikimedia.org [07:48:27] yes! [09:06:26] jeremyb_: so? should that be filed? actually I see endless "Loading..." on http://etherpad.wmflabs.org too [09:07:23] But that must be NoScript, while wm.o version fails on its own where it usually didn't [13:05:59] Could somebody run the follow on the WMF commonswiki cluster? [13:06:01] SELECT * FROM transcode WHERE transcode_image_name="VOA_News_in_Russian_2ai_talks_in_Washington.ogv"\G [13:07:08] <^demon> Dispenser: Empty set. [13:09:19] Looks like (dedicated) commonswiki is corrupt on TS [13:13:01] where I can translate sitenames for Relatedsites ext? [13:14:37] Пов'язані сайти [13:14:37] Wikipedia [13:14:37] Wikimedia Commons [13:14:38] but i want Wikipedia and Wikimedia Commons be localized too [14:25:03] Error: 1205 Lock wait timeout exceeded; try restarting transaction (10.64.16.27) https://commons.wikimedia.org/w/index.php?title=File:2012_State_Of_The_Union_Address_%28720p%29.ogv&action=purge [14:26:35] db1038... [14:26:49] s4 master [14:27:59] @replag all [14:28:01] jeremyb_: [s1] db1017: 0s, db1042: 0s, db1043: 0s, db1049: 1s, db1050: 0s, db1051: 0s, db1052: 0s; [s2] db1034: 0s, db1002: 0s, db1009: 0s, db1018: 0s; [s3] db1019: 0s, db1003: 0s, db1010: 0s, db1035: 0s [14:28:02] jeremyb_: [s4] db1038: 0s, db1004: 0s, db1011: 0s, db1020: 0s; [s5] db1039: 0s, db1005: 0s, db1026: 0s, db1021: 0s; [s6] db1006: 0s, db1022: 0s, db1027: 0s, db1040: 0s; [s7] db1041: 0s, db1007: 0s, db1024: 0s, db1028: 0s [14:28:24] hah, it reports seconds for masters too? :) [17:20:24] AaronSchulz: You have a second to help with a git issue? I'm trying to get a deploy out, but my local checkout keeps telling me the relevant change is empty when I try to cherry pick it over to the right branch. [17:20:54] git log disagrees, though. [17:28:11] what cherry-pick command are you using? [17:30:20] Just git cherry-pick, like always. [17:30:30] Eh, it's okay. I'm taking a later window. [17:30:52] K4-713: you have a conflict.... [17:31:07] you need to resolve your conflict [17:31:30] Yeah, actually, it _finally_ started saying that instead of trying to tell me the thing was empty. [17:31:52] heh [17:32:29] * Aaron|home was curious what the exact command was, but meh ;) [17:32:59] git cherry-pick 9d7e9c31415975 [17:36:58] K4-713: did you figure it out? [17:37:30] I think so. But not with enough time to do the required manual resolution and still make the window, so I'm taking a later one. [17:41:50] K4-713: I didn't know if it was a local cherry-pick or one directly from a gerrit patch or something [17:42:14] Ah, I see. No, it's in master. [17:42:49] All merged in and everything. [17:47:40] Huh. It seems to be only wmf/1.21wmf11. 12 was perfectly happy to do the cherry-pick and report conflicts. [17:48:02] erm? [17:50:32] mwalker: Any chance that original language change only went out to 1.21wmf12, and not 11? [17:51:06] yes -- that's true [17:51:15] this was deployed on the train; so only 12 [17:51:17] * K4-713 headdesks [17:51:24] That would explain... everything. [17:51:35] whoo! :p [17:51:37] sorry [17:52:22] It's okay. I'm just happy I haven't lost my mind. ;) [17:59:50] Reedy: just so I know what to keep track of and when: the wgCategoryCollation bugs (eg !bug 45776) are ideally merged before a wmfXX and done on the normal deploy cycle (eg, that one would go out with 1.22wmf1)? [18:00:19] or are they higher priority than that? [18:01:50] They're not high priority [18:02:01] This is stuff people have been waiting multiple years for... [18:07:59] Reedy: so, other than the settings change, is there a big resource impact? Or should they just be committed/pushed out with the normal flow as we get them (and they're deemed appropriate)? [18:08:18] The code can go in whenever, and then the appropriate config changes made afterwards [18:10:41] so each one needs code changes? I didn't realize that. [18:11:42] I'm not actually that familiar with teh code [18:12:32] Most of it depends on an external library... But then for some of them, they require a couple of extra config lines in MW [18:12:51] Compared to the chinese collations, these are all easy [18:13:04] gotcha [18:13:31] so, what prompted me asking you these questions was me thinking I should add the Ukranian one to the "new wmf deploy" tracking bug, yes/no? ;) [18:13:55] nah [18:14:17] most of the prep work is in now, so they aren't tied to a specific deployment [18:15:00] Reedy: I just thought that if it was ready, why not 'just do it' [18:15:39] Most of them are. Just need a maintenance script running at the same time [18:15:55] have smb answered my question at 13:13 PM (UTC) ? [18:17:13] Reedy: ok, so should I just stay out of your way on these and trust you'll get them done when needed/appropriate? [18:17:16] :) [18:18:00] Base-w: I don't think anyone has [18:18:37] Base-w: No. By the look of the code, they don't support anything else [18:18:47] 'text' => Language::fetchLanguageName( $title->getInterwiki() ) ?: $site, [18:26:31] hey people - especially greg-g … did AFTv5 go live 3/19 ? [18:26:38] because since then there has been a huge increase in 500's [18:26:46] and that's the one thing i see that could be causing it [18:27:02] (unless mobile front end is 500'ing every single page they serve…. in which case, bad mobile ;) ) [18:27:35] LeslieCarr: both AFTv5 AND mobile did a deploy on 3/19 :) [18:27:43] https://wikitech.wikimedia.org/wiki/Deployments#Recently_completed [18:27:51] um, i think we need to revert because holy shit the 500's [18:28:08] yeah? let's see if we can pin down the cause of it [18:28:26] well i would rather revert and then have them work more on aftv5 [18:28:30] don't want to rollback all of wmf12, though man, it has turned out to be a not so hot release [18:28:42] yeah, just mean make sure it is aft not mobile [18:28:46] we already know that it has been an unstable version of code [18:28:58] well i was joking, i doubt that every single mobile front end request has been 500'ing [18:29:29] k [18:29:42] going to it to make sure ... [18:29:58] nope, looks pretty and 200's [18:30:12] actually the mobile looks nicer than the desktop version [18:30:20] thanks mobile, making everyone else look bad! :p [18:30:44] i am going to request that aftv5 does some more testing before rolling out again [18:30:57] and that the team keep a close close eye on 500 count [18:31:00] LeslieCarr: cool, I'm emailing you and them right now [18:31:05] thank you :) [18:31:08] thank you [18:31:09] could that be/is that somehow related to stuff like http://commons.wikimedia.org/w/index.php?title=MediaWiki:Filepage.css&action=raw&maxage=2678400&usemsgcache=yes&ctype=text%2Fcss&smaxage=2678400 in https://bugzilla.wikimedia.org/show_bug.cgi?id=46612 ? [18:31:10] me too [18:31:20] (the 5** errors) [18:31:22] it's nice to have someone knowing when things are deployed instead of wondering "what happened" [18:31:36] possibly ? [18:31:40] i am unsure [19:02:35] I need help for user account(s) in wiki projects. [19:07:17] xJaM: Tell what you want, it'll be more simple to help you in this case [19:08:12] Automatik: Today I've created user account in Wikidata (with different name (as usual)). [19:08:48] Automatik: I am also not sure if this the right place to ask. [19:08:49] Is someone knows why I can't write on Wikimedia sites? My keyboard works only on other sites than Wikipedia, Mediawiki, Wiktionary, etc. It's strange, isn't it? [19:09:23] xJaM: You can ask here, why not [19:09:40] I want one global user account - but now I have two in different wiki projects. [19:10:06] And secondly I want to change my current user acount. [19:10:28] Is this possible without too much nerving? [19:11:03] xJaM: Maybe you have to rename your account (asking a bureaucrat) [19:11:36] xJaM:Then, you'll can merge them [19:14:42] Automatik: In which project should I ask bureaucrat? I am bureaucrat in one project (where I usually work). [19:15:32] xJaM: On the project where you have the name you want to change [19:16:49] My keyboard doesn't work on Internet. If someone know how to resolve the problem, I would be happy to know how to fix it [19:17:38] Automatik: Yes, look in this case. I have NAME1 in one project. In second project (namely wikidata) I've created NAME2, and this account was created globaly also in previous project, so I can't rename it (as I am bureaucrat there) as NaME2 alredy exists. [19:25:46] Automatik: Perhaps here they can help me? http://meta.wikimedia.org/wiki/Steward_requests/Username_changes [19:26:00] xJaM: please ask #wikimedia-stewards [19:28:30] xJaM: understand that "This page hosts requests for a username change on a Wikimedia wiki with no active bureaucrats." [19:29:54] Automatik: thanks - I understand now. [19:40:40] Error: ERR_CANNOT_FORWARD, errno (11) Resource temporarily unavailable at Thu, 28 Mar 2013 19:40:14 GMT [19:40:42] Panic [19:42:19] same error here (CA US) ERR_CANNOT_FORWARD [19:42:38] * Hazard-SJ says hello [19:43:24] ah grr [19:43:25] We broke Wikipedia [19:43:36] Yup, it's down. [19:43:46] And Wikidata I guess? [19:43:48] Page just correctly refreshed [19:43:54] Request: POST http://www.wikidata.org/w/index.php?title=Wikidata:Project_chat&action=submit, from 10.64.0.134 via cp1010.eqiad.wmnet (squid/2.7.STABLE9) to () [19:43:56] Error: ERR_CANNOT_FORWARD, errno [No Error] at Thu, 28 Mar 2013 19:40:08 GMT [19:43:56] ok good [19:44:03] "self healing" :) [19:44:05] I got that a short while ago [19:44:16] ...and now it's back up. [19:44:22] anything after x:43 I guess we want to know about [19:44:38] what did we do? [19:44:44] Did we upset the server kitties? [19:44:55] Did you guys need to toss another ball of yarn in there? [19:45:05] more hamsters [19:46:24] server kitties needed more beagles [19:46:25] https://commons.wikimedia.org/wiki/File:Beagle_and_sleeping_black_and_white_kitty-01.jpg [19:46:59] mindspillage: your pic ^ [19:47:25] marktraceur: was there too. [19:47:30] LeslieCarr: Or other puppies https://fbcdn-sphotos-b-a.akamaihd.net/hphotos-ak-snc7/480419_10151774504755385_1184986905_n.jpg [19:48:03] http://commons.wikimedia.org/wiki/File:Server-kitty.jpg [21:17:28] Hey, any techies who are admins can view a threat here, and since the user's now blocked, they may be able to/want to formulate a similar attack on WMF servers. Not that it matters, I doubt that we'll have too much trouble, but just notifying all of the tech gurus http://en.wikipedia.org/w/index.php?title=Glendale_Jr/Sr_High_School&oldid=547496064 [21:20:36] gwickwire: what kind of threat? [21:20:44] (I'm not an admin, but work for WMF in engineering) [21:20:54] They were threatening basically a huge takedown on the servers of a whole city. [21:21:02] Glendale to be exact. [21:21:48] Their username is "TeamSentaSecurity", and I just wanted you all to be aware of it, since if they're willing to openly advertise/threaten an entire city they may be willing to attempt the same on servers of WMF once they're blocked. [21:22:07] It's kinda out there, but I got a feeling that made me want to notify ya'll :P [21:22:07] oh, I see [21:22:12] thanks [21:22:25] LeslieCarr: due diligence here (on my part, you do what you will) ^^^ [21:22:29] lolol [21:22:37] Anytime, if you'd like me to have an admin forward you the threat I can attempt to find one, but I'm sure someone here will be a sysop on enwp. [21:22:40] yeah, notskeered [21:22:43] or be able to access it backdoor like :P [21:22:57] greg-g: "Xirzon: are the servers up for slashdotting ? brion: we get more traffic than /. usually we don't even notice the bump on the traffic graphs" [21:22:59] ;) [21:23:13] yeah [21:24:13] For a lot of them the squid really don't even notice :D [21:24:13] Reedy: of course, talking about a slashdotting anymore is really a low-grade ddos, something any old laptop could take now that /. isnt' relevant ;) [21:26:15] hrm, can't see what the content was [21:26:23] can someone forward it to the ops list ? [21:29:41] LeslieCarr: Was the WMF affected by the recent DDOS fallout? (last Wednesday) [21:30:29] nope [21:30:43] other than having more than the usual numbers of emails to read and irc discussions about it [21:31:08] and also thinking that cloudflare sounded like they singlehandedly defeated ddos'es -- that had to have been written by a PR person [21:31:20] I got about 2Gbit of extra traffic :P [21:31:34] hehe cool [21:32:21] multichill: per second or in sum? [21:32:32] Per second [21:32:47] comparing the numbers it looks like it may ahve been about a 500mbit peak difference --- but that also is often just random [21:33:01] so nothing worth noticing [21:33:03] Netflow showed a lot of dns crap [21:33:11] (udp 53) [21:33:33] :) [21:33:41] ah i wish i had an arbor box -- they make netflow display so pretty! [21:34:44] ann* arbor ;) [21:35:22] hehehe [21:36:14] * greg-g went to grad school in Ann Arbor [21:36:32] cool [21:36:38] and didn't wind up working for arbor? ;) [21:36:53] nope ;) there's a decent amount of good tech stuff there, though [21:37:19] i have heard it's the best city in michigan [21:37:24] bar none [21:52:47] greg-g: I've got the 3pm window, right? [21:53:54] K4-713: indeederoonie, I just pinged siebrand to check the bug that ori-l thinks might be related to the one you're deploying to fix [21:54:01] (see -staff) [21:54:42] K4-713: and, just to be clear, spagewmf is deploying right now, still cleaning up it looks like [21:55:34] K4-713 hi, we're in scap [21:56:06] Cool. :) I'll just be standing by, then. [22:34:32] What now? [22:34:35] * Hazard-SJ sighs [22:34:47] PHP fatal error in /usr/local/apache/common-local/php-1.21wmf12/includes/cache/MessageCache.php line 669: [22:34:48] Call to undefined method Language::getFallbacksIncludingSiteLanguage() [22:34:50] (on Wikidata) [22:34:58] crap, sorry. Should be fixed now. [22:35:09] Thanks :D [22:39:27] greg-g: Finished, actually. [22:41:15] K4-713: sweet, things look good? siebrand ? [22:41:32] mwalker poked it and verified. [22:41:34] (he may be asleep) [22:41:38] cool, thanks mwalker