[00:42:21] Brooke: exercises in logic clashes? https://bugzilla.wikimedia.org/show_bug.cgi?id=36096#c3 [00:43:10] Nemo_bis: What? [00:43:34] Nemo_bis: Why did you re-open that bug? [00:44:01] We dupe down unless there's a really good reason to dupe up. In this case, you duped up from a bug with a lot of explanation to a bug with a single sentence. [00:45:32] Brooke: you didn't explain this supposed convention [00:45:48] Which part is confusing? [00:45:51] I keep the most useful bug open, surely not the most confusing one [00:46:04] It's about courtesy to the bug filer. [00:46:05] Brooke: none, perhaps the part which was missing [00:46:18] Saying "oh, your bug is a duplicate" when it was filed months or years earlier isn't very polite. [00:46:27] I don't see how. [00:46:56] When someone closes my bug as a duplicate my reaction is "oh, nice, someone else has noticed my bug elsewhere!" [00:47:05] "Oh, he even described it better!" [00:47:13] "Hey, this summary is wonderful" [00:47:15] etc. etc. [00:47:24] If you really liked Niklax's description better, you can copy and paste to the old bug. [00:47:35] Niklas's * [00:48:21] And if there are indepedent problems with the original bug, they can be fixed. [00:48:23] no [00:48:29] uh? [00:49:25] Duplicate usually means "you filed a bug that has already been filed." Duping up doesn't make much sense. [00:49:36] I disgree with your definition [00:49:43] {{citation needed}} [00:49:54] It's only ever done when it's too onerous to copy the comments from the newer bug. [00:50:03] {{citation needed}} [00:50:30] > The problem is a duplicate of an existing bug. [00:50:35] That's what our docs say. [00:50:52] Yes and it's different from what you said [00:51:50] anyway part of my confusion was cause by the ignorance of the verbs "dupe up" and "dupe down", I admit [00:52:01] I've never had much luck with English phrasal verbs [00:52:10] It's also kind of made-up English. ;-) [00:52:19] indeed [00:52:38] I think there's a judgment being made on the bug filer when we resolve their bug. [00:52:58] In the case of duplicates, I think it's unfair to say "well, you made a copy of a bug that came after yours." [00:53:06] marking as duplicate is just moving the bug, not trashing it [00:53:07] It's distorted. [00:53:14] it's NOT what it says [00:53:17] don't repeat it [00:53:28] You mean don't make a copy of it? [00:53:38] «say "well, you made a copy of a bug that came after yours."» [00:53:54] ok, as I sais, this is a logic problem [00:53:56] That is what it's saying. [00:54:11] You're labeling the old bug as a duplicate (a copy) of a future bug. [00:54:19] https://en.wikipedia.org/wiki/Equivalence_relation [00:54:27] Hey, Thehelpfulone or anyone, I mistakenly created a SUL as "SPage (WMF)". I requested a username change to "S Page (WMF)" on en wiki. That's gone through, but it seems only on en wiki (doesn't say "Logging you in to Wikimedia's other projects"). What's my next step? [00:54:28] you're messing up with tense [00:54:55] spagewmf: Global renames are pretty painful. [00:54:59] lead section is horrible; better https://en.wikipedia.org/wiki/Equivalence_relation#Definition [00:55:45] Nemo_bis: Are you saying a duplicate is not a copy? [00:55:59] Brooke: I've no idea how the two terms differ [00:56:14] Hmmm. [00:56:20] I think I see what you're saying. [00:56:31] light came [00:56:33] :p [00:56:41] But even with a copy, you still have the original. [00:56:54] There's always the one that came first. It's timestamped, even. [00:57:08] I see duplicates just as another way to mark templates. [00:57:15] templates? [00:57:22] It's different from blocker vs. blocked [00:57:24] bugs, sorry [00:57:39] It's different from blocker vs. blocked only in that it doesn't have nice graphs, but that's a bugzilla issue [00:57:39] Right, but there's something to be said about the sequential nature of the bugs. [00:57:48] naaaaaaaaaaaaaah [00:57:48] They're chronologically sequential. [00:58:03] So to say "39000 is a dupe of 2400" is just wrong. :-/ [00:58:04] spagewmf, hey, so what do you want your user name to be across all projects - "S Page (WMF)"? [00:58:20] Brooke: doesn't make any sense [00:58:30] What doesn't? [00:58:32] That view? [00:58:58] "The problem is a duplicate of an existing bug." != "The problem was a duplicate of a bug existing before its creation" [00:59:24] That it doesn't make sense is what I've been saying. ;-) [00:59:37] aka you're adding unstated assumptions somewhere [00:59:45] spagewmf, we don't actually have a global renaming solution, so that means that if you want a rename you need to be renamed on each of the wikis that you have an account [00:59:52] They're not unstated. The bugs are chronologically sequential. [00:59:55] They're ordered. [00:59:59] Brooke: the correct definition is the first, you're trying to impose the second [00:59:59] You're viewing them as unordered. [01:00:08] Which doesn't make much sense, because they're ordered. And timestamped. [01:00:13] ordering doesn't mean that the first has some entitlemenet [01:00:25] It does when you're determining which to keep open. [01:00:26] for the username "SPage (WMF)" - https://toolserver.org/~pathoschild/stalktoy/?target=SPage%20(WMF) shows all of your accounts [01:00:32] Brooke: tautology [01:00:43] whilst for https://toolserver.org/~pathoschild/stalktoy/?target=S_Page%20(WMF) - that shows you've only got one account as "S Page (WMF)" [01:00:55] Nemo_bis: https://en.wikipedia.org/wiki/First-come,_first-served [01:01:02] It's a fairly old principle... [01:02:36] Thehelpfulone: https://toolserver.org/~pathoschild/stalktoy/?target=SPage%20(WMF) eh? [01:02:44] Thehelpfulone Is there any way to get a Global account for "S Page (WMF)"? [01:03:04] You can leave your username as "SPage (WMF)" and continue to use that, or if you want your username to be "S Page (WMF)", then you need to request a rename on those wikis at https://toolserver.org/~pathoschild/stalktoy/?target=SPage%20(WMF) where you have edits. I can do that for metawiki, testwiki and outreach wiki - and I can find someone to do the rename for mediawiki.org [01:03:38] Brooke, hmm? [01:03:49] Thehelpfulone: Oon't mind me. [01:03:51] Don't [01:03:56] Brooke: which is not stated anywhere in our bugzilla; besides, a dupe is not unserved, the issue it is about is just moved elsewhere [01:04:28] Nemo_bis: I kind of always assumed that everyone else looked at the numbering and felt similarly. [01:04:35] But, sure, it could/should be better documented. [01:05:01] Thehelpfulone I'll do that, but it sounds like they'll all be local non-SUL accounts. Hmm, can I go to a wiki where I don't have edits, sign up as "S Page (WMF)" and have that be a unified login? [01:05:35] spagewmf, actually they will be SUL account [01:05:40] accounts* [01:06:17] Thehelpfulone great, thanks. [01:06:36] if I rename you to be [[User:S Page (WMF)]] then there's a way that you can link all the accounts to make them a SUL account [01:07:58] spagewmf, sure, so can you just make an edit from [[User:SPage (WMF)]] perhaps on my user talk page on meta - https://meta.wikimedia.org/wiki/User_talk:Thehelpfulone confirming that you'd like the rename for metawiki, testwiki, outreachwiki and mediawikiwiki? [01:08:53] and make sure that you are not logged as [[User:S Page (WMF)]] on any wikis [01:09:53] Brooke: actually, when someone closes my bug as duplicate I receive an email which says I've been kindly added to the cc of the other bug. I've always taken that as a kind redirect, a pointer, a "let's continue here". [01:10:21] Nemo_bis: I personaly don't like having my bugs marked as duplicates. [01:10:34] Brooke: why [01:10:39] It usually means to me that I was stupid and couldn't find the original bug. [01:10:42] Or the original bug was vauge. [01:10:43] vague [01:10:58] Brooke: it doesn't if the other bug has been filed afterwards [01:11:11] so this case doesn't make anyone feel stupid [01:11:21] It can also make me as a bug filer feel ignored. [01:11:32] I dunno. [01:11:38] I view it as a principle of fairness. [01:11:44] I too feel stupid in the other case, but I think "oh well, I've added some sort of redirect for people searching my definition of this bug" [01:12:15] Except resolved bugs rarely show up in searches. ;-) [01:12:27] Brooke: duplicates are shown by default [01:12:37] Fair enough. [01:12:47] which is another reason why bugzilla surely doesn't treat them as trash as you seem to say [01:13:07] maybe we should call it RESOLVED #REDIRECT [01:14:41] and a bug like that receiving any sort of change after years is a relief for the reporter (at least when I'm the reporter), whatever the change [01:16:18] hmm discussed with you instead of reading this medieval philosophy vs. Wikipedia article, now time to bed [01:30:36] Thehelpfulone, done and both accounts logged out. Thanks x 10. [02:26:58] !log LocalisationUpdate completed (1.20wmf10) at Wed Sep 5 02:26:57 UTC 2012 [02:27:06] Logged the message, Master [02:44:09] is there a way to get around the throttle of number of e-mails which can be send? [02:51:43] !log LocalisationUpdate completed (1.20wmf11) at Wed Sep 5 02:51:43 UTC 2012 [02:51:52] Logged the message, Master [03:23:57] Romaine: Anything's possible. What are you trying to do? [03:24:18] Admin accounts (and maybe bots?) usually have a higher limit for e-mail per day, I think. [13:11:20] http://enwp.org/File:Basilique_Saint-Pierre_et_Saint-Paul.jpg is a recently upload (3 July 2012) Wikimedia Commons file that's gone missing [13:14:33] Dispenser: hi [13:14:38] hi [13:14:39] thanks for the report, I'm looking at it [13:16:22] FYI, I'm sending a bot through Commons fetching all thumbnails at an usual size (48x48), I've posted other problems at https://commons.wikimedia.org/wiki/Commons:Village_pump [13:17:46] not on ms7 either [13:18:45] oh good morning apergos [13:18:51] still jetlagged [13:19:03] very badly [13:19:06] woke up at 4 am again [13:19:15] lemme check snapshots on ms7, see what's around [13:20:03] hm, swift says not found [13:20:11] I wonder if that's checking the actual object or the container [13:21:16] we have it in the july 2 snapshot, now to find out when it disappeared [13:22:06] sept 3 snapshot it's missing, aug 27 it's there [13:25:06] apergos: so you said you've listed containers before? [13:25:47] sec [13:25:59] ../partial-2012-08-31_23:15:00 has it [13:26:05] ../partial-2012-08-31_23:30:00 doesn't [13:26:18] so that;s the time frame of the disappearance [13:26:25] utc times [13:26:43] wow, I'm impressed :) [13:27:02] yeah well when solaris/zfs goes, that will go too, it's too bad cause it's very helpful [13:27:07] have you listed swift containers before? do you know how expensive is it? [13:27:08] but it's not worth the aggravation [13:27:16] yes I have and it isn't very [13:27:31] that's how I got those three container lists for the data consistency check [13:27:48] yeah that's why I'm asking [13:27:50] list wikipedia-commons-local-public.0a | grep Basilique_Saint-Pierre_et_Saint-Paul.jpg [13:27:53] archive/0/0a/20120610031342!Basilique_Saint-Pierre_et_Saint-Paul.jpg [13:27:56] archive/0/0a/20120823035721!Basilique_Saint-Pierre_et_Saint-Paul.jpg [13:28:06] archive [13:28:22] yes [13:28:32] so maybe a failed delete? [13:28:51] or a failed upload of a new version or something? [13:29:02] I guess if it was deleted it would have been moved to privatae [13:29:15] there commons page has a 06-10 revision, but not a 08-23 [13:29:24] the current one is 06-10 in fact [13:29:32] but it seems to be missing 08-23 and the 08-30 deletion [13:30:52] hmm the history page there is interesting all right [13:31:05] Dispenser: the original is at http://upload.wikimedia.org/wikipedia/commons/archive/0/0a/20120823035721!Basilique_Saint-Pierre_et_Saint-Paul.jpg [13:31:12] Dispenser: clearly a bug, we'll have a more in depth look [13:31:13] you can see n aug 31 where they noticed that it was 404ing [13:31:33] I can't see an aug 31? [13:31:46] http://commons.wikimedia.org/w/index.php?title=File:Basilique_Saint-Pierre_et_Saint-Paul.jpg&action=history [13:32:47] not much in the logs for the page, that's bad [13:32:50] oh yay, we lost the thumb now too [13:33:07] my bad, I action=purge'd it [13:33:40] woops [13:36:30] * apergos looks at the recent changes table for those 15 minutes [13:36:38] maybe something will have made it in there [13:37:37] we have 08-23 in swift and 08-31 in ms7, that's kind of strange [13:43:29] nada, rats [13:45:18] there's some deployments etc but they are a few hours later [13:46:40] and nothing good from theirc logs of the wm ops channel either [13:46:44] bummer :-( [13:49:00] so the things I see in archive have Jun 9 and Jun 10 dates on em (as they should)) [13:50:42] btw there are thumbs on ms5 still for it :-D [13:51:19] n big ones though [13:51:23] *no [14:02:14] I wish we had a way for people to automatically let us know when a move or delete or whatever fails [14:02:50] because then we could at least go look at the fail reports and track down the user, see what they were doing and what the symptoms were... this way though there's nothing [14:03:35] "Mediawiki had crashed during your action. Do you wish to report to the developers?" ;) (I wonder what MS does with the millions messages every day…) [14:04:15] I don't care about failed reads, but failed actions that change things? yeah it would be nice to know [14:09:18] there's no reasn any sort of sync between ms7 and swift would have been happening on aug 31 is there? [14:09:40] as in "remove from ms7 if it's not in swift" or something [14:14:41] hmm there are some failed heads from swift from 29/Aug/2012:18:06:00 [14:14:56] 404s [14:18:37] * apergos wishes yet again we kept syslogs longer than a week [14:28:10] so this is interesting, I see on ms-be2 in /var/log/messages.2.gz, some heads that succeed [14:28:13] and then a delete [14:28:51] [23/Aug/2012:03:57:25 +0000] "DELETE /sdi1/61208/AUTH_XXX /wikipedia-commons-local-public.0a/0/0a/Basilique_Saint-Pierre_et_Saint-Paul.jpg" 204 - "-" "-" [14:29:00] that's why the file is gone anyways [14:29:05] paravoid: [14:29:21] * paravoid nods [14:30:54] so it's interesting that the delete didn'tmake it over to ms7, I guess that there was a failure in the middle someplace [14:31:01] I'm going to pack it in and go to the office [14:48:50] !log reedy synchronized php-1.20wmf11/extensions/CheckUser 'CheckUser to master' [14:49:00] Logged the message, Master [16:00:34] Bonjour: Fehler bei Datei-Löschung: Snowcreekwater3web small - West Virginia - ForestWander.jpg [16:00:42] Fehler bei Datei-Löschung: Die Datei „mwstore://local-multiwrite/local-public/3/36/Snowcreekwater3web_small_-_West_Virginia_-_ForestWander.jpg“ befindet sich, innerhalb des internen Speicher-Backends, in einem inkonsistenten Zustand. [16:01:04] =Can't delete https://commons.wikimedia.org/w/index.php?title=File:Snowcreekwater3web_small_-_West_Virginia_-_ForestWander.jpg&action=delete [16:01:32] But I would like ;-) [16:24:15] rillke: Could you report this incident on https://bugzilla.wikimedia.org/show_bug.cgi?id=39221 please? [16:24:30] It's been happening fairly regularly [16:36:15] {{done}}. Thank you. [16:37:54] RoanKattouw: are exceptions included in fatalmonitor? [16:40:29] I don't think so.. [16:41:02] Even more so now, they'll be on fluorine.. [16:46:39] I don't think they are [16:46:48] They'll be in exceptions.log [17:00:24] is there a way to get around the throttle of number of e-mails which can be send? [17:00:42] Multiple accounts [17:00:46] :-) [17:01:13] How is your ratelimit? https://commons.wikimedia.org/w/api.php?action=query&meta=userinfo&uiprop=ratelimits [17:01:30] Or are you just spammed by someone? [17:01:49] no, I want to send e-mails for a project [17:03:05] With my sysop-account, I don't see any limits there ... [17:04:01] >12 hours ago I send a bunch of mails from our project account (which isn't sysop) as promised to our users, but got stuck with the e-mail throttle [17:04:48] Right [17:05:00] Most likely you'll need to have a bot flag or be a sysop for this [17:05:03] * RoanKattouw looks up the rate limits [17:05:07] Ah, on dewiki it is [17:05:49] Yeah 200/day [17:06:01] for "normal users" [17:06:03] sysop, bureaucrat, bot and accountcreator are exempt [17:07:44] bug triage now in #wikimedia-dev [17:08:37] the account is not supposed to do edits, neither actions as sysop [17:09:01] (it is on Commons by the way, but same limits so far I see) [17:09:14] is therea way to get around the throttle? [17:09:23] flag it as bot [17:09:26] Yeah [17:09:37] ask 99of9 [17:09:43] (e.g.) [17:10:52] https://commons.wikimedia.org/w/index.php?title=Special:ListUsers&group=bureaucrat [17:12:13] someone mentioned yesterday: https://gerrit.wikimedia.org/r/gitweb?p=operations/mediawiki-config.git;a=blob;f=wmf-config/InitialiseSettings.php;hb=HEAD#l6389 [17:13:04] Oh, there's an IP exemption list too? I didn't know that [17:15:09] guys, is there a way to add additional "view x" links to Whatlinkshere pages? like 1000? [17:25:59] anyone? [17:27:23] Vito-Genovese: This is probably something to ask at #mediawiki or at bugzilla [17:27:29] !log aaron synchronized php-1.20wmf11/includes/StreamFile.php 'deployed 8a63ff8c8558b1ca9f2265da71376e494053c234' [17:27:37] Logged the message, Master [17:27:53] thank you rillke, will do [17:28:01] !log aaron synchronized php-1.20wmf11/includes/filerepo/file/File.php 'deployed 8a63ff8c8558b1ca9f2265da71376e494053c234' [17:28:10] Logged the message, Master [17:28:38] !log aaron synchronized php-1.20wmf11/thumb.php 'deployed 8a63ff8c8558b1ca9f2265da71376e494053c234' [17:28:47] Logged the message, Master [17:51:54] !log catrope synchronized wmf-config/InitialiseSettings.php 'Enable CORS for the API (347f386e2144688bcd4c80d24bd9f3be7d135d74)' [17:52:04] Logged the message, Master [17:52:13] !log catrope synchronized wmf-config/CommonSettings.php 'Enable CORS for the API (347f386e2144688bcd4c80d24bd9f3be7d135d74)' [17:52:22] Logged the message, Master [17:54:26] RoanKattouw: Will CORS be available on meta? I am just concerned about small wikis that add evil JS code that would make visiting stewards performing bad things under their account... [17:55:12] or simply have a security vulnerable in their js [17:55:20] CORS is available from all wikis to all wikis, except to private wikis [17:55:26] Yes, this did come up [17:55:49] OTOH meta and commons are probably also going to be the most popular destination for legitimate CORS usage [17:56:39] I just tested CORS from enwiki to mediawiki.org and it works, yay! [17:56:49] Nice! [18:08:11] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: Misc wikis to 1.20wmf11 [18:08:20] Logged the message, Master [18:14:49] Do mailman command such as unsubscribe sent to -owner acutally work/ [18:14:51] ? [18:14:55] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikinews, wikiversity and wikimedia to 1.20wmf11 [18:14:57] commands* [18:15:04] Logged the message, Master [18:16:11] RoanKattouw: timestamps! [18:16:35] Reedy: Yeah did we end up merging your change into wmf11? [18:16:57] Looks like we didn't [18:17:01] I don't think it was approved by anyone into trunk [18:17:11] OK doing so now [18:18:33] Reedy: https://gerrit.wikimedia.org/r/22743 [18:19:33] Suppose we should ask asher nicely to get the schemas fixed up [18:19:40] !log reedy synchronized php-1.20wmf11/includes/Timestamp.php '0000...' [18:19:49] Logged the message, Master [18:22:29] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: wikiquote and wiktionary to 1.20wmf11 [18:22:39] Logged the message, Master [18:22:41] oh god.. is that garbage left behind from some ancient mysql 4.0 -> 5.x varbinary migrations? [18:23:17] It's only affecting 10s of wikis... [18:23:17] https://rt.wikimedia.org/Ticket/Display.html?id=3514 [18:24:23] binasher: Probably that *plus* an issue with creating new wikis for some length of time, see the RT ticket Reedy referenced [18:24:32] There are two different ways in which the schemas deviate [18:25:39] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: Everything else non wikipedia to 1.20wmf11 [18:25:49] Logged the message, Master [18:25:57] yuck [18:26:06] at least those are fairly small wikis [18:26:12] when is 1.20wmf11 going out? [18:26:20] heh [18:26:26] A minute or so ago for everything non wikipedia [18:26:39] We've got the === "\0.... "hack" in wmf11 now [18:27:37] !log reedy synchronized wmf-config/CommonSettings.php 'Use an updated version of cortado for testwiki and test2wiki' [18:27:46] Logged the message, Master [18:29:17] apergos: looks like updated cortado works fine on test2wiki [18:29:17] https://test2.wikipedia.org/wiki/File:WikipediaBlackoutTwitterStream.ogv [18:29:25] Just getting a lot of 404s for thumbnails for it.. [18:30:13] i'll take care of the schema fixes so the \0 workaround can be backed out before wmf12 [18:30:33] binasher: Thanks man [18:30:58] has the "wiki creator" script been fixed? (i have no idea how we make new projects) [18:31:12] It uses the schema in maintenance/tables.sql [18:31:24] Which has user_last_timestamp varbinary(14) NULL default NULL [18:32:38] !log reedy synchronized wmf-config/CommonSettings.php 'Go back to older cortado to check thumbnail isue' [18:32:44] binasher: It would appear so because wikimania2013wiki is fine [18:32:47] Logged the message, Master [18:33:05] And what he said [18:33:11] good good [18:38:34] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: commonswiki back to 1.20wmf10 for testing [18:38:44] Logged the message, Master [18:40:22] !log commonswiki to stay on 1.20wmf10 for the meantime [18:40:31] Logged the message, Master [18:43:30] https://upload.wikimedia.org/wikipedia/commons/thumb/d/d8/WikipediaBlackoutTwitterStream.ogv/mid-WikipediaBlackoutTwitterStream.ogv.jpg [18:43:38] https://upload.wikimedia.org/wikipedia/commons/thumb/d/d8/WikipediaBlackoutTwitterStream.ogv/240px--WikipediaBlackoutTwitterStream.ogv.jpg [18:44:33] Reedy: good to know [18:44:35] what's next? [18:45:22] changing the jar breaks the thumbnailing of videos seems a bit weird [18:45:40] Oh, wait, which issue? ;) [18:46:31] cortado off of ms7 :-D [18:46:58] video playback is fine on the newer version [18:47:06] that's cool [18:47:28] on test/test2, you updated the wgCortadoJarFile and you got 404s? [18:47:38] On thumbnails, yeah [18:47:46] urk [18:48:01] I've changed it back, but it doesn't seem to want to revert (dunno if it's cached) [18:48:12] !log reedy synchronized wmf-config [18:48:21] Logged the message, Master [18:49:54] ohboy [18:50:48] It's weird, using chrome to search through all the included files, on commons it can find cortado in the sources, in chrome it can't [18:50:48] wtf [18:51:02] weird! [18:51:38] how about cortado.jar? [18:51:55] as opposed to the new one [18:52:12] (I see where you made the config file change and changed it back, sure looks harmless) [18:53:32] were you testing on test or test2? [18:54:39] I was using test2, but it seems to have stuck on test too [18:54:51] hm [18:55:35] It looks like the config hasn't been synced [18:56:03] well if you didn't sync it... [18:56:14] I mean I thought test doesn't need it synced [18:56:14] !log reedy synchronized wmf-config/CommonSettings.php [18:56:23] Logged the message, Master [18:56:37] but test2 does but it wants a conditional? [18:57:00] yeah, otherwise it'd be sitewide [18:57:09] right [18:58:28] so now you've synced it with nothing happening for test/test2 [18:58:35] since that line is commented out [18:59:03] I wonder what the default value is, probably nothing good [18:59:28] !log reedy synchronized wmf-config/CommonSettings.php [18:59:34] $wgCortadoJarFile = "cortado-ovt-stripped-0.6.0.jar"; [18:59:36] Logged the message, Master [18:59:48] ah changed. good [19:00:35] !log reedy synchronized wmf-config/CommonSettings.php 'revert all back to old cortado.jar' [19:00:36] well I'm staring at it, perms are just like the rest of the files [19:00:44] Logged the message, Master [19:01:08] reedy@fenari:/home/wikipedia/common$ mwscript eval.php test2wiki [19:01:08] > var_dump( $wgCortadoJarFile ); [19:01:08] string(39) "//upload.wikimedia.org/jars/cortado.jar" [19:01:22] uh huh [19:09:12] what error does it give you anyways? [19:10:52] Which error where? [19:11:16] when you try to look at an ogg thumb [19:11:42] I'm getting them spewed into teh console [19:11:42] Failed to load resource: the server responded with a status of 404 (Not Found) https://upload.wikimedia.org/wikipedia/commons/thumb/d/d8/WikipediaBlackoutTwitterStream.ogv/240px--WikipediaBlackoutTwitterStream.ogv.jpg [19:12:01] Commons is using https://upload.wikimedia.org/wikipedia/commons/thumb/d/d8/WikipediaBlackoutTwitterStream.ogv/mid-WikipediaBlackoutTwitterStream.ogv.jpg [19:13:11] "GET /sde1/12169/AUTH_43651b15-ed7a-40b6-b745-47666abf8dfe/wikipedia-commons-local-thumb.d8/d/d8/WikipediaBlackoutTwitterStream.ogv/mid-WikipediaBlackoutTwitterStream.ogv.jpg" 304 - "-" "txf1bdbb14b22e482fb140b6132a7d4218" [19:13:17] grrrrrr [19:13:20] * apergos stabs self [19:13:28] heh, back in a few [19:13:31] this is what i get for trying to do real work when jetlagged [19:14:16] I think that string is in puppet but still [19:19:00] I wonder what the difference between the ov and ovt versions is.. [19:19:48] Oh, old was ovt (in git), so don't need to worry about [19:23:09] !log krinkle synchronized docroot/default/wmf_logo_with_text.png 'Ia7ca3eaf' [19:23:18] Logged the message, Master [19:23:34] !log krinkle synchronized docroot/default/index.html 'Ia7ca3eaf' [19:23:43] Logged the message, Master [19:24:11] ok well you are the only one who gets to see the errors as they seem not to be logged anywhere, sadly [19:27:16] can you test it on test? because that's srv193 and we can actually sit on the host and watch what it thinks it's doing (maybe) [19:34:54] It presumably must be cached JS somewhere.. [19:36:55] well scaling of that is working on the general wikis now anyways [19:36:58] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: test2 back to wmf10 for a few mins.. [19:37:07] Logged the message, Master [19:38:03] !log reedy rebuilt wikiversions.cdb and synchronized wikiversions files: test2 back to wmf11 [19:38:13] Logged the message, Master [19:39:07] !log reedy synchronized wmf-config/CommonSettings.php [19:39:16] Logged the message, Master [19:39:28] !log catrope synchronized php-1.20wmf11/includes/ImagePage.php 'Fix for bug 40018 (ac25a9d1127848269b6c9766262c23f961b35c5b)' [19:39:37] Logged the message, Master [19:39:48] Sorry! This site is experiencing technical difficulties. [19:39:50] Reedy: ---^^ With that, it should be safe to put Commons back on wmf11 [19:39:54] Can't contact the database server: Unknown database 'enwiki' (10.0.6.21) [19:39:59] :O :O :O [19:40:03] (Can't contact the database server: Unknown database 'nlwiki' (10.0.6.35)) [19:40:04] RoanKattouw: I went to http://meta.wikimedia.org/w/index.php?title=File:Berlinagreement1.png&action=render&uselang=en to see if it was fixed [19:40:05] ah, so its not me, food :) [19:40:09] Guys - any clues what's up? [19:40:14] nvm, just saw [19:40:16] Query: SELECT tmi_value FROM `translate_messageindex` WHERE tmi_key = '6:berlinagreement1.png' LIMIT 1 [19:40:17] Function: DatabaseMessageIndex::get [19:40:19]