[00:13:39] gn8 folks [02:57:22] errors with WIkivoyage, something has gone wrong it seems [04:44:19] gerrit atom feed what what [04:44:21] http://schmerrit.wmflabs.org/gerrit/changes/mediawiki/core.atom [06:08:28] Hey, how did Wikipedia servers do through election day? Are there any realtime stats or graphs for the public to see? [06:13:38] lexein: http://ganglia.wikimedia.org/ ? [06:15:19] well I hope those 5xx errors are unrelated? https://gdash.wikimedia.org/dashboards/reqerror/ [06:21:30] Nemo_bis Wow. I had never seen those graphs. I didn't even know the image scalers were _live_. I know so little. [06:21:47] heh [06:22:15] the second URL *might* be showing you a lot of timeouts served to people looking for Obama/Romney articles [06:22:31] and other stuff hard to digest (for the parser; and perhaps the reader too) [06:22:57] [but I'm just speculating for the purposes of narrative/fiction] [06:24:10] Hee! I like the sole spike in "HTTP 5xx Responses -1day" [06:24:47] nothing, then 400k 5xx within, what, a minute? [06:27:43] yes, that's definitely an outlier [06:53:14] Thanks again [07:19:11] hey mark? [10:05:13] Reedy: can you sync TMH to master in wmf3 please [10:39:34] i can make fatal errors on meta… :( [10:39:36] PHP fatal error in /usr/local/apache/common-local/php-1.21wmf3/extensions/Translate/tag/PageTranslationHooks.php line 367: [10:39:36] Call to undefined method TPException::getWikiText() [10:39:47] if you want I can send the text that caused it [10:55:11] Denny_WMDE: file a bug against translate and include all details please [10:55:51] because of travels probably nobody in language team will be able to look into it soon [10:56:42] ok [11:22:21] done https://bugzilla.wikimedia.org/show_bug.cgi?id=41844 [11:57:19] Not 100% if this is the right place for this. Donated some and after entering credit card info and making the payment, got this page http://i.imgur.com/zXh4p.png [11:57:30] empty special page :P [11:58:10] bank website says that payment went trough, so that's all good :) [11:59:02] Leevi: i'm not sure anyone from the fundraising team is around, but thanks for reporting. i'll email them, just in case they miss this. [11:59:30] thanks [12:01:40] Leevi: (and, thanks for donating!) [12:03:06] np [12:22:52] oh and since I'm already here. Is there any technical reason why all the irc links on https://en.wikipedia.org/wiki/Wikipedia:IRC are like this irc://irc.freenode.net/#wikipedia [12:23:02] that # makes the links not work [12:29:21] would have fixed them myself, but can't because the page is protected somehow :P oh well [13:02:53] j^: lol, sure [13:03:39] Leevi: if your irc client has registered protocol handlers, it will open straight into your irc client [13:03:43] It should possibly have both [13:09:55] hmm, looks like you are right [13:10:39] * Leevi goes to bug Opera developers about it [13:33:33] apergos: I presume you received my mail? [13:46:09] Brooke: I've always been told that easy and shell are incompatible [13:46:48] although apparently 40 bugs have both [13:46:49] Not in the age of Git! [13:47:01] it still doesn't make sense [13:47:01] easy --> any developer can work on it. [13:47:06] all config changes are easy [13:47:15] <^demon> I thought easy meant it was easy to do? [13:47:19] It does. [13:47:29] It isn't very difficult to add a changeset for most shell changes. [13:47:33] non-shell users can't fix shell bugs [13:47:44] They can half-fix them. [13:47:49] Which usually brings us closer to a full fix. [13:47:56] allegedly [13:48:04] Basically, [13:48:14] if you add "easy" right now, Krenair will come along and add a changeset. [13:48:24] And then the bug can move forward. [13:48:24] he does anyway [13:48:29] he knows shell keyword [13:48:33] Maybe. [13:48:38] No, for sure. [13:48:52] Maybe someone other than Krenair will work on the bug. [13:49:05] Anyway, the two keywords aren't really incompatible in my view. [13:49:05] And, [13:49:05] and then he will be commit-conflicted by Dereckson [13:49:19] that was a pretty bad bug description on your part. [13:50:37] ?? [13:59:02] Nemo_bis: What's confusing? [14:00:01] Technically most developers can't "fix" any bug. Their code still has to be reviewed by someone and then merged and deployed, just as shell requests must be. [14:00:12] I don't see an incompatibility between the two keywords. [14:01:50] * Nemo_bis shrugs [14:02:47] From my experience (anecdotal, no hard evidence unfortunately), marking a bug as easy dramatically decreases time to resolution. [14:02:57] So I try to mark as much as I can. [14:03:27] I'm pretty sure you're wrong. :) [14:03:32] (for shell) [14:03:53] Hmmm. Maybe. Shell requests have their set of efficiency/resolution issues, though. [14:04:07] ^demon: How's that configuration database coming along? (o: [14:04:50] <^demon> Oh, I finished that like 4-5 months ago. [14:04:50] <^demon> It's just so awesome, I don't want to share ;-) [14:05:07] To me "easy" refers to making the patch, not necessarily to getting it in (having shell access). So to me it's not incompatible per se [14:05:19] just useless ;) [14:05:23] andre__: I think https://bugzilla.wikimedia.org/show_bug.cgi?id=26992 should be higher priority. [14:05:33] It has the power to resolve dozens (maybe hundreds?) of bugs. [14:06:44] Brooke: I agree, but I think that Chad is too busy with Gerrit stuff still. I wonder if anybody else could take a look. [14:07:22] It's unclear to me what next steps are. [14:07:29] <^demon> Someone needs to just sit down and start coding. [14:07:36] Is there a design in mind? [14:07:39] A design architecture, I mean? [14:07:45] <^demon> Roughly. [14:07:54] Is that posted anywhere? :-) [14:07:55] if there is it would be great to share it. [14:08:04] so others might be able to pick up work. [14:08:11] * Brooke flees to work. [14:08:19] <^demon> Brooke: http://www.mediawiki.org/wiki/Requests_for_comment/Configuration_database [14:08:22] <^demon> It needs more work. [14:08:33] So that's step one, I guess. [14:08:49] We should make that clearer on the bug, I think. [14:08:50] oops, that's mentioned in the report, true [14:08:50] <^demon> Anyone can feel free to work on that--it's been on my eternal back-burner and andre__'s right, with Gerrit I just haven't had the time to touch it. [14:09:19] Mentioned by me, no less. [17:02:53] ping AaronSchulz [17:22:17] Reedy: could you check which template causes the currently high job queue on zhwiki? this morning (19 hour ago) the job queue was low (40000 entries) [17:23:00] quit [17:23:26] array( 'transcode_key' => $key ), [17:23:30] wrong channel [17:23:36] Merlissimo: I can in a little bit [17:51:08] Merlissimo: at the top of the queue are jobs from AM yesterday [17:52:01] 10 | Country_data_ITA [17:52:05] 20121106012635 [17:54:30] my bot stopped editing on zh main templates because of high job queue from 31 oct. to 5 nov. since nov 6th 1 am it started editing three location maps and many babel templates. since 1 pm on 6 nov. the script stopped again because of high job queue [17:55:28] 877,000 or so jobs [17:55:35] <^demon> Yeah, I just looked that up. [17:55:38] <^demon> Ouch [17:55:52] so moving also location map langlinks (as done with country data) to subpage would help i think [17:56:01] enwiki has 75,000 [17:57:04] job runners look reasonably active [17:57:08] on nov 6th 0:41 am the jobqueue on zhwiki was about 43000 [18:00:11] Merlissimo: babel templates?? [18:01:43] Nemo_bis: on ckbwiki many new babel templates were created (because of the enabled babel lextension i don't know why, too) [18:02:04] http://ckb.wikipedia.org/wiki/%D8%AA%D8%A7%DB%8C%D8%A8%DB%95%D8%AA:%D8%A8%DB%95%D8%B4%D8%AF%D8%A7%D8%B1%DB%8C%DB%8C%DB%95%DA%A9%D8%A7%D9%86/CalakBot [18:03:00] Merlissimo: makes no sense, why did they do this? [18:03:35] you have to ask the bot operator Calak [18:03:35] (I mean, in their contorted mind) [18:03:48] and what about zh.wiki? [18:04:27] they have babel templates because these exist before there was a babel extension [18:04:49] but why do you still need to edit them [18:05:08] do they have any particular reason to use them rather than disable them? [18:05:23] adding ckb langlinks [18:05:43] *facepalm* [18:06:20] can't a sysop on zh.wiki disable them? [18:06:32] if they're unused they should kill the job queue I hope? [18:06:37] babel templates are not causing job queue problems, i think [18:06:53] ah ok [18:07:11] as Reedy wrote i think it is the location map [18:07:23] thet's surely a worse offender [18:43:06] hello [18:43:11] hey sumanah [18:43:24] so I was talking about https://gerrit.wikimedia.org/r/#/c/32192/ [18:43:26] average_1rifter: ok, so, https://gerrit.wikimedia.org/r/#/c/32192/ - what is it blocking? [18:45:44] average_1rifter: also you might want to read https://www.mediawiki.org/wiki/Git/Code_review/Getting_reviews [18:46:41] and average_1rifter you can also look at #wikimedia-operations and look for who's mentioned as being on RT duty this week -- that person is sort of on the "interrupts" rotation and can often help you get questions answered and get blockers fixed [18:48:08] alright, thank you sumanah [18:48:19] Daniel just sent some comments on the review [18:48:20] I shall read them [18:48:27] ok! [18:48:33] glad I could help. [18:48:35] off to a meeting :) [21:55:38] Reedy: still around? [21:55:42] yup [21:55:59] should we reschedule the Monday deployment (US Veterans day) or keep it where it is? [21:56:38] it might make more sense to do it earlier in the day on Tuesday [21:57:29] morning TimStarling! [21:57:50] morning [21:58:43] so nice to have our days overlapping a little more sanely [21:59:21] yes [22:00:33] Reedy: lemme know what you think....gotta run now [22:00:51] robla: just so you know, I'm on my way out the door for the night [22:00:56] (if I drop off of IRC, I'll read the log) [22:01:04] sumanah: k...enjoy your evening [22:01:13] :)