[13:19:52] 1.20wmf6 is to be branched within a couple of hours [13:41:06] * apergos thinks about space issues: snaps ok? check. hume ok? check [13:41:19] we didn't have anything else cropping up did we? [13:42:19] errm, I don't think so [13:42:37] On which note, let me kill the wmf4 l10n folder (properly this time) [13:42:43] woo hoo [13:43:48] -rw-rw-r-- 1 l10nupdate l10nupdate 1626122 2012-06-18 02:39 .nfs00000000000d262300000145 [13:45:33] !log Killed php-1.20wmf4/cache/l10n from mediawiki-installation hosts [13:45:49] yay [13:49:06] for wmf5 its 771M [13:50:37] <^demon> Are we cleaning up the old installs, eg wmf1? [13:51:05] we don't have wmf1 still checked out [13:51:31] <^demon> Ah ok [13:51:38] we need to keep the folders around for last day of version being used + number of days squid caches stuff for [13:52:17] I log'd in the wrong channel [13:52:18] fail [13:53:55] <^demon> If we consolidated every one and every bot in the same channel, then we'd never mislog or miss any discussions :) [13:54:47] * Reedy high fives ^demon [14:02:12] ^demon: do you know if gerrit can show diffs of SVG files? Or XML? [14:02:27] <^demon> Should be doable. What's the mime-type? [14:02:52] <^demon> I added image/* to the diff rules. [14:03:30] Good question. I know SVG is XML, but... [14:03:53] are there any known issues with thumbnail generation for logged-out users currently? [14:04:15] The MIME type for SVG is "image/svg+xml" [14:04:23] says w3.org [14:04:31] <^demon> Hmm, it should be showing up. Can you give me an example where it's not? [14:05:29] ^demon: Nope. I'm going to need a new repo for that :) I created one on github to work on it until you were online [14:05:49] But github is stupid and doesn't diff SVGs [14:06:04] So I was wondering if gerrit did [14:06:13] <^demon> Well I told image/* to diff, and anything it can detect as text will be diffed. [14:06:16] if you look at http://de.wikipedia.org/wiki/Fußball-Weltmeisterschaft_1982/Qualifikation#Qualifikation_zur_Fu.C3.9Fball-Weltmeisterschaft_1982 and are not logged in, the flags of algeria and peru aren't displayed, it only says "30px". [14:06:18] <^demon> So I hope it works :) [14:06:26] ok, so I'll add a new repo request. Thanks :) [14:10:50] ^demon: You probably have other things to do, but when you get a chance: https://www.mediawiki.org/w/index.php?diff=554893&oldid=554628 [14:11:38] * guillom is going to make a few calls to schedule apartment visits in the meantime. [14:12:01] <^demon> guillom: May get to it today. If not then tomorrow. [14:12:39] hmm, okay. In that case, I'll continue to work on github and I'll add the URL to the request so you can clone it when you get to it [14:12:50] pill, an old version? [14:13:57] Platonides, you mean you can't reproduce? [14:14:18] I can [14:15:02] then i don't quite get what you mean by "and old version?" ;) [14:15:13] but the page source just contains 30px [14:15:22] I wondered if an old version of the page contained it [14:16:16] ah, ok, i'll have a look. [14:16:31] hehehe [14:16:42] compare http://de.wikipedia.org/w/index.php?title=Fu%C3%9Fball-Weltmeisterschaft_1982/Qualifikation&oldid=99999309&uselang=en with http://de.wikipedia.org/w/index.php?title=Fu%C3%9Fball-Weltmeisterschaft_1982/Qualifikation&stableid=99999309 [14:17:47] File:Flag_of_Algeria.svg was edited in commons after Fußball-Weltmeisterschaft_1982/Qualifikation review [14:17:51] yeah, that works. [14:18:20] it was approved 22 February 2012, reuploaded (several times) on 17. Mär. 2012 [14:18:41] I don't think FlaggedRevs should do that, though [14:19:22] what can we do to fix it? [14:19:47] i tried to purge the image pages, but that apparently didn't do the trick ;) [14:21:14] unflag the article? [14:21:25] flagging the current version should work [14:21:44] it probably needs an approval with the new flag [14:23:14] I'd talk about it with Aaron_Schulz [14:24:12] ok, i have now flagged the current version and purged the article. that seems to have worked. [14:24:22] ok. [14:24:39] thanks for your help. [14:30:11] you're welcome [15:08:06] Reedy, when was 1.20wmf4 deployed to enwiki? [15:08:12] at the same time as testwiki and mediawikiwiki, I guess [15:08:21] but there's no entry at the Server_admin_log [15:08:30] https://www.mediawiki.org/wiki/MediaWiki_1.20/Roadmap [15:09:43] why is it in such weird order? :S [15:10:05] make it ASC or DESC but not DESC sections with ASC entries inside [15:10:23] The information you most likely want is going to be at the top [15:10:25] I didn't create it [15:10:58] Fix it if it offends you [15:15:01] done [15:22:59] https://www.mediawiki.org/wiki/MediaWiki_1.20/wmf6 <-- that one kick asses [15:23:25] really great to see that the properly formatted commit messages are actually used for something :-]] [15:25:40] :) [15:33:13] New patchset: Hashar; "ts submit: make $verbose a simple boolean" [integration/testswarm] (master) - https://gerrit.wikimedia.org/r/12887 [15:33:45] New review: Hashar; "Picked from Ib0d81753 which had unrelated changes :-]" [integration/testswarm] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/12887 [15:33:47] Change merged: Hashar; [integration/testswarm] (master) - https://gerrit.wikimedia.org/r/12887 [15:36:00] Reedy: Is there a (public) script that produces those wikitext commit logs? [15:36:16] make-deploy-notes ? [15:36:20] it's in the tools/release repo [15:36:23] ok [15:37:01] Reedy: svn ? [15:37:05] git [15:37:39] ah, I see it now [15:37:40] mediawiki/tools/release.git [15:37:41] https://svn.wikimedia.org/svnroot/mediawiki/trunk/tools/ still has some make-* things too [15:37:44] git clone ssh://gerrit.wikimedia.org:29418/mediawiki/tools/release.git [15:37:49] New patchset: Hashar; "(bug 34815) Use QUnit ?module= query parameter." [integration/testswarm] (master) - https://gerrit.wikimedia.org/r/11830 [15:37:53] yeah, we don't tend to delete stuff from svn [15:38:24] by we you mean you? I see many things being deleted there recently when they are moved to git (unless they are very public oriented like extensions + core) [15:38:33] o_0 [15:38:35] Oh [15:38:37] Ask Chad then [15:38:39] <^demon> Deleting stuff from SVN is just likely to anger people who still have checkouts. Read-only is fine :) [15:38:44] <^demon> I haven't been deleting stuff from svn. [15:38:58] I guess "we" is those with rights to set read-only without having to wait. [15:39:15] I think it's been done as new git repos are created [15:39:23] in nearly all cases [15:39:23] <^demon> I haven't been doing it. [15:39:31] New review: Hashar; "Patchset 4:" [integration/testswarm] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/11830 [16:05:27] I wonder sometimes if I get unreasonably excited for jenkins-bot to tell me that my code is verified. [16:05:33] heh [16:07:11] https://www.mediawiki.org/wiki/MediaWiki_1.20/wmf6 [17:03:27] hey ^demon, I've got a do not submit from you on one of my changes... [17:04:21] Is there something I need to do to get that removed? [17:04:22] <^demon> Ah, let me remove that. [17:04:38] ah. looks like talking to you was the solution. :) [17:04:53] <^demon> Yeah, solution to a -2 removal is "ask the person who wrote it" [17:04:54] https://gerrit.wikimedia.org/r/#/c/11137/ [17:05:17] yeah [17:05:22] thanks for taking care of that! [17:05:38] I was a bit worried I was still missing something in my code. [19:07:23] I pushed a new patch set shortly after pushing a patchset to the same change id today [19:07:34] jenkins-bot verified all of my previous patchsets in minutes [19:07:41] but hasn't verified this one yet and it's been several hours. [19:08:39] is there a queue or something that I'm waiting on or did I hork it by pushing another patchset before it could finish building the first one? [19:10:21] <^demon> nullspoon: Which one wasn't verified? [19:10:33] 11137 [19:11:17] <^demon> hmm. [19:11:53] <^demon> hashar: http://integration.mediawiki.org/ci/ is timing out for me. [19:12:01] jenkins dead so [19:12:12] poor jenkins [19:12:15] <^demon> nullspoon: There's your answer. [19:12:18] (lerooyyyyy jeennnniinnnnns) [19:12:21] yup did not receive much love for the last 2 months or so :( [19:12:23] ahhah [19:12:30] haha [19:12:34] ^demon: box is dead :/ [19:12:47] <^demon> srsly? [19:12:52] looks like leroy jenkins rushed early. [19:13:22] switching to the guild channel [19:13:39] thanks guys [19:23:41] nullspoon: actually thank me whenever I setup the stuff you are waiting from me :-D [19:24:22] hashar: heh. Well thanks for looking into it at least. [19:24:23] :) [19:24:51] any idea why jenkins crashed? [19:24:59] waiting for it to finish fsck [19:25:17] must be some test that gone wild and rocketed jerkins to 100% usage [19:25:18] you're waiting for it to finish? [19:25:21] that must be a big partition [19:25:25] though I am unsure why ssh was unresponsive [19:25:33] 1TB I think [19:25:36] at least for a build server to take time. [19:25:39] wow [19:25:44] so we can checkout ton of stuff hehe [19:26:07] man. I hate it when my laptop home paritition hits 30 mounts and my box runs fsck on boot [19:26:28] whimpy little core 2 duo 1.6 Ghz and a little 5400 rpm drive [19:26:30] 320 gigs though [19:26:33] takes a while [19:26:43] can you sneak a SSD in there? [19:26:53] I could, but I'm still hesitant about those. [19:27:03] THey have limited writes [19:27:08] and they slow down as they're written to more [19:27:11] yeah like a million write per cell [19:27:19] eh. [19:27:23] i'm a little paranoid probably [19:27:24] but they get algorithm to dispatch the write evenly over all the cels [19:27:24] lol [19:27:30] (don't quote me on the million write) [19:27:37] I dad works at Seagate. [19:27:43] been working on those projects for hybrid drives [19:27:47] if you don't do a lot of write, I am sure you will be fine for sometime [19:27:50] now those I'm interesting in [19:28:00] well that is a project [19:28:01] heh. I've been using this drive for probably 7 years now. [19:28:10] and they need to make money out of us by selling ton of SSD first [19:28:18] same story with CRT -> LCD -> OLED screen [19:28:29] mmmm oled [19:28:46] i'm stoked about the amoleds [19:28:47] anyway, definitely consider a SSD for next laptop [19:28:51] black is actually off [19:28:51] it changed my life [19:28:57] haha [19:28:58] oh yeah? [19:29:01] really [19:29:04] boot my comp in a few seconds [19:29:06] I've got a bunch of friends who use them [19:29:10] every app just appear [19:29:21] (i just use browser, terminal , small apps) [19:29:24] actually, a buddy of mine runs a minecraft server that has at any point in time around 100 people on it. [19:29:25] but even iPhoto is really fast [19:29:37] they use an SSD for it [19:29:44] gave them like a 10x speed boost or something. [19:29:52] apparently the minecraft server software is super write heavy [19:29:57] wikipedia must have an article comparing access time [19:30:01] he says it's still performing great though. [19:30:19] writing to ram is good too :-D [19:30:25] (tmpfs ?) [19:30:25] heh [19:30:28] that's actually what he does [19:30:32] he set it up with a ram drive [19:30:37] 32 gigs of ram [19:30:53] that runs and rsyncs to the ssd every five minutes [19:31:23] I really want to put 16 or so gigs into my laptop [19:31:35] on boot it syncs / to the ram drive and boots off that [19:31:41] just never turn it off [19:31:42] lol [19:31:53] I got 4GB on my laptop, never use them :/ [19:31:59] yeah, same here [19:32:07] linux isn't very resource hungry [19:32:11] http://en.wikipedia.org/wiki/Solid-state_drive#Comparison_of_SSD_with_hard_disk_drives [19:32:12] what distro you on? [19:32:27] random access time is 0.1ms on SSD compare to 5-10ms [19:32:34] lol [19:32:40] that's a pretty sweet decrease [19:32:45] plus no noise [19:32:52] and you can move your laptop while it does write haha [19:32:53] :-D [19:32:56] little heat no doubt too [19:32:56] well I am convinced [19:33:08] I just run a backup from time to time just to be safe [19:33:15] as well as lower electricity cost no doubt [19:33:17] for the distro, I use a Macbook [19:33:21] so some Darwin kernel [19:33:23] heh [19:33:34] I recalled the iphoto remark [19:33:37] dumb question [19:33:41] sorry. :) [19:33:55] I love opensource [19:34:08] but I have been eventually tired of fixing my fonts and the GUI under Debian [19:34:17] so I gave up and bought a MacBook pro a few years ago [19:34:26] then upgraded to an Air for the high resolution and SSD [19:34:34] if there's one thing Mac is good at it's fonts and display [19:34:37] might consider a retina display next, will have to try to be sure [19:34:40] nifce [19:34:41] nice [19:34:54] retina display is beauuuutiful. [19:34:58] given I spend like 10 hours per day on the screen .. that is worth it :-] [19:34:58] are you worried about all the customizations causing issues with Apples TOS? [19:35:01] not everything works in high-res yet though [19:35:14] (and brion uses a Mac, which is a strong argument all by itself) [19:35:36] forgive me for not knowing, I'm new to this channel and project but who are you brion? [19:35:41] :) [19:35:51] nullspoon: i'm one of the original devs of mediawiki from back in the day [19:35:57] wow [19:35:58] <^demon> brion: You need to hang out more often :) [19:36:08] brion: It's a pleasure to meet you. [19:36:12] * brion bows [19:36:13] if you can call this meeting. [19:36:17] hehe [19:36:19] * hashar roll on the floor [19:36:27] I miss your humor brion :( [19:37:04] my wife's out of town for a few weeks so i'll probably be on irc more off-hours :) [19:38:00] another warning on the retina display mac -- x11 is not high-resolution yet, so that's not good for gimp or inkscape [19:38:08] they work, just pixelated [19:39:22] brion: My wife says that she approves that you control your irc time when your wife is home. [19:39:23] haha [19:39:33] :) [19:50:28] nom nom time, back laters [20:12:41] Do we have a formal schema change process now? [20:14:39] not really [20:14:50] commit it, and make sure asher knows about it [22:16:08] I had a bug about adding asher as a reviewer automatically [22:24:42] it should support multiple "gates" [22:25:04] if a commit touches php files, it should require a verified-by-jenkins that they contain no php errors [22:25:17] if a commit touches sql files, it should require a verified-by-asher that they won't melt the db [22:25:24] we could surely add more [22:32:18] did the purge [22:32:22] still metadata null [22:32:27] need to investigate that further [22:32:31] well for another day [22:32:56] Platonides: there is some universal linting job [22:33:06] feel free to enhance it to take care of python / sql etc :) [22:33:27] I am off btw [22:33:30] have a good night