[10:20:30] could somebody run a batch delete for me, please? [10:26:09] btw, Danny_B|backup, did you write to that designer whom I emailed (& cc'd you) a few weeks back? [10:26:42] Danny_B|backup: subject: "volunteering in design" [10:27:49] yup, no reply yet. i'll poke it in case something got lost on either way [10:28:37] ok, thanks Danny_B|backup [10:31:45] yw, sumanah [16:33:30] RoanKattouw-BSc: Congrats! [16:33:48] Thanks [16:34:07] So you got your official title today ? [16:34:50] Yeah, I now have a degree [16:35:03] Interestingly both the Dutch and English versions are translations, the original is in Latin [16:35:37] *g* [16:43:14] siebrand: Poke re i18n deployment [16:43:33] RoanKattouw-BSc: congrats! [16:43:37] Thanks! [16:43:43] RoanKattouw-BSc: can you please take it up with Nikerabbit ? [16:43:51] Sure [16:44:00] RoanKattouw-BSc: Basically we tagged versions 1.18 and we'd like thiose in. [16:44:31] Well the 1.18 tag list is empty [16:44:38] So everything that was tagged has been merged and untagged [16:44:47] don't let RoanKattouw-BSc get a PHd... He'll make us call him "Doctor" ;) [16:44:56] haha [16:44:59] RoanKattouw-BSc: oh, and congrats [16:45:00] Not to worry [16:45:02] Thanks [16:45:06] I don't aspire to get a PhD [16:45:28] I want to get an MSc but there are limits to how much more time and effort I want to spend on academics [16:46:06] I'm here [16:46:55] congratulatons on getting done! [16:46:59] +i [16:47:00] RoanKattouw-BSc: congrats from me too, you did it before me :) [16:47:31] Hah well you get an MSc, don't you? [16:47:34] (Or MA) [16:48:05] Hey RoanKattouw-BSc [16:48:08] Congrats! [16:48:25] I think the 1.18 tags were removed from Narayam and WikiEditor [16:48:31] Note, 1.18wmf1 *has* been branched [16:48:37] Whoa, yay! [16:48:49] It's been copying to /home for 40 minutes so far [16:49:01] You're running it on *fenari* ? [16:49:03] RoanKattouw-BSc: MA yes [16:49:05] Or is that svn checkout? [16:49:13] I ran the svn checkout to /tmp [16:49:44] Took 15 seconds to checkout [16:50:29] oh it has? wow [16:50:41] hmmm I wonder if that will brok the dumps :-D [16:50:44] (bork [16:51:15] I guess it hasn't been synced or deployed [16:51:20] so I can stop panicking :-D [16:51:52] It's going live to test2 wiki in sometime after 6 hours [16:51:57] apergos: It'll be het-deployed, anyway [16:52:00] vvv: Thanks! [16:52:22] real 45m55.731s [16:52:25] Raymond_afk: will it spam ausrufer? ;) [16:52:26] 45 minutes to copy [16:52:34] yes, but that means that it will go live to certain wikis [16:52:37] Nikerabbit, siebrand: OK so this is about WikiEditor and Narayam changes? [16:52:43] apergos: Only test2wiki for nwo [16:52:51] sure [16:52:58] I mean when het deploy actually starts to go around [16:52:59] surrur [16:53:06] 300 meg or so [16:53:11] RoanKattouw-BSc: essentially yes, and some WebFonts updates that haven't been reviewed yet. [16:53:17] Aha [16:53:18] oh right I have to fix https://bugzilla.wikimedia.org/show_bug.cgi?id=30846 [16:53:19] I guess I should be around for that (not test2, actual roll-out) [16:53:24] RoanKattouw-BSc: because of the many changes, I removed/rebranched already. [16:53:39] siebrand: Oh have you put the required changes in 1.17wmf1 already? [16:53:43] (I'm doing a deployment somewhere else; may be a little absent) [16:53:50] RoanKattouw-BSc: no, 1.18. [16:54:17] RoanKattouw-BSc: It's OK if all of those go in the 1.18 deployment; no need for *NOW*. [16:54:27] gifti: spam-ausrufer????? [16:54:28] OK [16:54:30] Well in that case... [16:54:48] RoanKattouw-BSc, there'd be no point just merging the narayam revisions that were tagged [16:54:56] Raymond_afk: =much entries in de:wp:neu [16:55:06] The WebFonts rebranch is in 1.18wmf1 [16:55:27] WebFonts isn't in 1.18wmf1 [16:55:34] so how does 118wmf1 differ from the plain 118 branch? [16:55:39] Oh right [16:55:45] gifti: oh yes.. a lot :=) [16:55:48] http://www.mediawiki.org/w/index.php?title=Special:Code/MediaWiki&path=%2Fbranches%2Fwmf%2F1.18wmf1 [16:55:55] WebFonts isn't even *reviewed* let alone deployed [16:56:03] ok, will shorten it :) [16:56:51] siebrand: OK so everything you put in REL1_18 was in by the time Reedy cut the 1.18wmf1 branch, I just verified that. Which means those revs will be part of the 1.18 deployment, except for the WebFonts revs (that extension isn't deployed, or even reviewed, at this point) [16:57:06] So it sounds like you don't actually need anything from me this time? [16:57:26] where is stylize.php located? [16:57:35] Nikerabbit, /trunk/tools/code-utils [16:57:38] thx [16:57:44] RoanKattouw-BSc: Those narayam and wiki editor revs are not in there, or are they? [16:57:57] siebrand: If you put them in REL1_18, they're in [16:58:16] ok this isn't so much... [16:58:16] If not, go find them and tag them with something like i18ndeploy [16:58:23] RoanKattouw-BSc: I only put WebFonts in there (I thought that was deployed, but apparently I'm mistaken). [16:58:27] *apergos dreams of deploying from trunk again. some day. [16:58:41] srv169: rsync: send_files failed to open "/php-1.17-test/includes/.swp" (in common): Permission denied (13) [16:58:45] *Reedy beats Aaron [16:58:55] Did Aaron STILL not fix his umask? [16:59:02] Or is this just a broken script [16:59:02] Not sure [16:59:07] No wait [16:59:09] but php-1.17-test is his test [16:59:10] You created that one [16:59:16] Oh he did [16:59:18] Right [16:59:27] Loads of errors [16:59:30] RoanKattouw-BSc: too much happening too fast. Looks like the revs I tagged 1.18 have already been processed. [16:59:37] RoanKattouw-BSc: I goess we're done. [16:59:38] MW 1.18 => php-1.18 [16:59:40] guess [16:59:43] siebrand, hexmode untagged some of them [16:59:49] Reedy: oh? [16:59:52] *siebrand curses. [17:00:20] gvd. [17:00:32] I don't have time for this now. [17:00:41] I'd rebranch Narayam from trunk to 1.18 and 1.18wmf1 to be honest [17:00:46] Then there is only your wikieditor revision [17:01:02] why did hexmode untag without letting me know? [17:01:06] 46.6%wa, [17:01:08] RAWR [17:01:14] Fenari is slloooowww [17:01:28] now I have no clue what I tagged and no way to find what happened to those revs. [17:01:34] siebrand: sorry [17:01:53] yeah, don't buy me anything. [17:01:54] I re-tagged them post1.18, though [17:02:02] hexmode: what's the tag? [17:02:11] post1.18 [17:02:19] not even flowers? [17:02:28] Reedy: -rw--w---- 1 midom wikidev 200704 2011-09-17 14:00 /home/wikipedia/common/php-1.17-test/includes/.swp [17:02:37] Now you know who to blame :) [17:02:47] what's the meaning of post1.18 tag? [17:03:03] RoanKattouw-BSc: 97466, 97385, 97252, 97250, 96944, 96943 are the ones. [17:03:09] Nikerabbit: if you find out, please update https://secure.wikimedia.org/wikipedia/mediawiki/wiki/Code_review_tags [17:03:16] RoanKattouw-BSc: all in http://www.mediawiki.org/wiki/Special:Code/MediaWiki/tag/post1.18 [17:03:27] OK [17:03:28] Suiseiseki: thanks. [17:03:33] Let me load the dishwasher and I'll get right back to you [17:03:33] sumanah: thanks. [17:03:37] Suiseiseki: sorry. [17:03:48] Yeah for Narayam I'll probably just rebranch [17:04:01] Nikerabbit: post1.18 is stuff to be merged after 1.18 deployment [17:04:08] hexmode: Why is r74966 tagged post1.18??? It's in the 1.18 branch, now it's just sitting there unreviewed [17:04:19] hexmode: how does it differ from 1.18wmf1? [17:04:24] (Or well, the IE issues unaddressed I should say) [17:05:23] Nikerabbit: I don't know what 1.18wmf1 means... perhaps they should be the same [17:06:14] *hexmode puts on his lawyer suit so he is more prepared to defend his tagging [17:06:47] RoanKattouw-BSc: you're right, r74966 should be back to 1.18 [17:07:07] I thought Trevor was going to review it [17:07:11] He was gonna [17:07:20] And someone (RobLa?) was gonna make him [17:07:50] Yeah, robla said he was aware of it [17:18:25] TrevorParscal: speak of the devil! [17:18:42] *TrevorParscal marks everything as ok and hides [17:37:30] siebrand: rawr, none of those revs are reviewed [17:38:13] *RoanKattouw-BSc reviews them [17:50:06] siebrand: Nikerabbit: OK so I'm gonna rebranch Narayam from trunk and include that one WikiEditor fix, sound good? Is either of you around to help me verify the deployment? [17:52:26] RoanKattouw-BSc: I'm going home now to eat and have a call again really soon. [17:52:29] RoanKattouw-BSc: with verify you mean testing that wikieditor and narayam works? [17:52:32] RoanKattouw-BSc: waiting for Nikerabbit to reply... [17:52:45] ok, I'm off. [17:53:16] Nikerabbit: Yes [17:54:24] I don't really know how to test most of the latest changes to narayam. I can mostly play around to see if it there is anything obvious. [17:54:44] hey, can I get anyone to review 3 small JS changes? I'd like to be able to deploy these... [17:56:17] RoanKattouw-BSc: oh and when is the deployment? in two hours? [17:56:38] It's like now [17:56:46] I have to yield the floor in 4 mins, theoretically [17:57:27] I'm gonna run over [17:58:41] TrevorParscal: can you take a look at something for me at my desk [17:59:06] in a sec [17:59:12] TrevorParscal: thanks [18:01:50] Rawr, r96944 does not apply cleanly at all [18:03:12] robla: Can you hear me? [18:03:25] meh [18:03:29] *RoanKattouw-BSc grabs other phone [18:03:47] Now? [18:04:00] rawr [18:04:15] I'll just listen then [18:04:18] Cause I can hear you [18:04:27] OK [18:04:47] Still can't hear me? [18:27:05] RoanKattouw-BSc: r74966 looks reverted (see r78175) no? [18:27:10] No? [18:27:13] !r 78175 [18:27:13] --elephant-- http://www.mediawiki.org/wiki/Special:Code/MediaWiki/78175 [18:27:22] TrevorParscal: /branches/REL1_17 [18:27:49] ok, yes - so you only backed it out there... [18:28:22] so, this still has tons of issues in 1.18 as well [18:28:29] it needs to be backed out, again.. :( [18:28:30] IE6 is likely completely broken [18:28:37] yes, it's quite bad [18:28:41] Can't we just, like, *fix* it [18:28:55] All that happens is the IEFixes*.css files aren't being applied [18:29:45] http://www.mediawiki.org/wiki/File:Screen_Shot_2011-09-19_at_11.28.45_AM.png [18:30:21] my god [18:51:28] jorm: http://shop.theoatmeal.com/products/unicorn-tetherball-shirt [18:51:53] TrevorParscal: Is that just IE6 being broken or the other ones too? [18:52:14] IE7 too [18:52:22] TrevorParscal: I'm willing to pull it out again if it's a lot of work to fix, but AFAIK it's not /that/ much work, just cherrypicking of rules from a deleted file that SVN still has history for [18:52:36] In fact just reinstating IE*Fixes.css would owkr [18:52:37] no, it's going to be a long tail [18:52:41] Meh [18:52:57] Maybe I'll just reinstate the IEFixes stuff [18:53:21] could do [18:53:34] That should work [18:53:48] What I *wanted* was to port the IEFixes files to browser hacks in the main CSS file [18:54:08] yeah [18:54:38] the problem is that mono book is a mess, and it needs to be rewritten by someone who knows what they are doing so we can have a clean version [18:54:56] in the mean time, not that many people even use mono book (relatively speaking) so??? meh [18:57:14] Did we have any problems with thumbnail cache recently? [18:57:32] http://commons.wikimedia.org/wiki/File:Symbol_confirmed.svg -- this file is broken at 180px [18:57:48] (and it does not rerender on thumb.php call, as it used to) [19:02:46] so scapping -- does that copy the state of fenari, or the state of SVN? [19:02:58] there is an committed modified file in fenari's /p/ [19:03:04] s/committed/uncommitted/ [19:03:11] "an"? [19:03:18] State of fenari:/home/wikipedia/common [19:03:32] Holy crap just ONE modified file?!? [19:03:34] *RoanKattouw-BSc gasps [19:03:51] wtf... [19:04:27] RoanKattouw-BSc: ...I don't know what you are trying to say [19:04:46] RoanKattouw-BSc: are you saying that the uncommitted changes are definitely good live hacks or what [19:04:50] It used to be way, way, worse [19:04:52] Tim did go through and clean the rest of it [19:04:56] We used to have a dozen changed files [19:04:59] Right now there are two [19:05:01] there are a ton of '?' files [19:05:03] ~ StartProfiler.php [19:05:05] <^demon> What status is ~ ? [19:05:14] ^demon: Ignored file that disappeared or something? [19:05:21] All I see is the file became a symlink [19:05:22] demon: uncommitted and svn doesn't know what this is [19:05:43] i.e. not svn added [19:05:45] <^demon> If it it's ignored, I'd think svn wouldn't care. [19:06:01] ignored files are indeed ignored. [19:06:02] <^demon> That should be '?' if it's uncommitted and not in svn:ignore [19:06:09] neilk_: Short version: it deploys what's on fenari. Our SVN checkout there is a mess but you don't need to worry about that (but we do) [19:06:17] *RoanKattouw-BSc looks up what ~ means [19:06:21] apparently "I" am not part of "we" [19:06:31] '~' versioned item obstructed by some item of a different kind [19:06:35] guh [19:06:38] how weird [19:06:42] I didn't mean to exclude you [19:06:45] <^demon> "Item is versioned as one kind of object (file, directory, link), but has been replaced by different kind of object." [19:07:04] I'm just saying that some things here are a half-maintained mess, but it's been like that forever [19:07:18] In fact it's a lot better now than it was a few months ago [19:09:39] neilk_: Again, sorry for sounding dismissive or excluding if that's how you read it [19:10:30] RoanKattouw-BSc: no, I understand, my point is that it's probably better to spread knowledge than say "don't worry about it" [19:10:40] True [19:10:51] I just didn't want you to fret over it regarding today's scap [19:11:23] But yeah there are still some unversioned files in extensions/SecurePoll [19:11:58] <^demon> We could just throw those in SVN for now and let Andrew sort them out? [19:13:21] Please don't [19:13:27] It's SecurePoll, it might be private data [19:13:47] It's an infrequently used extension, we can just keep the files in php-1.17 and let Tim or Andrew sort it out properly [19:14:33] <^demon> Well I wasn't going to *blindly* check them in without looking at them first? [19:14:40] robla: Hey do you still want me to take a post-deployment shift tonight (your time; tomorrow morning my time, I'm pretty tired already and feeling like going to bed early for an engineer, so I can get up engineer early too if you need me to) [19:14:43] Right [19:14:45] <^demon> I'm not that XXhardcoreXX [19:14:59] Still, I see no compelling reason why you should spend time on it [19:15:12] I think it's reasonable to say that it's Tim and Andrew's mess and that they can clean it up [19:15:30] RoanKattouw-BSc: yeah, get your rest :) [19:15:54] Reedy should be available for the cleanup shift [19:15:58] Oh, cool [19:16:07] *robla is about to disappear into an interview [19:16:12] *^demon will be around for the deployment and immediate fire-putting-out [19:16:51] *RoanKattouw-BSc will be asleep then [19:17:33] Yeah [19:18:11] <^demon> I think I'm going to take a bit of a nap now so I'm fresh before we start the process. [19:18:46] Reedy: Alright so I probably won't be around all that early, graduating is tiring :) I intend to be around post-deployment for the others, but this is test2 [20:57:35] Ryan_Lane: thank you for that blog post! [20:59:03] you're welcome :) [20:59:14] I imagine we'll get some press over that [20:59:40] we'll see if it's "they're crazy" or "that's awesome" or both [21:04:47] LOL [21:05:10] As it happens, I was one of the original managers at Red Hat of the project that became Spacewalk. [21:05:24] Ryan_Lane: maybe I can make some personal intros. [21:05:46] I used to use red hat satellite server [21:06:10] they are apparently integrating puppet [21:06:22] I'd love to have full ubuntu server/client support for spacewalk [21:06:28] with puppet integration [21:06:35] we very badly need something like it [21:07:19] What's missing at this point? [21:07:29] proper patch management [21:07:47] How is "patch" defined? [21:07:52] new packages [21:07:58] security and bug updates [21:08:11] we handle security updates well, but it's way more work than it should be [21:08:22] *gregdek wonders if it's useful to map the "errata" concept in this case. [21:08:33] Honestly, I'm not even sure if the Spacewalk code honors errata. [21:09:10] Ohwait -- maybe I meant "what is Spacewalk missing to support dpkg". [21:09:20] Did I make that clear? [21:09:37] yeah. I've seen the current support for ubuntu in spacewalk :) [21:09:56] meta-data support is missing, I believe [21:10:05] I may just end up writing this myself [21:10:13] Ryan_Lane: Hey two random questions before I disappear to bed [21:10:15] I only need package management [21:10:22] not the other million things that spacewalk does [21:10:24] RoanKattouw-BSc: sup? [21:10:27] First, it appears that Gerrit does not have inline diffs, you have to click to get a diff and it opens in a new tab. Am I missing something? [21:10:38] I really, really, really like inline diffs (collapsible is OK) as a feature in Extension:CodeReview [21:10:44] hi gregdek [21:10:45] I also prefer that [21:10:53] RoanKattouw-BSc: we need to send some changes upstream [21:10:53] Hi sumanah :) [21:11:02] Second thing is something I just posted as a blog comment too [21:11:09] oh? [21:11:12] Why don't we make boilerplate versions of the private puppet files available? [21:11:28] we kind of do... [21:11:34] So that people can replace some placeholder passwords etc with theirs and make it Just Work [21:11:37] Oh, we do? [21:11:40] but it's meant to be the private repo for labs [21:12:03] I don't quite understand that last bit [21:12:30] it won't be boilerplate for too long [21:13:03] I... still don't understand [21:13:13] we can probably release a boilerplate too [21:13:24] One way or another there are gonna be files that contain private things, and the public config is gonna refer to private files [21:13:33] basically whats currently the empty files without passwords, is going to have passwords [21:13:49] Then we either release boilerplate versions of those files with passwords stripped or replaced with placeholders, or we don't [21:13:57] it's going to be available to Labs users, but not the general public [21:14:01] Right [21:14:53] replied :) [21:15:10] OK [21:15:21] Just making sure I understand the current situation correctly [21:15:36] I thought I was reading something like "we do have those but I didn't communicate that in my blog post" [21:17:20] But it seems to be something more like "we don't have them, plan to have them for Labs users, didn't plan to have them for the general public but now you mention it that's a good idea" [21:19:16] yeah [21:19:32] the hard thing is keeping the boilerplate accurate [21:19:54] it's easy enough for production, it's a PITA for test, but it's going to be forgotten ofter for the boilerplates [21:20:00] True, you don't have automatic revision control there [21:20:12] Unless you structure it in a somewhat roundabout way [21:21:06] e.g. by having a versioned boilerplate that contains $password = $realPassword; for every var, then having another file that contains $realPassword = 'foo'; [21:21:54] Then you still need a socially enforced convention not to use $realFoo anywhere, and update two places (the versioned boilerplate and the real config) when you change something. The latter is the case anyway but this scheme (kinda hacky, maybe) ensures things won't work unless you're a good boy and update the boilerplate too [21:23:21] I guess we could make people check into the boilerplate, merge to private [21:23:31] then add the real password in private [21:23:50] Ah, right [21:23:59] Have private pull the rev that adds the var from the boilerplate [21:24:21] and we could mark a revision bad if it didn't come from boilerplate and added something [21:49:28] yuvipanda: hey, looking forward to this blog entry :) [22:00:04] brion: am I right in assuming I should avoid asking you about parser progress till next week? :) [22:01:24] :) [22:01:37] yeah i got partway through some updates last week but ain't finished [22:01:49] working on template expansion [22:02:47] k [22:03:16] brion: I'm not clear on whether Inez/Trevor are waiting on stuff from you or not, and whether anyone can help with the stuff you're doing [22:26:34] okay. time to go get Into the Cave and finish this design. [23:04:34] Morning Tim [23:05:06] morning [23:57:17] how fun is it that both of brion's current babies are deploying upgrades at the same time