[01:07] elmo: looks vaguely reasonable; please don't think i know anything about practical email handling [01:13] mwhudson, hi [01:13] mwhudson, I'm about to head out to lunch... [01:14] jml: call when you get back then? [01:14] mwhudson, k. [01:14] o [01:14] sorry i was out for so long, that wasn't really the plan... [01:14] mwhudson, no worries. it happens. [03:23] jml: hello? [04:00] mwhudson, hi [04:00] jml: today has been a day of hhgttg-style lunch breaks, it seems [04:02] mwhudson, haha [04:02] mwhudson, yeah. [04:02] mwhudson, still up for a call? [04:03] jml: yeah [04:03] jml: let's see how the gods of linux audio treat me today [04:03] mwhudson, heh heh [04:03] jml: assuming skype works for you? [04:04] mwhudson, it should. [04:04] jml: i can hear you! [04:05] no [04:06] grr [04:07] mind you [04:07] hotel internet [04:07] probably shouldn't have expected miracles [04:32] mwhudson: you're in wlg ? [04:38] lifeless: yes [04:49] home of the rubber boot. [05:19] mwhudson, would it be very hard to show the lines changed on the active merges page? [05:19] mwhudson, the same data seems to be rendered synchronously on the merge proposal page [05:41] hmm. [05:42] maybe what I should ask is, if I made such a change, how could I test it locally? [09:19] bigjools: Did you end up working out what's wrong with gina? [09:19] I didn't [09:20] what exactly is it doing wrong? [09:20] Not importing 3.0 stuff. [09:20] eg. dpkg, quilt, ftplib [09:20] ok [09:21] Plus the other stuff which has been inexplicably ignored for several weeks, even though it was 1.0. [09:26] I'll try and get it sorted today [09:27] Thanks. [09:27] it's my last working day until next year, so I might have to pass the buck^Wbaton [09:27] Otherwise, all looks good. [09:27] Ah. [09:27] ok cool [09:29] I think that whoever wrote _dominateBinary must have had something against encapsulation. [09:37] morning [10:05] E: goobox_2.1.1.orig.tar.bz2 (from goobox) is in the DB but isn't an orig.tar.gz. (Probably published in an older release) [10:06] eh? [10:06] That's not an LP error, is it? [10:07] sync-source.py [10:07] Oh. Crap. [10:08] * wgrant checks the code. [10:08] sync-source.py is not like anything else in the world :( [10:12] wgrant: gina problem found - /tmp on the machine it runs on has ~32k entries and the filesystem refuses to add more, so I guess it's not cleaning up after itself [10:12] bigjools: Ahh. [10:12] bigjools: ext3 ftw. [10:13] If you don't care about uploading the orig.tar.* multiple times, the sync-source.py fix is trivial. [10:13] Otherwise it's a bit of work. [10:14] wgrant: http://pastebin.ubuntu.com/344022/ [10:15] wgrant: you ok fixing that? [10:15] bigjools: Fixing which? [10:15] wgrant: gina is not cleaning up its temp files [10:15] [dpkg-source output:] dpkg-source: warning: -sn is not a valid option for Dpkg::Source::Package::V3::quilt [10:15] I can take it if you can't, it just won't get fixed for a few weeks if so [10:16] bigjools: I'll do it. [10:16] james_w: I noticed that a couple of days ago. It seems harmless, but I can't see why it's even passing that during extraction. [10:16] wgrant: great, thanks. [10:17] * bigjools also notices that gina is not a LaunchpadCronScript ... :/ [10:17] gina isn't a LaunchpadAnything. [10:17] It's not even Zopey. [10:18] But I guess the problem here is that it doesn't log its actions to scriptactivity? [10:18] correct [10:18] # We only rmtree if everything worked as expected; otherwise, [10:18] # leave it around for forensics. [10:18] shutil.rmtree(tempdir) [10:18] we can do it manually [10:18] ew [10:19] I bet it just tried to import 3.0 packages slightly under 32000 times before we upgraded it. [10:19] [dpkg-source output:] dpkg-source: error: File /home/lp_archive/syncs/ttf-sil-scheherazade_1.001.orig.tar.gz has size 126665 instead of expected 128385 [10:19] james_w: Uh... [10:19] seems to be a short download or something [10:19] james_w: Any informative log output before that? [10:19] I: ttf-sil-scheherazade [main] -> ttf-sil-scheherazade_1.001-3build1 [main]. [10:19] * command was 'dpkg-source -sn -x /home/lp_archive/syncs/ttf-sil-scheherazade_1.001-6.dsc' [10:19] [dpkg-source output:] dpkg-source: warning: -sn is not a valid option for Dpkg::Source::Package::V3::quilt [10:19] [dpkg-source output:] gpgv: Signature made Thu Dec 3 21:14:41 2009 GMT using DSA key ID C0143D2D [10:19] [dpkg-source output:] gpgv: Can't check signature: public key not found [10:19] [dpkg-source output:] dpkg-source: warning: failed to verify signature on /home/lp_archive/syncs/ttf-sil-scheherazade_1.001-6.dsc [10:19] [dpkg-source output:] dpkg-source: error: File /home/lp_archive/syncs/ttf-sil-scheherazade_1.001.orig.tar.gz has size 126665 instead of expected 128385 [10:19] Which is ok behaviour provided IS knows a /tmp cleaner is needed.... [10:20] grabbing the package from the debian server shows that it is fine with the declared size [10:20] * wgrant adds that to a list of things to fix tonight. [10:20] wgrant: so the real problem is that we need it writing to scriptactivity [10:21] Bug 449408, that is. [10:21] Bug #449408: Need scriptactivity monitoring of "gina" [10:22] mthaddon: there's the bug number :) [10:22] He filed it himself! [10:23] thx [10:23] james_w, bigjools: Do you have any objections to sync-source.py reuploading origs unnecessarily? [10:23] wgrant: the only issue is client bandwidth [10:24] I don't really have an opinion [10:25] * wgrant sets up gina and sync-source.py. [10:31] bigjools: Is gina running reasonably happily now that /tmp is clean? [10:31] mthaddon: --^ ? [10:32] seems to be, from what I can see [10:32] Aha, yes. [10:32] It has imported stuff. [10:32] 3.0 stuff. [10:32] yay [10:32] Thanks. [10:33] wgrant: did you say you were going to do that scriptactivity bug? I'll get it sorted if not. [10:33] bigjools: I've not dealt with that sort of stuff before, so I'd prefer if somebody else did. [10:33] wgrant: not a problem [10:38] E: webkit is trying to override gir1.0-webkit-1.0_0.6.5-3ubuntu2 without -f/--force. [10:39] it used to be built by gir-repository, but sync-source.py -a tries to sync then fails [10:41] james_w: Is that a problem? [10:41] It looks quite deliberate. [10:42] well, it's autosync [10:42] it's not very auto if you have to handhold it because it thinks it can sync something and then changes its mind [10:43] Oh, that error kills the whole process? [10:44] yes [10:44] Ah, that is stupid. [10:44] * wgrant fixes that too. [10:46] thanks [11:02] Morning, all. [11:47] james_w: Is that size difference anything more than an orig conflict between Debian and Ubuntu? [11:47] it's at dpkg unpack time [11:47] I assumed it grabbed the whole thing from Debian [11:48] but if it is taking the orig from Ubuntu then that could explain it [11:48] It'll download the tarball from the librarian if it already exists in Ubuntu. [11:51] But that should have failed earlier... [11:51] Ah, no. [11:52] If it's downloaded from the librarian, it skips all of the internal size/hash checks. [12:46] al-maisan: I'm going through HowToUseSoyuzLocally but running into trouble... any idea what I could do when dpkg-buildpackage fails with "unmet build dependencies"? [12:59] jtv: looking.. [13:00] jtv: won't it say what the missing dependencies are ? [13:01] al-maisan: it does—it's debhelper >= 4 [13:01] This is while trying to build buildd [13:01] jtv: ..and what is the debhelper version on your system? [13:02] on my box it's: Version: 7.3.15ubuntu3 [13:02] al-maisan: hmm... not installed. Shouldn't that be in our developer dependencies? [13:03] jtv: don't think so .. [13:03] dephelper is typically needed for packaging etc. [13:04] al-maisan: in retrospect it should [13:04] ve been obvious I guess... I blindly assumed I had debhelper installed [13:05] :) [13:05] jtv: most things are obvious in retrospect ;) [13:05] very true [13:06] yup, working now. thanks al-maisan! [13:06] jtv: de nada :) === salgado is now known as salgado-afk [13:41] I can't get devel to run. css combine script offers a ton of warnings/errors and then make exits. [13:41] is this just me? (I changed my local setup to use lightweight checkouts and store trees in a different path, so maybe I've just not updated something I should.) [13:43] deryck: fwiw, i had some problems with css combine as well. i think a make clean solved it for some reason, i think. [13:43] BjornT_, ah, ok, I'll try that. [13:49] deryck, BjornT_: i'm surprised that css combine works in devmode. isn't this something we want to do on production only? debugging concatenated files is harder because you don't get a good reference to the file where something originated. [13:51] make clean did seem to work. [13:52] and yes, intellectronica, I would agree that we shouldn't have this in devmode. But certainly defer to BjornT_'s judgment about this. [13:52] intellectronica: well, the combine-css script is called in the build process, which is ok for both production and devel. not sure whether we use the combined css file in dev mode [13:52] ah ok [13:58] hi - how is https://launchpad.net/+apidoc/ generated? [14:01] intellectronica: it seems like we're using the combined css in dev mode, though. i'm not sure it's such a big problem. maybe better to wait and see, rather trying to fix something that might not be a problem? [14:03] BjornT_: it's not a huge problem. not as much as with executable sources. but it might make it harder to debug css that is being included from lazr-js for example. i sort of agree that there's no point doing anything about it until we find out whether it really is a problem. === salgado-afk is now known as salgado [14:44] flacoste: do you know if someone other than gary knows a lot about builout recipes? [14:45] BjornT_: on this team, i suspect not [14:47] flacoste: yeah, i thought so. btw, i'm ready for our call whenever you're ready [14:50] BjornT_: i'll call you in 10 minutes [14:51] ok === salgado is now known as salgado-lunch [14:56] does Launchpad stall anyone else's browser when loading a page? [14:57] it keeps my CPU through thr roof for about 20 seconds on each page load [15:06] it's fine here [15:08] beuno: I get the impression its slower than usual here as well [15:09] jelmer_, is it slower to load for you, or does it seem to increase the CPU load? [15:12] beuno: the page is shown pretty quick but there's an hourglass for longer than usual after that [15:13] beuno: at least that's my impression [15:13] jelmer_, that seems to be what happens for me as well === ursula_ is now known as Ursinha-lunch === danilos changed the topic of #launchpad-dev to: Launchpad Development Channel | Week 0 of 3.1.13 | PQM is open | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting | On-call review in #launchpad-reviews | Use http://paste.ubuntu.com/ for pastes [15:46] is there any way to *disable* blueprints on a per-project level? [15:47] dobey, not at the moment, no [15:47] if I get my way, we will eventually be able to disable apps per project [15:48] unfortunate... (the "not at the moment") === salgado-lunch is now known as salgado [16:31] beuno: I've set up a project on lighthouse for the UI [16:31] beuno: http://launchpad.net/rocketship [16:43] A branch of mine triggered a failed build of jscheck (2 'Looking up name loginpage_submit_login, failed.' windmill failures in utterly unrelated tests) [16:43] https://lpbuildbot.canonical.com/builders/jscheck/builds/274 [16:43] what should I do about this [16:43] ? === Ursinha-lunch is now known as Ursinha [17:26] beuno: flacoste: We now have a css tag for our bugs [17:27] sinzui: thanks! mars will also be interested === deryck is now known as deryck[lunch] [17:50] * bigjools is off until new year, enjoy the break everyone === beuno is now known as beuno-lunch === deryck[lunch] is now known as deryck [18:53] \o/ I broke his page: https://edge.launchpad.net/~buy-viagra-39- === beuno-lunch is now known as beuno [19:46] good morning [19:46] evening ;) [20:01] an interesting theory === salgado is now known as salgado-afk [22:48] * mwhudson afk for a few