[01:12] Can someone verify the beta 1 arm images for me? I'm still getting failed checksums. This is after zsyncing locally. Twice. http://paste.ubuntu.com/875432/ [01:46] GrueMaster: ubuntu-12.04-beta1-preinstalled-desktop-armhf+ac100.tar.gz: OK [01:46] GrueMaster: Though I do have the server failures. [01:53] GrueMaster: Mirrors should be correct now. Someone should poke me violently tomorrow about finding the root cause of this issue, though. [01:56] Oh, and I note that there's a bug, and it's been assigned to me. Lovely. [02:15] Your welcome. [04:21] pitti: The stack of KDE related demotions that component mismatches wants to do look bogus to me. For example, kcolorchooser is on list. kdegraphics (which is in Main) depends on it: https://launchpad.net/ubuntu/precise/i386/kdegraphics/5:71~pre15ubuntu11 so why does component mismatches want to demote it (most if not all of the others are similar). [05:48] ScottK: ok; it's possible it was because of the kde-mdeta/kubuntu-desktop uninstallability? they are gone from c-m as well now, anyway [10:01] * pitti rebuilds ubuntu desktops and alternates, uninstallability is fixed now [10:56] pitti: did you promote maas-enlist? [10:56] Daviey: no? I think I might have demoted it as part of cleaning up component-mismatches [10:56] but I'm not sure any more which package that was [10:56] (well, it was several) [10:57] pitti: you ACK'd it after jdstrands review, 4 hours ago. [10:57] right, the FFE [10:57] but it needs a reason to stay in main [10:57] otherwise it'll keep falling out again [10:57] pitti: it's seeded [10:57] it's on c-m [10:58] it was this morning [10:58] but for main -> universe [10:58] http://people.canonical.com/~ubuntu-archive/component-mismatches.txt [10:58] * pitti re-runs it [10:58] c-m doesn't currently auto-update [10:58] haven't had time yet to find out why [10:58] pitti: it was added to the seed a few days ago :/ [10:58] well, all I can say is that this morning it wanted to go back [10:58] but perhaps it was another package [10:59] Daviey: oh, right, that was there before [10:59] Daviey: now I remember [10:59] Daviey: I demoted cobbler-enlist [10:59] cool [10:59] * pitti promotes maas-enlist [10:59] that was correct [11:00] thanks === greyback is now known as greyback|lunch [12:05] could someone score ghc/ppc up please? Looks like we might have a full arch set this time and I would like to avoid any possibility of skew. [12:05] Took 1h49 last time, so it's not too chunky. [12:06] done [12:06] cheers [12:07] cjwatson: also, could I ping about deploying ben? [12:07] mehdi grabbed me about it the other day [12:07] gcc-4.4 is nearly finished so it should start after that [12:07] oh yeah, ok - I should write up my release meeting stuff first, then I'll have a look [12:07] cool [12:08] note that i haven't merged the new config branch with what we currently have for some time [12:08] we should be able to simplify the configs using the new global.conf stuff [12:13] urg, there is a new dep if we want to merge upstream again [12:13] libtyxml-ocaml-dev [12:36] pitti: Thanks. === greyback|lunch is now known as greyback [13:45] does the release-team meeting have a new format this week? [13:51] I believe so [13:51] pitti: so do I send in a report or do something else? [13:51] Riddell: the report format hasn't changed [13:52] just that the meeting won't be a standup round any more, but free form questions AFAIR [13:52] pitti: and we read all the e-mails before then ask questions in channel [13:52] right [13:52] I wonder if skeat is planning to do a "poke read e-mails now" before the meeting [13:53] We should decide in advance who we're going to pick on and everyone have a question to fire at that one person. More fun that way! [13:53] ;-) [14:01] fortunately IRC has a natural way of serialization :) [14:21] how do I log into a live CD if it puts me at a login manager? [14:22] oh the user changed, got it [15:34] skaet, cjwatson, apw, ogasawara: sorry, forgot to send email about yesterday's gcc-4.6 upload. changes affect ARM only. will send email later today [15:35] doko: ack thanks, I'll hold our upload till gcc's finished building [15:35] doko, prewarning of a day before the upload would be appreciated at this stage in the release cycle. [15:36] doko, any more bug fixes on the horizon for gcc/eglibc/binutils between now and beta freeze? [15:37] skaet, just a gcc-defaults upload to make gccgo (universe) point to 4.7 once it's built [15:38] doko, thanks. :) [15:39] afk now, getting breakfast [15:42] skaet, wait, there are still two eglibc issues left. jodh looks at 508083, and I have to look at 929219. not sure if before beta2 [15:44] doko, gotcha, thanks. keep me posted please. :) [16:04] Okay if i do a server (i386|amd64*) respin? [16:06] inprogress [16:08] infinity: your metalink fix didn't seem to work for some reason... http://cdimage.ubuntu.com/edubuntu/dvd/20120309/ [16:09] Daviey: sure [16:14] stgraber: Curious. I'll look at that later today when I'm also looking at hash sum weirdness. [16:14] infinity: thanks [16:20] infinity, yeah, its intresting that ac100 has a proper one ... iirc cjwatson fixed that on my request in the last alpha, i looked but coundt find a commit related to it though [16:21] *couldnt [16:21] I'm sure it was fixed by hand, like I just fixed the beta by hand. [16:21] But I'm going to look into the root issue later today. [16:21] I wittered about it on IRC at the time [16:22] i thought that was about the bashism in our -boot script [16:22] 2012-02-01 [16:22] pitti: where did you get the '8 mins' per cd build from? [16:22] not just that no [16:23] http://irclogs.ubuntu.com/2012/02/01/%23ubuntu-release.html#t11:54 [16:23] note that I went down a blind alley there for a while so read the whole thing rather than believing my initial comments :) [16:24] I thought that the checksum-directory change I made had fixed this [16:24] and I think also the publish-daily change; but evidently not [16:25] cjwatson: It's not ac100-specific, I think it's more to do with carrying over old builds into new directories. [16:25] right, that was what the checksum-directory change was supposed to be about, though [16:25] cjwatson: Which is only rarely done in the case of arch-only respins, but now ARM does it on every run. [16:25] anyway, just a place to start [16:25] cjwatson: But yeah, I'll look later today, it's on my TODO. Purge it from your mind. ;) [16:25] I might've screwed up the -nt test or something [16:30] Daviey: if you start the buildlive program on cdimage, you can see the start and end time from the buildds [16:45] pitti: right, but seems to take *much* longer for non-live :) [16:45] oh, not *much* longer, that was an exaggeration [16:46] Daviey: alternates should take less than 10 (didn't measure) [16:46] Daviey: cron.daily-live also take a bit, but it runs in parallel with the next buildlive [16:48] Daviey, about time to produce live server images then :P [16:55] ogra_: please drive a blueprint at UDS for that :) [16:55] (i'm not kidding.) [16:56] heh, only if NM goes onto the server-live image :P [16:56] (are you actually thinking about live builds ?) [16:58] Riddell,Daviey,ogra_,pitti,etc.: https://wiki.ubuntu.com/ReleaseTeam/CDImageSetup [16:59] ogra_: Considering a squashfs into d-i. [16:59] whee [16:59] cjwatson: lovely, thanks, can you link to it from the description of this team? https://launchpad.net/~ubuntu-cdimage [16:59] I expect that's where I'll look when I've forgotten [17:00] cjwatson, pretty clear, thx [17:00] Riddell: sure, done [17:01] hmm, i have never sudoed to cdimage before deploying ... good that you wrote that up :) [17:03] that would be why there are lots of files owned by ogra then? ;) [17:03] heh, sorry [17:04] Tsk. [17:04] can actually cause problems sometimes - best to always sudo to cdimage before doing anyway [17:04] anything [17:04] yeah, i think the distinction of production/development helps there [17:04] that way i will always sudo right after login [18:07] pitti: gilir: skaet: lightdm-gtk-greeter just needs to be consolidated, I have two source packages at the moment that need to be merged and I"ll endeavor to do that over the weekend [18:09] thanks micahg :) Let me know if you want help for this [18:10] gilir: I'll need a review of the combined package before upload [18:13] thanks micahg. :) [19:56] pitti: FYI, the eglibc package in maverick-proposed to fix bug 605042 has been superceded; also, for some reason that bug was not showing up on the SRU tracking page that I can see. [19:56] Launchpad bug 605042 in eglibc "[armel] java fails to start with eglibc-2.12-0ubuntu4" [High,Fix committed] https://launchpad.net/bugs/605042 === cody-somerville_ is now known as cody-somerville [23:09] is anyone available to process bug #951206? [23:09] Launchpad bug 951206 in haskell-deepseq "RM: Provided by GHC" [Undecided,Confirmed] https://launchpad.net/bugs/951206 [23:12] Laney: what does 'provided' mean? Is this a source-only removal because the binary should come from somewhere else, or does ghc now ship it in the main binary package? [23:14] n/m, see it in the ghc binary package [23:14] removing [23:14] it means that ghc now exposes that library [23:15] and Provides it in the packaging sense too [23:16] cheers