[02:53] Hey release-ers. I was working with someone in Debian, and he requested a FFe (LP: #849167) which closes RC bug LP: #820983. It's a bugfix only upload, but it requires dropping the ubuntu1 patch (which failed to fix what it was intended to fix). Can someone triage, please? [03:17] If it's bugfix, it doens't need release team review. [03:18] ScottK: Is there any chance you could fill me in on some of the process? I was just helping a friend, and I don't really know how to handle the paperwork [03:18] Subscribe ubuntu-sponsors to the bug and someone will review. [03:19] ScottK: cheers === maxb_ is now known as maxb === doko_ is now known as doko === lool- is now known as lool === chrisccoulson_ is now known as chrisccoulson [09:34] we ask for FFe for dh_python2 transitions, but what about others? (CDBS → dh in this case) [09:54] Laney: I guess it could be argued that they need an FFE, but personally I don't care much [09:54] it's a matter of debdiffing the binaries to check the result [10:14] Would someone be able to tell me what is unusual about ubuntu/pool/universe/libc/libcdio/libiso9660-7_0.81-4_amd64.deb ? It appears to be a 404 on archive.ubuntu.com, a 403 on gb.archive.ubuntu.com, but accessible on de.archive.ubuntu.com [10:15] pitti: thought as much, ta [10:15] maxb: the only unusual thing I can see is that it's a symlink into main [10:16] Hmm... so that would explain the 403 if gb.a.u.c is configured with a http server that won't follow symlinks [10:17] It's a 404 on archive.ubuntu.com even changing universe to main [10:18] Last sync failed. [10:24] maxb: OK now? [10:26] Yes, that works - thanks [11:12] Anyone know when skaet usully arises from her slumbers? [11:16] jussi: Somewhere around when Texas wakes up :) [11:25] Laney: I think build system changes are feature changes that should be reviewed, but OTOH if someone forgets to ask, I'm not going to go hunt them down unless they break something. [11:26] jussi: I'm gonna say approximate 14:00-ish gmt [11:26] davmor2: excellent, thank you [11:27] jussi: most of the states are online between 14:00-16:00 so somewhere around then at any rate :) [11:27] ok [11:27] I guess I should go back to bed then. [11:28] I definitely should [11:28] ScottK: you slangasek and persia ar eall androids so use UTC I'm sure of it :D [11:28] see ^ [11:29] Sleep is for the weak. [11:29] sleep is something you do when you die right? [11:29] There's even a song about that. [11:30] ScottK: there are several with the line in :) [11:30] so it must be true the song said so [11:30] No doubt, but it's the title of a Frank Zappa song. [11:32] \o/ [11:34] ScottK: Indeed, and also sung by Meatloaf, bon jovi, guns'n'roses, some bond theme has it I'm sure oh and others [11:37] I'm glad I'm not the only person awake all night. [11:40] nigelb: you're a rank amateur at it ;) [11:41] I'm not sure if I wwant to be pro :P [12:20] slangasek: cjwatson skaet (anyone)... any armel livecd images planned in the next hour or 2? I need to do some work on annonaceae [12:21] speaking only for myself and the crontab, no [12:21] slangasek: ta [12:21] me neither [12:22] buildd's ssh auth keys file is currently disabled, fwiw [12:22] would like the x86 ones to keep working though [12:22] purely an armel activity [12:22] total of 5 arm boxes are getting better disks, and I need to migrate things, starting with 4 of the lp-buildds [12:22] *aceae.buildd [12:23] (for many, but not all, values of "*") [12:30] lamont, well, i was just fiddling with mx5 images but i will wait for infinity anyway [12:31] and worst case i can switch to sycamore for the moment [12:34] ogra_: once the rsync finishes, it's a couple commands and a reboot [12:34] yeah, no hurry, i'm stuck anyway [12:35] (somehow mx5 is convinced it builds a DVD and not a preinstalled image and i cant find out why) [12:41] lucky you [12:42] heh, yeah [12:42] in a different question, wtf does the lucid arm redboot world think that fsck can be omitted from the initramfs? [12:42] muight be cjwatson's fault though i think there was something about the usb key stuff he fixed last night ... and that i end up with 1.6G ext3 files from live-build is suspicious [12:43] lamont, how do you mean ? [12:43] gonna need some bigger CDs [12:43] redboot doesnt do anything with the initrd but loading it [12:43] ogra_: buildd reboots with errors on /dev/sda1, fails to boot, because fsck.ext2 is missing [12:43] everything else is kernel [12:43] seems to be localized to the bbg3 machines [12:43] uh, oh [12:43] ergo "armel redboot world" [12:44] yeah [12:44] but yeah, absolutely nothing to do with redboot itself [12:44] i thought you mean redboot does anything with the initrd it shouldnt [12:44] it unpacks it. that's not nothing. :p [12:44] shouldnt mountall pull that into the initrd ? [12:44] fsck i mean [12:45] one would expect something to. I'll check into lucid's mountall [12:46] zcat < /boot/initrd.img-2.6.35-24-omap | cpio -t | grep fsck [12:46] 24033 blocks [12:46] iz fact on bbg3 boards and beaglexm [12:47] wow [12:47] ogra_: mountall is not run from the initrd [12:47] i thought mountall is what ships the fsck parts into initrd [12:47] no, nothing does [12:47] oh [12:47] there is a wishlist somewhere for making the initramfs generally more useful for recovery which would include fsck [12:47] but it's a feature [12:47] mountall runs from the real system not the initramfs [12:48] bug 512349 [12:48] Launchpad bug 512349 in baltix (and 2 other projects) "include filesystem repair facilities in initramfs (affects: 3) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/512349 [12:48] ah, k [12:48] cjwatson: it's a personally annoying feature when you have to have someone physically in the data center to reboot the machine [12:48] lamont, what are the exact fs errors you see ? [12:48] /sbin/fsck.ext2 /dev/sda1 [12:48] e2fsck 1.41.11 (14-Mar-2010) [12:48] ext2fs_check_if_mount: Can't check if filesystem is mounted due to missing mtab file while determining whether /dev/sda1 is mounted. [12:48] /dev/sda1 contains a file system with errors, check forced. [12:48] and no actual errors [12:49] during the run [12:49] anyway, buttercup and kaylaberry are back in the game now that I've fscked them [12:50] hmm, sad, i was hoping for RTC relations [12:51] normally the initramfs can still do a read-only mount and the lack of fsck should not block it [12:51] perhaps there's a 'rw' argument telling it not to do that? [12:52] hmm, could be that there is no ro on the cmdline [12:52] but that should still default to ro, no ? [12:52] yes [12:52] though I haven't checked lucid, only oneiric [12:53] having a machine that refuses to boot until someone manually gets fsck into the initramfs and runs it strikes me as something closer to a bug than a feature [12:53] we surely never set rw explicitly [12:53] lamont, whats your cmdline ? [12:54] console=ttymxc0,115200n root=UUID=ff4d3b1f-c1f2-4a63-859c-71395c87e61c [12:55] looks fine [12:55] not that then [12:55] lamont: point is that the absence of fsck from the initramfs should not cause that bug [12:55] maybe I'll see if I can reproduce it on the home-panda [12:55] fsck is not meant to be required in the initramfs for booting [13:00] oh, *finally*, a working zh_CN livefs build [13:00] only about ten attempts later [13:00] now I can see what the output looks line [13:00] *like [13:03] GRRR ... why is *every* armel package on the ftbfs list i look at a GL issue [13:10] ogra_: I see a lot of /usr/include/GL/glext.h:5072:19: error: conflicting declaration 'typedef ptrdiff_t GLintptr' - Is that what you're hitting? [13:10] that's standard GL vs. GLES, AIUI [13:10] ScottK, yes, i saw that in several packages [13:10] right [13:11] /usr/include/GL/glew.h has the same typedef. [13:11] but often needs more than just changes of the build dep sadly [13:11] I have no frickng clue why there's anything armel specific in there. [13:11] GLES vs GL [13:12] Hmmm. Is mesa-common-dev GLES on armel? [13:12] not armel specific ... GPU specific :) [13:13] I see. [13:13] there is a specific gles lib iirc [13:13] Yes. [13:13] In the case I'm looking at it's not installed. [13:13] Oh, wait. It is. [13:14] well, usually you need to change the build-dep, change configure to use GLES if possible and then clean up the fallout for missing or different declarations [13:16] it'd be helpful if https://wiki.ubuntu.com/ARM/FTBFS discussed this for more than the Qt case [13:30] pitti: I have a merge proposal for a update-manager httplib2 change, but it would add python-httplib2 to update-manager-cores and via that to ubuntu-standard - is that something acceptable at this point? the caching is greatly improved with that branch but its not OMG critical [13:34] mvo: that sounds like it would restructure some code from e. g. urllib to python-httplib2? [13:34] mvo: as for the dependency, that's mainly a question for the server guys (Daviey) [13:36] pitti: yeah, the code itself is fine, I'm mostly concerned about the new dependency at this point, I will eventually merge it for P, I'm pretty certain about that [13:40] pitti: you should've rejected my dh_python2 conversions due to the extraneous X-Python-Version field since there was no minimum specified before :) [13:41] * micahg just got yelled at in #debian-python... [13:43] * micahg guesses that goes beyond the call of duty for the release team review though [13:43] micahg: They are not required, but they don't actually hurt beyond making the source very slightly larger. [13:45] ScottK: that was the impression I had, but #debian-python said otherwise [13:45] cjwatson: for tracking, where should I file the (lucid/maverick at least) bug about needing ext2 for boot? (note that the machines in question were not installed via the normal install processes, so it's likely pebcak) [13:45] s/ext2/fsck/ [13:45] micahg: I saw. I'm not quite as pedantic as some other people. [13:59] lamont: not sure, I guess initramfs-tools for starters [14:02] ok [14:11] * ogra_ wonders if febootstrap is what he suspects [14:26] Can we try a respin of Xubuntu Alternate images. There are no errors in the log, but the install fails and there was an email notiification for a fail [14:27] what did the notification say? [14:27] ah, well, if nothing else report.html is non-empty [14:28] micahg: sorry, missed that [14:28] xubuntu/daily: Uninstallable packages: [14:28] has abiword installability been fixed then? [14:28] abiword 2.8.6-0.3ubuntu2 produces uninstallable binaries: [14:28] * abiword (amd64 i386) [14:28] * abiword-plugin-grammar (amd64 i386) [14:28] * abiword-plugin-mathview (amd64 i386) [14:28] * libabiword-2.8 (amd64 i386) [14:28] xubuntu-meta 2.137 produces uninstallable binaries: [14:28] (http://cdimage.ubuntu.com/xubuntu/daily/current/report.html) [14:28] * xubuntu-desktop (amd64 i386) [14:29] * cjwatson checks in chdist to see if that's actually been fixed [14:29] yes, that is the notification [14:29] pitti: well, according to ScottK it's YMMV, so not your fault [14:29] Why doesn't http://people.canonical.com/~ubuntu-archive/cd-build-logs/xubuntu/oneiric/daily-20110914.log show anything? [14:29] cjwatson: it installed fine on my ubuntu-desktop oneiric setup [14:29] because it's not a CD *build* failure [14:29] *abiword [14:30] the images built successfully, they just contain uninstallable pieces [14:30] it's often useful to have the output anyway [14:30] So, file a bug with the logs? [14:31] no [14:31] micahg: agreed, it seems fine in chdist [14:31] I'll respin now [14:31] Thank you [14:31] cjwatson: thansk [14:31] *thanks [14:52] slangasek: cjwatson skaet (anyone)... annonaceae is back over to all y'all, fwiw [14:55] thanks :) [15:16] mucking about with archive/i386 - manualized [15:16] should be back in a couple min [15:17] and we're back [16:41] I need yet another respin. Something is wrong with the mirrors or server because it refuses to allow abiword to work. [16:41] Please respin Xubuntu Alternates again [16:50] phwoar, nbs looks good [16:50] nice [16:52] charlie-tca: it's not clear that respinning will help, then [16:52] I don't want to respin if nobody understands the problem === davidm` is now known as davidm === davidm is now known as Guest56598 [16:53] cjwatson: okay [17:12] bug 850172 for Xubuntu failing to install [17:12] Launchpad bug 850172 in debian-installer (Ubuntu) "Xubuntu Oneiric Alternate images fail to find abiword dependencies when building images (affects: 1) (heat: 8)" [Undecided,New] https://launchpad.net/bugs/850172 [17:32] failure on two different hardware systems [18:19] hm, no local copy of xubuntu-oneiric-alternate-i386.iso [18:19] UNLEASH THE JIGDO [18:22] hi. is it feasible to sync a libgpod 0.8.2 from debian at this point? [19:18] hi, could someone please ack https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/849994 ? [19:18] Launchpad bug 849994 in network-manager (Ubuntu) "FFE: add NetworkManager DUID support (affects: 1) (heat: 8)" [Undecided,New] === jjohansen is now known as jj-afk [23:33] * cjwatson goes to see if he can decipher the Xubuntu failure before falling asleep [23:35] thank you very much, but sleep might be more important [23:36] I have fairly frequent insomnia, it may be important but it's not always an option :-/ [23:36] might as well do something useful with it [23:39] Then again I will say "Thank you very much". Your help is very much appreciated.