[07:50] bah, oneiric schroot broken [07:50] stupid /run [08:55] So, a quick bit of advice is sought. I am editing debian/rules to get a package to build. The changes I make allow the package to build using pdebuild. [08:56] However, once I turn those changes into a patch use edit-patch the package no longer builds using pdebuild. [08:56] Any suggestions as to what might cause this? [09:04] aonyx: it would help to see a buildlog. [09:05] aonyx: and for changes to files in debian/ you don't have to make a patch anyway. [09:05] only for changes to upstream files (outside of debian/) [09:07] Ampelbein: Thanks, well now that I don't have to make a patch my life just got a whole lot easier [10:17] aonyx: we use patches (quilt etc) for changes to the upstream code, not the contents of debian/. Those we just change directly [10:17] broder: still looking for a sponsor? [10:19] oki-doki, should have it uploaded soon, jtaylor seemed to think that your initial suggestion was fine so I am doing that. [10:21] cool [10:39] aonyx: I'd also remove the -lc LDFLAGS addition [10:40] it should be tested that bug 205783 is not reintroduced somehow [10:40] Launchpad bug 205783 in libpam-encfs (Ubuntu) "pam-encfs fails on upgrade to Hardy" [Undecided,Fix released] https://launchpad.net/bugs/205783 [10:45] tumbleweed: can you ack this sync? bug 808288 [10:45] Launchpad bug 808288 in unison2.27.57 (Ubuntu) "Sync unison2.27.57 2.27.57-4 (universe) from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/808288 [10:47] tumbleweed: good point, getting to it now. [10:52] jtaylor: acked [10:59] tumbleweed: they're removed now [11:00] aonyx: I'd reuse the "Explicitly link to C library" comment for DEBIAN_PAMLIB [11:00] sounds like a good idea [11:08] tumbleweed: done [11:10] aonyx: LGTM, thanks. Please forward that to Debian [11:11] woohoo, will do [12:26] RainCT, if you don't want me to NMU mistelix, i'd speak up. [12:29] directhex: I'm fine with NMUs (or with mistelix, I'm OK with having it team-maintained) [12:30] it's one of the final blockers on completing the 2.10 transiiton [12:47] directhex: well, I've no idea what the error is about [12:49] ... i just worked out an error in mistelix. it's cringeworthy [12:49] it uses AC_PATH_PROG to get the MCS variable... but uses hardcoded CSC variable as the compiler instead [13:02] what is the CMAKE equivalent of configure LIBS=... ? [13:09] noo my experiment has invalidated my ccache :( === Quintasan_ is now known as Quintasan [14:54] Any idea why the baltix package was added to bug 791950 ? [14:54] Launchpad bug 791950 in Baltix "Please merge clipit 1.4.1-1 (universe) from debian unstable" [Undecided,New] https://launchpad.net/bugs/791950 [15:07] evaluate: baltix is a distribution, not a package. And on +activity, you can see who added it [15:08] Hmm? Never heard of baltix... [15:08] Will look it up, thanks! [15:27] tumbleweed: yeah, i'm still looking, although i'm going to be out for the next several hours, so i might have a delayed response to feedback [15:27] broder: only feedback I have: README is probably worth including, and please bump the changelog timestamp [15:28] oh, also I wasn't aware that there was a format for using branch names in Vcs-Git. Are you sure that there is? [15:28] no! :) [15:28] i think it's totally underspecified for VCS's that can hold multiple branches at a single URL [15:30] hmm...i also forgot to check my watch file now that nelson's started tagging releases [15:30] broder: I found a devscripts bug on the subject that was wontfix [15:30] ah, yeah I seem to recall some mangling was necessary with githubredir [15:30] Hi guys, I want to start helping with FTBFS again but have forgotten how to pull source for oneiric into natty....HELP brain not working! [15:31] xteejx: pull-lp-source [15:31] tumbleweed: when i originally packaged this up, nelson hadn't tagged any releases yet, so i was using awesome 0.0.0.git+TIME~etc+more+crud version numbers :) [15:31] tumbleweed: That's the one! Thanks :D [15:34] tumbleweed: do you think i should drop the branch name and assume people will be able to guess? [15:35] broder: ah, debian bug 508433 [15:35] Debian bug 508433 in devscripts "[debcheckout] Impossible to specify branch to checkout" [Wishlist,Open] http://bugs.debian.org/508433 [15:36] no, ubottu, it's won'tfix, closed, and archived. Not open at all [15:37] heh. ok, i'll drop the branch [15:37] it'll still be in vcs-browser so hopefully that's enough of a pointer [15:45] tumbleweed: ok, new packages at http://web.mit.edu/broder/Public/reptyr/ [15:45] but i do actually have to run now [15:46] i'll check backscroll when i get back, and thanks for looking :) [15:46] np [15:49] can anyone tell me why this merge is still on hold? https://code.launchpad.net/~psusi/ubuntu/oneiric/dmraid/drop-64bit-pdc/+merge/60136 [15:49] psusi: because commenting with a "resumit" review isn't the same as resubmitting the merge proposal [15:50] it isn't? [15:50] it's ambiguous and confuses people :) [15:50] what is a resubmit comment then? [15:50] as I understand it, it's for a reviewer to say you should resubmit this [15:50] isn't that what needs information is for? [15:51] I've never seen the point for it, but I think different projects use it differently. [15:51] at any rate, what you want is to either change the merge proposal's status back to "Needs review", or click the "resubmit this merge request" button [15:52] k, got it [15:54] How do I test build a package? Sorry guys I lost my reference material :S [15:54] debuild -S ? [15:55] that will build the source package... leave off the -S to build both source and binary [15:55] cool thanks :) [15:56] Hmm, it's not auto grabbing the build-deps [15:57] xteejx: it won't. Are you maybe thinking of pbuilder / sbuild? [15:57] Well I just wanted to test build a package, rule out autobuiler gremlins [15:58] *autobuilder [15:58] then you probably want to use pbuilder [15:58] doh I just remebered [15:58] pbuilder build *.dsc [16:34] why do people not read the warning dh emits :( [16:34] libosl.so has no links against its deps [16:34] dh_shlibs detects that [16:36] jtaylor: people suck :) (seriously though, dh_shlibdeps was always pretty noisy esp. before --as-needed) [16:37] yes, those "uncessary link" messages mask the important ones [16:37] unecesary link is really only a warning, but missing link should be an error [16:38] jtaylor: it didn't matter much without --as-needed, if you mean gpsshogi FTBFS. Before --as-needed you just had to pass the libraries to the linker to "fix" this, which is why in debian the build succeeds. [16:38] missing dependency is a RC bug or? [16:39] Ampelbein: it does matter as dependencies are missing, that the rdeps build due because they declare there b-d properly does not make it better [16:40] thats all libosl1 depends on: http://paste.ubuntu.com/641403/ [16:40] and its a 150mb source package ^^ [16:41] jtaylor: just saying why this problem didn't show up in debian, not saying it's right. [16:43] and yes, I think missing dependendy is RC bug. [16:48] Hi guys, I'm having problems with pbuilder-dist with a debootstrap error, any way to fix this? [16:51] which error? [16:52] http://paste.ubuntu.com/641410/ [16:52] ^^ pasted here [16:54] I issued "pbuilder-dist oneiric create" btw [16:56] Any idea what the problem is? [16:56] xteejx: that's a known bug [16:57] tumbleweed: Oh right, any way around it for the moment? [16:57] bug 807974 [16:57] Launchpad bug 807974 in eglibc (Ubuntu) "debootstrap fails to install libc6 installing oneiric from natty" [High,Confirmed] https://launchpad.net/bugs/807974 [16:58] easy workaround is to mkdir -p /var/run in your target directory before debootstraping [16:59] Ah ha! I'll try that now :) [17:04] tumbleweed: It didn't work, I did "mkdir -p /var/run" from ~/sources (where I keep MOTU type stuff) and redone the pbuilder-dist oneiric create but the same error came up [17:04] Not to worry I'm sure it'll be sorted - I'll be back ont eh scene in a couple of weeks and hopefully itll be fixed :) [17:04] Thanks again btw [19:01] Hello. I'm following http://people.canonical.com/~dholbach/packaging-guide/html/getting-set-up.html and getting an error when I run "pbuilder-dist oneiric create" [19:02] ap0th: known problem, I'm afraid [19:03] oh, ok [19:03] you do raise a point that ubuntu dev week is going to be painful until that's fixed :) [19:03] in the meantime, you should be able to create a natty chroot successfully [19:04] I tried "pbuilder-dist natty create" as well w/o luck (on Oneiric alpah 2) [19:05] ap0th: what was the error you got? [19:05] oh, that could be the linux 3.0 issue [19:06] "W: Failure trying to run: chroot /var/cache/pbuilder/build/14464/. dpkg --force-depends --install /var/cache/apt/archives/libc6_2.13-0ubuntu13_i386.deb" then "E: debootstrap failed" [19:07] that's the oneiric debootstrap issue, bug 807974 [19:07] (sorry...running in tty1 in Virtualbox because can no longer login after dist-upgrade...so can't really copy-paste) [19:07] Launchpad bug 807974 in eglibc (Ubuntu) "debootstrap fails to install libc6 installing oneiric from natty" [High,Confirmed] https://launchpad.net/bugs/807974 [19:08] so I need to create a /var/run directory? [19:09] in the directory that debotstrap create the chroot in [19:10] pardon my ignorance...but how do I know what directory that is? [19:12] your ignorance isn't out of place here, it can't be predicted :) [19:13] should I just skip the pbuilder step then? [19:14] or wait for this issue to be resolved [19:14] hopefully tomorrow morning... [19:15] I installed Oneiric alpha 2 in Virtualbox and did an update/dist-upgrade and now I am no longer able to graphically login to all 3 options (Unity/Unity-2D/Gnome)...getting "failed to load errors at GDM" - any ideas on that? [19:16] no idea. but you should probably switch to lightdm which will be the default login manager for oneiric [19:16] maybe it was lightdm...whatever was default [19:17] Is there a "proper" time to do dist-upgrades? thinking that bonked my system [19:18] no. There is occasional breakage, so if you see people mentioning massive problems on IRC / the planet, maybe don't upgrade for a couple of days. Otherwise any time is good [19:19] Would there be anything wrong with setting up an Oneiric pbuilder environment on my Natty system or would I run into the same bugs? [19:19] you'll run into the same bug. natty on natty should work, though [19:21] would that help me if I just want to get started working on simple bugs? [19:21] it'll certainly help you get the feel of how to do things [19:21] ok. thanks for your help! === yofel_ is now known as yofel === Quintasan_ is now known as Quintasan [20:26] are there any release managers around that could approve a natty task in bug #675108? [20:26] Launchpad bug 675108 in gnome-power-manager (Ubuntu) "gnome-power-manager adds battery to indicator twice when battery hot-unplugged and plugged back in" [Medium,In progress] https://launchpad.net/bugs/675108 [20:35] psusi: accepted for natty [20:35] stgraber, ty [20:36] psusi: I'm not sure it's relevant for oneiric as default ubuntu will be using the power indicator and I'm not even sure of what's the state of g-p-m with gnome 3.x [20:37] but if you're sure it affects Oneiric too (in a gnome 3.x fallback session for example), I'm happy to accept the oneiric nomination too [20:38] g-p-m seems to have undergone some serious weirdness in oneiric... gnome-power-preferences seems to have gone away, there is no icon, and the new system settings thing shows some sort of configuration that is a much more stripped down version of gnome-power-preferences [20:38] I can't confirm that it affects oneiric or not since it doesn't show up at all... so should it still get a task and just leave it new until it can either be confirmed, or invalidated? [20:39] well, I'm guessing that if you have a task opened for both oneiric + natty the SRU team will expect you to fix it in oneiric first then backport to natty [20:39] which isn't really doable in this csae [20:39] *case [20:40] good point [20:40] ok, I declined it for Oneiric for now. The task can still be re-approved later if needed [20:43] which reminded me I wanted to install indicator-power on oneiric :) [20:45] wow, and it even works! [20:45] you sound surprised at this [20:45] does that bring back the battery icon? isn't that a third party thing? shouldn't gnome-power-manager just work? [20:45] well, considering how difficult it was to get some kind of battery indicator before, I'm :) [20:46] as I understand it (from some mailing lists post) g-p-m essentially dropped the applet. Gnome shell somehow pokes g-p-m for battery info to display in their UI [20:46] and now unity has an indicator doing the same thing [20:47] http://permalink.gmane.org/gmane.comp.gnome.powermanager.devel/2144 was the one I was thinking about