[14:36] <xnox> can libimobiledevice be denewed ^ it's api/abi transition which i'd like to complete whilst patch piloting
[14:37] <seb128> xnox, looking
[14:37] <xnox> seb128: thanks a lot!
[15:22] <bdmurray> cjwatson: How would you find the publishing history here? https://launchpad.net/ubuntu/raring/i386/apport  I'm specifically interested in the Superseded record with p-u-p set to 10%.
[15:22] <bdmurray> using the Launchpad API
[15:27] <slangasek> infinity: precise ISOs are all listed as oversized... do we have too many kernels again?
[15:27] <ogra_> cant have anough kernels
[15:28] <cjwatson> bdmurray: You mean something like this?  http://paste.ubuntu.com/5962969/
[15:31] <bdmurray> cjwatson: yep, but is there anyway to get there from getPublishedSources?
[15:31] <tseliot> any admins around who can reject some packages for me?
[15:31] <stgraber> tseliot: sure
[15:31] <tseliot> stgraber: nvidia-prime fglrx-pxpress and jockey in precise-proposed, thanks
[15:32] <stgraber> tseliot: done
[15:32] <tseliot> stgraber: thanks again
[15:33] <wgrant> bdmurray: spph.getPublishedBinaries()
[15:33] <cjwatson> bdmurray: You could go through each published source and call getPublishedBinaries on it, I suppose, but it would be slow if you had to go back very far
[15:37] <slangasek> tseliot: reject> hmm, so are you prepping a reupload?
[15:38] <tseliot> slangasek: yep, I wanted to add a fix for bug which has already landed in saucy. I'll reupload in a minute
[15:38] <tseliot> *for a bug
[15:39] <tseliot> slangasek: I can give you the relevant diffs if you want
[15:39] <slangasek> tseliot: ok
[15:48] <tseliot> slangasek: you can find the patches in the three directories here: http://people.canonical.com/~amilone/1198942/
[16:55] <cjwatson> ^- one-line fix for a v-failed SRU; should be a quick review for somebody
[17:03] <stgraber> cjwatson: I'll take a look
[17:08] <stgraber> bdmurray: any idea why ./sru-review -s lucid debootstrap doesn't work?
[17:08] <cjwatson> works for me
[17:09] <stgraber> http://paste.ubuntu.com/5963264/
[17:11] <bdmurray> Did you get the warning about it already being published?
[17:11] <stgraber> nope, just what I pasted above
[17:11]  * stgraber tries on another machine
[17:12] <stgraber> bdmurray: ok, on another machine it seems to work (and indeed I'm getting the already published warning there)
[17:13] <stgraber> so maybe some caching problem on my laptop, will investigate later
[17:14] <stgraber> cjwatson: there you go
[17:14] <bdmurray> stgraber: okay, weird
[17:15] <cjwatson> thanks
[19:57] <bdmurray> Does anybody have an idea why ogmrip is in the SRU queue for raring?
[19:58] <bdmurray> the changelog just says "New upstream release"
[20:58] <Ampelbein> Ups. I accidently uploaded to precise-proposed, can that upload of tboot be removed please. Sorry sorry sorry.
[22:29] <infinity> slangasek: We shouldn't be overly kerneled, the fix last time around should still be valid.  Stuff might have just gotten bigger again.  Time to drop English? :P
[22:30] <infinity> slangasek: Or, more realistically (if it's not something obvious we can fix), time to say "if you want CD-sized ISOs, install from 12.04.1" and stop guaranteeing it for point releases?
[22:35] <slangasek> infinity: wasn't 12.04 already oversized for CDs?  I don't remember now
[22:35] <infinity> slangasek: Nope.
[22:36] <infinity> slangasek: http://releases.ubuntu.com/precise/
[22:36] <slangasek> anyway, if everything has just gotten bigger, then yeah we should raise the limit and call it good; but I think we want to check first that we don't have spare packages on the images
[22:36] <infinity> slangasek: Those are all CD sized.  But easy enough to diff some manifests and see if something went obviously awry or if it's just that firefox/tbird/etc grew again.
[22:36] <slangasek> infinity: would you mind having a look at that?
[22:37] <infinity> slangasek: I think we're down to a point where we can no longer remove any langpacks, so we're running shy on options to shrink precise.
[22:37] <infinity> slangasek: Yeahp, I can haz look.
[22:37] <slangasek> ok
[22:48] <infinity> slangasek: Nope, doesn't look like we've added anything (though, we've removed a few packages), so it's just that stuff got bigger.
[22:48] <slangasek> infinity: ok, thanks for checking
[22:48] <infinity> slangasek: I'll poke at it and see if there's any sane path to shrinkage before we decide to give up and tell people precise point releases ain't gonna fit on CDs anymore.
[22:49] <slangasek> cjwatson: ^^ any objections to just letting 12.04.3 spill over beyond CD size?
[22:49] <slangasek> hmm, this may be my favorite typo ever: DEB_BUILD_POTIONS
[22:51] <xnox> slangasek: https://plus.google.com/+FlorianRohrweck/posts/KAgNuNKQHgJ
[22:52] <slangasek> wgrant, infinity: hey, so do you guys know if launchpad is ok with :any build-dependencies, or if there's more work needed to support that?
[22:52] <slangasek> xnox: hah
[22:55] <wgrant> slangasek: Our sbuild will break on that.
[22:56] <wgrant> I believe.
[22:59] <slangasek> wgrant: fiddlesticks
[22:59] <cjwatson> slangasek: spill over> I have minimal ability to care just now, I think
[22:59] <cjwatson> people can always use .2.  of course it would be nice to avoid it
[23:00] <slangasek> wgrant: so, ah, how about that plan to upgrade away from a fork of a nine-year-old version of sbuild? ;)
[23:04] <slangasek> wgrant: the reason I'm asking is that we're at the point where this blocks having stock packages in the archive be cross-buildable, for things like unity: https://code.launchpad.net/~saviq/unity8/prepare-cross-build/+merge/179250
[23:04] <slangasek> cjwatson: ack
[23:06] <wgrant> slangasek: The main blocker is getting Soyuz ddeb support switched on, which is waiting on librarian space, which is waiting on a physical SAN move.
[23:06] <wgrant> We went through it all in London, and it looks like the upgrade should be pretty painless.
[23:06] <slangasek> wgrant: ok.  Do you have an ETA from IS on this?  (Is there as RT I can play along with at home?)
[23:10] <wgrant> slangasek: I'm not sure of an ETA.
[23:13] <slangasek> ok
[23:13] <slangasek> I guess I'll pester IS then :-)
[23:39] <slangasek> wgrant: btw, how does soyuz ddeb support relate to the sbuild upgrade?
[23:41] <wgrant> slangasek: ddebs are currently implemented by a hack in sbuild which copies them to public_html.
[23:41] <slangasek> aha
[23:41] <wgrant> ITYM eww
[23:42] <slangasek> so... supposing we needed :any support sooner than IS was going to get us the newshiny librarian... how much work would it be to forward-port that patch?
[23:42] <wgrant> It's possible that we can do it from outside sbuild.
[23:42] <wgrant> Or hack it back in
[23:43] <wgrant> But we also either need to backport precise's sbuild to hardy, or get the remaining buildds off hardy.
[23:43] <wgrant> And we can't do the latter yet.
[23:44] <slangasek> we... can't?
[23:44] <wgrant> It's complicated :)
[23:44] <cjwatson> needs builderstack
[23:45] <slangasek> muttrgrmbl
[23:45] <cjwatson> we should draw a diagram, the dependency graph is non-trivial
[23:51] <infinity> slangasek: Doing any support in our sbuild is trivial.  And may already work.
[23:51] <infinity> slangasek: Because the right answer when you're always building natively (as we are) is just to stip off :foo and pretend you wanted native.
[23:51] <slangasek> infinity: right, that's a much better shortcut
[23:52] <infinity> And we already do.
[23:52] <infinity>         # For the moment, we treat multiarch-annotated build-dependencies as
[23:52] <infinity>         # the same as any others because we're not implementing a
[23:52] <infinity>         # cross-buildd.
[23:52] <slangasek> ah, spiff
[23:52] <infinity>         $deps =~ s/:any//g;
[23:52] <infinity> So, sbuild supports it, as long as LP doesn't choke elsewhere.
[23:52] <infinity>         $deps =~ s/:native//g;
[23:52] <slangasek> so unity isn't going to explode suddenly :)
[23:53] <infinity> Wait, didn't we already have a bunch of :any build-deps before?  (gettext?)
[23:53] <infinity> So, this is provably tested and known working.
[23:54] <slangasek> ah, yes; I think those wound up being backed out and handled differently because *Debian* wasn't ready to cope with them yet at the time
[23:54] <slangasek> but they did go through once already in Ubuntu
[23:54] <wgrant> Oh, I just remembered that we got rid of some in the not too distant past, but it could quite possibly be that it was a Debian issue... oops.
[23:54] <infinity> So, yes, the "use a sane sbuild" thing is on the roadmap, but not urgent.
[23:55] <slangasek> infinity: right, thanks for setting me straight before I started burning the curtains
[23:57]  * infinity wonders why lib32asound2 and libc6-i386 are on the precise CDs.
[23:57] <infinity> Not that this is new, they appear always to have been.
[23:58] <slangasek> interesting question
[23:59] <infinity> sl-modem-daemon, apparently.
[23:59] <slangasek> horrifying answer