[05:27] <alkisg> Ah ok the missing headers bug I mentioned yesterday is already reported in launchpad: https://bugs.launchpad.net/ubuntu/+source/linux-lts-backport-maverick/+bug/824080
[08:56] <AnAnt> Hello, is there any info about making themes for lightdm ?
[09:05] <AnAnt> are GTK2 themes compatible with GTK3 ?
[09:45] <doko> ev, fyi, accepted the gst upload
[12:07] <bluefoxicy> 1570 bluefox   20   0 1818m 929m 4436 S    3 25.1 228:10.18 compiz
[12:07] <bluefoxicy> lol compiz using 2 gig
[12:15] <AnAnt> Hello, are GTK2 themes compatible with GTK3 ?
[12:52] <flecha> Hello! I am developing an App Indicator for Unity (those that appear in the menu bar). How do I make it binded to a hotkey?
[13:45] <cjwatson> slangasek,jelmer: do either of you have a clue what's going on with the cyrus-sasl2/i386 build failure?  configure is failing to include gssapi/gssapi_ext.h, and there appear to be conflicts between /usr/include/gssapi/gssapi_ext.h and /usr/include/heimdal/gssapi/gssapi.h; I don't know whether that's the root cause
[13:48] <jelmer> cjwatson: looking..
[14:24] <flecha> is there a way to map a key to open a hidden gtk window?
[14:59] <infinity> sgnb: Do you feel like convincing me to upgrade ocaml to 3.12.1 today?  I feel like listening. :P
[15:08] <jelmer> cjwatson: sorry, got distracted. Found the issue - cyrus-sasl2 is hardcoding the library path rather than using krb5-config, which broke now heimdal switched to multi-arch.
[15:09] <sgnb> infinity: are you aware that doko has uploaded a fixed (3.12.0) ocaml? (thanks, btw)
[15:09] <sgnb> infinity: with the switch to 3.12.1, (almost) everything will have to be recompiled, too... and it will happen soon in Debian, too
[15:10] <sgnb> infinity: it is mostly a bugfix release, and (almost) everything from testing compiles with no changes
[15:10] <sgnb> it will also be easier to track, too, via http://people.canonical.com/~ubuntu-archive/transitions/ocaml.html
[15:10] <sgnb> (everything is currently uselessly green)
[15:16] <sgnb> infinity: the updated ocaml package is in my ppa, I can also provide ocamlduce... and everything else (except jocaml, mingw32-ocaml, janest-core) will be recompilation with no changes (or sync with Debian)
[15:17] <sgnb> it would probably be too intrusive for a Debian freeze, but I don't know what is acceptable for Ubuntu feature freeze
[16:08] <cjwatson> jelmer: aha - are you fixing, or do you need sponsorship / other help?
[16:09] <cjwatson> jelmer: (distracted> no problem, it's kind of a weekend)
[16:09] <jelmer> cjwatson, sponsorship would be great
[16:09] <jelmer> cjwatson, https://code.launchpad.net/~jelmer/ubuntu/oneiric/cyrus-sasl2/ftbfs-825872/+merge/71456
[16:10] <infinity> sgnb: I'm not concerned about compilation time (since the slowest arch we have has to recompile everything anywa), I'm just concerned about potential impact and breakage.  If it really is "mostly bugfixes", that would be convincing enough, I think.
[16:10] <cjwatson> hm, wonder why I didn't already get mail about that
[16:10] <infinity> sgnb: Well, if you can define what isn't bugfixes that makes us say "mostly" instead of "entirely". :P
[16:10] <cjwatson> oh, "4 minutes ago"
[16:11] <cjwatson> jelmer: will review once bzr co gets round to giving me a useful working tree (modulo baking in progress)
[16:18] <jelmer> cjwatson: great, thanks :)
[16:19] <cjwatson> done, thank you
[16:25] <sgnb> infinity: http://anonscm.debian.org/gitweb/?p=pkg-ocaml-maint/packages/ocaml.git;a=blob;f=Changes;hb=HEAD
[16:25] <sgnb> upstream is very careful about not breaking stuff in minor releases
[16:31] <blueyed> wow.. cannot believe bug 658865 is still there.. (been hit by it on the second machine now)
[16:37] <doko> how can you that compiz is broken?
[16:38] <doko>   ... you are able to see the lower right character in a terminal ;-P
[16:56] <infinity> sgnb: That does, indeed, look fairly conservative.
[16:58] <infinity> cjwatson: Care to give the changelog that sgnb referenced above a once-over and give me a second opinion on FFeing a new ocaml before I have to reupload/retransition/rebuild the ocaml world (which I have to do anyway, regardless...)
[16:59] <infinity> cjwatson: The few small non-bugfix changes look pretty conservative and isolated to me (well, isolated in functionality, at least), and he seems confident in them.
[17:00] <infinity> sgnb: If we give this a go-ahead, when would I be able to steal your time and expertise to help with any speedbumps and get it done quickly?
[17:19] <sgnb> infinity: I'm available during the next 8 days
[17:24] <sgnb> FWIW, I recompile everything in < 6 hours on my (amd64) own machine...
[17:25] <sgnb> infinity: packaging for jocaml and mingw32-ocaml are not updated and most likely broken, is that a problem?
[17:25] <sgnb> (I don't take care of those myself)
[17:25] <sgnb> (no rdeps)
[17:42] <jelmer> cjwatson, still there?
[18:45] <infinity> sgnb: It would be nice if we could get those fixed too.
[19:06] <sgnb> well... I would shamelessly have them removed from testing if they are not fixed by the time the transition is done in Debian
[19:06] <sgnb> I would understand if this is a blocker for Ubuntu
[19:07] <sgnb> infinity: ^^^
[19:08] <infinity> Yeah, intentionally breaking other software at this point in our release cycle would be bad.  So, we either need to have a clear plan about how to make sure everything works, or not do it at all. :P
[19:13] <sgnb> infinity: ok, let's recompile everything as it is, then
[19:44] <sgnb> infinity: after looking, in a first round, you should start with: camlidl camlp5 camlzip cothreads cryptokit facile findlib lablgl llvm-2.7 llvm-2.8 llvm-2.9 menhir mlgmp ocamlagrep ocamlduce ocamlpam perl4caml pycaml react uuidm xml-light graphviz jocaml
[20:07] <sgnb> infinity: actually, jocaml might need the same patch
[21:29] <sveinse> When changing runlevels in with upstart, how is start and stop ordering handled? E.g. if I'm in runlevel 2 and enters level 3, will the "stop on runlevel [!2]" services stop before the "start on runlevel [3]" are started?
[22:22] <cjwatson> jelmer: kind of, best to leave a message :-)
[22:55] <penguin42> Does anyone else find the 'not directly subscribed to this bug's notification' that lp has these days to be entirely non-obvious when you look at it?
[23:04] <mdeslaur> penguin42: yes, it had me confused as well when I saw it the first time
[23:05]  * penguin42 finds it takes him a few seconds to read it each time