[00:01] <bdmurray> bryceh: hmm, not too many it looks like
[07:02] <broder> slangasek: looks like there are a handful of packages that have popped up on <http://lintian.ubuntuwire.org/tags/preinst-uses-dpkg-maintscript-helper-without-predepends.html>. any chance i can pawn those off on you to follow up on?
[07:20] <broder> anybody on ~ubuntu-branches around? looks like https://code.launchpad.net/~evfool/ubuntu/oneiric/brasero/fixlp848340/+merge/85782 should be marked as merged (http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/oneiric/brasero/oneiric-proposed/revision/91)
[11:37] <AlanBell> hi all, anyone know who is involved in the new sound theme project?
[14:12] <cjwatson> broder: dpkg-maintscript-helper> fair number of false positives in there since it doesn't understand things being guarded with 'supports' checks
[14:13] <cjwatson> though arguably it's better to just pre-dep these days
[14:24] <cjwatson> broder: marked that MP as merged, thanks
[15:58] <rhorstkoetter> hi. I hope this an appropriate channel for my question ;) I'm wondering how to use casper-rw persistent overlay ext2/3/4 partition for live-usb with the new hybrid iso that ships with oneiric+X
[15:58] <rhorstkoetter> so far, we used f32 and ext2/3/4 partition labeled casper-rw with syslinux as a workaround
[15:59] <rhorstkoetter> problem is that the hybrid iso dumped to usb doesn't actually recognize the casper-rw ext2/3/4 partition appended to the end of the stick
[15:59] <rhorstkoetter> is there any conf option I'm missing, i.e telling grub to actually use that partition as an overlay?
[15:59] <rhorstkoetter> thanks in advance for your help
[16:01] <rhorstkoetter> sidemark: I actually tried asking in #ubuntu but they doesn't seem to understand my actual problem, thus I decided to ask here. I hope this is sufficient
[16:06] <kklimonda> rhorstkoetter: does hybrid iso actually have a partition table? I was wondering that myself lately
[16:06] <rhorstkoetter> kklimonda: um, I'm actually not sure
[16:07] <rhorstkoetter> good question )
[16:07] <rhorstkoetter> ;)
[16:07] <rhorstkoetter> I know though that something similar is possible with the opensuse hybrid isos, where the ubuntu tech is (most likely) adopted from
[16:08] <rhorstkoetter> but suse even does the persistence step automatically on first boot
[16:09] <kklimonda> yeah, that's what I've done myself before hybrid image
[16:09] <rhorstkoetter> I actually never had any issues with the f32 plus casper-rw approach/syslinux workaround, but you may have hit the nail .. the partition table
[16:09] <kklimonda> I have a small initrd script that creates additional partitions and mount them
[16:10] <kklimonda> (but it obviously depends on an existing partition table)
[16:10] <kklimonda> I'd be interested how to do that with the new hybrid image myself
[16:11] <rhorstkoetter> kklimonda: we're two then
[16:12] <rhorstkoetter> I think I'll ask some people I know at suse and if they know what to do, maybe a feature request would help?
[16:12] <rhorstkoetter> i.e. to adopt the rest of the suse automatism?
[16:14] <rhorstkoetter> btw, before the hybrid iso, I mean with ubuntu, there isn't even a initrd script needed
[16:15] <rhorstkoetter> you just need to create the live usb, add a second partition labeled casper-rw (ext2/3/4 formatted), plug in and you're set
[16:15] <rhorstkoetter> kklimonda: not sure though if I misunderstand something about your initrd approach
[16:16] <kklimonda> yes, but I don't want make users create partitions themselves
[16:17] <rhorstkoetter> kklimonda: oh man, you just put in one sentence why I like ubuntu that much more compared to suse. they understand user-convenience
[16:17] <rhorstkoetter> i.e. they know what that actually means ;)
[16:17] <rhorstkoetter> anyway, this is another topic
[16:26] <kklimonda> it's a great question, and it would be a good idea to check how SUSE does that
[16:26] <kklimonda> I'd come back this monday when more of the foundation developers are here and talk with them about it
[16:27] <rhorstkoetter> they're usually here all weekdays?
[16:28] <rhorstkoetter> it may be difficult to get an answer over the weekend. but I can certainly contact some people to find out and then get back to you guys
[16:28] <kklimonda> yes, most (if not all) of them work for Canonical
[16:28] <rhorstkoetter> great
[16:28] <kklimonda> so they are here during the week, and don't work on weekends :)
[16:28] <rhorstkoetter> I'll get back once I know better
[16:28] <rhorstkoetter> sure, I understand
[16:28] <rhorstkoetter> well deserved spare time (as is with suse devs)
[16:29] <rhorstkoetter> ok, thx so far. see you
[16:29] <rhorstkoetter> have a good weekend
[17:38] <jbicha> webkit i386 build ran out of memory again :(
[17:40] <tumbleweed> some of the i386 builders have more ram than others
[17:42] <Ampelbein> Wasn't webkit supposed to build on a more powerful builder?
[17:44] <Ampelbein> Oh, it did but failed due to -c1 instead of -c0.
[17:46] <tumbleweed> we have a way of pointing apckages at buildds?
[17:48] <Ampelbein> Yeah, I guess a buildd admin just sets the other buildds to manual
[17:48] <tumbleweed> oh, that way :P
[18:07] <stgraber> tumbleweed, jbicha: so does the new webkit (-ubuntu3) need some magic to use the right builder for i386?
[18:09]  * stgraber prepares to do buildd magic if that's the case
[18:09] <tumbleweed> stgraber: dunno, I don't know the webkit package
[18:09] <tumbleweed> but I have had trouble with an i386 bulider not having as much memory as I'd like, before
[18:09] <stgraber> k, I'll force it to build on roseapple and see if that helps
[18:09] <jbicha> stgraber: thanks
[18:10] <Ampelbein> stgraber: roseapple should help, see the -ubuntu2 build.
[18:10] <Ampelbein> That one failed only due to symbols mismatch.
[18:11] <stgraber> ok, roseapple is the only buildd still in auto mode and I bumped the score of webkit, so whenever roseapple is done with the current build, it should start building on it
[18:11] <stgraber> building now
[18:11] <stgraber> and moved the rest back to auto mode, hopefully that'll work
[18:34] <ingwa> hello
[18:34] <ingwa> I searched for a unity channel but found none.  Is this the right place for unity questions?
[18:35] <jussi> ingwa: for normal tech support type questions, I would say #ubuntu
[18:35] <jussi> ingwa: what sort of question do you have?
[18:36] <ingwa> I have a general question: are applicaitons for unity going to be developed with a certain look&feel?  Perhaps using a certain toolkit?
[18:36] <jussi> ingwa: ahh, I would say #ubuntu-unity for that
[18:37] <ingwa> jussi: that sounds like a good tip.
[18:37] <jussi> :)
[18:37] <ingwa> But it would be even better if it could be found on https://wiki.ubuntu.com/IRC/ChannelList :)
[18:37] <ingwa> jussi: that channel seems well populated so thanks
[18:38] <jussi> ingwa: Ill zap them to update that - its a fairly new channel replacing #ayatana
[18:38] <ingwa> jussi: excellent
[18:38] <ingwa> and thanks again
[18:38] <jussi> yw
[18:53] <AlanBell> #ubuntu-unity is now in the list
[18:55] <broder> cjwatson: a supports guard isn't actually sufficient, is it? it prevents the maintainer script from erroring out, but doesn't actually handle the conffile mv/rm/etc, no?
[19:18] <cjwatson> broder: yeah, that's what I thought after my initial comment
[19:19] <cjwatson> broder: I've changed a few packages already on the grounds that a pre-dep creates more reliable behaviour
[19:19] <cjwatson> pretty sure pcmciautils had a latent upgrade bug due to this, since update-rc.d would've failed if the old conffile hadn't been removed, for instance
[19:21] <broder> cjwatson: https://code.launchpad.net/~l3on/ubuntu/oneiric/gdevilspie/fix-for-783568/+merge/89718 should also be marked as merged, if you've got a sec
[19:27] <cjwatson> broder: done
[19:27] <broder> thanks
[19:35] <Pjotr> Hello, I have a question about a Ubiquity bug that I reported on Launchpad:
[19:35] <Pjotr> https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/930676
[19:36] <Pjotr> My question is: should I file a bug report against migration-assistant as well?
[19:37] <Pjotr> @cjwatson: maybe you know?
[19:37] <udevbot> Error: "cjwatson:" is not a valid command.
[19:38] <Pjotr> cjwatson: maybe you know?
[19:45] <Pjotr> ev: if I'm correct, you're the maintainer of migration-assistant. Should I file this Ubiquity bug, against migration-assistant as well?
[19:55] <Pjotr> Well, I suppose a separate bug report against migration-assistant is best. So I just filed it:
[19:55] <Pjotr> https://bugs.launchpad.net/ubuntu/+source/migration-assistant/+bug/930754
[19:58] <Pjotr> https://bugs.launchpad.net/ubuntu/+source/migration-assistant/+bug/930754
[22:28] <haled> how do I get HUD running in 12.04 alpha?
[23:08] <lamont> do we think webkit is going to successfully build on zirconium?
[23:14] <stgraber> lamont: I moved it to roseapple earlier but the build was killed and the builder disabled for some reason
[23:15] <stgraber> lamont: we need a builder with a lot of RAM for it to build (AFAIK, I'm not webkit expert)
[23:15] <stgraber> lamont: so if the current builder doesn't have plenty of RAM, we should kill the build and move it back to another builder like roseapple
[23:16] <elmo> umm
[23:16] <elmo> guys
[23:16] <elmo> requiring > 3Gb to link isn't reasonable or sane
[23:16] <elmo> and is either a webkit or binutils regression
[23:16] <elmo> moving this package around buildds is not a sane long term strategy
[23:16] <elmo> and eventually you're going to run into limits you simply can't get round - this is 32-bit after all
[23:16] <stgraber> elmo: as I said, no webkit expert here, I just want to be able to install an amd64 system ;)
[23:16] <stgraber> elmo: I definitely agree that requiring >3Gb to link is insane ;)
[23:27] <lamont> stgraber: who's owning fixing it to not need >3GB to link?
[23:30] <stgraber> lamont: desktop team I guess?
[23:31] <stgraber> well, ultimately upstream I'd think but desktop team should be responsible of nagging them to do so ;)
[23:32] <cjwatson> elmo: we know
[23:32] <cjwatson> elmo: but it's breaking all amd64 image builds so we're trying short-term as well as long-term solutions
[23:32] <lamont> restarting/ed on roseapple.  This should be fixed before it's uploaded again
[23:32] <stgraber> lamont: thanks
[23:32] <cjwatson> I already suggested a couple of ld options to seb128 that might help
[23:33] <lamont> cjwatson: do you have a name for me to point people at if someone uploads ubuntu4 without it fixed?
[23:33] <cjwatson> but it takes six hours for him to build it locally so it does take a while to fix this kind of stuff
[23:33] <cjwatson> lamont: seb128 was the last person I was aware of who was working on it
[23:34] <lamont> I'll go with that then
[23:34] <cjwatson> how actively that was, I don't know
[23:34] <lamont> I'll poke him on monday
[23:35]  * lamont files the bug against webkit, just for completeness
[23:37] <lamont> bug 930805
[23:38] <lamont> and marked critical
[23:39] <lamont> stgraber: cjwatson: any time we decide to jockey a build to a particular builder, there needs to be a bug filed, since there clearly is one somehow (or an RT if it's truly something that needs to be fixed on the builder, in which case the builder should be disabled)
[23:40] <lamont> I wonder... is this going to break my upgrade to precise that I'm doing now?
[23:40] <cjwatson> agreed
[23:40] <cjwatson> quite possibly, if you're on amd64
[23:40] <lamont> fact
[23:40] <lamont> too late to stop now, though
[23:41] <cjwatson> which is why we've been trying to fix this for two or three days now ...
[23:41] <lamont> I wish I had a "go faster" button on roseapple that I could push
[23:41] <cjwatson> we know it got past the link on roseapple before, but then ran into a later failure
[23:41] <cjwatson> so it should manage it again if only it keeps building
[23:42] <lamont> roseapple is the only 64-bit kernel i386 builder
[23:42] <lamont> ah, so ubuntu3 is the fix for the later failur?
[23:48] <cjwatson> lamont: yes
[23:53] <lamont> well, as long as precise works with webkit 1.4.3... this does not fill me with hope.