[13:39] <pitti> Daviey, Laney, ScottK, slangasek: do you still consider multiarching libraries FFE material? (I'd like to merge poppler with Debian to get multiarch)
[13:39] <ScottK> At this point I'd say the more the better, but the big question is how it affects rdepends.
[13:40] <pitti> my feeling was that it was FFE material for natty, but now we are confident enough with it
[13:44] <pitti> ScottK: how could it affect rdepends? It's not a special case like gstreamer or gio, it's just a simple .so
[13:45] <ScottK> I know when Qt was multiarched since the libraries were it different locations, some of the rdepends installed files in different locations.
[13:45] <ScottK> This meant .install files needed updating.
[13:45] <ScottK> Usually it was minor (changing a / to /*/), but the trick was finding them.
[13:46] <ScottK> I'm not sure if that would be a concern for other libraries or not.
[13:46] <seb128> it's not for standard libs
[13:46] <pitti> we certainly did have this case for e. g. gstreamer plugins
[13:46] <seb128> qt and gtk are special that they have optional .so they can load
[13:46] <seb128> or gst
[13:46] <pitti> or input methods
[13:46] <seb128> but popple doesn't have anything like that
[13:47] <ScottK> OK.
[13:48] <ScottK> I think it's ok then.
[13:48] <pitti> thanks
[13:51] <Daviey> pitti: A core library, *maybe*.. an optiional library doesn't concern me.
[14:53] <Laney> pitti: are you ok with doing GHC ~
[14:53] <Laney> ~now?
[14:53] <Laney> builders seem as quiet as they'll get
[14:56] <pitti> Laney: seems fine, yes
[14:56] <pitti> the test rebuild is done
[14:56] <Laney> k
[14:56] <Laney> iulian: want to fire ze missiles?
[14:56] <Laney> i'll mail letting everyone to know that it should be done by syncs whenever possible
[15:29] <bjf> pitti: (FYI) bug 931813 reports firewire modules copied to universe instead of main (i verified)
[15:29] <ubot2`> Launchpad bug 931813 in kernel-sru-workflow "linux-lts-backport-oneiric: 3.0.0-16.29~lucid1 -proposed tracker" [Undecided,In progress] https://launchpad.net/bugs/931813
[16:11] <pitti> bjf: fixed, thanks
[16:22] <slangasek> pitti: even for standard libs, revdeps that are Doing It Wrong can FTBFS because of the path change
[16:23] <pitti> slangasek: certainly
[16:24] <pitti> slangasek: so would you think we rather stop multiarchification at this point?
[16:28] <slangasek> pitti: I think it's subject to FFe :)
[16:28] <pitti> ok
[16:28] <slangasek> for things like merges where it's already done in Debian and it would otherwise go in, if it's done early I think we can reasonably make an exception
[18:21] <iulian> Laney: I'd appreciate it if you could take care of uploading GHC. If you're busy, then I can deal with it on Thursday when I get back home.
[18:56] <Laney> iulian: OK I will see what I can do, but I do not think I will have the energy to carry the transition completely through so help is appreciated. I'll mail devel probably.
[18:56] <Laney> the Debian graph is still a fair bit more red than I expected
[18:57] <Laney> http://pkg-haskell.alioth.debian.org/haskell-pkg-graph.pdf
[19:03] <iulian> Laney: I will do as many uploads as I can in the weekend. I just don't have time to work on GHC right now.
[19:04] <Laney> the weekend is pretty bad for me
[19:05] <Laney> running these transitions is pretty emotionally tiresome; I wish computers could just take care of it already
[19:06] <Laney> of course I should put up and implement binNMU in Launchpad rather than complaining
[19:06] <iulian> Indeed. :)
[19:07] <iulian> No worries, we'll be done in no time.
[19:07] <Laney> until the next time
[19:07] <iulian> Only if the buildds can keep up with me.
[19:07] <Laney> fish and chips for dinner will perk me up
[19:07]  * Laney bikes off into the sunset
[19:07] <Laney> (bbl)
[19:07] <iulian> Later!
[19:37] <infinity> Laney: I'm sure I can help with a GHC transition (and I'm not sure how binNMUs make it particularly simpler, except when someone screws up timing)
[20:09] <Laney> infinity: not having to rev the source revision, build, sign, upload manually
[20:09] <Laney> infinity: smarter depwait would probably help more, granted
[20:10] <infinity> Laney: Yeah, I've never found the uploading bit onerous.  Maybe I'm weird.
[20:12] <Laney> having looked at the Debian graph, I think that would be a good place for me to focus my efforts if somebody else will do the merge and initial sync pass(es)
[20:12] <infinity> Laney: If you have a list of what needs merging/uploading, I can certainly help.  I've not really been paying attention to GHC since the last transition.
[20:14] <Laney> infinity: merge ghc, haskell-devscripts and then sync everything else in the standard order
[20:15] <Laney> it should really all be syncs in this case because we needed arch:all rebuilds this time
[20:15] <infinity> Do we actually need a ton of syncs, or will rebuilds do?
[20:15] <infinity> Anyhow, the "merge GHC" bit might be touchy (unless it's a simple merge?), but happy to do the rest.
[20:16] <Laney> some will have required patches, but in general either is fine
[20:16] <infinity> If it's not a rocket surgery merge, I'll poke at that this afternoon.
[20:16] <Laney> the GHC diff should not be too hard, some linker stuff from doko
[20:17] <infinity> Oh, the linker stuff is right up my alley. :P
[20:17] <Laney> quite :-)
[20:17] <infinity> It's mostly a "do I have to speak fluent haskell?" thing.  Cause Haskell gives me aneurysms.
[20:17] <Laney> oh, no, it's m4 I think
[20:17] <infinity> Sold!
[20:40]  * micahg can help with rebuilds also
[20:41] <Laney> you saints!
[20:53] <iulian> Uh oh!
[22:34] <cyphermox> can someone please confirm that the fix for bug 176125 is alright to upload now that we're in FF -- I already included the necessary patch but had forgotten to reset the default value for enabling privacy extensions to TRUE
[22:34] <ubot2`> Launchpad bug 176125 in network-manager "Ubuntu should activate the IPv6 privacy extension by default (echo 2 >/proc/sys/net/ipv6/conf/all/use_tempaddr)" [Wishlist,Confirmed] https://launchpad.net/bugs/176125
[22:35] <stgraber> it's a regression from what we had a few weeks ago isn't it?
[22:35] <ScottK> cyphermox: Please fix it.
[22:35] <cyphermox> stgraber: yes
[22:36] <ScottK> Either way, I think lack of privacy extensions is a bug.
[22:36] <slangasek> cyphermox: I agree, bug -> squash it
[22:36] <cyphermox> it is indeed a regression, with just procps this was already set to use privacy extensions since at least january, perhaps earlier
[22:36] <cyphermox> cool, just being thorough
[22:37] <cyphermox> so this I'll upload in a minute with a few other fixes