[11:02] <allenap> ttx: The tests all passed for lp:~ttx/launchpad/lp690712 on Friday, but the submission to PQM got lost it seems :-/ I am submitting it again.
[11:05] <ttx> allenap: great, thanks
[11:39] <wgrant> allenap: We were in testfix for most of the day.
[11:39] <wgrant> And PQM doesn't send failure emails any more.
[11:40] <allenap> wgrant: Interesting. Was the silent PQM a decision, or did it just get the grump one day?
[11:41] <wgrant> allenap: The upgrade broke it.
[11:41] <wgrant> I don't know if it's been investigated.
[11:58] <maxb> There are some branch scanner malfunction questions that have been open for a long time - could someone take a look and route them to whoever is appropriate to get and analyse the production logs?
[11:58] <maxb> https://answers.launchpad.net/launchpad/+question/142589
[11:58] <maxb> https://answers.launchpad.net/launchpad/+question/142142
[14:10] <SevenMachines> hi, i was wondering i some nice person could kill a stalled ppa package build for me? amd64 and i386 both stuck for over a day https://edge.launchpad.net/~sevenmachines/+archive/testing/+builds?build_state=building
[14:12] <bigjools> SevenMachines: yes, one moment.  Is it a package bug?
[14:12] <SevenMachines> No, it was a bug with the source, i meant to upload the fixed version
[14:13] <bigjools> ok
[14:14] <SevenMachines> great, thanks bigjools
[14:30] <bigjools> SevenMachines_: did you upload the new version already?
[14:35] <bigjools> nm you deleted the source
[14:36] <SevenMachines_> i tried to delete the source before the build started once i realised that it was the bad version and was going to hang, is that the wrong thing to do?
[14:36] <SevenMachines_> Haven't uploaded the new version yet
[14:37] <SevenMachines_> if only there was a ctrl-c link :)
[14:38] <bigjools> SevenMachines_: that was the right thing, you were just unlucky it got dispatched first
[14:38] <bigjools> detecting bad builds is hard :/
[14:39] <SevenMachines_> Thats the pain, i new it was bad and i had the fixed one right there but rushed it and mistakenly uploaded the broken one. Less haste more speed i suppose
[14:41] <bigjools> yup :)
 There are some branch scanner malfunction questions that have been open for a long time - could someone take a look and route them to whoever is appropriate to get and analyse the production logs?
 https://answers.launchpad.net/launchpad/+question/142589
 https://answers.launchpad.net/launchpad/+question/142142
[15:30] <kklimonda> hey, I have a package that has stopped building recently, it's sitting waiting for libcurl4-dev which is a virtual package provided by libcurl4-gnutls-dev - this dependency was there for as long as remember and it started failing only now, built fine even in december, and I can't reproduce it on my pbuilder. Any idea what's going on? package link is
[15:30] <kklimonda> https://launchpad.net/ubuntu/+source/transmission/2.13-0ubuntu2
[16:02] <maxb> kklimonda: 'apt-get install libcurl4-dev' in an uptodate natty pbuilder fails.
[16:08] <kklimonda> maxb: and yet the package build fine :)
[16:08] <kklimonda> (and pulls the right libcurl dev dependency)
[16:09] <maxb> That's down to different dependency resolution methods employed by pbuilder vs. sbuild
[16:10] <maxb> Essentially, if a virtual package has multiple possibilities, you have to hint at the one you want used by writing libcurl4-gnutls-dev | libcurl4-dev
[16:11] <kklimonda> so is it sbuild being stupid, or pbuilder to smart for its own good?
[16:11] <kklimonda> too smart* even
[16:12] <tsimpson> probably them just being different
[17:00] <geser> maxb: while I agree with you that pure virtual build-dependencies should be avoided, I still don't understand why it stopped working between the build in December and now: it both cases libcurl4-dev gets passed to "apt-get install" but once it gets resolved and once not and apt didn't change inbetween
[17:01] <geser> maxb: see http://launchpadlibrarian.net/60523825/buildlog_ubuntu-natty-amd64.transmission_2.13-0ubuntu1_BUILDING.txt.gz for comparison
[17:27] <maxb> geser: The key is that on 2011-01-26 the curl source package changed to produce multiple binaries which Provide: libcurl4-dev
[17:31] <geser> ah, missed that
[19:59] <fta> danilos, yt?