[00:53] <thumper> where'd my damn code go....
[01:13] <thumper> :-(
[01:14] <poolie> thumper, ?
[01:14] <thumper> installing repoze.what causes Launchpad not to build
[01:14] <thumper> no freaking idea why
[01:15] <thumper> from the import site part of our scripts
[01:16] <thumper> bin/py to get an interactive interpreter
[01:16] <thumper> then "import site" blows up
[01:16] <poolie> ouch.  i'm glad it's not my fault though :)
[01:19] <thumper> apt-get remove python-repoze.what fixed it
[01:19] <thumper> for now
[01:19] <thumper> but we should look into it
[01:19]  * thumper files a bug on foundations
[01:22]  * thumper afk to get Maia and have lunch
[06:28]  * thumper goes to make dinner, back to talk to some europeans
[08:03] <adeuring> good morning
[09:08] <bigjools> good morning
[09:09] <jelmer> 'morning bigjools
[09:10] <bigjools> hey jelmer
[09:17] <mrevell> Hello!
[09:18] <jelmer> 'morning Matt
[09:25] <bigjools> jelmer: hey you're not on the internal server
[11:03] <deryck> Morning, all.
[11:06] <adeuring> morning deryck
[12:42] <SmartViking> Launchpad power
[12:43] <SmartViking> Hi everyone! :D
[12:48] <SmartViking> No one want to say hi to the lonely viking :(
[12:50] <SmartViking> sidnei hi! :D
[12:50] <sidnei> hi
[12:53] <SmartViking> Why are people in this channel if they dont talk? :)
[16:28] <mars> rockstar, ping
[16:32] <rockstar> mars, pong
[16:33] <mars> hi rockstar, did you have some time to chat today about the windmill tests?  You said on Friday that you may have a local reproduction of the hang.
[16:34] <rockstar> mars, I may have some time, but not for the next few hours.
[16:34] <rockstar> mars, the basic idea is to run windmill without a $DISPLAY
[16:34] <mars> rockstar, that's fine, I'll be around :)
[16:35] <mars> rockstar, interesting
[16:35] <rockstar> I suspect that whatever fake X11 server we're running is broken.
[16:36] <mars> maybe.  I did not check for an xvfb.log on the hung servers.  That would have been useful.
[16:37] <mars> rockstar, unfortunately the test_on_merge.py issues obliterated the crime scene - before this, we would not have been able to see any X server issues, because the server was being forcibly killed
[16:38] <mars> rockstar, when I reproduce this on EC2, that will be one of the first things I look for then.  That is a good hypothesis.
[16:41] <rockstar> mars, I know for a fact that it hangs here, and it looks exactly like it looked on ec2.
[16:41] <mars> rockstar, did you just unset $DISPLAY to get it to happen?
[16:42] <rockstar> mars, I run launchpad in a chroot, so I have to set $DISPLAY to run the windmill tests.  It was just pure coincidence that I forgot to set it.
[16:48] <mars> rockstar, ok, thanks for the insight
[17:36] <deryck> BjornT, ping.
[18:00] <mrevell> Night all :)
[19:12] <Guest99246> hello how do you remove everything installed by rocketfuel (launchpad's source)
[19:20] <matthew_> does any body know how to remove it?
[20:47] <sinzui> deryck, ping
[20:48] <deryck> hi sinzui.  Was just about to log out for EOD, though.
[20:48] <sinzui> deryck, I extracted a mixin class from bug supervisor that will be used in 3 different modules...
[20:49] <sinzui> deryck, Do you want this mixin in lp.bugs.brower.__init__ or lp.bugs.browser.bugroles
[20:50] <deryck> hmmm, bugroles kind of seems nice.
[20:50] <deryck> sinzui, yeah, let's put it in bugroles
[20:50] <sinzui> I will create it and move the mix there
[20:51] <deryck> sinzui, thanks!  Looking forward to this refactor.
[22:51] <thumper> sinzui: ping
[23:06] <krkhan> in a locally running launchpad, i'm getting the 'unknown consumer' error on all REST reqs. do i have to add known consumers somewhere in lp?
[23:19] <bryceh> what happened to lib/canonical/launchpad/doc/security-teams.txt ?  It seems to no longer be in the current tree?
[23:19] <sinzui> hi thumper
[23:20] <thumper> sinzui: hey, got some time for a call?
[23:21] <sinzui> I do not for the next hour. I am interviewing in 9 minutes
[23:21] <sinzui> do you want my rs to remove something I despise?
[23:21] <thumper> heh
[23:21] <thumper> project groups?
[23:22] <sinzui> They are on my list of papercuts
[23:22] <sinzui> thumper, do you want the go work or go away
[23:24] <thumper> sinzui: I don't get that last statement
[23:26] <sinzui> Users use PGs to show affiliations. Drivers use PGs to manage permissions and see milestones. The problem is that since the user controls the linking, the driver has not control over what he controls. eg any user can add their project to Launchpad-Project, forcing the drivers to see unwanted milestones
[23:32] <thumper> sinzui: I want project groups to be entirely different
[23:33] <thumper> sinzui: I want project groups to be controlled by the project group creator, easier to create, and they can add whatever projects they want to their group
[23:33] <thumper> sinzui: it allows a user to provide a unified view over projects they care about
[23:33] <sinzui> ...and create series and milestones that propogate to the projects!
[23:33] <thumper> in some cases maybe
[23:34] <thumper> it it just happens to be a group of projects that I depend on, then no
[23:34] <sinzui> landscape and OEM have asked for that several times. So has mrevel
[23:34] <sinzui> l
[23:34] <wgrant> If you need your series and milestones to propogate, you probably want components rather than projects.
[23:35] <sinzui> Yes I want components. I am not sure I will ever get them
[23:35] <thumper> I want components
[23:35] <thumper> I'd like launchpad to use launchpad like we tell others to
[23:35] <wgrant> YES.
[23:35] <thumper> if we can't dogfood it for what we want,
[23:35] <thumper> we should fix it
[23:36] <sinzui> lets start a list of papercuts and pimp that for the epix
[23:40] <thumper> sinzui: sounds good to me
[23:43] <thumper> sinzui: I'm heading out briefly(ish) for a coffee, but I'd love a talk - I have some sekrit plans