[01:25] <axw> axw> wallyworld: my modem keeps resetting itself, may not be able to make standup
[01:25] <wallyworld> ok
[02:03] <waigani> axw, wallyworld sorry guys, connection problems
[02:03] <axw> waigani: nps, we just called it a wrap
[02:04] <waigani> ah cool
[10:48] <rogpeppe> mgz: standup?
[10:51] <mgz> rog, whoops., thanks
[10:52] <mgz> ...and g+ decides it's time to play up
[11:26] <rogpeppe1> my net connection is v dodgy currently
[12:45] <rogpeppe1> mgz: what's the standard procedure for changing the deps on the bot?
[12:45] <rogpeppe1> mgz: just ssh in and run godeps -u?
[12:46] <rogpeppe1> dimitern: do you know?
[12:47] <mgz> rogpeppe1: pretty much
[12:47] <mgz> I can do it if you like
[12:47] <rogpeppe1> mgz: thanks
[12:47] <mgz> only funny if you want to change the older branches as well
[12:48] <rogpeppe1> mgz: i think it must be using the old version of mgo, which is probably why tests haven't been failing all the time there
[12:48] <mgz> (I manually pulled and built last time, probably using godeps wouldmake more sense)
[12:48] <mgz> rogpeppe1: that's likely
[12:48] <mgz> I should have been the one to bump and didn't
[12:51]  * dimitern is away for 2h
[15:40] <dstroppa> JoshStrobl: have a look at the debug-hooks command, it really helped me
[15:40] <dstroppa> https://juju.ubuntu.com/docs/authors-hook-debug.html
[15:42] <lazyPower> Greetings, I'm looking over charm-tools and I found a rare condition that the template generator breaks on adding charm tests. I'm trying to get my environment working so i can patch the bug, however following along in Hacking.txt results in the following output when running make, or make test: http://paste.ubuntu.com/6909757/
[15:42] <lazyPower> I get this in other projects too, is this something I"ve got fouled up in my environment, or am I missing something obvious to seasoned python hackers?
[15:45] <natefinch> marcoceppi: ^^ any ideas?   It looks like it has the right permissions to run
[15:46] <natefinch> lazyPower: I'm guessing it's probably not running that python, but some other python... but I'm not familiar enough with the charm tools to know what might cause that
[15:46] <marcoceppi> lazyPower: one sec
[15:52] <lazyPower> marcoceppi, natefinch,  I foudn the issue. It was on a NFS hosted symlinked directory. Moving it onto my primary for $HOME seems to have worked.
[15:52] <lazyPower> so, it was my env. ta
[16:12] <TheMue> rogpeppe1: I proposed https://codereview.appspot.com/58510045/ again. I removed the passing of the location. So far I have no better idea than passing at least juju-home and env name.
[16:13] <TheMue> rogpeppe1: Or storing it during bootstrap in the Environment instance in state.
[16:13] <rogpeppe1> TheMue: ok, thanks. will take a look soon.
[16:13] <rogpeppe1> TheMue: did you consider my suggestion of a method on environs.EnvironProvider?
[16:13] <TheMue> rogpeppe1: At least CLI now works for local provider.
[16:14] <TheMue> rogpeppe1: The method would/could be needed to access. But the API server runs as root but the log location is determined by the bootstrap user.
[16:15] <rogpeppe1> TheMue: the API server is started by the bootstrap user, isn't it?
[16:15] <TheMue> rogpeppe1: During this time rsyslogd is configured and that's it.
[16:15] <TheMue> rogpeppe1: After sudo as root. ;)
[16:16] <TheMue> rogpeppe1: Simply try a bootstrap and do a ps -ef. It runs as root. *sigh*
[16:17] <rogpeppe1> TheMue: i'm not sure how the fact that it's sudo'd precludes it from knowing stuff about the user that started it. but don't worry about that for now.
[16:18] <TheMue> rogpeppe1: Maybe I'm only missing the path to the information I need. Any good idea is welcome.
[16:20] <TheMue> rogpeppe1: The next bigger step will be the unit tests. Need to export internal parts again to mock real behavior.
[17:52] <hazmat> this sound familiar to anyone "0 17:51:45 ERROR juju runner.go:220 worker: exited "authenticationworker": adding current Juju keys to ssh authorised keys: generating key fingerprint: invalid authorized_key"
[17:52] <hazmat> my log keeps filling up with those
[17:54] <hazmat> is there a reason why juju grabs all public keys it can find?
[17:56] <mgz> not to me I'm afraid
[17:56] <natefinch> hazmat: just in case? :)   I don't actually know.
[18:41] <rogpeppe1> hazmat: axw__ is the one to talk to about those errors. i think i might know what's going on, but i have to stop for the day, sorry
[18:41] <rogpeppe1> g'night all
[18:41] <natefinch> night rog
[21:12] <thumper> wallyworld: hey there, just letting you know I'm not working tody
[22:23] <wallyworld> thumper: yeah, figured as much
[22:23] <wallyworld> good sprint hopefully
[22:24] <thumper> wallyworld: yeah pretty good
[22:24] <thumper> wallyworld: oh fyi, I'm not working today (and axw_)
[22:24] <wallyworld> thumper: for when you are back tomorrow, i have an upgrade frameworkin review
[22:24] <thumper> awesome
[22:24] <thumper> looking forward
[22:24] <thumper> to it
[22:24] <wallyworld> i'll add the first 1.18 plugin today
[22:41]  * thumper signs off