[08:36] <JamesTait> Happy Friday, everyone! :-D
[08:54] <rye> joshuahoover: 2012-07-05 23:38:20,081 - twisted - ERROR - u'Host name does not match certificate. Expected fs-2.ubuntuone.com but got fs-2.one.ubuntu.com.'
[08:57] <rye> ping webops, what do we need to do for _https._tcp.ubuntuone.com to start returning fs-{1,2}.one.ubuntu.com instead of fs-{1,2}.ubuntuone.com
[08:57] <gnuoy> rye, let me take a look
[08:58] <rye> because we have certificates issued for fs-1.one.ubuntu.com and for at least one person that became an issue
[08:58] <rye> and I am not sure why it became an issue for only one person
[08:59] <rye> gnuoy: continuing in #u1-internal
[09:42] <mandel_> morning from florence :)
[12:28] <mandel_> ralsina, hello!
[12:29] <mandel_> ralsina, where is everyone!
[12:52] <mandel_> rye, where is everyone?
[12:54] <rye> mandel_: python conference I suppose
[12:54] <mandel_> rye, ah..
[13:06] <dobey> yeah
[13:06] <dobey> pycamp is this weekend
[13:06] <dobey> so all of .ar is there
[13:34] <mandel_> dobey, buh… I had no idea, now my code is blocked :(
[13:34] <dobey> what code?
[13:35] <mandel_> dobey, fsevents, not to worry, I wanted to bully gatox :)
[13:35]  * mandel_ restarts
[14:20] <dobey> hmm
[14:20] <dobey> mandel: are you working today?
[14:21] <dobey> no
[14:22] <dobey> i guess it's just thisfred, mmcc and me today
[14:22] <thisfred> sí
[14:24] <dobey> so let's not do the standup
[14:30] <dobey> brb
[14:34] <mmcc> sure, no problem here
[14:35] <mmcc> btw, no one gave me any reviews on Tuesday so if you have any today let me know, I'll make time today
[15:01] <thisfred> TODO: immanentize the eschaton
[15:45] <dobey> lunch time, bbiab
[16:53] <dobey> thisfred: https://code.launchpad.net/~dobey/ubuntuone-infrastructure/u1db-nightlies/+merge/113757
[16:54] <thisfred> on it
[16:54] <thisfred> +1
[16:54] <dobey> thisfred, mmcc: can you also review https://code.launchpad.net/~dobey/ubuntuone-installer/core-cache/+merge/113286 ?
[16:54] <mmcc> sure
[17:00] <mmcc> dobey, re: your question about the sso branch you just looked at - that branch is old, I should cancel the merge proposal or something. what's the right way to do it? I was going to ask after finishing the replacement- wasn't expecting anyone to look at it in the meantime, sorry
[17:00] <dobey> mmcc: ah ok, i'll reject it then
[17:00] <mmcc> ok
[17:01] <dobey> mmcc: what about https://code.launchpad.net/~mikemc/ubuntu-sso-client/fix-1018924-use-buildout-python/+merge/112813 ?
[17:02] <mmcc> also reject
[17:02] <dobey> done
[17:12] <mmcc> hm, looking at this windows installer xml file, most paths have the backslash, but the path to the uninstaller has a forward-slash. anyone know if that's a problem?
[17:13] <dobey> don't know
[17:13] <dobey> i guess the bitrock thing does the right thing regardless
[17:13] <dobey> as it does support osx, windows, and linux
[17:14] <mmcc> guess so. sometimes forward slashes work on windows too, but I don't understand when or why
[17:14] <dobey> i think they work more often since MS made explore.exe and iexplore.exe be the same thing
[17:15] <dobey> because the internet made people more used to typing / than \
[17:15] <mmcc> hooray internet
[17:17] <mmcc> Todays moment of zen: the title of the menu item I select to create a new console window with Console2 on windows is "Main console module"
[17:29] <mmcc> dobey, trying to run dirspec tests on windows, I don't have trial.py or pyflakes.py where it wants-- I guess run-tests.bat isn't expecting  a buildout-style python, huh.
[17:32] <mmcc> …and running trial.py from within the egg in the buildout just silently does nothing :(
[17:33] <dobey> mmcc: hrmm, i don't know. afaik the tests work. i guess we'll have to wait to ask alecu about that? :-/
[17:35] <mmcc> I guess so. care to review my exe_finding branch as code only? tests work on mac and linux
[17:37] <dobey> well, it /is/ my review day :)
[17:37] <mmcc> oh right. just a sec, new bug -> commit --fixes -> dance dance
[17:41] <mmcc> and if you're light on reviews to do, I still have no idea what I'm doing with bugs affecting projects/ distributions, etc. Be nice to stop offloading that onto you.
[17:45] <mmcc> dobey, ok - https://code.launchpad.net/~mikemc/dirspec/add-exefind/+merge/113782
[17:45] <dobey> oh jenkins, thou art a harsh mistress
[17:46]  * dobey bets precise is offline again
[17:46] <dobey> and it is
[17:46] <mmcc> that branch is definitely still open to suggestions as to API design or where it should go or whatever…
[17:48] <dobey> right. i haven't even had time to read through your e-mail yet either
[17:48] <dobey> wasn't expecting a branch already :)
[17:50] <mmcc> heh. well, I've had all that code arranged in different places a couple of times already so it was quick to put it together
[17:51] <mmcc> and wherever it ends up, it'll be most of that same code in one place so I figured it couldn't hurt to get a jump
[18:18] <dobey> mmcc: can you run the fsevents daemon tests?
[18:24] <mmcc> dobey, I should be able to now, yes
[18:24] <dobey> mmcc: maybe you could review https://code.launchpad.net/~mandel/ubuntuone-fsevents-daemon/allow_user_readd/+merge/113749 also then?
[18:24] <mmcc> ok, sure.
[18:29] <mmcc> lunchtime...
[19:30] <dobey> hmm
[20:05] <dobey> brb
[20:11] <dobey> meh, having vaguely supported hardware is no fun :-/
[20:15] <mmcc> dobey, just saw your comments, thanks - do we ever run frozen on platforms other than mac & win?
[20:16] <dobey> i don't know what frozen means exactly there
[20:16] <mmcc> well, specifically it means sys.frozen is not None, but I also just need to know more about how we package things on linux
[20:16] <dobey> but even the "unfrozen" path there seems to be the wrong way to do things on Linux
[20:17] <dobey> AttributeError: 'module' object has no attribute 'frozen'
[20:17] <dobey> is what Linux does apparently :)
[20:18] <mmcc> ah, yes - that's true. that path is only correct for development, and since this code is really only for mac and win at first, I was using 'frozen' == packaged
[20:18] <mmcc> that's why it uses getattr , I guess
[20:23] <dobey> mmcc: also, what came of the discussion i see between you and alecu about the variable definitions for each project?
[20:23] <dobey> do you need to change the code as a result of that?
[20:24] <mmcc> no, this code incorporates that feedback.
[20:25] <dobey> ah ok
[20:26] <dobey> i think it does need to be a bit more generic as well
[20:26] <mmcc> I had originally made the "app_names" dict a parameter to the get_cmdline function, but since it'll be specific to each project (i.e., only u1client needs to know the sub-app-name for syncdaemon), it seemed ok to just get it from the module argument
[20:26] <dobey> but i'm not exactly sure where/what that means at the moment
[20:26] <mmcc> sorry, what needs to be more generic? the app names part of it or the whole thing
[20:26] <dobey> the whole thing
[20:27] <mmcc> definitely if we want it to be able to be used for installed linux , it needs a change… although we always use dbus for launching stuff there, afaict
[20:28] <dobey> i don't think we launch the sso gui with dbus. those have to be installed in the specific sso libexecdir
[20:28] <mmcc> hmm
[20:35] <mmcc> yes, that's right - that uses ubuntu_sso.utils.get_bin_dir, which looks in path-to-__file__/../../bin/ and falls back to using  ubuntu_sso.constants.BIN_DIR, which is generated at install time… so on linux we 'compile in' the right path
[20:36] <mmcc> that fall-back to a constants.py is one way to add packaged linux support to my get_cmdline branch, but it only works for sso. (although only sso would use it)
[20:39] <mmcc> that is, only sso would use get_cmdline on linux
[20:42] <dobey> right
[20:42] <dobey> well
[20:42] <dobey> maybe we could provide a primary_path kwarg or something, that if is not None, look there first
[20:42] <dobey> not sure exactly yet
[20:43] <dobey> need to think about it a bit
[20:43] <dobey> brb, need to reboot again
[20:47] <dobey> hmm, secondary still all red, and lightdm was behaving weird
[20:49] <dobey> and gl stuff is still crashy :-/
[20:49] <dobey> doh
[21:00] <mmcc> arg, mandel's integration tests aren't working for me...
[21:01] <mmcc> they have to be run as root and running with sudo changes your environment. the weird buildout $u1trial stuff is hitting us here
[21:02] <mmcc> and Xcode says "Unexpected internal inconcistency: No recorder for buildTask: "
[21:02] <mmcc> hey great speling, XCode
[21:03] <dobey> oh fun
[21:03] <dobey> heh
[21:03] <dobey> need to reboot again
[21:03] <dobey> whee
[21:09] <dobey> back
[21:11] <dobey> hmm
[21:13] <mmcc> I note that the OCMock license on the website still hasn't changed
[21:14] <dobey> :-/
[21:14] <dobey> should we ping about it?
[21:27] <mmcc> I guess so, I'll ping him now.
[21:27]  * mmcc pokes his head up out of trying to run mandel's Xcode tests
[21:30] <dobey> heh
[21:48] <dobey> ok, have a good weekend mmcc. i'm out of here :)
[21:49] <mmcc> alright, you too dobey.