=== axw_ is now known as axw [06:09] trave: o/ did you figure out your vagrant issue? [14:57] marcoceppi, lazyPower: ping [15:00] jose: lots of US folks on holiday today, anything I can help with? [15:00] tvansteenburgh: there is a meeting scheduled for... now [15:00] I guess I am going to have to cancel that? [15:01] jose: dunno, guess you could wait a bit and see if anyone shows up [15:02] Hi guys... i've sent an email to the list, regarding amulet failing tests due to an ssh REMOTE HOST IDENTIFICATION ChANGED. Any idea how can i solve this? [15:02] despite i destroy the environment many times per day, it gets happening over and over again (in local environment) [15:02] tvansteenburgh: I'll just cancel, it's an on-air meeting and we're already delayed [15:03] thanks for the heads up [15:03] nicopace: clear the offending entry from your known_hosts file [15:03] jose: np [15:03] nicopace: the error output gives a command to fix it [15:03] just run that and you'll be fine [15:04] tvansteenburgh, jose: sure, i've been there in the past (with ssh), but the problem is that it gets repeating again and again because of the tests [15:04] nicopace: which provider? [15:04] i may remove the know_hosts file each time i run a test [15:04] local [15:04] jose: local [15:04] eh, I don't use local. won't be of much help. [15:04] that was happening with HPCloud a couple months ago [15:04] I filed a bug and now it's gone, but don't know how it'll work for Local [15:05] jose: there is an open bug now: https://bugs.launchpad.net/juju/+bug/802117 [15:05] Bug #802117: juju ssh/scp commands cause spurious key errors

[15:05] jose: I just got the calendar invite [15:05] it is regarding juju [15:05] and as amulet uses juju well... [15:05] jose: we really suck at planning these meetings [15:05] marcoceppi: it's just 3 of us atm, we won't meet quorum [15:05] meh [15:05] happens [15:06] nicopace: yeah, I've escalated the bug to core team [15:06] nicopace: there is a work around. On my phone atm but will reply tonight to you email about it [15:06] thanks marcoceppi, but i need to find an alternative fix, as it is blocking me cc:arosales [15:07] And will update the big with work around [15:07] Bug* [15:07] thanks marcoceppi, i'll apreciate that :) [15:07] * marcoceppi bounces [15:07] marcoceppi: hope it is not happening in automate testing also === kadams54 is now known as kadams54-away [15:17] What is the current process to get a fresh automated testing run? I got my box turned black end of last week, which makes me think there is still some manual process involved. [15:17] stub: in your own box, bundletester [15:18] jose: Nah, I want the bot for a second opinion [15:18] stub: you can click the box again [15:19] (if you are in ~charmers?) [15:19] I can fire them for you if you want [15:19] tvansteenburgh: yeah, did that. It changed from red to black end of last week [15:19] then it's on the wait queue [15:19] from it it's all automated (afaik) [15:19] jose: oh, ok. Didn't think it would be that long. [15:20] sometimes it takes long, I know :( [15:20] stub: i inadvertently dequeued a bunch, do it again [15:20] there is nothing queued right now [15:21] * stub wrestles with his browser [15:22] tvansteenburgh: done. Black box spun around, remained black. Queued now? [15:22] stub: http://juju-ci.vapour.ws:8080/job/charm-bundle-test/11027/console [15:22] in progress [15:22] ta [15:26] tvansteenburgh: Is this running on a precise host by any chance? I think I'm getting python3.3 instead of python3.4 [15:27] stub: no it's trusty [15:28] DEBUG:runner:tests/test_helpers.py:1275:28: F821 undefined name 'ProcessLookupError' [15:28] Unless I'm getting python2 somehow? [15:28] jenkins@charm-bundle-slave:~$ python3 [15:28] Python 3.4.0 (default, Apr 11 2014, 13:05:11) [15:30] stub: check shebangs in tests i guess? [15:32] yeah, its getting a python2 interpreter. print(*args, **kw) is a syntax error. [15:36] Which I think means /usr/bin/nosetests is being invoked, rather than using the one in my venv. My makefile sets PATH, but that must be getting overridden. Or nosetests not installed in that venv at all. [15:41] yer, my Makefile runs virtualenv fine, but the following pip commands are not run so none of the pip dependencies get installed. Hmm... === kadams54 is now known as kadams54-away === ejat- is now known as ejat === kadams54-away is now known as kadams54 [16:11] oh, make -s so maybe they are? [16:13] stub: doh, yeah === kadams54 is now known as kadams54-away === roadmr is now known as roadmr_afk === kadams54-away is now known as kadams54 [16:50] So 'pip install flake8' won't install flake8 if the libraries are already there, and doesn't create the 'flake8' script in the venv's bin. So the system one runs, which is python2 and fails on python3 syntax. === kadams54 is now known as kadams54-away === roadmr_afk is now known as roadmr === kadams54 is now known as kadams54-away [19:39] lazyPower: are you arround? [19:40] i'm starting the zookeeper tests, and wanted to know if you had any input on that, as it seems you have been the last one pushing on that charm :) [19:57] nicopace: o/ [19:58] which zookeeper charm, the hdp-zookeeper? === roadmr is now known as roadmr_afk === kadams54 is now known as kadams54-away === kadams54 is now known as kadams54-away === roadmr_afk is now known as roadmr