=== chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === vrruiz_ is now known as rvr [10:16] Hi, anyone know how to crate a none hybrid ,img === ubot5` is now known as ubot5 [10:18] ill try a difrent chanel === _salem is now known as salem_ === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === alexabreu is now known as alex-abreu === jpds is now known as Guest86457 [17:20] rhuddie: ping. Can you help me with the test output discussion that you had with jibel? [17:21] or jibel if you are around. [17:22] elopio, sure [17:22] rhuddie: I see that the default autopilot output will print all he's requesting. [17:22] do we need to do something other than printing the pyunit results? [17:24] elopio, well what was discussed was, run-rate, pass-rate, duration of each test, and a list of failures (if i remember correctly) [17:24] rhuddie: well, runtime is the only thing that this doesn't give us. [17:25] http://paste.ubuntu.com/9444367/ [17:26] elopio, I was also about to say a summary of the error, so that looks good [17:28] elopio, does this tell you how many tests are expected to run? [17:28] rhuddie: ok, so I'll do that task as it's easy from where I am. [17:28] rhuddie: well, it tells us the number of tests it ran. [17:29] which should be the same. There is no reason for it to miss any. [17:31] elopio, what was discussed was that if a test is aborted or a test suite terminates early, the run rate might be reported higher, because the remaining tests were not executed [17:32] elopio, can we get that number from autopilot list command? [17:33] rhuddie: we can. But I think it would be better to use the autopilot timeout command so all tests that run longer are reported as failures. [17:35] elopio, well, main thing is to make sure the results are accurate :) === roadmr is now known as roadmr_afk [18:46] never before have i had such a difficult time using xvfb :-/ === roadmr_afk is now known as roadmr [20:11] Letozaf_, doug5 buonasera [20:11] balloons, buonasera :) [20:12] hey Letozaf_ :) [20:13] elfy, hey, howzit ? [20:15] Letozaf_, balloons hello guys! how is it going? [20:16] good thanks Letozaf_ - you? [20:16] hi doug5 [20:16] elfy, fine thanks [20:16] doug5, hi [20:16] * balloons waves to all [20:17] Just writing up some things and prepping for the demo tomorrow [20:19] * elfy is glad he realised that was all voodoery before watching :p [20:41] balloons, is mihir generally available here? [20:42] doug5, generally yes, but he isn't around for a couple weeks at the moment [20:43] I *think* anyway [20:43] balloons, ok, I will try to contact him via email [20:43] doug5, ahh yes, starting dec 7th he was/is away [20:49] balloons, I found something strange: when we open the new event page with the bottom edge button sometimes the default calendar is not "personal" [20:49] balloons, same is not true if we open the new event page with the standard button [20:49] doug5, hmm.. I filed a bug about that once, and I believe it's now in the default test to prevent a regression [20:50] it may have regressed. Let me find the bug [20:59] doug5, found it: https://bugs.launchpad.net/ubuntu-calendar-app/+bug/1347841 [20:59] Launchpad bug 1347841 in Ubuntu Calendar App "Creating a new event preselects a random calendar" [Medium,Fix released] [21:00] elopio: well, i got the autopilot tests running inside autopkgtest finally, without any weird display or unity8-dash issues [21:00] elopio: unfortunately, qemu crashes while the tests are running [21:01] dobey, what exactly are you up to? [21:01] dobey: that sucks. [21:01] balloons: trying to get autopilot tests running for click scope, in the autopkgtest infrastructure [21:02] dobey: is it qemu's fault, or something weird we are doing in the test? [21:02] hmm.. so you need unity8 then, and you are running it in qemu. Are you trying it in the emulator then I take it? could you run it in a desktop env instead? [21:03] no, this is in qemu under xvfb [21:03] elopio: i think just qemu being unstable afaict [21:04] http://pastebin.ubuntu.com/9446765/ [21:04] dobey, ahh, then the opposite question holds (running it in the emulator) :-) [21:05] balloons: no, this is autopkgtest :) [21:05] balloons, ok, I raised the point with mihir [21:05] if it was in the emulator, i wouldn't need xvfb [21:05] as mir would be running there [21:06] dobey, right.. I'm curious though why you can't use it. I get it's autopkgtest, but hmm.. [21:06] balloons: why i can't use what? [21:07] in theory you just want to test the click scope right? And for the moment actually that test makes the most sense in the phone env, or is this not the case? [21:08] balloons: no. i don't think it makes sense to use the phone emulator here [21:08] it'd be too slow, and would result in a vm running inside a vm [21:10] ok, just sanity checking [21:10] dobey: balloons: better on a vm during MPs. But the test class is ready to run in a phone or emulator. [21:10] and qemu would probably still segfault [21:11] balloons: we are forced to use qemu, right? [21:11] I would prefer an lxc. [21:12] "forced"? autopkgtest does support containers [21:12] i would prefer not an lxc [21:13] problem with lxc is that $HOME in the container is $HOME outside the container [21:13] good evening qua [21:13] at least, without plenty of modification to make it otherwise [21:13] which causes problems [21:13] +lity. [21:13] good evening dkessel [21:14] would be nice to run it in sbuild or something [21:14] dobey, lxc is the recommended way to run.. That said, schroot and chroots also work, but aren't nearly as nice [21:15] let me think.. there's also ssh, heh. You could use a cloud vm or something [21:16] balloons: well, the way i need to run it is the same way that it is run in jenkins [21:17] and jenkins is running with qemu [21:20] tell jenkins to let you in on the secret! [21:21] there's no secret [21:21] the autopilot tests are not currently being run [21:21] so qemu doesn't crash in that case [21:22] running the autopilot tests results in a crash [21:22] ahh.. well then [21:26] balloons, I think that when this problem is gone, we should be ok [21:29] doug5, that's awesome. you have an MP? [21:29] balloons, should I create one even if the tests are failing? [21:30] doug5, sure if you code you believe is complete [21:30] you could even add a comment about it on the MP. I like to see MP's [21:30] personally speaking :-) [21:30] ok, I'll do [21:37] elopio: might not be able to land this branch any time soon, thanks to qemu crashing, but might be able to use a little bit of what i learned from making those tests run, to perhapse get the autopilot tests more stable when being run directly for an MP in jenkins [21:37] assuming those aren't also going through qemu [21:37] dobey: yes, they are not qemu. but they have ubuntu-desktop installed which makes some other problems. [21:38] dobey: that plan sounds good, of course. [21:38] dobey: and maybe you can also do autopkg for ubuntucredentials, which should be more simple. [21:39] ubuntuone-credentials I mean. [21:40] dobey: we can tell jfunk to put pressure on somebody to get qemu fixed. Can you report a bug? and do you know who can work on that? [21:40] no idea [21:41] i don't see a crash report [21:41] and i'm not sure if it's really qemu at fault or if it's autopilot or something else [21:41] dobey: we can ask pitti for some help when he's back. [21:42] sure [21:42] i don't know that ubuntuone-credentials will be any simpler [21:42] well, i mean it's not running unity8-dash of course [21:43] but still [21:48] balloons, https://code.launchpad.net/~acerisara/ubuntu-calendar-app/bottom-edge-autopilot/+merge/244229 [21:51] doug5, thanks. Is this intended to merge into mihir's branch [21:51] ? [21:52] balloons, I don't know which is the standard normal flow in this case...I just took mihir's stuff and added the changes on top of that [21:53] doug5, right it gets a bit more murky. I would change the proposal to merge into lp:~mihirsoni/ubuntu-calendar-app/bottomEdge so it will cleanly show what you added [21:53] make sense? [21:54] how can I do that from launchpad? resubmit proposal? [21:55] doug5, yes === roadmr is now known as roadmr_afk [21:58] doug5, hmm, maybe not much cleaner eh? :-) [21:58] doug5, ohh I see, mihir's branch is behind trunk [21:58] makes sense [21:58] balloons, :) I guess it's because I took master, then I merged [21:58] yep yep [21:59] yes exactly, his branch is not updated [21:59] k, so I think it's all in his hands for the moment [22:00] good work === salem_ is now known as _salem [22:04] balloons, yes, I sent him an email; I think the problem is that one because I saw the issue by running manually the application, will see [22:04] balloons, I have to go now, I'll come back as soon as I have news [22:04] ahh right, understandable [22:04] balloons, cheers [22:04] doug5, cheers and thanks! [22:12] elopio: hrmm, i think the ubuntuone-credentials ap tests need updated [22:13] dobey: last time I ran them was like a month ago. And I haven't seen any changes since then. [22:13] but I could be wrong. [22:14] huh [22:14] autopilot.exceptions.StateNotFoundError: Object not found with name 'NewAccount'. [22:14] well it is apparently not finding the thing [22:15] wonder why that is [22:15] the component is definitely there [22:15] dobey: is that running from the cmake task? === roadmr_afk is now known as roadmr [22:15] elopio: no, this is running from autopkgtest inside qemu :) [22:21] file:///usr/lib/Main.qml: File not found [22:21] hmm [22:21] that's probably not good :) [22:22] why is it trying to load that file [22:26] i see [22:35] it's using the wrong include path for qmlscene [22:38] hello? :) Anyone there? [22:42] I would like to help you with writing autopilot tests? Since I'm still learning Python can anyone direct me how I can be usefull to you guys? :-) [22:43] nikjov, hello [22:43] hi [22:44] how can I help? [22:44] nikjov, might I recommend having a look at http://developer.ubuntu.com/api/devel/ubuntu-14.10/python/autopilot/tutorial/tutorial.html to get familiar with autopilot a bit. Then I can walk you through running some tests, and finally writing some tests for the core apps :-) [22:45] Great! :) Getting right on it, thanks. :) [22:46] nikjov, in case I'm not around and you want to keep going, there is a walkthrough on the wiki that should help: https://wiki.ubuntu.com/Touch/CoreApps/Testing/ContributeAutopilotTestcase. If you've never used launchpad before see this page to get setup: https://wiki.ubuntu.com/QATeam/ContributingTestcases/Launchpad [22:47] ok that is great, I actually already have Launchpad set up I made an account but If I get stuck anywhere I will get by wiki to see if I can un-stuck myself :-) [22:54] nikjov, dinner time for me, I'll be back later. Feel free to leave questions if you get them. [22:58] Hello there [22:59] howdy Retrotime1987 [22:59] Great. Don't You worry. I would probably have questions. It's the timezone issue but I'm going through the documentation. [22:59] Hey can any of you be of assistance in the adding of my account to who we are and hardware sections i cannot figure it out [23:00] Retrotime1987, what's not working for you? [23:00] * balloons really does have to run [23:00] bbl [23:00] Adding myself to the who we are section [23:01] I can't seem to find where to add myself [23:01] Retrotime1987, ahh, if you login to the wiki, you simply edit the page [23:02] Thanks [23:06] ok where is the edit button? [23:12] Retrotime1987, top left. make sure you hard refresh (ctrl+shift+R in firefox for example) the page after logging in if it isn't visible [23:16] welcome spineau [23:17] thx Retrotime1987 :) [23:17] no problem [23:20] I seem to be having a problem with virtualbox and testdrive [23:23] Virtualbox keeps complaining about this: ernel driver not installed (rc=-1908) The VirtualBox Linux kernel driver (vboxdrv) is either not loaded or there is a permission problem with /dev/vboxdrv. Please reinstall the kernel module by executing '/etc/init.d/vboxdrv setup' as root. If it is available in your distribution, you should install the DKMS package first. This package keeps track of Linux kernel changes and rec [23:23] y [23:23] Retrotime1987: that's probably better suited for a different channel, but have you run that command? [23:23] (with sudo if you aren't logged in as root) [23:26] ok got it fixed [23:28] thinking about it now i'll just test it on a spare old laptop [23:41] should this beta say ubuntu gnome? [23:41] and something about 14.10