[08:12] HELLO DESKTOP [08:14] desrt: stop screaming, please [08:15] * ogra_ rasies his ear trumpet [08:15] what did he say ? [08:18] * davmor2 shakes walking stick and shout "Get Off My LAWN!" [08:18] * ogra_ rasies his ear trumpet again [08:18] what did he say ? [08:20] ogra_: peskie kids muttering to themselves all the time [08:24] http://pad.ubuntu.com/N4aJtsz2V7 [08:26] Good morning [08:26] yay, back to a mostly sane TZ :) [08:26] pitti: hi! How's London? [08:28] hey pitti, had a good trip back to Europe? [08:28] https://docs.google.com/a/canonical.com/spreadsheets/d/18-enLvvZScjIlkO0VhHG-CuTdK1TUzlNMkdevDqjMB0/edit#gid=0 [08:28] http://pad.ubuntu.com/N4aJtsz2V7 [08:29] seb128: yeah, I did; all went well [08:29] seb128: how was your's? you took the Eurostar, I figurE? [08:29] larsu: you aren't here? for the desktop sprint? [08:29] pitti: trolling isn't what it used to be in here ;) [08:29] pitti, mine was fine, yeah, eurostar, easy and uneventful ;-) [08:29] pitti: I am. It's nice and sunny! [08:30] larsu: sorry, too distracted/jetlagged for subtle jokes still [08:31] pitti: no worries. Enjoy your day! Hope to see you some time this week [08:31] larsu: yeah, I was too tired last night for getting to your hotel again, sorry [08:31] but definitively let's meet for some dinner/beer this week! [08:32] :) [08:36] pitti, we are going to be in the office on thursday [08:36] seb128: oh, nice! for the release party? [08:36] + beer event in the evening for vivid [08:36] yeah [08:36] rockin' [08:37] :-) === hyperair is now known as m4 === m4 is now known as hyperair [10:36] http://www.whizzy.org/2015/02/snapping-mosquitto-mqtt-broker/ [10:36] robert_ancell, ^^ [10:44] for more than an hour I'm watching systemd hanging at "A start job is running for Enable s...y formats" during boot... anyone with an idea how to fix/workaround that... or what's causing this? [10:44] oh... and "Login Service" and "Network Manager" failed to start too === MacSlow is now known as MacSlow|lunch [11:55] willcooke, inspiron 11 3000 series 3138 [11:55] reg model P19T === alan_g is now known as alan_g|lunch === MacSlow|lunch is now known as MacSlow === alan_g|lunch is now known as alan_g [13:15] Does someone in the desktop team want to take bug 1445358? Looks like it should be considered release critical to me. [13:15] bug 1445358 in pulseaudio (Ubuntu) "Pulseaudio fails to run when system's language is in certain non-English locales" [High,Confirmed] https://launchpad.net/bugs/1445358 [13:15] heh, so shiny [13:21] rbasak, can you mention it to diwic on #ubuntu-devel? [13:21] rbasak, TheMuso is off this week and we don't have anybody else in desktop working on pulseaudio [13:26] mterry: hey, we are on a sprint, and we were wondering if you got Unity8 running with a snappy image. In that case, what device and procedure have we used? We tried an usb stick, but of course, most of the time, no wifi drivers on the image… [13:27] didrocks, no, I didn't get as far as the entirety of unity8 [13:27] mterry: did you get at least a working machine, which isn't a qemu device? [13:28] like a laptop, with ssh, and such… [13:28] didrocks, yeah, I had a snappy laptop that I ssh'd into [13:28] didrocks, did you see my email about deb2snap? I got mir and xmir clients working [13:28] didrocks, but didn't try something as complicated as unity8 [13:28] I bet there are some gaps (dbus? gsettings caches?) [13:28] mterry: that's exactly why I'm pinging you :) But to get mir, we need to use some intel based laptop [13:29] and we still have to find one with a running ubuntu core image :p [13:29] (where we have network) [13:29] didrocks, I had an nvidia laptop [13:29] oh, mir supports it? [13:29] didrocks, it seems to... using nouveau [13:29] interesting [13:29] do you know about qemu? Maybe that can help iterating [13:29] (well kvm) [13:31] didrocks, yeah well I tried kvm [13:31] didrocks, but mir needs real hardware (/dev/dri/card0 access) [13:31] mterry: ok, thanks for confirming. We'll try harder to find a way to at least have ubuntu core running on real hardware first :) [13:31] mterry, doesnt the new HW access framework now make that available ? [13:32] ogra_, I don't know... It didn't last time I tested on vivid [13:32] yeah, that was still in development back then ... i think it is there now [13:35] * didrocks tries in a vm [13:48] mterry: ogra_: doesn't work FYI [13:48] (with kvm) [15:35] kenvandine, hey, are you going somewhere with those flacky tests? I see you had a bunch of (re)tries, do you understand what's going on? [15:36] yes [15:36] i'm trying to beat on it :) [15:36] ignore the noise for now [15:36] i have it reliably passing the specific tests we had failing before [15:36] but i introduced another problem... [15:36] fixing [15:39] kenvandine, what was the issue? [15:40] I'm curious, your mp doesn't explain much [15:40] * larsu waves to seb128 [15:40] * seb128 waves back [15:42] seb128, we think it was because we had a test calling another test [15:43] so i refactored it to not need to call that test to setup [15:43] but i also added more wait_selects, etc [15:43] but that wasn't the real issue [15:49] kenvandine, it doesn't make sense, for me the failing test was clearly a tap from autopilot not resulting as a tap/generated as an onClicked [15:49] seb128, elopio says that's because it wasn't ready when the click was sent [15:49] so it didn't get it [15:49] the qml side wasn't ready [15:50] kenvandine, what does that mean? [15:50] the UI is on screen [15:50] and built [15:50] and the listitem is stable/displayed and get something happening to it [15:50] that's what he said :) [15:50] e.g you get the selected bg color changed [15:50] but was it ready before autopilot called the click [15:50] like if it was dnd-ed rather than clicked [15:51] define ready [15:51] it was on screen stable displayed without changing for some seconds [15:51] and the click made the line bg change [15:51] the text just doesn't turn red and the onClicked handler doesn't trigger [15:51] yeah [15:51] to me it looks like the tap is doing a dnd [15:51] hmm [15:51] or something which is not seen as a proper click [15:52] so removing the call to the other test, did seem to fix that [15:52] i got a couple passes in a row in CI [15:52] doesn't make sense to me [15:52] not sure how that caused it though [15:53] but he said we should never call a test from a test [15:53] but [15:53] why not? [15:53] that code as been there for 15 months [15:53] and it wasn't a problem before [15:53] but it is the only place we do that [15:53] he didn't really say why [15:53] he also said we shouldn't have more than one assertion [15:54] which i don't really agree with [15:54] he said each test should only have one assert [15:55] k, weird [15:55] but for a UI test, i think it makes sense to assert on multiple things [15:55] I think people just suggest workarounds there [15:55] what i'm still puzzled by is why this just started happening [15:55] but let's see if that's enough to make things stable, I somewhat doubt it === alan_g is now known as alan_g|EOD === rickspencer3_ is now known as rickspencer3 === qengho is now known as CardinalFang === CardinalFang is now known as qengho