=== veebers_ is now known as veebers === robru-sick is now known as robru-sick-eod === _salem is now known as salem_ === salem_ is now known as _salem [03:26] hi, I heard unity 8 is faster. will that be faster than lxde? thanks. === Zhenech_ is now known as Zhenech [06:59] Saviq: hey, mind doing a special unity8 delivery containing the cleanup socket fix? === popey_ is now known as popey [09:16] Saviq or tsdgeos, ping [09:16] davidcalle: hi [09:17] tsdgeos, hey ! Do you have an ETA for header attributes and emblem? [09:17] you're going to need Saviq for that sorry :D [09:17] i mean for the ETA [09:19] tsdgeos, thanks :) [09:25] didrocks, will do [09:26] Saviq: thanks! please stick that in your landing :) [09:36] hi, how can i update unity to the most recent release? [09:37] vini, how do you usually update software in ubuntu? [09:37] hi Saviq [09:38] i mean beta and early versions, not final [09:38] vini, you'll have to update to the ubuntu development release (trusty) [09:38] ok [09:39] need fresh install? [09:39] vini, no, http://askubuntu.com/questions/12909/how-do-i-upgrade-to-the-development-release-aka-ubuntu1 [09:40] mzanetti, are you in a state to review / test https://code.launchpad.net/~saviq/unity8/drop-stale-socket/+merge/196917 ? [09:40] thanks Saviq [09:40] didrocks, icanhassilo for row 28? [09:41] didrocks, actually, since you wrote I can get the other things... [09:41] let me [09:42] didrocks, or actually... you didn't comment on row 17? does the comment for 16 apply? === rachelliu_ is now known as rachelliu [09:49] Saviq, there is an option to update only unity? not the whole distro [09:50] vini, you could try, but that's not guaranteed to work - unity has quite some dependencies that would have to be pulled in anyway [09:51] ok thank you [09:51] vini, https://wiki.debian.org/AptPreferences#Pinning has some info on how to do that [09:52] vini, or https://help.ubuntu.com/community/PinningHowto for more ubuntu-style docs [09:53] 2nd link is more preffered, thanks [10:10] Saviq: so, yeah, if you want to stich more in, it's now or never! :) [10:10] didrocks, !!! [10:10] * Saviq does [10:10] Saviq: line 17 can land, after discussing with xnox, it's not activated until line 16 lands [10:10] didrocks, ah, so let me stick all that into line 16 [10:11] erm [10:11] 17 [10:11] Saviq: sure [10:11] Saviq: please do take care of the landing as per my email [10:11] when testing [10:11] didrocks, will do [10:11] we really need to be extra cautious [10:11] thanks :) [10:15] didrocks, ok, row 17 ready [10:15] sil2100: I'm assigning it ^ [10:15] Saviq: "exterminating" the other line? [10:16] didrocks, already did [10:16] or not [10:16] now dii [10:16] did [10:16] Saviq: mind being a little bit more descriptive than "Unity8 update"? [10:16] didrocks, right [10:17] didrocks, done [10:18] Saviq: silo 008 is yours [10:18] didrocks, \o/ [10:18] :) [10:18] didrocks, no it isn't :/ [10:18] #REF? [10:19] oh now it updated [10:19] hum? [10:19] ignore [10:19] ah, be patient :) [10:19] I even refreshed! [10:19] GAS are taking some times to recompute ;) [10:25] guys [10:25] anyone not using qt5.2 [10:25] can you run unity8 on the desktop [10:25] and tell me if you get [10:25] file:///home/tsdgeos_work/phablet/unity8/unity8/qml/Shell.qml:761:5: QML Binding: Binding loop detected for property "target" [10:25] on startup? [10:26] tsdgeos, upgrading [10:26] Saviq: upgrading? i said "not" using 5.2 [10:26] tsdgeos, ah [10:27] ok, ignore, that's not the binding that is breaking the phone [10:28] tsdgeos, yes, it's there [10:28] ok, tx [10:28] huh, btw [10:28] target: i18n [10:29] binding loop! [10:29] ? [10:29] i guess i18n is not defined by whoever defines it until a later stage [10:29] maybe should be in Component.onCompleted [10:29] as a Qt.binding() thing [10:29] tsdgeos, maybe [10:30] dednick, you around? [10:30] Saviq: yarp [10:31] dednick, you in a state you could test https://code.launchpad.net/~saviq/unity8/drop-stale-socket/+merge/196917 ? [10:31] i.e. do you have a device running unity8? ;) [10:31] Saviq: think i can manage it [10:31] dednick, would be enough to copy /usr/share/upstart/sessions/unity8.conf into ~/.config/upstart/ [10:32] dednick, applying the patch [10:32] dednick, stop unity8; start unity8; pkill -9 unity8 [10:32] if it recovers, we're good [10:49] Saviq: mako test failures normal on jenkins? [10:49] dednick, yeah, random? [10:50] dednick, we've a Qt crasher that got worse with android 4.4.2 - we're running away from it to Qt 5.2... but then there's two other crashers tsdgeos is looking into [10:50] Saviq: ok. [10:50] dednick, if it says that "process not found" or so - that's it [10:51] Saviq: mmm. doesnt say that. but unity8 did crash [10:51] you can quickly get the crash [10:51] and check that frame0 address finish with d8f8 [10:51] it's the signature :) [10:52] dednick, which job on jenkins? [10:52] Saviq: https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-mako/5574/console [10:52] dednick, "NoSuchProcess: no process found with pid 6296" [10:53] ah. looking for wrong text. [10:54] Saviq: approved [10:54] dednick, thanks! [10:55] yup, d8f8 [11:10] hi, there is a plugin for file manger to "open terminal" in specific location? [11:13] didrocks, hum, silo failed on armhf https://launchpadlibrarian.net/167968584/buildlog_ubuntu-trusty-armhf.unity8_7.84%2B14.04.20140228-0ubuntu1_FAILEDTOBUILD.txt.gz any idea? [11:13] shall I just kick it again? [11:13] Saviq: ah, I guess it's time to chdist [11:14] let me check if I can get that easy [11:28] * didrocks wonders why it's trying to fetch i386 on his new created chdist armhf [11:28] Saviq: meanwhile, retry to see if that was transient [11:28] I keep looking [11:29] didrocks, can you kick "retry" on https://launchpad.net/~ci-train-ppa-service/+archive/landing-008/+build/5645715 [11:29] didrocks, I don't have the rights [11:29] oh right [11:30] done [11:30] thanks [11:33] no daniel? [11:37] tsdgeos, yeah, national holiday for him [11:38] the whole week, according to cadmin ;D [11:38] ah [11:38] carnival [11:38] makes sense [11:43] Saviq: was an archive skew [11:43] didrocks, whatever that means, but transient, yeah :) [11:43] yeah [11:56] xnox, hey, I tried to x-build unity8 with qt5.2 from canonical-qt5-edgers/qt5-beta2, got http://pastebin.ubuntu.com/7009939/ - any ideas? [11:58] Saviq: that spec file is gone in 5.2, no? [11:58] (or the other way around, this _is_ the new spec) [12:01] xnox, hmm wait, I might've confused my chroots [12:01] Saviq: then, you know you can just relaunch "build" with "watch only" to only watch the ppa and not rebuilding [12:02] didrocks, ah, doing [12:05] xnox, hmm seems legit http://pastebin.ubuntu.com/7009976/ [12:05] * Saviq is worried they broke x-compilation again [12:09] xnox, so... seems set(_qt5_corelib_extra_includes "${_qt5Core_install_prefix}/lib/arm-linux-gnueabihf/qt5//mkspecs/linux-g++") is a new addition to the cmake module [12:10] xnox, looks like qt5-base-dev would have to start depending on qt5-qmake to get those... or we need to install the mkspecs in some other way [12:10] Mirv, ↑ === _salem is now known as salem_ [12:12] Saviq: Mirv: oh, that looks wonderful! if mkspecs are not in multiarch locations it's execellent! and knowing _qt5Core_install_prefix also helps a lot! [12:13] Saviq: so we need more deps. and qt5-qmake should be multiarch-same, i think it is/was. if not that's easy to fix. [12:13] xnox, yup, that's what I thought [12:14] I have no idea of the CMake speficics you're talking about, but mkspecs moved from /usr/share/mkspecs to /usr/lib/{multiarch}/qt5/mkspecs in 5.2 and some sources might need changes accordingly (known so far and with patches maliit-framework, pyqt5) [12:14] xnox, fyi https://qt.gitorious.org/qt/qtbase/commit/6fa9394c [12:14] larsu: https://code.launchpad.net/~aacid/qmenumodel/fix_uninit/+merge/208783 [12:15] Mirv: i'm so happy for that move! =) [12:15] * xnox schedules some time to enable cross-compilation with qmake [12:15] xnox: ah, you meant s/not/now/, so it's good news like I though :) [12:15] +t [12:16] uh oh [12:16] xnox, that's worse http://pastebin.ubuntu.com/7010023/ :/ [12:16] Saviq: Mirv: correct =))) bad typpo. It's all good news all around! =) [12:16] Saviq: that binary is probably somewhere in tools [12:17] qtbase5-dev-tools [12:17] tsdgeos: the patch looks good to me, but I didn't write that code... [12:17] * larsu approves anyway [12:17] Mirv: Saviq: when cross-compiling, that's the wrong path. But that's easy to fix. [12:17] larsu: who do you want me to nag? [12:18] larsu: ah thanks :) [12:18] Saviq: i'll fix things up for 5.2. Let me just finish up with android i686 cross-toolchain. [12:18] xnox, ok thanks [12:18] tsdgeos: it's trivial enough I think [12:18] tsdgeos: would be dednick or renato I guess === MacSlow is now known as MacSlow|lunch [12:20] oki [12:21] tsdgeos: um, dont think thats right... [12:21] dednick: why not? [12:21] tsdgeos: oh, it's in constructor. no, it's fine then [12:22] thought the signal wasn't getting emitted [12:22] it won't anyway [12:22] you can't emit signals from a constructor [12:22] well you can [12:22] but nobody will be listening [12:22] tsdgeos: yeah :) i didn't realise the code was in the constructor [12:22] since noone has had time to connect to you yet [12:23] oki :) [12:28] Mirv, did you notice on qt5.2 that you have to press twice to wake the screen when it's timed out? [12:29] Mirv, ignore [12:29] Mirv, I've powerd-cli in a terminal ;) [12:31] Saviq: hmm, no I have not. maybe I've just ignored it but I don't remember having to press twice. [12:31] Mirv, ignore me, I had it locked with powerd-cli [12:31] Saviq: the good thing was that the waking started to work with 5.2.1. maybe it was even some qtbase 5.2.0 bug or such, the original one there were. [12:31] ok :) [12:31] Mirv, :) [12:34] didrocks, can you reconfigure silo 008 please? we've a fix for the unity8 5.2 crash in store [12:34] tsdgeos FTW! [12:34] tsdgeos, looking good, halfway through ap run and /var/crash clean! [12:35] \o/ [12:35] one down! [12:35] the other should not be that hard [12:35] hopefully [12:35] seems one of those "dude you removed the stuff under my feet" [12:36] Saviq: do you have bluetooth in your desktop and qt 5.2? does the crash happen in there? [12:36] tsdgeos, upgrading to 5.2 [12:36] maybe using dednick's branch of desktop indicators === alan_g is now known as alan_g|lunch [12:37] Saviq: waaaaaaaaaai [12:37] Saviq: new scopes won't work [12:37] in case you need that [12:37] tsdgeos, nw [12:37] though you can always rollback [12:37] tsdgeos, should work without dednick's branch I think [12:38] tsdgeos, and why wouldn't new-scopes work, btw? [12:38] tsdgeos, they're rebuilt for 5.2 no? [12:38] sure, it's just the fact i don't have bluetooth [12:38] but if you can get it to crash [12:38] tsdgeos, yup, gimme 10 [12:38] i'll find a bluetooth dongle [12:38] Saviq: were are they rebuild for 5.2? [12:38] tsdgeos, in the qt5.2 repo, no? [12:38] it's regular unity8 i think [12:38] no new-scopes [12:39] wait [12:39] tsdgeos, https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta2/+packages?field.name_filter=scopes&field.status_filter=published&field.series_filter= [12:39] new scopes may be [12:39] what is not there [12:39] is the toolkit stuff [12:39] tsdgeos, ah [12:39] Cannot assign to non-existent property "nativeOrientationAngle" [12:39] tsdgeos, ah right, so it's not new-scopes, but mwc [12:39] anyway [12:39] it's fine [12:40] running valgrind on the phone is a party of warnings :D [12:40] ;0 [12:41] sil2100, can you reconfigure silo 008 for me please? [12:41] i haven't even started [12:41] and i have this [12:41] http://paste.ubuntu.com/7010119/ [12:42] yay [12:43] the dlopen ones may be "ok" [12:43] Saviq: should we delay the landing for that? maybe 2 landings will be better? [12:43] but there's some uninit memory that definitely may be worth fixing [12:43] Saviq: as I want to kick an image with the unlock screen fix asap [12:43] didrocks, ok [12:43] thanks :) [12:44] didrocks, will take me twice the time to land, but OHKAY :P [12:45] Saviq: well, I'm sure you will have also good fixes on Monday :) [12:45] didrocks, Monday!? [12:46] and it's taking double time for us as well, do not forget :) [12:46] didrocks, we want the two crashers in today! ;P [12:46] ah, well, if you are quick enough, that's doable :) [12:46] (fixes for them, even) [12:46] Saviq: sure [12:46] sil2100, nope [12:46] sil2100, didrocks overridden my request [12:46] Saviq: un-sure [12:47] sil2100, thanks [12:47] ;) [12:51] tsdgeos, +1! [12:51] \o/ [12:52] tsdgeos, you get sixpack of beer ;D [12:53] but how do i fix the other bug if i'm drunk? [12:53] luch time now, after taht i really need a way to create an indicator that mimics the bluetooth behaviour for the desktop [12:54] dednick: if you have any pointers in that ↑↑↑ i'll be terribly happy [13:02] tsdgeos, yeah, crashes on desktop [13:02] bregma, do we have any other indicator than bluetooth that hides itself? [13:03] didrocks, 212 doesn't boot, known? seems the container doesn't start [13:05] Saviq: it does boot for popey though? [13:05] * didrocks flashes [13:05] maybe it failed to flash correctly, /me tries again [13:05] looks I'm missing system.img [13:06] hard to boot without it :p [13:06] !wfm [13:06] Common Sense: Just because you can, does not mean you should (and especially recommend to others). Think before you do. "Works for me" does not mean it is ok. The latest version of everything is not always useful if you aim for stability. Please see http://geekosophical.net/random/worksforme/ [13:06] ☻ [13:06] ahah :) [13:06] ☺ [13:07] popey, obviously you spend more time in the sun... [13:07] slacker! === MacSlow|lunch is now known as MacSlow [13:08] :D [13:08] Saviq: You know popey is in England right? There is no sun :P [13:09] There might not be in the smelly north where you live. [13:09] davmor2, right, just tar-like rain? ;) [13:09] But down here in the tropical south it's lovely ☻ [13:09] http://popey.com/webcam/ ignore this [13:09] Saviq: beautiful starting system.img here :) [13:09] didrocks, k, let's see [13:09] popey, ouch ;D [13:10] popey: haha yeah there are no massive storms hitting the uk again at all :P [13:10] sunny 13° here :) [13:10] in FEBRUARY, no less [13:12] larsu, maybe you know - any indicator other than bluetooth that would hide itself? or maybe a way to trick the bluetooth one to think there is a bt device when there isn't? [13:14] Saviq: sound has a gsettings key to show/hide itself, but that seems to be broken right now [13:14] uh [13:14] Saviq: bluetooth has a visible key as well [13:14] but you're not seeing it because you don't have any devices... [13:14] Saviq: keyboard works [13:14] gsettings set com.canonical.indicator.keyboard visible false [13:15] * Saviq tries [13:17] didrocks, yeah, booted now, not sure what happened there :/ [13:17] Saviq: do you have anything in /cache/recovery? [13:17] there are the logs of different flashing, would be interesting to see if something failed [13:18] didrocks, only one install log it seems, let me pastebin them all [13:19] larsu, hmm, keyboard not showing up in unity8 under the desktop profile :/ [13:19] Saviq, larsu: battery and datetime also has a show key [13:19] have [13:19] Saviq: what seb128 says :) [13:20] larsu, btw, I was pondering doing https://bugs.launchpad.net/ubuntu/+source/unity-control-center/+bug/829648 next week [13:20] Ubuntu bug 829648 in unity-control-center (Ubuntu) "No easy way to turn the sound menu on/off" [Undecided,In progress] [13:21] seb128: apparently that gsettings key doesn't work right now. I can fix that though, should be quick [13:21] larsu, there was design disagreement over those UIs back then but we ended up adding the option for the other indicators, so for consistency we should probably do it for sound as well [13:22] I agree [13:22] thanks [13:23] ok so... it must be that the state is changing while it's being queried or something [13:23] as no disappearing indicator causes the crash - not unless the change comes from the indicator UI itself [13:25] http://paste.ubuntu.com/7010320/ [13:25] http://paste.ubuntu.com/7010321/ [13:25] http://paste.ubuntu.com/7010324/ [13:25] http://paste.ubuntu.com/7010325/ [13:25] http://paste.ubuntu.com/7010325/ [13:25] http://paste.ubuntu.com/7010329/ [13:26] didrocks, ↑ all the logs from /cache/recovery, I'm not sure what I'm looking for but nothing stands out [13:26] Saviq: 2 links are the same, wanted? [13:27] didrocks, dupe, sorry [13:27] didrocks, 5 logs is all I have [13:27] ok ;) [13:27] * didrocks looks [13:29] Saviq, do you know if zanetti's right-edge branch also works on tablet mode? [13:30] (with side stage) [13:30] dandrader, the branch should work, yes [13:30] dandrader, not right edge, of course - it "switches" to old behaviour on tablet [13:30] -"" [13:30] dandrader, I haven't tried though - it is meant to work [13:32] /food, too === alan_g|lunch is now known as alan_g [13:44] Saviq: it does? awesome! [13:44] * tsdgeos starts looking for a bluetooth dongle [13:45] tsdgeos, data point: it has something to do with you interacting with the indicator panel when it's going away [13:45] Saviq: what do you mean? [13:45] tsdgeos, or maybe the checkbox still wants to animate but is destroyed [13:45] Saviq: i think it's the checkbox thing [13:45] tsdgeos, I mean that if you hide the indicator from outside [13:46] larsu: you do indicator stuff, right? [13:46] tsdgeos, it will not crash [13:46] hey folks, any sbuild user ever get this fail on creating an armhf chroot: http://pastebin.ubuntu.com/7010432/ [13:47] i wonder what's easier, finding someone close to me with bluetooth adaptor or coding a similar indicator :D [13:47] i guess i can just fork indicator-bluetooth and fake the backend part... [13:47] * tsdgeos tries [13:48] tsdgeos: what are you trying to do? [13:48] and why don't you have a laptop with bluetooth :P [13:48] larsu: i'm trying to get an indicator with a checkbox that when i click on it the indicator disappears from the indicators, like the bluetooth one does [13:48] larsu: that is indeed a good question :D [13:49] tsdgeos: datetime and power have such a check (see the discussion we jut had a bit earlier in this channel) [13:50] larsu: where is that checkbox? i can't see it [13:50] tsdgeos: settings / time and date / clock / "Show a clock in the menu bar" [13:51] larsu, he wants a control in the indicator menu to disable the indicator I think [13:51] yes [13:51] larsu, they seem to think their bug happen when you turn off the indicator from its own UI [13:52] what bluetooth does [13:52] wtf? [13:52] larsu, you can turn bluetooth off from the bluetooth indicator [13:52] bluetooth does this? [13:52] yes [13:52] when I turn it off, I still see it [13:52] it has a on/off toggle [13:52] oh, this is phone [13:52] right [13:52] unity8 issue [13:53] let's call it QML issue for now ;-) [13:53] it should be easy to patch e.g power to add a checkbox that toggle the gsettings show key [13:53] I can hardly believe that the reason for it crashing is that the toggle is in the indicator ui [13:55] larsu: well, you can find it hard to belive, but it is what it is [13:56] you can trust my informed opinion or not [13:56] that's up to you [13:56] i'm asking you not for help in fixing my bug but in creating an indicator that behaves like the indicator one [13:56] if you can do that, cool, if not i'll do it myself [13:58] tsdgeos: fair enough. Quickest way is proabably to add a fake device in the bluetooth indicator [13:59] yep, i'm trying that [13:59] tsdgeos: something like this should be enough: http://paste.debian.net/84568/ [14:00] did you try asking pitti if he has some mock you could use to fake a bluetooth device? [14:00] tsdgeos: oh wait, which profile are you using? [14:00] the phone one I guess [14:01] not sure tbh [14:01] larsu, seb128, we have a patch to use the desktop profile optionally [14:01] tsdgeos, yeah, phone is default [14:01] problem is, it bails out earlier [14:01] the desktop profile doesn't hide the indicator when bluetooth is off iirc [14:01] since it can't find anything in bluez [14:01] need to find out where that's happening [14:01] easier option "go out, buy a bluetooth dongle" [14:03] seb128: on the phone it seems to be always visible when bluetooth is enabled [14:03] ah, I guess the toggle disables bluetooth instead of just hiding the indicator? [14:04] right [14:04] the indicator hides by a side effect of bluetooth turning off [14:06] didrocks, so, only failure I got was the StateNotFound one, is that ok? [14:07] Saviq: can you check on that one with elopio? He told StateNotFound is normally because of badly written tests [14:07] seems to be working [14:07] and can help fixing them :) [14:07] didrocks, ah no [14:07] didrocks, I had my bluetooth disabled [14:08] oh ;) [14:08] that can explain [14:10] didrocks, yeah, it's fine [14:10] great ;) [14:10] sil2100: can you test unity8 as well? Maybe that can still make it for that image ^ [14:13] crash \o/ [14:14] larsu: thanks for the pointers :) [14:14] * tsdgeos starts debugging in a proper environment (i.e. !phone) [14:22] Saviq, if you have a sec - https://code.launchpad.net/~mhr3/unity-scopes-shell/category-order/+merge/208806 [14:23] didrocks: once I finish frying I'll move on to unity8 ;) [14:23] sil2100: taking a break? lucky you :p [14:25] didrocks, kick the button on silo 008 please! [14:26] Saviq: we are doing a second check to protect any bad things happening as long as we are in this flaky world, which is what sil2100 is going to do [14:27] didrocks, sure [14:33] @unity: standup [14:39] Mirv, I could try and help from unity PoV for the landing, unless you need someone from the landing team === salem_ is now known as _salem === _salem is now known as salem_ [14:45] dednick: easy one https://code.launchpad.net/~aacid/unity8/importcomponents/+merge/208822 [14:45] Saviq: ↑↑↑ [14:46] tsdgeos, yeah, saw that yesterday [14:46] tsdgeos, seems like 5.2 cleans the scope somewhat? [14:47] tsdgeos, that doesn't fix the crash, does it? [14:47]