=== evanwang is now known as evanwang-afk === evanwang-afk is now known as evanwang [08:25] sil2100: ping [08:25] morphis: pong [08:25] sil2100: can you have a quick look on https://code.launchpad.net/~morphis/libhybris/+git/libhybris-ubuntu/+merge/288238 ? [08:25] morphis: looking [08:25] sil2100: thanks [08:48] morphis: ok, commented [08:48] sil2100: thanks [08:50] sil2100: fixed (but the diff didn't refreshed yet) [09:05] sil2100: can you approve if you are fine now? [09:26] davmor2: ping [09:26] davmor2: Hello! [09:26] jin: hello [09:26] davmor2: I am checking the dashboard: https://trello.com/b/AE3swczu/qa-testing-requests-for-questions-ping-ubuntu-qa-on-ubuntu-ci-eng [09:26] jin: it will be hit today 3rd in the queue [09:27] davmor2: oh yes thanks for your answer [09:27] jin: I figured that might be the question :) [09:27] davmor2: hahahaha you smart, mate [09:28] okay, let's wait for the result... :) [09:30] jin: check it tomorrow morning your time, I assume you are based in Asia right? It should be covered by then but if the ticket is still in the queue and red label on it it means there was an issue and it got blocked waiting on a response from you. If you fill that in, in the comments on the ticket we can look at it when we start. [09:30] jin: other wise the silo and ticket will both be marked as passed [10:15] morphis: btw.! Did you by any chance pick up LP: #1517597 ? [10:15] Launchpad bug 1517597 in android (Ubuntu) "[regression] mouse pointer support on emulator is broken" [Critical,In progress] https://launchpad.net/bugs/1517597 [10:16] sil2100: I did [10:16] sil2100: https://requests.ci-train.ubuntu.com/#/ticket/1081 [10:19] morphis: excellent, I suppose in that case emulator would only use the mouse cursor as a touch device, right? [10:20] sil2100: not sure, anpok_ should know [10:20] just releasing his fix :-) [10:20] Ok ;) [10:22] sil2100: yes [10:23] anpok_, sil2100: can you help me verifying the emulator is really fixed with that silo? [10:24] yip [10:26] anpok_: great! [11:24] brb.. just recreating an u-e from rc-proposed which should have the right mir verison... [11:49] morphis: not xenial build? [11:49] oh still building [12:01] anpok_: not for the android package yet [12:07] jibel, hey, re: [1], this has had some unfortunate delays due to infra, could you maybe heighten the priority on that one? [1] https://trello.com/c/sxgt75nk/2855-1052-ubuntu-landing-006-ubuntu-system-settings-jgdx === morphis_ is now known as morphis [12:28] anpok_: sil2100 did android get updated with the input fix for the emulator? [12:28] or is there a new bug? [12:32] kgunn: landing is in progress [12:33] kgunn: https://requests.ci-train.ubuntu.com/#/ticket/1081 [12:40] kgunn: it's in a silo [12:47] sil2100: btw. packaging of the android package is now at https://code.launchpad.net/~phablet-team/android/+git/android [12:51] Oh [12:55] * sil2100 notes that down === timp is now known as t1mp [13:02] sil2100: feel a lot better with that === alan_g is now known as alan_g|lunch [13:54] davmor2, are you testing silo 50 on a bootstrapped/wiped device? did you get through the wizard after having installed it? [13:55] Saviq: not had a look at that yet but will shortly [13:56] davmor2, I think you can stop looking now [13:56] davmor2, it's in a crash loop [13:56] zsombi, just drop ~/.config/ubuntu-system-settings/wizard-* and restart unity8 [13:57] Saviq: \o/ [13:57] zsombi, the wizard sends unity8 into a crash loop with silo 50 [13:57] http://pastebin.ubuntu.com/15320881/ [13:58] davmor2, can I bribe you to put https://trello.com/c/TpHXuLrK/2843-996-ubuntu-landing-057-unity8-unity-scope-click-dobey at the top of the list then? [13:59] unless both of you can confirm this is something wrong with my device [13:59] Saviq: no but you might be able to bribe rvr I'll be moving onto bluetooth next to upset morphis cause you know how I like to do that :) [13:59] Saviq: if the QQMlEngine is null, then we have a bigger problem... and it doesn't help if I have to do this with unity... what can happen so that the engine got null? is there a new view created meantime and the items were not cleaned somehow? [14:00] davmor2: you know I am not "responsible" for bluetooth anymore, do you? ;-D [14:00] morphis, that doesn't help with davmor2 [14:00] morphis: your names on the silo it's your fault :P [14:01] Saviq: i think it needs rebuilt first [14:01] Saviq: does that happen on the normal first time use? [14:01] davmor2: awe's name is on the silo too :-) [14:01] zsombi, that is "a normal first time use" [14:02] almost [14:02] morphis: yeah but I blame him for all the networking :D [14:02] I will wipe after installing silo 50 now [14:02] Saviq: well, almost is not teh first time use then :) [14:02] +normal [14:02] zsombi, still unacceptable [14:02] Saviq: is it almost because you "simulate" it to be? :) [14:02] hmm [14:03] zsombi, I don't simulate anything, I just tell the shell to not skip the wizard [14:03] Saviq: it woudl help me a lot if I could see which style loading fails... right now I can only see ... [14:03] zsombi, I'm flashing the phone fresh now and will wipe after installing the silo [14:04] zsombi, that's as many symbols as apport would give me, would need to not optimize probably [14:05] dobey, I don't think it needs a rebuild, the previous silo with unity8 is looking like a fail right now [14:05] Saviq: the one that landed last thurs/friday? [14:06] dobey, oh well, why's it in QA ready queue, then? [14:07] Saviq: it seems to me that it tries to create a style on an item that has no QML Engine assigned! [14:07] Saviq: because the one that landed on friday wasn't landed yet, and i have no control over when things actually land or end up in the qa ready queue === alan_g|lunch is now known as alan_g [14:07] Saviq: do you happen to create any StyledItem derived component in C++? [14:07] but this is to be expected and normal with silos, so eh [14:07] zsombi, not that I know of [14:08] and i wanted to ping pstolowski to see when the filters silo was going to land [14:09] sil2100: hey, can you give artmello permissions to create silos please? [14:09] urh, there's like 5 ongoing silos with unity8 now [14:09] Saviq: because that is what I actually see in line 9 of http://pastebin.ubuntu.com/15320863/: engine = 0x0 [14:09] yeah [14:09] zsombi, sure, that I get [14:09] Saviq: and that is due to QQmlEngine* engine = qmlEngine(parent); [14:10] Saviq: so the parent, in our case a StyledItem has no QML Engine assigned, so the style cannot be created [14:10] Saviq: and of course, the Q_ASSERT doesn't guard it well :/ [14:11] Saviq: ok, I'll provide a fix for that, but that would not make you rife better, as the style won't be loaded... there has to be something in your code that tryes to set a style for a non-QML created styled item [14:11] Saviq, we could combine social-actions with some other unity8 silo once we're happy with the impl [14:11] zsombi, we are creating views dynamically per screen [14:11] Saviq: do those views have QML Engine set? [14:12] zsombi, yes they do [14:12] ... [14:12] zsombi, in theory, on a phone, there should only ever be one screen on startup so that shouldn't matter [14:12] but more than that, how's the wizard affect that [14:12] zsombi, I'd say it must be a race, the wizard is smaller than the shell [14:13] zsombi, https://bazaar.launchpad.net/~unity-team/unity8/trunk/view/head:/src/ShellApplication.cpp#L76 [14:13] Saviq: it coudl be that the engine was always null there, and now that the default theme and other singletons are lkinked to a given QML Engine instance, it revealed that... [14:14] zsombi, FWIW it's real easy to reproduce... just start unity8 with that file gone (I'm confirming on a fresh-bootstrapped mako now) [14:15] and it came up in autopilot tests for unity8, for that matter, but no one paid attention it seems [14:15] Saviq: well, it woudl be nicer to see it repro on a smaller code so I can test it easier... [14:16] Saviq: actually we saw it and it ended up that it was something else.... kalikiana or t1mp may know more about that, about the unity8 startup we faced last time [14:16] zsombi, I'm afraid I can't spend time on that, I need to push at least one unity8 silo out today, so I'm jumping at 64 now [14:16] Saviq: but at that time the logs were not showing us any toolkit pain... [14:17] zsombi, what do you mean "no toolkit pain", that it crashed in a test? [14:17] Saviq: there were no signs of crashing in the toolkit [14:17] that's what I meant toolki tpain [14:17] zsombi, so it's fine if it crashes somewhere else, even though it didn't crash before? [14:18] and how is the traceback I showed you not "crashing in the toolkit"? [14:18] Saviq: as said, if you'd read the lines, it ended up that the crash was somewhere else... [14:18] it's a direct retrace of a crash from an autopiltot test [14:18] Saviq: hey, calm down, as said, we had not seen this so far [14:19] zsombi, do you agree that silo 50 is causing it to crash, and that it can't go in as is? [14:19] Saviq: have I said ever that it's not our fault? I'm sorry if you got it like that [14:19] regardless where the real issue is, the net result is that you can't use your phone when you first boot it up [14:20] All I'm trying is to get the cause of the crash [14:20] mzanetti: Hi. How can I install new sticker sets? Currently, I can only see the default set. Through telegram.me doesn't work. [14:21] rvr, when you receive a new sticker, you can install it from there [14:21] rvr, I didn't implement uri handling (yet) [14:21] mzanetti: How? I long tapped and nothing happened. [14:21] Saviq, I saw the unity8 crash loop too this morning with silo 50 after reenabling the wizard. I wasn't sure it was it and reflashed [14:21] rvr, swipe [14:22] rvr, or right click if you have a mouse connected ;) [14:23] mzanetti: I see, with the info icon [14:24] yes [14:25] Saviq: one more question, if you don't mind, can you point me to the code where the first time use is handled? [14:26] mzanetti: Is there any way to remove the sticker sets? (yet) [14:26] rvr, nope :/ [14:28] rvr, you are probably wondering why I "fixed" the avatar thing the way I did (disabling it). Here's the complete reasoning behind it: https://code.launchpad.net/~mzanetti/telegram-app/avatar-spinner/+merge/287868 [14:28] zsombi, https://bazaar.launchpad.net/~unity-team/unity8/trunk/view/head:/qml/Shell.qml#L581 it really just fills the shell and loads the wizard conditionally on those files [14:30] mzanetti: At least the avatar is still editable using the top bar button [14:30] rvr, yeah... that's always been the only way. the spinner is, and always was, only for downloading the high-res version of an avatar. [14:31] mzanetti: Ahh, I see [14:31] rvr, I totally see how you got tricked into thinking it would be for changing it... [14:31] :D [14:31] it is not ideal from a UX point of view [14:43] zsombi, so yeah, I just wiped mako, installed silo 50 on it, removed the files (could not wipe because I'd have lost adb/ssh) [14:43] Saviq: and you see the same... [14:43] zsombi, yes [14:44] Saviq: I'll provide a log to see where the hell are we failing to load the style. [14:44] Saviq: and fix the crash [14:44] zsombi, flahs a device, install silo 50, then reset the device from u-s-s [14:44] it'll trigger the crash [14:44] jibel: thx [14:45] just the end user side test case [14:47] jibel: zsombi Saviq just confirmed the bootloop and failed the silo [14:47] davmor2: I'm on it [14:47] zsombi, I'm pushing for a different unity8 silo then, should be tested in a few h [14:53] jibel, what do you do with https://trello.com/c/TpHXuLrK/2843-996-ubuntu-landing-057-unity8-unity-scope-click-dobey ? it's not actually ready to test since it needs a rebuild at least [14:54] it's rebuilding now [14:56] Saviq, if it's rebuilding, this card can be deleted and a new one will be created once it's marked ready again [14:57] usually we check the status of the request before starting the verification [14:58] something really needs to be done about that, so that cards are kept in sync with biletto status [14:58] or we need a QA view on biletto, and drop trello [14:59] agreed, and auto-ack should prevent conflicts or something [15:00] dobey, how long do you need after it rebuilds? [15:02] dobey, dropping trello would be my preferred option. The trello board exists just because the original spreadsheet didn't offer much flexibility. Now we just have to find the time to do it. [15:03] Saviq: how long do unity8 autopkgtests take? [15:04] dobey, ~2h [15:04] on a good day [15:04] today looks like a good day http://autopkgtest.ubuntu.com/running.shtml [15:05] mzanetti: telegram update approved. Please, it's important to keep the test plan up to date. Current one is quite complete. [15:07] dobey, oh actually just over an hour these days it looks like http://autopkgtest.ubuntu.com/packages/u/unity8/xenial/amd64/ [15:07] well, as long there isn't a qt, perl, pythom, etc upload eating up the queue i guess [15:09] rvr, can you please tell that to jin? He's the maintainer, I'm just a drive-by contributor. [15:09] rvr, and thanks for testing/approving it! [15:09] mzanetti: Sure! [15:10] * rvr is even happier now he can use stickers [15:10] hehe+ [15:10] many people will be [15:10] popey, seems it is approved :) can you publish please? [15:10] stickkaaas! [15:11] mzanetti: yay [15:12] mzanetti: rvr where's the click that was tested? [15:14] popey: https://requests.ci-train.ubuntu.com/#/ticket/1076 [15:19] mzanetti: changelog? * Stickasss, * Fix reconnect issue" ? [15:21] popey: Replies. [15:21] rvr: plus what I said? [15:21] popey, reconnect issue has been fixed in the one before [15:21] popey, yeah, replies [15:21] so "just" replies and stickers? [15:23] yeah... well, let me check the changelog to be sure [15:24] popey, yeah... mostly that's it... I've also upgraded the items to use the SDK ListItem instead of the old ListItemWithActions from ucs... [15:25] but that should ideally be invisible to the end user [15:26] ok [15:26] thanks [15:26] bfiller: done [15:26] sil2100: thanks [15:48] Saviq: we have the wizard... so we may be lucky to push one more build [16:11] kenvandine: ping [16:11] Oops, actually... [16:11] rvr pong [16:11] jgdx: ping [16:11] question about silo 6? [16:11] kenvandine: Yes [16:12] you should see nothing changed :) [16:12] kenvandine: No visible changes? [16:12] right [16:12] just switched from a private API to qtsystems [16:12] I have attached the bt and usb kb and mouse, and they work fine [16:12] and the affected plugin is actually disabled [16:12] kenvandine: Ahhh [16:12] :) [16:12] Ok [16:13] should be enabled soon, but that'll be a different landing [16:13] this mostly just removes a pile of code :) [16:13] Yeah, I saw it [16:13] Ok, then it's fine [16:13] Approving it [16:13] great, thx [16:14] when we enable the plugin, it'll actually work on the desktop too :) [16:14] it didn't before [16:17] zsombi, note https://requests.ci-train.ubuntu.com/#/silo/057 is already lander-acked, should land in QA's queue within an hour or two [16:43] Saviq: ok, so that means we'd need to pull unity8 again to silo50 before we go again with it [16:43] trainguards: could someone kick off a rebuild of silo 69 just for arm64 on vivid please? [16:43] Elleo: on it [16:43] sil2100: thanks :) [16:44] Elleo: ongoing, yw o/ [16:48] zsombi, no, you just need to wait for it, you can't land without unity8 because our autopkgtests will fail [16:50] mardy: Hi. I'm testing silo 36... I think there is a typo in the dialog's title, "Duplicate account". Shouldn't it be "Duplicated account"? [16:54] rvr, disagree, duplicate is better, it's an adjective instead of... whatever the other word is :) [16:54] yeah, and really the silo is quite needed so we can then build and land the other one [16:54] but not a native speaker so ;) [16:55] rvr: ^^ 77 is queued right after 36, and 77 has the critical fix i'd like to get in for ota-10; just that it needs that dependency update that is in 36 [16:55] ie https://bugs.launchpad.net/ubuntu-system-settings-online-accounts/+bug/1544063 [16:55] Launchpad bug 1544063 in webapps-sprint "UI freezes when trying to log in from youtube scope" [Critical,In progress] [16:56] and a couple of nice design updates [16:58] robru, hey, is it expected that it takes almost an hour for britney to start testing silos? https://requests.ci-train.ubuntu.com/#/silo/57 [16:58] Saviq: Ok :) [16:59] dbarth__: Ack [17:16] dbarth__: Silo 36 approved [17:35] rvr: thanks! [17:36] rvr: silo 77 is just next, i re-approved it, so it should pop up on the qa board [17:52] morphis_: ^ \o/ [17:53] morphis_: still needed 47 but no regressions on connections before installing the silo and with 47 installed most thing now work as expected again [18:58] jibel, any chance to expedite https://trello.com/c/QTDB4lAF/2857-996-ubuntu-landing-057-unity8-unity-scope-click-dobey then? :) [18:59] alesage, if you're testing silos today ↑ :) [19:00] Saviq, ack, appears to be next up [19:00] hmm [19:06] Saviq: yes, britney runs each silo in sequence, takes about an hour to get through them all [20:09] kenvandine, thanks for talking to victor for me. :) [22:06] jgdx, np === salem_ is now known as _salem