[08:27] Mirv: at Berlin you said you had a ppa following "stable", you're aware that "stable" doesn't exist anymore? [08:32] tsdgeos: it's not automatically following, but yes I noticed they're now merging to 5.3 [08:32] tsdgeos: the last build is from 20140605 [08:32] ok [08:33] tsdgeos: any change of looking at the 5.3 crash? https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1328485 Pat assigned to Saviq though but it can be reassigned if it's more your turf [08:33] Ubuntu bug 1328485 in unity8 (Ubuntu) "unity8 crashed with SIGABRT on Qt 5.3" [High,New] [08:38] tsdgeos, yeah, could you try and have a look at ↑? [08:38] tsdgeos, from the symbols I saw, it'd be an upstream bug around HTTP requests [08:39] brrrr [08:39] * tsdgeos checks [09:17] Saviq, new infographics are broken here [09:17] Saviq, I am testing... [09:17] don't review yet [09:17] something got broken between the split and unsplit merges [09:19] Saviq, will you be able to land the id hack today? [09:20] mhr3, I think so, yes [09:22] Saviq: Mirv: so that crash should just happen by scrolling up/down in the apps scope? [09:24] tsdgeos, I believe some "normal" usage is causing it [09:25] tsdgeos, but Mirv knows more [09:25] i can't get it to crash :/ [09:27] tsdgeos: Saviq: yes, but sometimes it seems pretty hard so I tend to also open several apps, continuing rapid swiping of apps scope up/down etc. [09:28] ok, let me open a few apps [09:28] see if that helps [09:28] tsdgeos: maybe expand the Available and then swipe up/down (as that probably generates http requests or so) [09:28] yeah i was doing that a lot [09:28] I actually just got a hang (?) that way, unity8 is consuming 100% CPU [09:29] :/ [09:51] hmmm [09:51] i just got it to crash [09:52] but with a different total bt than Mirv's [09:52] http://paste.ubuntu.com/7652231/ [09:52] uncool bt [09:53] tsdgeos, ouch [09:53] tsdgeos: http://people.canonical.com/~tjyrinki/qt53/unity8hang.mkv this was right after reboot (pure hang in this case, a crash earlier) [10:16] Mirv: is apport running at 99+ % in the background [10:17] Mirv: as in after reboot it started to report the earlier crash because it hadn't done so previously? [10:18] tsdgeos, you were supposed to fix the one found already, *not* find two new ones ;P [10:19] Saviq: i think it may still be just on [10:19] e [10:19] memory gets corrupted [10:19] crashes in a random place [10:19] tsdgeos, yup [10:21] biab [10:34] davmor2: no, the only case I saw apport was when it was reporting the currently happened crash (I tend to let it finish) === MacSlow is now known as MacSlow|lunch [11:21] Saviq, what was installing /usr/share/unity8/Greeter/graphics/infographic_circle_back.png ? [11:22] Cimi, > dpkg -S /usr/share/unity8/Greeter/graphics/infographic_circle_back.png [11:22] unity8: /usr/share/unity8/Greeter/graphics/infographic_circle_back.png [11:22] hah I don't have it [11:23] why... weird :D [11:26] Saviq, what does debian/control needs? [11:26] and what should go in unity-api? [11:26] https://code.launchpad.net/~cimi/unity8/infographics-new-lightdm/+merge/222953/comments/534954 [11:26] Cimi, in any case, that file should be gone now, the infographics svg needs to include that circle [11:27] Cimi, isn't there a new package with the qml bindings that your MP depends on? [11:27] mmm no [11:27] or yes [11:27] well yeah it does indeed [11:27] unity8 [11:28] Cimi, qtdeclarative5-infographics0.1 [11:28] Cimi, as per https://code.launchpad.net/~unity-team/libusermetrics/file-based-infographics/+merge/214020 [11:28] Saviq, adding [11:30] Cimi, make sure to add the dep in the correct place (unity8 vs. unity8-common, depending on where you put the QML files in) [11:31] Saviq, this dep is runtime for unity8 [11:31] Saviq, infographics are in qml/Greeter [11:31] Cimi, sure, but which package do the QML files that "cause" that runtime dep get installed in? unity8 or unity8-common? [11:32] Cimi, hint: debian/*.instsall [11:32] -s [11:34] Saviq, both [11:34] Saviq, Infographics in in qml/Greeter [11:35] Saviq, but uses CrossFadeInfographics that is in qml/Components [11:35] (will work in getting the mod in the sdk, but will come later I believe...) [11:35] Cimi, and both import the module? [11:38] nope [11:38] ok [11:38] just unity8 then === alan_g is now known as alan_g|lunch [12:03] Saviq, ok updated and using png instead svg [12:03] Cimi, png/svg of what? [12:03] Saviq, mocks [12:03] Cimi, kk [12:03] Saviq, diff easier to read [12:03] Cimi, indeed [12:08] Saviq: do you know what's the ETA for the split greeter revert? [12:08] mzanetti, already in trunk [12:08] oh.. [12:09] dandrader: ping [12:09] mzanetti, pong [12:10] dandrader: hey. do you think you could update the qtcomp unity8 to the latest unity8 trunk? [12:13] mzanetti, sure [12:13] cool === MacSlow|lunch is now known as MacSlow === davmor2__ is now known as davmor2 === alan_g|lunch is now known as alan_g [13:09] * kgunn wonders if mterry took his swap day today === zequence_ is now known as zequence [13:16] kgunn, last mention of him on the channel was 2200 UTC [13:17] kgunn, but you should be able to find out in hr c c, assuming he reported it there [13:17] good point [13:21] Cimi, QtCS notes are on warthogs for you [13:21] Saviq, thx [13:36] Saviq, people are getting nervous about departments, mind if i build u8+all the rest in a silo? [13:36] mhr3, ETA on your part? [13:37] mhr3, in any case, I'm landing u8 real soon, probably best to wait for that [13:39] @unity: anyone running a ~clear recent image [13:39] nope, lots' of 5.3 stuff in here [13:39] do you have all app icons in Installed? [13:39] ↑ to anyone that *does* [13:40] Saviq, will still take a while, i'd want it as a non-blocking silo [13:41] Saviq, we need to land all the api breaks in there, so will have to also update and rebuild all the scopes etc [13:42] mhr3, I'm fine with that, only worry I'd have is we'd be blocking silo, but if you can convince sil2100, be my guest [13:44] i'll promise sil2100 to not request any other silos while i'll have that one [13:53] seb128, every laptop suspend in utopic unity crashes and keeps crashing until I reboot... happening since few weeks. do we have a bug for that? [13:53] Cimi, not that I know about no [13:53] ok [13:53] seb128, strangely apport doesn't collect anything [14:11] Saviq, dednick texted me that dist-upgrade hosed his system... guess he might not make it to standup [14:12] mhr3, ouch [14:12] mhr3, thanks [14:16] Saviq: any idea who/what can be setting the Theme to Ubuntu.Components.Themes.SuruDark ? [14:16] Saviq, for the "changelog bump" in lp:unity-notifications, regarding the increased max. number of actions, what part of the package-version (x.y.z+ddmmyyyy-pkgrev) is meant the be increased? === dandrader is now known as dandrader|afk [14:17] MacSlow, just the part before + [14:17] MacSlow, and leave it UNRELEASED [14:17] grrr [14:17] new kernel hosed my system. [14:17] dednick, welcome back [14:17] tsdgeos, Shell.qml:74 [14:17] Saviq: no, that's SuruGradient [14:18] for some reason above 3.8 kernel doesnt work on my system. [14:18] tsdgeos, ah, then UITK probably [14:19] i did notice some errors go past on dist-upgrade though... [14:20] tsdgeos, MainViewStyle.qml [14:20] yeah [14:20] /usr/lib/x86_64-linux-gnu/qt5/qml/Ubuntu/Components/Themes/Ambiance/MainViewStyle.qml === dandrader|afk is now known as dandrader [14:20] found it [14:20] Saviq: you aware it's overriding our surugradient? [14:20] Saviq, ok... that's what's to be checked in unity8's debian/control afterwards. Still, I have yet to find the correct option of "dch" to create the correct changelog-entry or is that something one has to do manually? [14:21] also that weird construct is making valgrind and asan complain [14:21] obviously not what causes our crashing problems [14:21] but makes asan stop there [14:21] Saviq, is what it looks like atm http://pastebin.ubuntu.com/7653268/ [14:22] MacSlow, 'dch -v 1.2.3 "Changelog entry"' [14:22] MacSlow, just drop the part after + [14:22] MacSlow, but yeah, `dch -v 0.1.3-0ubuntu1` would work [14:23] Saviq, ok... well 0.1.2 is already the bumped version :) [14:24] MacSlow, then that, yes [14:25] tsdgeos, so we don't get SuruDark in the end? [14:26] erm SuruGradient? [14:26] Saviq: nope [14:26] *interesting* [14:26] or at least i'd say not [14:26] http://paste.ubuntu.com/7653295/ [14:27] first output is what you set, second what you had [14:27] the last ones are Dark, Dark anyway [14:27] seb128, where do I find unity 7 log? [14:27] seb128, keeps respawing === alan_g is now known as alan_g|tea [14:28] Cimi, ~/.cache/upstart/compiz.log or gnome-session-Unity.log [14:32] seb128, I have this https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/1311037 [14:32] Ubuntu bug 1311037 in xserver-xorg-video-intel (Ubuntu) "[mesa] intel_do_flush_locked failed: Invalid argument" [Undecided,New] [14:32] seb128, for unity7 [14:33] seb128, no nvidia card on this laptop [14:35] Cimi, try talking to the #ubuntu-x team if you hit that error === alan_g|tea is now known as alan_g [14:42] mhr3, had to recon the unity8 silo, will take a bit more to land :| [14:42] Saviq, np, i'm ready to rebuild it a few times... [14:43] mhr3, I meant the app id hack [14:43] erm, category id [14:43] right === dandrader is now known as dandrader|afk [15:00] hm. apparently my new kernel doesnt like nvidia [15:05] seb128, works with trusty kernel I think [15:05] seb128, why don't we have tests for those errors? [15:06] Cimi, because hardware specific issues require real hardware, and we can't easily own/manage every single hardware config that ever shipped in the world [15:06] seb128, it's a dell certified vostro [15:06] btw I see the point [15:06] hopefully we'll debug it [15:07] step one is to report a bug [15:07] seb128, I failed to get debug symbols [15:07] seb128, tried attaching gdb before suspending [15:07] resuming [15:07] boh, didn't get anything [15:08] I will try again another time === dandrader|afk is now known as dandrader [15:29] mzanetti, unity8/mirCompositor merge done === salem_ is now known as _salem [15:36] Saviq, thoughts how we'd do https://f966f709-a-c881af26-s-sites.googlegroups.com/a/canonical.com/unity8dash/toolkit/10-content-loading/mixed_bar_surface.png?attachauth=ANoY7cqrTVVU7UYpiae_0usTaP887PIG-9Qf90uiqJ9FRI0RUUCjLtx0jE9tpeJyimfW_3nRUiX0D3pDXqd4tncf7yep5FF7oaD8k-aRzkbl4Ntkto2VGgfvzxZu6uyNQbYaZk6mZnw5VLLv0b0sZbpqB6bAUiV4mz9_UWqkQqjVBm4HmvN1hgS0HhczULqx_5KEc93IMLJ0RfI6gZSl7RYY7aAeRmt-5xeWiJicT42ecs_Ck0KDLjGRss1jpp [15:36] S6PifqzmOUKdbg&attredirects=0 ? [15:36] mhr3, yeah, url shortener [15:36] :) [15:36] Saviq, http://goo.gl/1aixM9 [15:37] mhr3, we have two usecases there, no internet connection and no account configured [15:37] mhr3, (at least) [15:37] mhr3, so I'm thinking two strings + url coming from the scope [15:38] Saviq, no account needs to be different, cause no internet is always just one, missing account creds can be multiple [15:38] mhr3, multiple as in multiple accounts missing? [15:38] yea [15:38] mhr3, don't think you'd display multiple lines like this [15:39] indeed, i think design was thinking about an inline card button for account issues [15:39] mhr3, one button "accounts" would be just as useful I'd think, maybe with the left one listing supported accounts? [15:39] hm, maybe, that's still not designed apparently [15:40] mhr3, yeah, so... [15:41] i do remember mike promising me that design though [15:43] ping Saviq, tedg: this is an interesting bug, but I'm not sure who can investigate on it [15:43] https://bugs.launchpad.net/unity8/+bug/1329141 [15:43] Ubuntu bug 1329141 in Unity 8 "qmlscene crashed while running test_can_launch_multiple_applications" [Undecided,New] [15:44] can you assign it to somebody? [15:44] elopio, please trace the .crash file first [15:44] elopio, apport-cli on the .crash file on the device where it's from [15:45] elopio, otherwise the .crash file is missing details [15:45] dandrader: thanks [15:45] Saviq: that one comes from ci. So we should configure jenkins to do that? [15:45] elopio, I *thought* it was configured already (whoopsie-upload-all does that) [15:45] elopio, but apparently it doesn't [15:46] elopio, bug #1317968 [15:46] bug 1317968 in Ubuntu CI Services "Smoketesting doesn't preprocess .crash files" [Undecided,New] https://launchpad.net/bugs/1317968 [15:46] bug #1260384 [15:47] bug 1260384 in Ubuntu CI Services "Should preprocess .crash files on test devices" [High,Fix released] https://launchpad.net/bugs/1260384 [15:47] elopio, in any case, qmlscene crashing has nothing to do with unity8 [15:47] I'll ask on ci about the trace. [15:48] tedg: if qmlscene crashes during ubuntu-launch-app, is it on your plate? [15:49] elopio, unlikely [15:49] elopio, Not really, it'd be the application that it's running. [15:49] elopio, but I believe fginther saw qmlscene crashing in jenkins recently [15:49] elopio, maybe he knows the issue [15:49] (and fixed it for us, too) [15:50] Ah, it seems to be in one of the test applications. [15:50] Saviq: in this case, it's not an application. [15:50] it's the fake qmlscene that we use to test application life cycle on unity [15:50] elopio, so it is an application ;) [15:50] so it's just an empty dummy app. [15:50] elopio, empty, dummy, but still an app (coming from UITK autopilot, too) [15:50] (I think?) [15:50] Not a fake QML scene, the real QML scene binary [15:51] right, a fake qml that we launch with qmlscene. [15:51] elopio, and fginther had exactly the same in our qmluitests job, fixed it in some way, so maybe he has an idea [15:51] it's an app, but should have nothing that could cause the failure. [15:52] * elopio waits for him... [16:06] mzanetti, oups, I replied to Olivier but dropped you guys from CC to not spam (went to warthogs still), yes there was a QtWeb* session, there's notes on the wiki, apparently none of us went there === gatox is now known as gatox_lunch [16:11] Saviq: yep... just saw the mail [16:11] sounds like Germany just scored a goal [16:11] * mzanetti does not have any TV running [16:12] they just started though :D [16:12] well... 12 minutes [16:12] yeah, 1:0 [16:13] * Saviq runs out o/ [16:13] people were going wild around the neighborhood... so I assumed must be that [16:31] and again :D lol === dandrader is now known as dandrader|afk [16:37] Saviq: is the split greeter now reverted in image? [16:39] dednick: its in trunk now, yes [16:39] mzanetti: in phone image? [16:40] ah, tsdgoes gone? [16:40] dednick: afaik in trunk also means in the latest devel-proposed image. its not in a propagated image yet [16:40] mzanetti: i c [16:40] mzanetti: thanks [16:40] np === alan_g is now known as alan_g|EOD === _salem is now known as salem_ === gatox_lunch is now known as gatox [17:47] elopio, what's up? [17:47] I made a search lense couple of years ago. It worked in ubuntu 12.04. However, in 14.04 it dosen't show up in dash [17:47] how can I make it work with ubuntu 14.04 ? [17:47] * greyback eod [17:47] fginther: have you seen qmlscene crashes recently? === greyback is now known as greyback|eod [17:48] elopio, I have not looked at any recently [17:48] elopio, there was an issue in CI with qmlscene running on desktop last week, but that was resolved to running on the wrong release [17:49] fginther: ok. I've marked the bug as incomplete, waiting to see if we can gather more information. [17:49] the last daily run was impressively green. [17:50] elopio, if I read the scrollback though, it looks like back traces are missing (due to the switch to using the smoke test runner)? https://bugs.launchpad.net/ubuntu-ci-services-itself/+bug/1317968 [17:50] Ubuntu bug 1317968 in Ubuntu CI Services "Smoketesting doesn't preprocess .crash files" [Undecided,New] === dandrader|afk is now known as dandrader [18:09] fginther: yes, I think you are right. Andy said he was going to prepare a branch for it. === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem [19:18] mzanetti, ping === dandrader is now known as dandrader|afk === olli_ is now known as olli === dandrader|afk is now known as dandrader