=== balloons is now known as Guest90681 === nudtrobert1 is now known as nudtrobert [04:46] * Guest42341 What a fine day for science! === nudtrobert1 is now known as nudtrobert === nudtrobert1 is now known as nudtrobert [08:34] pstolowski, morning, any unity8 silo you're planning to land real soon? [08:34] (you got 4 ;)) [08:41] Saviq, hey, no, none of them is going to land anytime soon [08:41] pstolowski, ack [08:41] tsdgeos, we can drop use_sdk_12 right? [08:42] Saviq: if we're going for the use_sdk_13, yes [08:42] kk [08:43] Saviq: still needs approval though https://code.launchpad.net/~unity-team/unity8/use_sdk_13/+merge/271603 [08:43] tsdgeos, yup yup [08:43] tsdgeos, how about https://code.launchpad.net/~cimi/unity8/wide-preview-zoomable-videoplayback/+merge/273081 ? [08:43] * Saviq preps new silo [08:44] Saviq: it looks a bit weird to me but if it's what design wants it's good, somehow i forgot the top approve [08:44] ack [08:45] cimi, there's no bug for sharing from preview? [08:45] Saviq: we may want to get https://code.launchpad.net/~aacid/unity8/new_and_old_audio_role/+merge/275365 into the silo too if it helps Mirv [08:45] tsdgeos, yup [08:45] needs review too though [08:46] will take care, still need to wait a bit for migration so we'll get there [08:49] tsdgeos, does "use sdk13" do "use qtquick24" as well? [08:50] or do we need another branch on top for that [08:50] yeah it doesn't [08:50] Saviq: no, you need https://code.launchpad.net/~aacid/unity8/use_quick_24/+merge/271602 for that [08:50] ah [08:50] missed that one [08:51] tsdgeos, qml test fail though? [08:51] hmmm [08:51] needs a rebuild only [08:51] let me merge the other use_ [08:51] and push [08:52] tsdgeos, needs a resubmit [08:52] why? [08:52] tsdgeos, ~aacid vs. ~unity-team for the usd_sdk13 branch [08:52] *use, even [08:53] (and a merge) [08:53] ah right you stole my other branch :D [08:53] resubmitting in a moment [08:54] wonder why did I steal it [08:54] oh no I di'in [08:55] ok, resubmitted at https://code.launchpad.net/~aacid/unity8/use_quick_24/+merge/275671 [08:56] tsdgeos, https://code.launchpad.net/~unity-team/unity8/use_sdk_13/+merge/266078 [08:56] tsdgeos, it's mzanetti who stole your branch :P [08:56] "plural you" [08:56] :D [08:56] ;) [08:56] i'm not even sure the original sdk_13 branch was mine, it may have been cimi's [09:17] Saviq: Is more-smooth-less-lag in rc-proposed? [09:23] Oh, it was proposed to xenial 9 minutes ago though. \o/ [09:29] duflu, yeah, just published this morning, will be in the next rc-proposed image [09:30] Saviq: the branches are not merged yet because it waits for the package to be in xenial? [09:31] tsdgeos, yes [09:31] k [09:33] tsdgeos, if you need it, the branch is here https://code.launchpad.net/~ci-train-bot/unity8/unity8-ubuntu-xenial-landing-022/+merge/275672 [09:34] oh well, not the merge, but the branch [09:34] sure [09:39] Saviq: the use_sdk_13 branch has conflcits with that branch, can i merge that branch in and hope it'll work or wait for unity8 to be actually merged in? [09:41] tsdgeos, it will work, you could even make it a prereq, only case when it'd be a problem is if the silo didn't actually publish [09:41] i see [09:41] let's wait then [09:41] no rush i guess? [09:41] or is it better for the silo you're preparing? [09:41] tsdgeos, i.e. the only thing the train will do with this branch is to merge trunk (in case translations etc.) and push to trunk [09:42] tsdgeos, so merging it ~= merging trunk [09:42] tsdgeos, yeah, I'm solving conflicts one-by-one using MPs from the previous silo, but in this case it might make sense to actually just merge it [09:42] tsdgeos, so yeah, merge and resolve please, but don't make prereq [09:43] ok [09:43] we can always rewrite history if needed === _morphis is now known as morphis [09:53] Saviq: the problem with the merge is that if i do merge it won't compile because of missing /usr/include/unity/shell/application/MirMousePointerInterface.h [09:53] so do we prefer to have working CI or a merged branch? :D [09:54] tsdgeos, vivid will work because it's in overlay [09:54] ah right [09:54] tsdgeos, xenial would work if we enabled proposed (we probably should, not even sure if we don't already) [09:54] ok then pushing [10:19] mzanetti, can you please merge lp:~ci-train-bot/unity8/unity8-ubuntu-xenial-landing-022 to panel-button-fixes (can call it trunk merge) [10:19] ack [10:21] Saviq, can you give a quick explanation on what's going on with those xenial branches? [10:22] mzanetti, not much, we're waiting for unity8 to migrate to xenail [10:22] xenial [10:22] (merged my branch ) [10:22] mzanetti, and I'm prepping a silo while that's going on [10:22] mzanetti, that branch is basically what will be trunk when it merges [10:22] s/merges/migrates/ [10:22] mzanetti, I'm pre-empting conflicts on the new silo is all [10:23] ack [10:50] mzanetti, ah, the conflict seems to be with Josh's upsidedown after all [10:50] mzanetti, can you please check? and if so, uncommit the merge and resubmit on top of Josh's? [10:50] or actually [10:51] I could [10:51] since I already stole Josh's branch [10:51] right.... whatever works best for you [10:51] just let me know [10:51] mzanetti, ok, uncommit and overwrite your branch without the merge, then [10:51] ack [10:52] grr [10:52] no wait [10:52] no, it's fine [10:52] already typed the push... [10:52] you caught me before pressing enter [10:52] type the --overwrite, too ;) [10:52] and press enter [10:52] so what now? want me to push or not? [10:52] mzanetti, yes, push [10:53] done [10:53] sry [10:53] lol [10:53] for the noise [10:53] ah ok... thought you'd tell me to merge again now D [10:53] no worries [10:53] ;) [11:06] * Saviq officially hates bzr :[ [11:06] can't get it to merge cleanly without --weave [11:07] oh one more thing I could do [11:07] ok yeah [11:08] mzanetti, sry, you need to merge after all, otherwise bzr is dumb... [11:08] haha [11:08] Saviq, still this? lp:~ci-train-bot/unity8/unity8-ubuntu-xenial-landing-022 [11:08] mzanetti, yeah [11:09] Saviq, pushed [11:09] mzanetti, because bzr decides that if the history is 22 → yours → his+22, it's no good [11:09] but if it's 22 → yours+22 → his+22, it's fine! [11:09] grrr [11:09] :D [11:09] even though yours+22 is no-op [11:09] I mean no conflicts [11:09] yep... [11:09] oh well. Merge all the things! [11:14] plugins/Cursor/Cursor.qml UNKNOWN *No copyright* [11:14] :? [11:14] https://jenkins.qa.ubuntu.com/job/unity8-vivid-i386-ci/1143/console [11:16] uggh [11:16] how does that get through everything else :/ [11:16] tsdgeos, I'll MP a fix [11:18] tsdgeos, in the mean time, can you rebase sdk 13 on top of wide preview, or strongarm Cimi to do the opposite? [11:18] Saviq, I can do it [11:18] prolly the opposite makes sense [11:18] yeah indeed [11:21] Saviq: cimi: ok i'll rebase [11:21] tsdgeos, nowait [11:21] * tsdgeos waits [11:21] tsdgeos, the opposite == cimi rebasing on yours [11:21] this one better no? https://code.launchpad.net/~cimi/unity8/sdk1.3_newUbuntuShape/+merge/271610 [11:21] otherwise you'll be catching up everywhere [11:21] ok [11:21] can we merge this guys? [11:21] * tsdgeos does nothing [11:22] cimi, yeah, if the only thing they were waiting for was 1.3 [11:22] indeed [11:23] tsdgeos, have a look at that branch pls, I fixed the conflicts the other day [11:23] if you ok I rebase the preview branch on that one [11:28] cimi, let me know please when you've resubmitted wide videopreview (and let me know of any other branches that could land soon) [11:29] Saviq, i'd like to know if I should rebase on that branch ^^ or sdk 13 [11:30] cimi, well, whatever, do they conflict? [11:30] /usr/bin/ld: BFD (GNU Binutils for Ubuntu) 2.25.51.20151022 assertion fail ../../bfd/elf32-i386.c:5245 [11:30] ok that doesn't look good [11:30] * Saviq grabs a xenial chroot [11:31] * cimi looks === mpt_ is now known as mpt [12:02] cimi, status report? [12:03] Saviq, finishing rebasing wide videopreview on top of ~cimi/unity8/sdk1.3_newUbuntuShape [12:03] ack [12:03] so it does conflict [12:04] tsdgeos, https://code.launchpad.net/~saviq/unity8/add-cursor-copyright/+merge/275692, obviously it's + 22, so... [12:04] it seems we might need to force merge/clean anyway as there's a qtmir issue in xenial, might need to skip a release [12:05] ok [12:08] Saviq: you have bad tags on that branch [12:08] tsdgeos, grr [12:09] tsdgeos, actually, just one, and one that LP considers fine... [12:09] ah no, it's my dumb colo root again [12:09] tsdgeos, cleaned [12:09] ok [12:11] approved [12:24] tsdgeos, I'm actually gonna merge this manually after we publish to improve ci [12:24] Saviq: ok [12:26] Saviq, tsdgeos pushed [12:27] cimi, new MP link? [12:27] Saviq, same, I pushed on top? [12:27] oh yeah [12:27] I need to rebase :D [12:27] * cimi hides [12:27] yes, you need to resubmit [12:28] Saviq, here you go https://code.launchpad.net/~cimi/unity8/wide-preview-zoomable-videoplayback/+merge/275696 [12:31] cimi, there no bugs to link? [12:31] cimi, to the new UbuntuShape, for example? [12:32] cimi: can you register in summit for uos-1511 so I can add you as a track lead? [12:32] mhall119, sure [12:33] Saviq, mmm no [12:33] ack [12:33] everyone else, cimi and I need to get some convergence sessions planned for UOS, specifically we should have sessions on: [12:33] Saviq, I think it's a thing you do when you update to 1.3 [12:33] Saviq, otherwise it uses deprecated APIs [12:33] Unity 8 desktop mode, what works, what's different from Unity 7, when it will be usable for desktop users [12:33] Saviq, but there is the branch with shadows in app/launcher that might have a bug somewhere [12:33] Snappy Desktop, how it'll work, when it'll work, etc [12:34] Puritine & Libertine projects, why they're needed [12:34] cimi, sure, just thought we might now start closing some bugs re: shape [12:34] but ack, will come in time [12:34] anything about the Mir work and roadmap to make it replace Xorg on the desktop [12:35] and Adaptive UI patterns and how the UI Toolkit supports them [12:35] plus anything else you guys know of that needs to be planned for 16.04, and anything you want to demo and talk about that's working now [12:35] oh, silo 22 just landed? [12:35] mhall119, we can do the first, "when?" is not a question we'd like to give an answer to, snappy, *tine, Mir and UITK we're not the team to talk about [12:36] dandrader, yes, kinda, had to force it and skip the release to xenial as its binutils is broken [12:36] but nothing to concern you with, really [12:36] you can create session topics here: http://summit.ubuntu.com/uos-1511/create_meeting [12:36] Saviq: ack, I'll be poking those other teams too [12:37] dandrader, so we're out of sync between v+o and x, but trunks are all up to date now, I'm prepping silo for unity8 right now [12:37] might make qtmir hitch a ride, too [12:38] Saviq, what about wily? [12:38] dandrader, wily's old news already [12:38] done & dusted [12:38] tsdgeos, scope_settings_keyboard_scroll conflicts in Preview.qml, likely with cimi's preview [12:39] Saviq, should I move all my wily machines to xenial already? [12:39] dandrader, if you want, safer to just use a chroot for now, there's a lot of movement in xenial atm [12:39] will wait then [12:40] I can update that Saviq [12:40] @unity: silo 22 landed, wily is dead for us, xenial out of sync for now due to toolchain issue [12:40] Saviq, clashes with which branch? [12:40] cimi, your preview already depends on shape and 1.3 [12:41] cimi, you can't make it prereq tsdgeos's branch too [12:41] Saviq, I was going to send him a patch [12:41] owell, that you can do if you're bored ;) [12:41] wl,e [12:41] cimi, https://code.launchpad.net/~aacid/unity8/scope_settings_keyboard_scroll/+merge/274874 [12:41] woke up with good mood, no head [12:41] headache :) [12:41] no head? ;p [12:41] lol [12:41] :D [12:42] wonder where the good mood comes from, then [12:42] had some annoying migraine for the last two weeks, coming back every other day, I'm fine from yesterday :) [12:42] I might had mood/seasonal swings like girls :) [12:42] @unity: I'm prepping another unity8 silo: https://requests.ci-train.ubuntu.com/#/ticket/564, if there's a branch missing in it, let me know please [12:44] Saviq, tsdgeos that preview.qml conflicts just in imports [12:48] dednick, we could land occlusion, only unity8 is missing a top-ack, is it ready for a re-review? [12:48] Saviq, did "bzr pull" on my lp:qtmir copy and, surprise surprise: "742 tag(s) updated." [12:48] Saviq: let me confirm [12:48] dandrader, no surprise there [12:49] dandrader, we never paid enough attention to tags on qtmir (or well, only started recently) [12:49] dandrader, and someone's been infecting all projects out there with them [12:49] Saviq, dandrader: unity8 occlude ready for re-review [12:50] dandrader, I'm starting to feel it's not worth it, bzr/lp just need to get better [12:50] in that regard [12:50] Saviq, I don't even know if bzr is still being maintained [12:51] yeah, well, part of the problem indeed [12:51] dednick, ok [12:51] Saviq, can you pls merge this one too: https://code.launchpad.net/~lukas-kde/ubuntu-settings-components/langpack/+merge/275017 [12:52] ltinkl, ack [12:54] ChrisTownsend, seems the unity8-lxc-setup is broken since we removed some files from cdimage.. did you know? [12:54] pstolowski, tsdgeos do https://code.launchpad.net/~unity-team/unity-api/music_cards_extra_attributes/+merge/271084 https://code.launchpad.net/~unity-team/unity8/audioCardSupport/+merge/271605 depend on one another? === alan_g is now known as alan_g|lunch [12:55] could use a link in the description [12:56] ltinkl, btw, what's up with fixLogin1Tests ? [12:57] popey: Yes, I've been working on it since last week. I have a new version in https://launchpad.net/~unity8-desktop-session-team/+archive/ubuntu/unity8-preview-lxc/+packages and I'm working out details to get a release into xenial. [12:57] Saviq, ye no real progress :/ dunno how to make the exit cleanly and/or make the test runner wait until the DBUS processes are terminated [12:57] ChrisTownsend, thanks... [12:58] ltinkl, ack, I'll have a thinklook [13:00] ltinkl, fwiw I'd say we're using it wrong, as `dbus-test-runner --task sleep --parameter 5 --task sleep --parameter 10` does not leave stuff laying around [13:00] dandrader, qtmir stripped, fwiw, I'll try and pay better attention to them now [13:01] Saviq, I'll look into libqtdbustest meanwhile [13:01] ltinkl, ack [13:01] Saviq, http://bazaar.launchpad.net/~unity-team/libqtdbustest/trunk/view/head:/src/libqtdbustest/SuicidalProcess.cpp [13:05] kgunn, FYI, 22 landed in vivid+overlay, skipped in xenial due to toolchain issue bug #1510067 [13:05] bug 1510067 in qtmir (Ubuntu) "qtmir rebuild fails with "assertion fail ../../bfd/elf32-i386.c:5245"" [Undecided,New] https://launchpad.net/bugs/1510067 [13:07] yep, just saw that [13:10] Hey guys, I'm trying to start a Unity 8 session on my Wily desktop machine, and I'm just getting the black u-s-c screen, ie, no greeter. I'm pretty sure there is something up with my machine, but I'm hoping one of you can help figure out what may be up. unity8 and unity8-dash are up and running and seem ok except the unity8.log looks very sparse. Here is the unity8.log: http://pastebin.ubuntu.com/12969778/ H [13:13] Saviq, there's a failtrain on that link you posted [13:13] :) [13:14] ChrisTownsend, if you stop unity8, do you get u-s-c background? [13:15] mzanetti, failtrain? ../../? [13:15] * mterry upgrades to xenial [13:15] Saviq, https://requests.ci-train.ubuntu.com/#/ticket/564, [13:15] mzanetti, comma [13:15] the comman at the end causes a 404 with a nice gif [13:15] ;) [13:15] Saviq: If by u-s-c background, you mean the black-ish screen w/ cursor, then I always have that either with or without unity8 running. [13:15] mzanetti: i looked in the ppa anyhow ;) only way to be sure [13:16] ChrisTownsend, hmm right, not sure we have a spinner on desktop (the boot screen from phone) [13:16] Saviq: Oh, that. No, it's not on desktop. [13:17] ChrisTownsend, do you have overlay enabled? [13:17] Saviq: Not for Wily. Should I? [13:17] ChrisTownsend, yeah, otherwise you're quite a bit behind [13:17] Saviq: Really? Wily was just released:) But sure, I can enable the overlay and try it. [13:18] ChrisTownsend, but we've switched to overlay around beta freeze and have been working there since [13:18] yeah wily has it's own overlay [13:19] Saviq: Hmm, ok. I'll try it. Won't hurt. [13:19] which will now start getting stale with xenial [13:19] available to land in [13:19] kgunn, we should probably scrap the wily overlay as soon as we get xenial in working order [13:20] It's the same PPA, right? Just w/ Wily built packages? [13:20] Oh, I see that now. nm [13:22] BTW, looks like webbrowser isn't working on Vivd+overlay on the desktop. [13:23] ChrisTownsend, do you get a crash on startup? [13:23] mterry: My unity8 problem or webbrowser problem?:) [13:24] ChrisTownsend, webbrowser :) [13:24] ChrisTownsend, thought it might be bug 1508054, which was recently fixed [13:24] bug 1508054 in apparmor-easyprof-ubuntu (Ubuntu) "[desktop] Crashes on startup" [Low,Triaged] https://launchpad.net/bugs/1508054 [13:24] mterry: Yes, I do. [13:24] mterry: Yes, that's the issue. [13:25] mterry: Thanks [13:25] np! :) [13:25] Saviq: didn't you find that we did actually have this issue occurring on phone [13:25] https://bugs.launchpad.net/ubuntu/+source/notify-osd/+bug/694224 [13:25] Launchpad bug 694224 in notify-osd (Ubuntu) "Synchronous notifications briefly steal focus" [High,Confirmed] [13:25] sorry, lost context [13:26] we talked about putting on desktop team, but then i thot you did some testing [13:26] kgunn, no, that was unity7 only [13:26] and found it on phone [13:26] ok...moving then [13:26] Saviq: it's listed in "Stuff from https://requests.ci-train.ubuntu.com/#/ticket/29" [13:26] kgunn, what about desktop team? [13:26] tsdgeos, yeah, not on unity-api [13:27] tsdgeos, I mean unity-api does not have that link in its description === Guest90681 is now known as balloons [13:27] Saviq: Same issue w/ overlay installed. I think my system is borked, I just wish I had a clue what is borked. [13:27] Saviq: that's right [13:27] seb128: that bug 694224 [13:27] tsdgeos, and anyway I'd rather have explicit MPs, silo requests might change with no notice [13:27] bug 694224 in notify-osd (Ubuntu) "Synchronous notifications briefly steal focus" [High,Confirmed] https://launchpad.net/bugs/694224 [13:28] Saviq: otoh the dependency is not double i mean the unity-api code doesn't depend on the unity8 code [13:28] kgunn, that bug is from 2010 why does it become important now? [13:28] seb128: b/c bfiller had put it on the OTA milestone target list :) [13:28] so just filtering [13:28] tsdgeos, but it breaks it, i.e. if we land unity-api, u8 will stop building [13:28] hum, k [13:28] I guess [13:28] seb128: well...i think it was accident [13:29] because we do steal focus with the password notifications :) [13:29] TextField { [13:29] id: textfield [13:29] ltinkl, that's different [13:29] seb128, kgunn, I think it's because webbrowser-app showed up and noticed this issue [13:29] ltinkl, the bug is about volume notifications [13:30] Saviq, ah, I was missing context there, sry [13:30] Saviq, but still it doesn't sound like a good idea [13:30] ChrisTownsend, can you stop unity8 and try and run it manually, providing the right MIR_SOCKET [13:30] Saviq: sure, we could have a branch that just increases the version dependency is needed (whcih shows we have a weird way of specifying dependencies) [13:30] ltinkl, disagree, password notifications are modal atm [13:30] Saviq: Ok, I can try it though I'm pretty sure it's using the right MIR_SOCKET when it starts. [13:30] they steal the user's focus, might as well get ketboard [13:31] Saviq, ok then but do they have to be? [13:31] ltinkl, ah, so now that's a totally different question ;) [13:31] ltinkl, the code is fine, you're just questioning the design [13:31] Saviq: i agree that the ticket links can change but otoh it's a bit "wasteful" to specify the dependencies twice [13:32] tsdgeos, yeah ok, as long as both sides have it [13:32] tsdgeos, otherwise I have to hunt for related MPs [13:32] (even if not dependents, just related ones) [13:32] * ltinkl goes to test the volume notification + focus [13:32] ltinkl, I tested, video continues to play just fine on unity8/phone [13:33] ChrisTownsend, I'm thinking about whether you get some more interesting feedback rather than that being the issue [13:33] Saviq: done, i don't understand why it causes you extra work if pstolowski is handling that branch/silo though, what's the problem? [13:33] Saviq, ye, even keyboard focus stays with a volume notification [13:33] tsdgeos, I just go through all top-acked bits [13:33] tsdgeos, and see what I can land [13:33] Saviq: Ok [13:33] tsdgeos, need context [13:34] Saviq: you said " otherwise I have to hunt for related MPs", why you need to hunt for related MPs if someone else is handling the branches? [13:35] or is it that there's no clear info if someone else is handling the branches? [13:35] tsdgeos, because I don't know if they are or not [13:35] tsdgeos, the MP doesn't say so [13:35] tsdgeos, I was just going through the ACKed unity-api branches, and had to think if we can land any of them safely [13:36] ok [13:37] Saviq: How do you suggest starting unity8. I tried 'MIR_SOCKET=/run/user/1000/mir_socket initctl --session start unity8' and nothing new was printed in the log. I also tried 'MIR_SOCKET=/run/user/1000/mir_socket unity8' and that bailed. [13:40] ChrisTownsend, skip upstart [13:40] ah [13:40] bailed how? [13:40] @unity do we have a component (or design directions) to display a mouse tooltip? [13:40] ltinkl, doubt that [13:40] Saviq: $ MIR_SOCKET=/run/user/1000/mir_socket unity8 [13:40] QXcbConnection: Could not connect to display [13:40] Aborted (core dumped) [13:40] ChrisTownsend, QT_QPA_PLATFORM=ubuntumirclient [13:41] Saviq, thought so :/ Qt Quick doesn't have that either, only the Controls [13:41] Saviq: Ah, ok [13:41] ltinkl, sounds like a bug for UITK+Ubuntu UX [13:42] * ltinkl does his homework [13:42] Saviq: http://pastebin.ubuntu.com/12969978/ [13:43] ChrisTownsend, ok that looks as if it's working [13:43] And then it exited. [13:43] ah it did [13:43] grr [13:43] ChrisTownsend, ok let me try a few things here [13:43] will get back to you in a mo [13:43] Saviq: Ok, thanks! [13:43] ChrisTownsend, in the mean time, try with guest / new user? [13:43] Saviq: Sure [13:45] Saviq: Ah, another user works fine. [13:45] Saviq: So I f'd up my user somehow. [13:45] ah, there it is: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1428779 [13:45] Launchpad bug 1428779 in Ubuntu UX "[SDK] Tooltips" [Medium,In progress] [13:46] Saviq, ^^ [13:46] ack [13:46] ChrisTownsend, oh good, would be interesting to know what's the breaking bit [13:47] Saviq: Yeah, I don't even know where to begin to look. [13:47] ChrisTownsend, the log doesn't really say anything interesting? [13:47] -? [13:47] Saviq: Yeah, it just seems it kind of stops proceeding for some unknown reason. [13:48] tsdgeos, hey, did i break anything? [13:48] no no === alan_g|lunch is now known as alan_g [14:05] cimi: lp:~cimi/unity8/sdk1.3_newUbuntuShape doesn't merge cleanly, can you merge the use_sdk13 branch there? [14:32] Saviq: any idea when the landing fix is going to be fixed on xenial? we have an unbuildable unity8 on xenial atm, i guess i can always use the vivid chroot but it's a bit more cumbersome [14:45] tsdgeos, yes [14:48] tsdgeos, you can use proposed, but the issue is known and there's a fix upstream, so shouldn't be long [14:48] tsdgeos, it seems to merge fine here [14:49] tsdgeos, see https://wiki.ubuntu.com/Testing/EnableProposed for some tricks with priorities to avoid upgrading all of proposed [14:49] Saviq: oki [14:50] cimi: sure my bad, i meant was unmerged and makes the testing uising the CI generated .deb a bit "fake" [15:01] tsdgeos, all good now [15:01] ktx [15:01] cimi: see the review i made on the wide-branch [15:01] something's wrong with a binding [15:12] tsdgeos, try now [15:19] tsdgeos, got a resubmit of keyboard scroll branch on top of cimi's preview? [15:21] Saviq: does it conflict with stuff? [15:21] tsdgeos, yes, Preview.qml [15:22] tsdgeos, scope_settings_keyboard_scroll conflicts in Preview.qml, likely with cimi's preview [15:22] Saviq: okl [15:27] Saviq: done [15:27] mterry: i had to resubmit https://code.launchpad.net/~aacid/unity8/scope_settings_keyboard_scroll/+merge/275723 can you re-review/approve? [15:28] tsdgeos, k [15:28] tsdgeos, tx [15:28] tsdgeos, I see merge conflicts? [15:28] well, LP does [15:28] mterry: yes because cimi has not merged https://code.launchpad.net/~cimi/unity8/wide-preview-zoomable-videoplayback/+merge/275696 [15:29] ah k [15:30] dednick, https://code.launchpad.net/~nick-dedekind/unity8/lp1475678.surface-occlude/+merge/273427 conflicts with trunk for sure [15:30] dednick, as well as https://code.launchpad.net/~nick-dedekind/unity-api/lp1475678.surface-occlude/+merge/273425 [15:31] Saviq: ok. will get to it in a minute [15:31] dednick, I'll write on the MPs [15:32] dandrader, standup [15:33] @unity. otp. may not respond in mumble [15:33] ack === dandrader is now known as dandrader|afk [16:06] Saviq: You still around? [16:07] ChrisTownsend, yup [16:07] Saviq: So I'm still digging into my issue w/ my user and I can tell that unity8 is in the SIGSTOP phase. So what is it waiting on, ie, what's the next phase? [16:08] ChrisTownsend, SIGCONT from upstart [16:08] ChrisTownsend, any chance you have a weird unity8.conf in ~/.config/upstart/ [16:08] ? [16:08] or modified unity8.conf in /usr/share/upstart/sessions [16:08] Saviq: Nope [16:09] ChrisTownsend, `initctl status unity8` [16:09] ps aux | grep unity8 [16:09] initctl get-env unity8 | grep STOP [16:10] $ initctl --session status unity8 [16:10] unity8 start/spawned, process 3343 [16:10] yeah [16:10] so upstart is waiting for the SIGSTOP, but never received t [16:10] it [16:10] somehow [16:11] Saviq: Hold on lemme kill the upstart session as I had strace inserted in there to try to capture stuff. [16:12] ChrisTownsend, yeah, try and start stuff as normal, check what upstart status does unity8 have, and whether it's indeed STOPed [16:13] Saviq: $ initctl --session get-env unity8 | grep STOP [16:13] initctl: No such variable: unity8 [16:14] $ ps aux | grep unity8 [16:14] townsend 3545 22.0 2.1 1324676 75796 ? Ssl 12:12 0:00 unity8 [16:14] townsend 3563 1.0 0.7 401096 25892 ? Ssl 12:12 0:00 unity8-dash --desktop_file_hint=/usr/share/applications/unity8-dash.desktop [16:14] ChrisTownsend, sorry, list-env [16:14] ChrisTownsend, so it's not STOPed, it'd be T, not S [16:14] ChrisTownsend, looks like a red herring due to your debugging [16:14] Saviq: Ok, sorry for the noise [16:14] ChrisTownsend, but [16:15] ChrisTownsend, if upstart still says it's spawned (not running) [16:15] it means unity8 did not SIGSTOP [16:15] $ initctl --session status unity8 [16:15] unity8 start/running, process 3545 [16:15] yeah [16:15] that looks good [16:15] and oh yeah, unity8-dash wouldn't start otherwise [16:15] that's what the SIGSTOP is for in the first place [16:15] to let upstart know when to launch dependent jobs [16:15] Saviq: Ok [16:16] we sig it when Mir is ready for connections [16:16] Saviq: Are there any user specific config files or dcong/gconf keys I could try resetting? [16:17] ChrisTownsend, nothing comes to mind that would've caused what you're seeing :/ [16:17] Saviq: Grrr [16:17] ChrisTownsend, can you install debug symbols, attach gdb to unity8 and see if it's stuck somewhere [16:17] Saviq: I assume overlay has debug symbols built, right? [16:18] ChrisTownsend, in separate packages, yes, you need main/debug in sources.list for it [16:18] and install unity8-*dbgsym [16:18] likely libmir*-dbgsym, too [16:18] Saviq: Ok, I'll install that and try a bt. [16:26] Saviq: Seems its running: /usr/lib/x86_64-linux-gnu/gstreamer0.10/gstreamer-0.10/gst-plugin-scanner -l and is blocked by that. [16:28] ChrisTownsend, oh [16:28] ChrisTownsend, does it block for the other user? FWIW it blocks for me here, too [16:29] ChrisTownsend, can you show the thread bt where this happens? I'm interested to see where are we calling that [16:29] Saviq: I'll reboot and try. I see no other threads running, just that running from the main thread. [16:29] ChrisTownsend, ack [16:30] Saviq: http://pastebin.ubuntu.com/12971157/ [16:31] Saviq: Now I'll reboot the machine and try the other user. [16:31] * Saviq wonders why we're using gst 0.10 at all ;P [16:31] tsdgeos, have a look ↑, ChrisTownsend's unity8 on desktop seems to be hanging on that [16:32] Saviq: it's what qtmultimedia uses [16:32] Saviq: No, the other user doesn't have that process running. [16:32] we need to update [16:32] mhm [16:32] ChrisTownsend, I'll try in a sec if I have the same symptom, as this command hangs for me here as well [16:32] Saviq: I wonder if it's because my user has some music in the the Music dir. [16:33] Saviq: And the test user does not have music. [16:33] I don't :/ [16:33] and gst still hangs [16:33] Nevermind then [16:36] Saviq: ChrisTownsend: are you sure that gst-plugin-scanner command is supposed to return? [16:37] nope [16:37] but should probably print something if Qt's waiting on it [16:37] i mean [16:37] reading the code [16:37] it seems it starts a "socket" and waits for instructions [16:38] i.e it has a debug line saying [16:38] GST_DEBUG ("Plugin scanner child running. Waiting for instructions"); [16:38] and then [16:38] while (!l->rx_done && exchange_packets (l)); [16:38] ack [16:38] tsdgeos: Well, it seems the unity8 main thread is waiting for it to return. [16:38] some info here indeed http://lists.freedesktop.org/archives/gstreamer-devel/2011-November/033898.html [16:38] And the main thread just sits there and nothing else happens. [16:39] * Saviq tries [16:39] brb [16:43] ChrisTownsend, so at least it hanging on console doesn't prevent my session from starting [16:43] ChrisTownsend, try removing .gstreamer-0.10 from your home [16:43] Saviq: Ok, I'll try that. [16:43] maybe the registry got borked and the command hangs [16:44] Saviq: Yeah, possible [16:44] It's definitely specific to my user. [16:45] yeah so that could explain [16:45] Saviq: Nope, still hangs:/ [16:45] you can cp -R your user to the test user and start removing things one by one... === dandrader|afk is now known as dandrader [16:46] Saviq: Ok, good idea. Also, if I kill the gst-plugin-scanner process, it just gets spawned again by unity8. [16:48] ChrisTownsend, http://paste.ubuntu.com/12971272/ [16:49] check if this ran on your normal unity7 session behaves somewhat fine [16:49] ChrisTownsend, you can also comment out unity8 bits importing QtMultimedia to see if it helps... [16:49] Saviq: unity7 seems fine. [16:50] ChrisTownsend, this is really the only thread? that sounds weird [16:50] info threads / t a a bt? [16:50] Saviq: Oh, you mean run that qml you pasted in Unity 7 [16:50] ChrisTownsend, yes [16:50] Saviq: Ok, I'll try that in a sec. Lemme get teh thread info first. [16:51] ChrisTownsend, it won't do much, just initialize the multimedia engine and print shit out [16:51] but if it doesn't, it would indeed suggest something's wrong with this side of the equation [16:53] Saviq: Ok, I was wrong about threads:) [16:53] Saviq: http://pastebin.ubuntu.com/12971312/ [16:54] Thread #7 looks interesting [16:54] indeed [16:55] or 14 [16:55] ChrisTownsend, that said, #1 shouldn't look like that either [16:55] tsdgeos, thoughts before you EOD ↑? [16:55] Saviq: Looks messed up [16:57] Saviq: not really, looks like we're waiting for something that didn't happen [16:58] i wouldn't say 7 or 14 would be causing thread 1 to get stuck [16:58] at most 7 is maybe stuck because thread 1 is stuck [16:58] but first random guess i'd say not the other way around [17:00] Saviq: i'd invoke jhodapp maybe he has seen this before [17:02] tsdgeos, right [17:02] tsdgeos, but #1 looks weird [17:02] yep [17:02] i'd be interested in getting the debug symbols so that this line gets more info [17:02] #9 0x00007f4638653df0 in ?? () from /usr/lib/x86_64-linux-gnu/qt5/plugins/mediaservice/libgstmediaplayer.so [17:03] ChrisTownsend, http://paste.ubuntu.com/12971370/ would get rid of multimedia from unity8 (it's -p1 in /usr/share/unity8/) [17:03] tsdgeos, what's up? [17:03] ChrisTownsend, if that makes it go, at least we'd know that's it [17:03] jhodapp: see http://pastebin.ubuntu.com/12971312/ [17:03] jhodapp, ChrisTownsend's unity8 hangs on start [17:03] Saviq: Ok, I'll try it. [17:04] jhodapp, #1 looks weird there [17:04] that ?? is probably just [17:04] ./src/plugins/gstreamer/mediaplayer/qgstreamerplayerserviceplugin.cpp:87: gst_init(NULL, NULL); [17:04] ChrisTownsend: you said this only happens with a user but not with a different one, right? [17:05] tsdgeos: Yes, that is correct. [17:05] Saviq, this pastebin? http://pastebin.ubuntu.com/12971157/ [17:05] so we can rule out missing packages or stuff [17:05] jhodapp: the one i pasted http://pastebin.ubuntu.com/12971312/ [17:06] Right, it's something specific to my user, but trying to figure out what is painful. [17:06] tsdgeos, looks like gstreamer isn't happy with one of the local plugins [17:06] it's getting stuck trying to dynamically load the .so of a plugin [17:06] jhodapp: Are plugins user specific? [17:07] there's a core set, and then yes there's a local cache [17:07] jhodapp: Where is the local cache? [17:07] ChrisTownsend, I would start by trying to clear ~/.cash/gstreamer-0.10/* [17:08] jhodapp: Ok [17:08] jhodapp, there's nothing there, that's for 1.0, already asked him to clear ~/.gstreamer-0.10 [17:08] right, ok [17:08] didn't help [17:08] been a while since I've dealth with 0.10 [17:09] I have registry.x86_64.bin in there. Removing it... [17:09] check in ~/.local/share/ [17:10] there might be a gstreamer-0.10 dir [17:10] Nope, that wasn't it. I'll try that [17:10] No, no gstreamer stuff in ~/.local/share [17:11] search for anything *gstreamer* in ~/ [17:12] tsdgeos, good job: http://pastebin.ubuntu.com/12971447/ [17:12] will comment on respective MPs [17:14] cimi, https://code.launchpad.net/~cimi/unity8/preview-sharing/+merge/273233/comments/696721 [17:14] tsdgeos, before you go, please bump qml/Components/ListViewOSKScroller.qml [17:18] jhodapp: Saviq: Thanks for helping. I'll try to narrow this down some more, but at least I know that it appears gstreamer is the culprit here. [17:18] tsdgeos, before you go, please bump qml/Components/ListViewOSKScroller.qml [17:19] tsdgeos, good job btw: http://pastebin.ubuntu.com/12971447/ [17:19] ChrisTownsend, yes indeed, and if you can...turn on GST_DEBUG=*:4 (or even 5) to see what plugin it's hanging on [17:19] Saviq, pushed [17:19] cimi, tx [17:19] ChrisTownsend, set that environment variable with whatever process owns gstreamer [17:19] jhodapp: Ok, does the debug get sent to syslog? [17:20] no to stdout [17:20] or err [17:20] I forget [17:20] cimi, QtQuick 2.4 too? [17:20] jhodapp: Ok, but to the console nonetheless. [17:20] yes definitely [17:20] cimi, I think you misunderstood [17:22] cimi, I didn't ask for merging (whatever you did merge), but rather to bump UITK/QtQuick imports to 1.3/2.4 [17:23] jhodapp: http://pastebin.ubuntu.com/12971531/ [17:24] I don't see anything that says "error" to me, but I don't quite understand everything I see there either. [17:25] I have never seen that output before, that's messed up [17:25] ChrisTownsend, not sure what's going on, I'd ask in #gstreamer honestly [17:25] show them that log [17:25] jhodapp: lol, ok [17:25] jhodapp: thanks [17:26] np, sorry I couldn't be more of a help [17:26] mzanetti, I added a new key to unit8's schema file, how do I install it or update dconf to use it? [17:27] Saviq: BTW, your patch makes my session come up now. [17:27] dandrader, glib-compile-schemas /usr/share/glib-2.0/schemas [17:27] (assuming you've copied the changed file there) [17:27] mzanetti, wow, would never find it. was playing with gsettings and dconf commands [17:28] mzanetti, as root? [17:28] yeah... took me a while too [17:28] I think you need root, yes [17:28] dandrader, iirc I ended up reading through dpkg post-install steps to figure it :D [17:29] jeez [17:30] ChrisTownsend, ok, so confirms it's gst going apeshit [17:30] Saviq: Indeed [17:30] dodged it! \o/ [17:30] Saviq: I just don't know how I f'd up my user so badly. [17:31] Saviq: I've been reworking the unity 8 lxc stuff and have no clue how I messed this up. [17:31] mzanetti, hmm, https://developer.gnome.org/gio/stable/GSettings.html does mention glib-compile-schemas [17:31] Saviq: But yes, you dodged it:) [17:31] ;) [17:33] dandrader, I'm sure there's docs for it, yes [17:33] it was really just me going the crazy route [17:35] mzanetti, they're hard to find. it took me a while to sort out gfonc, dconf and gsettings [17:35] *gconf === alan_g is now known as alan_g|EOD [18:02] Saviq: bumped [18:03] tsdgeos, thanks, o/ [18:03] cimi, did you fix yours, too? [18:03] now gone for good good [18:29] Saviq: jhodapp: Hey guys, I fixed the issue. I copied over .gstreamer-0.10/ from the test user to my user and now the Unity 8 session works\o/ [18:30] ChrisTownsend, awesome [18:30] ChrisTownsend, did you ever figure out why gstreamer was choking? [18:31] jhodapp: I think it wants that directory there. [18:31] jhodapp: For some reason, it was removed for my user. [18:31] I'm not sure how it happened. [18:31] ChrisTownsend, oh ok interesting, so does the issue come back if you remove it? [18:32] what if you just mkdir .gstreamer-0.10...is that enough? [18:32] jhodapp: I can try that just to see. [18:32] ok great [18:34] ChrisTownsend, well, it just recreates the dir for me normally [18:34] but glad you got it working [18:34] jhodapp: It occurs again with that directory removed. It also occurs when the directory exists, but the registry is not there. [18:35] Saviq: For some reason, it does not create it automatically for my user. [18:35] interesting [18:35] ChrisTownsend, yeah it's supposed to automatically create it...wonder if it's a permission issue [18:36] ChrisTownsend, just running the qml code I pasted before gets me this dir again [18:36] as probably any gst-inspect call etc. [18:37] jhodapp: Saviq: Yeah, hmm, maybe permissions is the issue, but I wonder what permissions it wants. [18:38] ChrisTownsend, what user is the process that owns gstreamer running as, and then does that user have write permission to ~/ [18:39] jhodapp: gst-plugin-scanner is running under my user and yeah, my user has permissions to write to $HOME. [18:39] not sure then, that's odd [18:40] It is odd. [18:44] jhodapp: It's weird, in the debug output when it fails, it looks like it can't find the registry, but then turns around and thinks it's there anyways: http://pastebin.ubuntu.com/12972277/ [18:44] At any rate, at least I know what the failure is if this ever happens again:) [19:10] Saviq: BTW, I have the unity8-lxc updated to work on Trusty, Vivid, & Wily: https://launchpad.net/~unity8-desktop-session-team/+archive/ubuntu/unity8-preview-lxc [19:11] Saviq: Some things are still kind of broken like: sound, VT switching, some indicators not showing up in the Panel, some devices such as cameras. [19:30] ChrisTownsend, awesome [20:54] mzanetti, around? [20:58] mzanetti, nice on tech lead! :) [20:59] mzanetti, now I won't feel so bad pestering you with questions :) [21:01] mzanetti, unping [21:01] mterry, right, it's in his job description now! ;) [21:34] mterry, thanks :)