14:30 <willcooke> #startmeeting Desktop Team Weekly meeting - 2017-10-24 14:30 <meetingology> Meeting started Tue Oct 24 14:30:01 2017 UTC. The chair is willcooke. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 14:30 <meetingology> 14:30 <meetingology> Available commands: action commands idea info link nick 14:30 <willcooke> Roll call: andyrock, dgadomski, didrocks, duflu (out), jbicha, jamesh (out), jibel/heber (both on hols), kenvandine (out), laney, oSoMoN (out), seb128, tkamppeter, trevinho (out), robert_ancell (out) 14:30 <willcooke> lots of people out today 14:31 <seb128> hey 14:31 <willcooke> Since andyrock is here, let's start 14:31 <willcooke> #topic andyrock 14:31 <andyrock> hey 14:31 <andyrock> 1. Theme fix for osd progressbar: https://bugs.launchpad.net/ubuntu/+source/ubuntu-themes/+bug/1724808 14:31 <andyrock> 2. Fix for https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/1724703 14:31 <andyrock> 3. Fix for https://bugs.launchpad.net/ubuntu/+source/gnome-control-center/+bug/1720400 14:31 <andyrock> 4. Fix for https://bugs.launchpad.net/ubuntu/+source/indicator-printers/+bug/1703046 14:31 <andyrock> 5. Some debugging and some work for: https://bugs.launchpad.net/ubuntu-themes/+bug/1724651 14:31 <andyrock> 6. Started to work on an oem bug on unity-control-center/16.04 14:31 <ubot5> Ubuntu bug 1724808 in ubuntu-themes (Ubuntu) "osd progress looks odd using Ambiance or Radiance" [Undecided,In progress] 14:31 <ubot5> Ubuntu bug 1724703 in gnome-shell (Ubuntu) "gnome-shell-extension-prefs crashed with SIGSEGV in gtk_widget_translate_coordinates()" [Medium,In progress] 14:31 <ubot5> Ubuntu bug 1720400 in gtk+3.0 (Ubuntu) "/usr/bin/gnome-control-center:11:update_buffers:image_get_buffers:intel_update_image_buffers:intel_update_renderbuffers:intel_prepare_render" [High,In progress] 14:31 <ubot5> Ubuntu bug 1703046 in indicator-printers (Ubuntu) "indicator-printers-service crashed with SIGSEGV in __GI_____strtol_l_internal()" [Medium,In progress] 14:31 <ubot5> Ubuntu bug 1724651 in nautilus (Ubuntu) "Nautilus progress spinner indicator is covered by a white rect in the beginning" [Medium,Triaged] 14:31 <andyrock> + some bug triaging 14:31 <andyrock> eow 14:32 <willcooke> Thanks andyrock, how's the LP API working out? 14:32 <willcooke> LivePatch 14:32 <andyrock> I'm planning to work on this starting from tomorrow 14:32 <willcooke> oki thanks 14:32 <willcooke> #topic dgadomski 14:32 <andyrock> after I finish with the oem bug 14:32 <willcooke> sorry andyrock 14:32 <andyrock> np! 14:33 <willcooke> looks like dgadomski is off 14:33 <willcooke> #topic didrocks 14:33 <didrocks> * Debug with the community the ATI card issues. Found regression commits and revert those just in time for release (after a late respin). Probably prevented (from the feedback) every non wayland supported configuration to start with a headless display. 14:33 <didrocks> * Prepared SRU with real upstream fix. In proposed right now, confirmed to work on previously broken configuration (and they run wayland now!) 14:33 <didrocks> * SRU for gnome-shell-extension-appindicator & Qt apps icon disappearing after lock/resume. 14:33 <didrocks> * SRU fixing a g-c-c crash due to dock panel, enable another patch from seb 14:33 <didrocks> * Discussed desktop snap not working on Fedora and workaround review on desktop helper 14:33 <didrocks> * Pinged about extensions still blocked on extensions.gnome.org (for us owning our extension naming). Still blocked on some legal constraints. 14:33 <didrocks> * Wrote the last blog post for the 17.10 series: https://didrocks.fr/2017/10/18/ubuntu-gnome-shell-in-artful-day-16/ 14:33 <didrocks> * Answer to people on the hub + monitor french forum and discussions on the blog. 14:33 <didrocks> * Filed a RT for discourse integration not working on the blog. Willcooke: do you mind nudging this on IS side? 14:33 <didrocks> * Workarounded with "manual discourse link" implemented on my blog post 14:33 <didrocks> * The new Ubuntu Make maintainer asked for a little bit of help, did it. (reviews, suggestions, merges) 14:33 <didrocks> * Some bug reviews (theme on the Shell) 14:33 <didrocks> . 14:34 <willcooke> thanks didrocks, can you ping me a RT link and I will ping 14:34 <didrocks> sure 14:34 <willcooke> ping ping a ping ping 14:34 <willcooke> ta 14:34 <willcooke> #topic duflu 14:34 <willcooke> * Mutter/Xwayland 14:34 <willcooke> - Completed a few iterations of my fix to allow Xwayland to dump core when it crashes (instead of always hiding core/crash files): https://bugzilla.gnome.org/show_bug.cgi?id=789086 14:34 <willcooke> * Totem stuttering (https://bugs.launchpad.net/bugs/1698270) 14:34 <willcooke> - Made a few attempts at fixing: https://bugzilla.gnome.org/show_bug.cgi?id=733780 14:34 <willcooke> - But eventually proposed the simplest solution (one line). 14:34 <willcooke> - Now we have fixes for all the totem stuttering issues and just need to work on landing/releasing them (see patch statuses below)... 14:34 <willcooke> * Daily bug management across gnome-shell, mutter, gdm3, ubuntu-themes, bluez, pulseaudio, dkms and mir. 14:34 <ubot5> Gnome bug 789086 in wayland "Xwayland leaves no core dump when crashing" [Major,New] 14:34 <ubot5> Ubuntu bug 1698270 in totem (Ubuntu) "Totem pauses and stutters during video playback even when CPU usage is low" [High,In progress] 14:34 <ubot5> Gnome bug 733780 in Movie player "Saving playlist state sometimes hangs totem" [Normal,New] 14:34 <willcooke> - Multiple days spent answering bug mail, thanks to the release. 14:34 <willcooke> * Patches awaiting Ubuntu sponsorship (or needing refreshing for 18.04): 14:34 <willcooke> - https://code.launchpad.net/~vanvugt/gtk/fix-1698270/+merge/331846 14:34 <willcooke> - https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/1714459/comments/15 14:34 <ubot5> Ubuntu bug 1714459 in Mutter "Shell font is blurry under Wayland (panels, menus and login screen too)" [Medium,In progress] 14:35 <willcooke> * Patches awaiting upstream review: 14:35 <willcooke> - https://bugzilla.gnome.org/show_bug.cgi?id=645433 14:35 <ubot5> Gnome bug 645433 in wayland "gnome-shell's panel ignores font settings" [Normal,Assigned] 14:35 <willcooke> - https://bugzilla.gnome.org/show_bug.cgi?id=733780 14:35 <willcooke> - https://bugzilla.gnome.org/show_bug.cgi?id=787665 14:35 <ubot5> Gnome bug 787665 in .General "gdk_frame_clock_get_frame_time is irregular and causing stuttering" [Normal,New] 14:35 <willcooke> - https://bugzilla.gnome.org/show_bug.cgi?id=773453 14:35 <ubot5> Gnome bug 773453 in general "cluttersink: support GLTextureUpload for EGL/Wayland" [Normal,New] 14:35 <willcooke> - https://bugzilla.gnome.org/show_bug.cgi?id=787001 14:35 <ubot5> Gnome bug 787001 in GtkClutterEmbed "clutter-gtk apps like totem and gnome-maps are spending 50% of their CPU time redrawing GTK widgets" [Normal,New] 14:35 <willcooke> - and soon again: https://bugzilla.gnome.org/show_bug.cgi?id=789086 14:35 <willcooke> * PulseAudio news regarding 17.10 - No real news, only four new bugs reported: 14:35 <willcooke> - A USB headset is dropping connection, but the user has already verified Pulse 11 fixes it. 14:35 <willcooke> - One new crash, occurred only once for one person. 14:35 <willcooke> - One VIA audio chip not showing up. 14:35 <willcooke> - One apparent user error. 14:35 <willcooke> * Older PulseAudio bugs being talked about in 17.10: 14:35 <willcooke> - Kubuntu pulseaudio doesn't work. But the bug is really upstream, easy to fix and to workaround, plus a second redundant fix is coming when we disable the offending module in pulse 11: https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/1720519 14:35 <ubot5> Ubuntu bug 1720519 in pulseaudio (Ubuntu) "KDE/Kubuntu: Module "module-switch-on-connect" should be loaded once at most. Refusing to load." [High,Confirmed] 14:35 <willcooke> - HSP really was being prioritized over A2DP, still! But it seems it doesn't manifest for many people so didn't showed up in our testing. Already fixed upstream in Pulse 12 and we'll patch it into 18.04 soonish. https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/1720684 14:35 <ubot5> Ubuntu bug 1720684 in pulseaudio (Ubuntu) "Bluetooth speaker used the HSP/HFP profile by default rather than the higher-quality A2DP profile" [High,In progress] 14:35 <willcooke> * BlueZ news regarding 17.10: 14:35 <willcooke> - No bluez bugs reported at all since release. 14:35 <willcooke> #topic jbicha 14:35 <jbicha> • forwarded https://bugzilla.gnome.org/show_bug.cgi?id=789254 14:35 <ubot5> Gnome bug 789254 in general "Create videos with vp9 instead of vp8 for much smaller file sizes" [Normal,New] 14:35 <jbicha> • fix uploaded for LP: #1725288 14:36 <ubot5> Launchpad bug 1725288 in gnome-shell (Ubuntu Artful) "[artful] gnome-shell missing dependency on gir1.2-gdm-1.0" [Medium,Fix committed] https://launchpad.net/bugs/1725288 14:36 <jbicha> • Added Provides: libsane to try to fix LP: #1707352 but that didn't work with versioned deps. 14:36 <ubot5> Launchpad bug 1707352 in sane-backends (Debian) "the change from libsane to libsane1 broke many (all?) 3rd party plug-ins for sane" [Unknown,New] https://launchpad.net/bugs/1707352 14:36 <jbicha> I think we need to add a transitional package libsane depending on libsane1 instead. Maybe we'll revert the rename for Bionic. 14:36 <jbicha> • Uploaded gtk3 for LP: #1719043 but that caused a totem regression. Instead, we'll follow GNOME and cherry-pick the mutter revert to fix epiphany (!) 14:36 <ubot5> Launchpad bug 1719043 in epiphany-browser (Ubuntu) "Gnome web urlbar very slow to show typed input" [High,Triaged] https://launchpad.net/bugs/1719043 14:36 <jbicha> • Proposing LP: #1726093 14:36 <ubot5> Launchpad bug 1726093 in ubuntu-meta (Ubuntu) "ubuntu-desktop: Install gir1.2-gmenu-3.0 by default" [Low,Triaged] https://launchpad.net/bugs/1726093 14:36 <jbicha> • Helped a bit on forums with people upgrading to 17.10. 14:36 <jbicha> eof 14:36 <willcooke> thanks jbicha. Let's talk about that MIR in AOB at the end 14:37 <willcooke> #topic jamesh 14:37 <willcooke> gnome-software: tracked down cause of bug #1724316: root cause is that 14:37 <willcooke> SSO will refuse to refresh snapd's credentials if the user has changed 14:37 <willcooke> their password. To compound matters, snapd was forwarding this error 14:37 <willcooke> as a generic "internal server error", making it difficult to 14:37 <willcooke> intelligently recover. Snapd changes submitted and reviewed (but not 14:37 <ubot5> bug 1724316 in gnome-software (Ubuntu) ""Sorry, something went wrong cannot authenticate to snap store ..." without any way to fix" [High,Confirmed] https://launchpad.net/bugs/1724316 14:37 <willcooke> landed). Corresponding snapd-glib and gnome-software changes in 14:37 <willcooke> progress. 14:37 <willcooke> snapd: followed up on default apps question: confined apps shouldn't 14:37 <willcooke> be able to change defaults without user approval, so this is probably 14:37 <willcooke> a job for xdg-desktop-portal in future (upstream bug filed). mvo has 14:37 <willcooke> put together a PR to provide an interim solution based on snapd-userd 14:37 <willcooke> and zenity for the UI: will need to look at the security implications. 14:37 <willcooke> I touched base with z_yga about the state of the mount namespace 14:37 <willcooke> related work. He is going to be working on the new content interface 14:37 <willcooke> changes discussed at the sprint that will be needed for desktop theme 14:37 <willcooke> support. We also went over how I can get my user mounts branch to do 14:37 <willcooke> its work through snap-update-ns. I need to finish off those changes 14:37 <willcooke> though. 14:37 <willcooke> going to assume kenvandine[m][m] is not available 14:37 <willcooke> #topic Laney 14:38 <Laney> • Short week 14:38 <Laney> • Worked on some gnome-software small bugs/features 14:38 <Laney> ∘ Fixing apt:// to work again with the PK backend (not a big problem as we still have apturl) 14:38 <Laney> ∘ Looking at the "shell extensions" category being empty because there are too many results, might be something to do to reduce the number shown 14:38 <Laney> • Investigated / debugged the problem which was keeping gtk in artful-proposed, fixed in the software-properties test 14:38 <Laney> • Reviewed / accepted some release fixes and random changes for others 14:38 <Laney> • Some random release checklist things, mostly turning stuff off 14:38 <Laney> • Helped a bit on arm64 testing, should be able to turn that on fully this week 14:38 <Laney> 😈 14:38 <willcooke> thanks Laney 14:39 <willcooke> #topic oSoMoN 14:39 <willcooke> � prepared SRU for bug #1718446 and got it published 14:39 <willcooke> � promoted chromium snap (61.0.3163.100) from candidate to stable channel 14:39 <willcooke> � updated chromium beta packages to 62.0.3202.62 for all supported 14:39 <willcooke> releases, then stable (ready for publication, Chris currently 14:39 <willcooke> validating) 14:39 <ubot5> bug 1718446 in gdm3 (Ubuntu Artful) "Ensure wayland -> xorg fallback to the corresponding session" [High,Fix released] https://launchpad.net/bugs/1718446 14:39 <willcooke> � built chromium snap 62.0.3202.62, published to candidate channel and 14:39 <willcooke> issued call for testing 14:39 <willcooke> � updated chromium dev packages to 63.0.3239.9 (artful, zesty, 14:39 <willcooke> xenial), trusty requires more work 14:39 <willcooke> � next up: libreoffice 5.4.2 SRU and snap, bug #1725238 14:39 <ubot5> bug 1725238 in firefox (Ubuntu) "Clicking snap:// urls doesn't work " [High,Confirmed] https://launchpad.net/bugs/1725238 14:39 <willcooke> #topic seb128 14:39 <seb128> would be good to verify that" 14:39 <seb128> ups 14:39 <seb128> * had one day off 14:39 <seb128> * iso testing 14:39 <seb128> * helped Didier testing gnome-shell revert debs to debug the ati/wayland issue 14:39 <seb128> * loads of recently-reported-bugs triaging 14:39 <seb128> * cleaned up gnome-software bugs and tagged snaps related issues 14:39 <seb128> * sponsored the gdm/gnome-session/fallback SRU from Olivier 14:40 <seb128> </week> 14:40 <willcooke> thanks seb128 14:40 <willcooke> #topic tkamppeter 14:40 <tkamppeter> - UDEV/systemd: Got upstream answer from Lennart Poettering (original author of systemd?) and it seems that the fix of the escaping in the UDEV rules file does not solve the problem. There is still a problem with the escaping (https://github.com/systemd/systemd/issues/7109). 14:40 <tkamppeter> - Google Summer of Code 2018: Started planning. 14:40 <tkamppeter> - Got the flu and had some sessions at the dentist. 14:40 <tkamppeter> - Bugs. 14:41 <willcooke> hope you're on the mend tkamppeter 14:41 <tkamppeter> Now its all OK with me, and the dentist was a long-standing issue, not acute pain. 14:41 <willcooke> #toipic Trevinho 14:41 <willcooke> good 14:41 <willcooke> :) 14:41 <willcooke> #topic Trevinho 14:42 <willcooke> Notes for the upcoming meeting: 14:42 <willcooke> � Fixed regression in gtk2 theme 14:42 <willcooke> � Fixed theming for destructive-action buttons in headerbar 14:42 <willcooke> � Fixed theming for *-action buttons in headerbar when maximized 14:42 <willcooke> � c-lobrano theme fixes (thanks!) review 14:42 <willcooke> � Prepared landing for theme fixes and SRU bugs 14:42 <willcooke> � Debugging (and proposed fix, that I want to improve a bit yet) 14:42 <willcooke> the tricky st-icon blur/shadow crash 14:42 <willcooke> https://bugzilla.gnome.org/show_bug.cgi?id=788908 14:42 <ubot5> Gnome bug 788908 in st "gnome-shell crashed with SIGSEGV in clutter_actor_get_allocation_box (from _st_create_shadow_pipeline_from_actor)" [Normal,New] 14:42 <willcooke> � Fixed a memory leak in mutter 14:42 <willcooke> https://bugzilla.gnome.org/show_bug.cgi?id=789227 14:42 <ubot5> Gnome bug 789227 in general "MetaLogicalMonitor: free the monitors list on finalize" [Normal,Resolved: fixed] 14:42 <willcooke> � Renamed mutter API get_flatpack_id into 14:42 <willcooke> get_sandboxed_app_id (and updated gnome-shell) 14:42 <willcooke> https://bugzilla.gnome.org/show_bug.cgi?id=789277 14:42 <ubot5> Gnome bug 789277 in general "window: rename get_flatpack_id into get_sandboxed_app_id" [Normal,Resolved: fixed] 14:42 <willcooke> � Found a way to debug crashes with proper C and JS backtrace 14:42 <willcooke> got from a gdb session launched from a normal terminal and 14:42 <willcooke> updated the GNOME wiki accordingly (https://is.gd/tddLmH) 14:42 <willcooke> � Made patches to make gnome-shell to report better debugging 14:42 <willcooke> infos on crashes (pointing out the bad JS code too) 14:42 <willcooke> https://bugzilla.gnome.org/show_bug.cgi?id=789237 14:42 <ubot5> Gnome bug 789237 in general "Call gjs_dumpstack on aborts and traps and optionally on segfaults" [Normal,Resolved: fixed] 14:42 <willcooke> and added functions to toggle such debugging from looking glass 14:42 <willcooke> https://bugzilla.gnome.org/show_bug.cgi?id=789377 14:42 <ubot5> Gnome bug 789377 in general "global: make possible to set debug-flags dynamically" [Normal,New] 14:42 <willcooke> � Triaging/debugging lots of various top-bugs in errors.u.c and 14:42 <willcooke> some GNOME upstream patches reviews 14:42 <willcooke> � Got nextcloud-client to the ubuntu snap store with automatic 14:42 <willcooke> travis deploy (edge) on new git revisions 14:42 <willcooke> https://github.com/nextcloud/client_theming/pull/215 14:42 <willcooke> � Fixed dash-to-panel extension that was causing gnome-shell 14:42 <willcooke> to crash https://github.com/jderose9/dash-to-panel/pull/263 14:42 <willcooke> � Last week here before 2 weeks of offline-vacation in Cuba! 14:42 <willcooke> 🌴 14:43 <willcooke> #topic robert_ancell 14:43 <willcooke> - Short week - public holidays in NZ 14:43 <willcooke> - GNOME Software snap bug diagnosing / fixing 14:43 <willcooke> #topic AOB 14:43 <willcooke> jbicha, want to remind us about the new package and the background there? 14:44 <jbicha> LP: #1726093 doesn't need a MIR, it's a request for the ubuntu-desktop package to recommend a package to help traditional menu GNOME Shell extensions 14:44 <ubot5> Launchpad bug 1726093 in ubuntu-meta (Ubuntu) "ubuntu-desktop: Install gir1.2-gmenu-3.0 by default" [Low,Triaged] https://launchpad.net/bugs/1726093 14:45 <jbicha> (& sru that to artful) 14:46 <willcooke> I think adding it for 18.04 makes sense 14:46 <willcooke> anyone got any thoughts about SRUing it? 14:46 <willcooke> or general comments 14:47 <didrocks> no objection for me 14:48 <willcooke> oki, sounds like agreement then, but we can discuss more after the meeting if that's the not case 14:48 <willcooke> any more for any more? Going in 10... 14:48 <willcooke> thanks all 14:48 <willcooke> #endmeeting