14:34 #startmeeting Desktop Team Weekly Meeting - 2017-11-07 14:34 Meeting started Tue Nov 7 14:34:35 2017 UTC. The chair is willcooke. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 14:34 14:34 Available commands: action commands idea info link nick 14:34 Roll call: andyrock, dgadomski, didrocks, duflu (out), jbicha, jamesh (out), jibel/heber, kenvandine, laney, oSoMoN, seb128, tkamppeter, trevinho (hols), robert_ancell (out) 14:34 😺/ 14:34 hey! 14:35 hey 14:35 Sorry for the late start everyone 14:35 o/ 14:35 o/ 14:35 Looks like we've got enough folk to get started.... 14:35 #topic andyrock 14:35 hey 14:35 1. py-macaroon-bakery 14:35 1.1 MR to relaxing py-requests and py-protobuf dependencies in order to facilite the ITP's py-macaroon-bakery 14:35 1.2 testing the MR to make sure relaxing the deps was ok 14:35 1.3 python-macaroon-bakery is in bionic (0.0.3 in universe and 0.0.5 in universe-proposed) thanks to cjwatson 14:35 2. software-properties 14:35 2.1 Refactoring code around in order to make it work with new API, code is almost ready I just need to deal with some corner cases 14:35 3. my xps 15 decided to died, spent some time to open it, debug the issue and order the different replacement parts to fix it 14:36 *to relax 14:36 andyrock, OEM said you can keep that machine for a little while if you need it 14:36 cool 14:37 andyrock, EOF? 14:37 yup 14:37 thanks andyrock 14:37 #topic dgadomski 14:37 hey 14:37 o/ 14:37 * discussed TLS1.2 support in python 2.7.6 on Trusty (bug #1443704) with the security team, checking if it's supported by openssl underneath without Python knowing about it 14:37 * debugging a potential cups regression in 17.10 - will report a bug after I know what's going on there 14:37 * back to working on bug #1700827 (after security issues has been resolved upstream) 14:37 bug 1443704 in python2.7 (Ubuntu) "Support for TLS 1.2 not present (added in 2.7.9)" [Undecided,Confirmed] https://launchpad.net/bugs/1443704 14:37 bug 1700827 in pcp (Ubuntu) "[MIR] pcp package" [Undecided,Confirmed] https://launchpad.net/bugs/1700827 14:37 eof 14:37 thanks dgadomski 14:37 #topic didrocks 14:38 plop 14:38 * File Google Code In tasks on the forum. Need now to enter them in the system (Ken, you did volunteered as well, I think I'll get you in). 14:38 * Started GNOME Shell vs extension mode upstream discussion for security on their mailing list: 14:38 - Few feedbacks for now, tried pushing on ML, bugzilla and IRC multiple times. 14:38 - Implemented a POC patch set enabling that. Running this patch set happily for some days and all works (updates prevention, can't remove extension, running local one instead of remote one). 14:38 - Once this is sorted out, will need to work on Tweaks and chrome shell extension to reflect correct status. 14:38 - For reference: discussion at https://mail.gnome.org/archives/gnome-shell-list/2017-October/msg00034.html and bug athttps://bugzilla.gnome.org/show_bug.cgi?id=789852. 14:38 Gnome bug 789852 in general "Ensure extensions parts of a mode are from system, report back correct status for g-s-extension-prefs" [Normal,New] 14:38 * Fix a dash to dock vs ubuntu dock conflict (do not look at gsettings key but rather looking at internal running state) + SRU/backport some upstream fixes to artful. 14:38 * Linux Unplugged interview 14:38 * Fix a GNOME Shell crasher when installed with no recommends 14:38 * Iterate with GNOME Shell designers on "sounds above 100% feedback" 14:38 * Rebase fo bionic ubuntu-dock on latest dash to dock and uploadedit. 14:38 * More community interaction on the hub and french forum. 14:38 … 14:38 thanks didrocks 14:39 #topic duflu 14:39 Standby for large paste.... 14:39 * PulseAudio 11.1 for bionic: 14:39 - Completed merge with Debian, packaging fixes and testing. 14:39 - Now just awaiting sponsorship; see sponsorship list below. 14:39 - History: https://git.launchpad.net/~ubuntu-audio-dev/pulseaudio/log/?h=ubuntu 14:39 * Mutter/Xwayland crash handling (https://bugs.launchpad.net/bugs/1724185) 14:39 - Fix released upstream this week (v3.26.2) 14:39 Launchpad bug 1724185 in mutter (Ubuntu) "Xwayland leaves no core dump or .crash file when crashing" [High,Fix released] 14:39 - Fix released in Bionic already! 14:39 * Mutter: blurry shell fonts: (https://bugs.launchpad.net/bugs/1714459) 14:39 Launchpad bug 1714459 in mutter (Ubuntu) "Shell font is blurry under Wayland (panels, menus and login screen too)" [Low,In progress] 14:39 - Fix landed upstream for release in mutter v3.27.2 only. 14:39 - HELP: Need to decide how we will get this into 18.04 if not Gnome 3.28. 14:39 * Totem stuttering fixes (https://bugs.launchpad.net/bugs/1698270) 14:39 Launchpad bug 1698270 in totem-pl-parser (Ubuntu) "Totem pauses and stutters during video playback even when CPU usage is low" [High,In progress] 14:39 - GTK timing flaws: patches awaiting review/sponsorship below. 14:39 - Totem disk fsync: patch awaiting review/sponsorship below. 14:39 * Totem high CPU usage (https://bugs.launchpad.net/bugs/1698282) 14:39 Launchpad bug 1698282 in clutter-gst-3.0 (Ubuntu) "Totem uses dramatically higher CPU than any other video player" [Medium,Confirmed] 14:39 - On hold till upstream or myself find time to work on completing DMA-buf support in clutter-gst. 14:39 * Unresponsive touchpads (https://bugs.launchpad.net/bugs/1696929) 14:39 Launchpad bug 1696929 in libinput (Ubuntu) "X1 Carbon gen4 and gen5 touchpads are unresponsive and laggy" [High,Confirmed] 14:39 - Upstream fix approved by all testers and now landed. 14:39 - Due for release in libinput 1.9.2 (I guess) 14:39 * Incorrect %CPU in gnome-system-monitor (https://bugs.launchpad.net/bugs/1723370) 14:39 Launchpad bug 1723370 in gnome-system-monitor (Ubuntu) "Processes % CPU column is inaccurate (only ever an integer multiple of the number of CPUs)" [Medium,In progress] 14:39 - Upstream fix has now landed. 14:39 - ETA for inclusion in Ubuntu not yet known. 14:39 * Leaky/bloated Gnome Shell (https://bugs.launchpad.net/bugs/1672297) 14:39 Launchpad bug 1672297 in gnome-shell (Ubuntu) "gnome-shell uses lots of memory, and grows over time" [High,In progress] 14:39 - I got sucked into this by an unexpectedly helpful and eager user so have been working with him. Looks like there are at least two repeating causes of memory bloat/leak. 14:39 - Work in progress. 14:39 * HELP: Patches awaiting Ubuntu sponsorship: 14:40 - https://code.launchpad.net/~vanvugt/gtk/fix-1698270/+merge/331846 14:40 - https://bugs.launchpad.net/bugs/1698270/comments/18 14:40 - https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/1716700/comments/2 14:40 Launchpad bug 1716700 in pulseaudio (Ubuntu) "Upgrade to PulseAudio 11.1" [Wishlist,In progress] 14:40 - https://code.launchpad.net/~vanvugt/totem/fix-1502476/+merge/333195 14:40 * Patches awaiting upstream review: 14:40 - https://bugzilla.gnome.org/show_bug.cgi?id=787001 14:40 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] 14:40 - https://bugzilla.gnome.org/show_bug.cgi?id=787665 14:40 Gnome bug 787665 in .General "gdk_frame_clock_get_frame_time is irregular and causing stuttering" [Normal,New] 14:40 * Upstream patches now making progress with upstream help: 14:40 - https://bugzilla.gnome.org/show_bug.cgi?id=773453 14:40 Gnome bug 773453 in gstreamer-vaapi "vaapi: support GLTextureUpload for EGL/Wayland" [Normal,New] 14:40 * Daily bug management across gnome-shell, mutter, gdm3, ubuntu-themes, bluez, pulseaudio, dkms, mir, wayland, totem, mpv. 14:40 Desktoppers: Please could someone take a look at the sponsorship requests above ^ 14:41 #topic jbicha 14:41 (totem at least is blocked on the headerbar decision) 14:41 • gdm3 security update for zesty (LP: #1729354) 14:41 Launchpad bug 1729354 in gdm3 (Ubuntu) "17.04: GDM lock screen can be circumvented when autologin is set" [High,Fix released] https://launchpad.net/bugs/1729354 14:41 • Uploaded mutter SRU to artful queue (LP:#1730097) 14:41 • Uploaded nototools and fonts-noto-color-emoji to NEW queues. 14:41 • Uploaded libcloudproviders to NEW queue. It will enable Nextcloud support in GTK/Nautilus file browser (so it will need a MIR). 14:41 • Debian Testing finally has GNOME 3.26 14:41 • Continuing discussion about headerbar patches on community.ubuntu.com 14:41 Xubuntu is considering switching some of their default apps (like evince and file-roller) to the MATE versions 14:42 eof 14:42 jbicha, what about totem is blocked on headerbar? 14:42 duflu's fix is to drop the headerbar patches! :) 14:42 oh ok 14:42 (I'm going to look at duflu's sponsoring request on thursday if nobody does it before (I'm off tomorrow)) 14:42 thanks seb128 14:43 and thanks jbicha 14:43 np 14:43 https://code.launchpad.net/~vanvugt/totem/fix-1502476/+merge/333195 14:43 Skipping jamesh, no update received 14:43 #topic jibel / heber 14:43 Hey! QA updates: 14:43 * Deploying upgrade tests for bionic 14:43 * Maintenance of the testing infrastructure 14:43 * Going through ubiquity and upgrade failing jobs 14:43 * Move jobs that are waiting for nodes on lcy01 14:43 * Getting/parsing results for gnome-shell tests 14:43 EOF 14:44 thanks heber, hope you had good hols. Don't think I've said hi since you've been back 14:44 #topic kenvandine 14:44 * Backported some more fixes from artful to the gnome-3-26 backports PPA including glib with the per desktop overrides. 14:44 * Updated gnome-3-26-1604 snap to include fixes 14:44 * Fixed snapcraft-desktop-helpers to include copying of the schema overrides 14:44 * Rebuilt and tested all the GNOME snaps to include the gsettings override fixes as well as updated packages 14:44 * Tested the new gnome-online-accounts-service interface from jamesh with gnome-calendar and gnome-contacts 14:44 * Created contacts-service and calendar-service needed interfaces for evolution access from gnome-calendar and gnome-contacts. No PR submitted yet, needs some tweaking and tests but they do work with gnome-contacts and gnome-calendar with strict confinement. 14:44 I'm out friday, US holiday 14:44 that' sit 14:44 thanks kenvandine 14:45 #topic Laney 14:45 oh, jamesh did a gnome-online-accounts interface, nice 14:45 • 2 day meeting in London to review Trello etc 14:45 • helped bring up new cloud region for autopkgtest (lcy01 was redeployed) 14:45 ∘ it's got / hitting a bug where instances are failing to reboot (something like bug #1713751), spent a fair bit of time looking into that with IS (ongoing) 14:45 bug 1713751 in linux-hwe (Ubuntu) "soft lockup / stall on CPU when shutting down with hwe 4.10 kernel " [Undecided,New] https://launchpad.net/bugs/1713751 14:45 • got review on some gnome-software PRs, fixed / responded to those 14:45 • started working again on seeding of snaps 14:45 🎆 14:46 jamesh, shame you forgot to write your summary, you usually have interesting work :) 14:46 seb128, and it works great :) 14:46 nice 14:46 thanks Laney 14:47 #topic oSoMoN 14:47 hey ho! 14:47 = chromium = 14:47 • chromium 62.0.3202.75 security update published to all supported series, 62.0.3202.89 currently building in stage PPA 14:47 • updated chromium beta to 63.0.3239.30 in PPA and snap in beta channel 14:47 • updating chromium dev to 64.0.3253.3 14:47 • my upstream CL that fixes source tarball generation got merged and deployed: https://chromium.googlesource.com/chromium/tools/build/+/3eab95a2d59d2ed3d983079c2fb1a0e449b858f3 14:47 • bug triaging, closing many old bugs that are not relevant any longer, confirming a few ones 14:47 = libreoffice = 14:47 • snap: fixed a couple of issues found by candidate channel testers (bug #1729025, bug #1728713) and promoted 5.4.2 to stable channel 14:47 bug 1729025 in libreoffice (Ubuntu) "[snap] Cannot install 5.4.2 snap with core from beta channel: invalid exec command" [High,Fix released] https://launchpad.net/bugs/1729025 14:47 bug 1728713 in libreoffice (Ubuntu) "[snap] theming regressed in 5.4.2 on Ubuntu 17.10" [Undecided,Fix released] https://launchpad.net/bugs/1728713 14:47 • bionic: prepared 5.4.2 upload, failed to build on amd64 due to new unit test failures, backported a couple of commits from the upstream debian package and prepared an updated package which is now building 14:47 = other = 14:47 • prepared PPU application for chromium-browser: https://wiki.ubuntu.com/OlivierTilloy/PPUApplication, please endorse me if you can! 14:47 EOF ❄️ 14:47 nice, thanks oSoMoN 14:47 #topic seb128 14:48 • 1 day off for national holiday 14:48 • 2 days in London for planning 14:48 • bugs triaging from incoming artful reports 14:48 • some sponsoring 14:48 • replied to some delayed email and to mailing list discussions 14:48 14:48 thanks seb128 14:48 #topic tk 14:48 err 14:48 tk taks 14:49 guess Till isn't online 14:49 tab complete fail 14:49 We can come back 14:49 #topic robert_ancell 14:49 - Got basic snap support working in AppCenter (ElementaryOS) and forwarded patch to them. 14:49 - Implemented new / updated snapd APIs into snapd-glib 14:49 #topic AOB 14:49 jbicha, you wanted to talk about.... 14:49 * willcooke thinks 14:49 LP: #1666676 14:49 Launchpad bug 1666676 in ubuntu-meta (Ubuntu) "Install tracker by default" [Undecided,Confirmed] https://launchpad.net/bugs/1666676 14:50 that's the one 14:50 ah, back on that one :) 14:50 I invited csoriano (Nautilus maintainer) and garnacho (tracker maintainer) in case there were more detailed questions 14:50 hi everyone 14:51 hi garnacho, thanks for joining 14:52 sorry but I didn't see that discussion would be raised today 14:52 I didn't re-read enough about the topic nor prepared 14:52 so I don't have questions or specifics to discuss 14:52 can we punt to tomorrow on IRC? 14:52 sorry, I didn't think to ping seb128 specifically ahead of time 14:52 hello all 14:53 jbicha, well, it's not only me 14:53 maybe I should have emailed the list instead 14:53 in advance 14:53 jbicha, seems like you invited upstream, would have been nice to tell us to prepare questions if we had some :) 14:53 I guess latest stand we got on perf and battery was summarized in https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1666676/comments/6 14:53 Launchpad bug 1666676 in ubuntu-meta (Ubuntu) "Install tracker by default" [Undecided,Confirmed] 14:53 jbicha, could we start a thread on the hub? Sounds like a good place to do it 14:53 hey csoriano! 14:53 (I didn't read more, didn't know we would have that AOB and didn't prepare for that question) 14:54 well, those discussions went round but "on the spot" comments 14:54 community.ubutnu.com is nice because it doesn't require people to all be on IRC at the same time to give input 14:54 - the LTS cycle is a tricky time to do disruptive changes and that's one of those 14:54 it was a stealth request on https://trello.com/c/4gjeaE66/93-re-evaluate-tracker-by-default 14:54 - we would need to plan for manpower to do the change and ensure quality and we didn't account for that 14:54 and I didn't think that everyone wouldn't have seen the little note there 14:56 I think the hub is probably a good place for this then, people can take time over their responses 14:56 willcooke: I can start the discussion on community.ubuntu.com, how about we see where we are with that at next week's Desktop meeting? 14:56 think things through 14:56 jbicha, I think that the summary is that upstream believes that it's good, there is no issue and that we should ship it and that we tend to want data to make up our mind and don't have enough to decide 14:57 jbicha, that sounds good. Thanks. Maybe we can resolve everything on the hub with a bit of luck 14:57 seb128: yes, that's where we were last cycle 14:58 csoriano, garnacho, sorry I don't think that we have questions for you, the positions are been stated several time of the years on that launchpad bugs and other places and I don't think anyone really changed their position 14:58 seb128: oh ok, then I guess that's all :) 14:58 see you! 14:58 csoriano, I do know you would be in favor if seeing it included though :) 14:58 so let's see how the hub discussion goes 14:58 csoriano, thanks for coming! 14:58 sounds good 14:59 np! 14:59 alright, please point us to it 14:59 k 14:59 see you! 15:00 oki, jbicha please get the thread started then and we can get it sorted 15:00 Anyone got anything else? 15:00 sure 15:00 well, while we are on the topic, do people here install tracker and use it? 15:00 nothing else from me 15:00 I don't 15:00 does anybody has an opinion on how it's working for them 15:01 or a +1/0/-1 on principle? 15:01 I still hear bad things due to impact on battery/perf from the 2 persons working on GNOME around me 15:01 I got traker enabled for some reasons 15:01 seb128: every other GNOME distro includes it (including Ubuntu GNOME before 17.10) 15:01 just trying to get a feeling of what people thing 15:01 that's the only input, but nothing measurable unfortunately 15:01 and it's always 100% 15:01 think 15:01 andyrock, 100% what? 15:01 Ubuntu Budgie intentionally stopped installing tracker in 17.10 so I want to ask fossfredom how much feedback they had gotten before about that 15:01 CPU 15:02 :P 15:02 k 15:02 I cannot debug right now as the computer where it's enabled is dead 15:02 the anecdotic comments we just had from didrocks and andyrock don't give me much confidence :p 15:02 not using it, but I can install and observe how it affects my machine 15:02 ah, also on the french forum 15:02 hum 15:03 someone complaining that we installed tracker by default :p 100% of CPU usage 15:03 (2 days ago) 15:03 should we maybe ask people in the team to install/opt in for it 15:03 ofc, arguedwedidn't :p 15:03 and revisit/see feedback next week? 15:03 or on the hub :p 15:03 hub! 15:03 k 15:03 there are also lots of anecdotal comments that it works fine :/ 15:03 I guess asking for a call for participation on the hub is nice 15:03 * Laney has it installed, have done for ages 15:03 so my anecdote is that I never notice it 15:03 :-) 15:03 and gather data 15:03 I actually was not sure why the tracker was enabled 15:03 Laney, do you find it to provide something useful? 15:03 I can link from the french forum as well 15:04 I was going to ask here because I was a bit confused 15:04 andyrock, you probably installed the package to build/test nautilus or something? 15:04 seb128: we're also being forced in to enabling tracker… 15:04 it can be 15:04 jbicha, that has to been seen :p 15:04 I'm not saying it's not going to be incrementally harder not having it 15:04 csoriano says that Nautilus 3.28 hard-depends on tracker for the new Favorites feature and he says we would have to revert several commits to avoid the tracker dependency 15:05 but we could do without for the LTS 15:05 he already said that for 3.26 and the batch rename 15:05 and fts 15:05 for a non LTS, I guess we could give it a try, for a LTS, I'm less positive 15:05 well fts doesn't work without tracker and batch rename is a bit less nice without tracker 15:06 ("full text search") 15:06 right, but he also said in 3.26 that it wouid be an hard depends 15:06 anyway, let's give it a try 15:06 but what Didier said 15:07 even if it's good it needs a maintainer and somebody to look at issues after integration 15:07 and the team is busy at full capacity already 15:07 tracker is a hard build-depends in 3.26 and the UI is confusing because it offers fts 15:07 unsure what we are arguing over 15:07 I was just pointed out that we managed to ship 3.26 without enabling tracker 15:08 and of course, we could still ship nautilus 3.26 for 18.04 LTS if it is a hard dependency and we don't want tracker 15:08 right 15:09 I guess it's enough said on the topic 15:09 let's all install/try it and use the hub to discuss 15:09 it's tricky because we'll get criticism no matter what we do 15:09 yeah, to the hub! 15:09 gonna end the meeting here, carry on if needed 15:09 #endmeeting