[00:56] jbicha, new snapd-glib release, please upload to Debian... [07:21] good morning! [07:23] salut didrocks [07:23] comment ça va? [07:34] salut jibel ! Ça va bien (toujours mon acouphène, mais bon…) après quelques vacances, c'est cool de voir l'install minimal landée et testée sans trop de surprise :) [07:34] et toi ? [07:37] didrocks, ça va bien, on profite de la neige :) [07:43] ahah, idem ici :) === ricotz_ is now known as ricotz [09:02] yo [09:02] morning all [09:02] hi Laney [09:02] It's c c c c cold today [09:02] sup willcooke [09:02] yeah [09:02] Feels like -13 apparently! [09:03] hey Laney, willcooke [09:03] wb didrocks!!! [09:03] hi didrocks! Good hols? [09:04] yeah, holidays were good, thanks! How was this week and half for you guys? :) [09:06] busy busy busy [09:07] hey willcooke Laney, how is u.k today? [09:07] morning seb128 [09:07] Might move to the lounge today, it's too cold in my office [09:22] feels like -17° here ;-) [09:24] good morning desktoppers [09:28] salut oSoMoN, ça va ? [09:28] ça va, et toi? [09:28] ça va bien ! ff pressure today though [09:29] did I just accidentally clear the channel topic, or is it just my client? [09:30] just your client [09:30] good [09:30] :) [09:35] Morning desktoppers [09:36] hey duflu, how are you? [09:37] seb128, good! Unexpectedly great progress: https://bugs.freedesktop.org/attachment.cgi?id=137715 [09:37] seb128, you? [09:38] Laney, could you tell britney to not block libinput on "Depends: libinput glibc (not considered) " because I don't think it's true (the resulting deb depends on >= 2.9 [09:38] Laney, or is that something I can do (and how?) [09:38] no, not possible [09:38] duflu, nice! [09:38] Laney, bah :/ [09:38] why does it think it depends on glib? [09:39] c [09:39] noto sure [09:39] probably something like if it's in excuses at the same time then they get tied together [09:39] but you'd need to read proposed-migration's source code to be sure [09:39] ah [09:40] udeb has [09:40] Depends: libc6-udeb (>= 2.27), [09:40] suggest asking Adam if he wants to skiptest it now [09:40] k [09:42] hey duflu [09:42] but yeah, unsure what's the status of libc, maybe it should be forced in [09:42] Hi oSoMoN [09:43] RAOF, hey, you notice that the mir 0.30 update is failing to build on most arches right? [09:45] hey desktopers [09:45] jamesh, hey, are you still around? Did Ken talk to you about translations and platform snap? [09:46] seb128: he didn't mention it, no. [09:46] seb128: is this gettext-using libs/apps not finding their translations when binaries are copied from debs, by any chance? [09:46] jamesh, k, so I think we have an issue with translations, what I wrote to Ken yesterday was [09:46] (I can see why that'd could be a problem) [09:47] is the Microsoft Hyper-V guy actually collaborating with Canonical? their install scripts scare me! [09:47] https://community.ubuntu.com/t/enhanced-ubuntu-vm-in-hyper-v/4363 [09:48] jamesh, [09:48] we do that hack in the launcher [09:48] # workaround the issue described in https://launchpad.net/bugs/1583250 [09:48] configflags: [--prefix=/snap/gedit/current/usr] [09:48] organize: [09:48] snap/gedit/current/usr: usr [09:48] Ubuntu bug 1583250 in Snapcraft "upstream use of build-time defined PREFIX or DATADIR incompatible with snaps relocation" [High,Triaged] [09:48] but that means it finds its own translationsn, not the xdg-users-dirs-update ones [09:48] nor any of the platform ones like gtk :/ [09:48] but yeah, it basically goes down to the issue you mention [09:48] gtk in the platform comes from the deb [09:48] which is build with a prefix of /usr [09:48] so it's looking for its .mo there [09:49] and not under /snap/gedit/current/gnome-platform/usr/share [09:49] jamesh, is that something we could solve with the new layout feature? [09:49] seb128: so, I don't think there is a solution for the platform snap without modifying every library: you've generally got an absolute path in each bindtextdomain() call [09:49] darkxst, hey, they are at least talking to us [09:49] seb128: and the platform snap will be mounted to a different path for each snap [09:50] jamesh, well, there is that ld-preloader hack :p [09:50] darkxst, yeah we said we would make it in to PPA for them, but they wanted to get it out there yesterday. They're focusing on 18.04, so that 16.04 still will all go away real soon. (I still think we should do a PPA though) [09:50] seb128: what would work, and still let us copy binaries from debs, would be to create a base snap for desktop apps [09:51] apps using the base snap would see it as the root file system rather than the core snap [09:51] right [09:51] that's not on the current plan for 18.04 though [09:51] so we could have glib, gtk, etc appearing under /usr/lib, with their translations in /usr/share/locale [09:52] do you think we can make that happen this cycle? [09:52] willcooke, ^ btw that issue is a serious problem for "snaps by default", just found that out yesterday :/ [09:52] well at least for the GNOME one [09:52] ones [09:53] if we wanted to recompile all of GNOME, with a custom prefix, we might be able to get the content interface to mount the platform somewhere other than $SNAP/ now [09:54] jamesh, I guess that's worth a forum discussion and I should post there? [09:54] seb128: yeah. I think the base snap would probably be easier and more maintainable long term though [09:54] jamesh, k, let's discuss on the forum then [09:54] thx [09:55] seb128: I don't think it would be too difficult to build a first go at the base snap, but the feature is still fairly new [09:57] seb128, willcooke: the entire scripts look hacky, but I drew the line at them installing bionic-proposed on peoples systems, as per my comment [09:57] one concern is that we'd now be shipping the entire base system runtime, so couldn't rely on core snap updates for security bugs in e.g. libc [09:58] since we'd be shipping the libc in the base snap [09:59] they clearly need some help on the packaging side ;) [10:00] jamesh, that would be annoying :/ [10:00] actually, OpenSSL is probably the more scary example [10:03] oddly enough I have a few Ubuntu server VM's running on Hyper-V but none of them have the desktop installed [10:08] willcooke, jamesh, kenvandine, https://forum.snapcraft.io/t/translations-in-snap-using-the-gnome-platform/4293 [10:09] seb128: thanks [10:09] jamesh, yw, thanks for the feedback [11:05] ricotz, can you confirm the apparmor-senddoc-fixes patch is not in 1:6.0.1~rc1-0ubuntu0.16.04.1~lo1 (re bug #1748895)? [11:05] bug 1748895 in libreoffice (Ubuntu) "LO unable to find a working email configuration" [Medium,Fix released] https://launchpad.net/bugs/1748895 [11:07] oSoMoN, yes, it is not included [11:07] oSoMoN, I didn't want to waste resources, while 6.0.2~rc1 was already built and ready [11:08] 6.0.2 will be released shortly, so I will copy the new release soon [11:16] ricotz, cool, thanks for the confirmation [11:22] since a recent update (I first noticed the issue on Monday) my bt headphones don't show up anymore in the output tab of the sound control panel in g-c-c [11:22] I see them in the bluetooth panel though, and am able to connect to them, but can't get sound to go to them [11:23] have we gotten similar reports? [11:25] this is from my apt log, when I suspect the issue started happening: https://paste.ubuntu.com/p/Y2TQsYfWfP/ [11:28] koza, did you see anything like that? ^ [11:28] oSoMoN, we did some testing on the new BlueZ and eveerything looked good - but maybe not. [11:29] oSoMoN, I was going to say ask duflu, but he's off tomorrow and can't make it to the rally. So maybe just ping him on IRC first thing on Monday [11:29] indeed bluez was upgraded on 2018-02-22, that might be when my issue started [11:29] yeah [11:29] jibel, koza ^ sounds like we need another line in the test plan [11:29] I'll see if I can downgrade [11:29] oSoMoN, if you can try the old ^H^H^H^H^H thanks [11:33] sure enough, downgrading to 5.46-0ubuntu4 gets me my sound through headphones back [11:34] upgrading again to see the regression in action [11:35] now that is weird, after upgrading I still get sound through my headphones [11:35] let's try a reboot [11:37] still there after a reboot [11:37] so that was a very strange transient error that fixed itself after downgrading/upgrading again [11:38] willcooke, is that worth a bug report? [11:39] wow, another interesting issue/new behaviour: after rebooting I had bluetooth on while airplane mode was on too [11:40] IIRC that was not allowed, one had to turn off airplane mode to be able to turn on bluetooth [11:40] {+libmp3lame0 (>= 3.100), libmpg123-0 (>= 1.6.2),+} {+libtwolame0 (>= 0.3.10),+} [11:40] some new deps for gst-good [11:41] good morning [11:42] good morning jbicha [11:51] oSoMoN, @ bug report - check in with duflu, he might know whats going on [11:51] ack, will do [11:52] I'll send him an e-mail just so I don't forget to ping him on Monday [11:52] a shame he's not coming to Budapest [11:52] oSoMoN, @ airplane mode - not sure, maybe they changed somthing to allow it? (Since you can now use things like that on planes maybe???? Bit of a reach, that one ) [11:52] yeah, will miss him [11:57] seb128: https://bugs.launchpad.net/ubuntu/+source/gnome-online-accounts/+bug/1752472 [11:57] Ubuntu bug 1752472 in gnome-online-accounts (Ubuntu) "Implement an Ubuntu Single Sign-on Provider " [Medium,In progress] [12:00] willcooke, g-c-c won't let me turn the bluetooth on without disabling airplane mode, so it looks like what I experienced was a bug [12:02] ack, thx oSoMoN [12:03] also good morning! [12:03] hey andyrock [12:03] seb128: robert_ancell last night said the GsAuth was created just for us [12:03] so we should able to modify it [12:04] upstream did not answer back [12:04] what does it mean when someone as an [m] next to his/her nickname ? [12:19] willcooke, duflu is not off tomorrow afaik [12:19] willcooke, he was going to swap that day due to travel but since he's not travelling he's not swapping iirc [12:20] andyrock, hey [12:20] andyrock, good for GsAuth [12:21] andyrock, [m] means they are using matrix.org [12:21] seb128: I'm talking with hughsie on #gnome-software right now [12:21] good [12:21] they want to use g-o-a too [12:21] let's see if they agree on what I've in mind [12:22] seb128: did you see the debdiff? I used jbicha trick in order to put binaries in the debdiff [12:27] andyrock, yeah, the debdiff looks code (I didn't do a code review though) [12:27] kk [12:28] andyrock, do we still need the patch from https://bugzilla.gnome.org/show_bug.cgi?id=792865 ? [12:28] Gnome bug 792865 in general "Enable introspection for backend API" [Normal,New] [12:28] just in ubiquity [12:28] k, so not now [12:28] thx [12:28] if we do SSO in ubiquity yes [12:29] not now [12:29] good [12:57] I complained about the Qt transition before it started (but KDE is a bit different) [12:58] we could maybe get pre-approval for the Evolution transition FFe to allow us to just wait until mesa finishes? [13:01] we should have forced the existing transition yesterday despite the clutter issue [13:02] nope [13:02] fixing it was the right way forward [13:02] uploading a KDE transition was the error, it's not fair to criticise fixing bugs for that problem [13:03] right, it just put us in a tricky position now [13:03] being punished for doing the right thing! [13:04] I wonder if we can delete the KDE transition, clear out the existing ones and restore it after [13:05] we can probably skiptest on mesa once some more armhf tests come in [13:05] seb128: I'm not sure how feasible that is since some (many?) of the KDE packages are needed for the Qt transition [13:05] Qt needed a single patch for mesa, not a transition but… [13:06] jbicha, qt was not in that position yesterday? [13:06] You are about to do something potentially harmful [13:06] To continue type in the phrase ‘Yes, do as I say!’ [13:06] hmm, no thanks [13:06] Kubuntu wanted LP: #1749472 before letting mesa migrate [13:06] Launchpad bug 1749472 in mesa (Ubuntu Bionic) "mesa 18.0.0 will cause rendering errors in Qt applications" [Undecided,New] https://launchpad.net/bugs/1749472 [13:07] but they had a qt upload with a fix that didn't involve a transition no? [13:07] like yesterday afaik the only issue was that clutter problem [13:07] otherwise we were good to go [13:08] I don't remember the exact state yesterday, maybe Qt was ready to migrate once mesa was ready [13:09] maybe kdepim is unrelated to the Qt transition, I don't know. Sorry [13:09] no worry [13:09] https://people.canonical.com/~ubuntu-archive/transitions/html/qtbase-abi-5-9-4.html [13:10] oh, qtbase-opensource-src has a lot of red autopkgtests [13:11] I wish Qt had listened to my suggestion 🤷 [13:14] what a mess /o\ [13:15] I'll look favourably on transitions which couldn't go in due to this, and expect others will too [13:15] :/ [13:15] lunch brbbrbr [13:16] Laney, thanks, enjoy lunch! [13:26] Laney: Thank for your help regarding seed snaps. Ubuntu MATE has 3 seeded snaps now. [13:26] All working fine. [13:26] And the Minimal Installtion is hooked up in Ubuntu MATE. Works a treat :-) [13:27] nice! [13:27] flexiondotorg, you don't want to write the patch to make minimal install deal with removing snaps by any chance? ;) [13:28] #careerlimiting [13:28] flexiondotorg, also do you see session timeout or slowdowns as well due to the seeded snaps? === Nafallo_ is now known as Nafallo [13:28] Session timeout in live or installed system? [13:29] Not that I'm aware of. I've only tested clean install on a couple of computer over the last couple of evenings. [13:29] seb128: How are you handling the transition from gnome-calc deb to snap. [13:30] For example, removal of the deb. [13:31] we don't [13:32] for the moment it's only new install [13:32] we need changes to the dist-upgrader [13:32] willcooke, ^ was anyone checking with foundation about that? [13:32] flexiondotorg, timeout on the live session [13:32] flexiondotorg, but could be vm related and increased memory usage hitting configured available ram or something [13:34] I've only tested on real computers. Not noticed any session timeouts. [13:34] k [13:34] thx [13:34] However, snap don't execute in the live session at all. [13:35] seb128, reading [13:37] seb128, we talked about it a few weeks back in the meeting, but nothing since then. I'll add it to the list [13:37] thx [13:38] flexiondotorg, snaps not working in the live session is known and j_dstrand was working on it [13:38] flexiondotorg, jdstrand and zyga are on active work solving that [13:38] flexiondotorg, https://github.com/snapcore/snapd/pull/4714 [13:41] Thanks seb128. I knew it was being worked on. Nice to see a pr. [13:42] yw! [13:52] Someone can sponsor the upload of brlaser for me: bug 1752579 Thanks. [13:52] bug 1752579 in brlaser (Ubuntu) "Needs sponsoring: Upload brlaser 4" [High,New] https://launchpad.net/bugs/1752579 [13:55] Especially I do not know whether I succeeded to subscribe ubuntu-sponsors as LP is running on reduced service, not showing subscriptions and not using pop-up dialogs. [13:56] tkamppeter, you are quite pushy about this one, you asked like 3 times on different channels in an hour [13:56] tkamppeter, and yes sponsors are well subscribed [14:00] seb128: We said in the NYC rally that upgrades were going to be out of scope for 18.04. [14:01] Laney, good then :) (though I wouldn't be surprised if we were asked to reconsider :/) [14:01] but that's a willcooke's thing t figure out [14:02] seb128, I asked on -release first and jbicha asked me to put it on -devel and -devel looked somewhat dead for me no one talking there, not even about other things, so I tried -desktop as I know that there is more talking. And in addition, today is FF. [14:02] right, well people are busy [14:03] and you don't start asking the day of ff usually, plan bit in advance! [14:04] k, upgraded my artful laptop to some parts of bionic and some parts of bionic-proposed [14:04] WCPGW [14:04] seb128, I have checked through possibly overlooked packages already some days ago, but discovered only yesterday that there was a new version of brlaser. [14:06] flexiondotorg: oh, good to know, glad those are generic enough to work for you [14:07] I had to add a few more package for LibreOffice to fully remove itself. libreoffice-math was left dangling on my first attempt. [14:07] flexiondotorg, is that the case on Ubuntu as well? [14:08] I have tested Ubuntu yet, but it is likely. [14:08] jibe_l said he tested that everything got removed [14:08] also libreoffice-math is in the list so that would be weird [14:09] https://bazaar.launchpad.net/~ubuntu-mate-dev/ubuntu-seeds/ubuntu-mate.bionic/view/head:/desktop.minimal-remove#L81 [14:09] That is everything I needed to list for LibreOffice to remove itself completely. [14:10] flexiondotorg, you have -math in your list before and that still installed after install? [14:11] -math was missing from my list first time around. I also needed to remove `uno-libs3` and `ure`. [14:12] https://bazaar.launchpad.net/~ubuntu-mate-dev/ubuntu-seeds/ubuntu-mate.bionic/changes?filter_file_id=desktop.minimalremov-20180226120841-01ibacm71lk1rsd0-1 [14:16] * Laney uninstalls bad ugly libav and manages to play an mp3 [14:16] weeeeeeeeeee [14:20] :) [14:23] flexiondotorg, thanks for giving that feedback, would be even better next time if you sent some merge requests/patches back, you know enough how things work to be able to send your fix back the proper way.. [14:23] also [14:23] cheese / user-accounts is fast with new gstreamer [14:23] flexiondotorg, like you have a commit there that remove shotwell-common in addition of shotwell, that should be pretty obvious that would be something to "upstream" back [14:23] unless this got fixed in some other way [14:25] * didrocks remembers to check everything prefixed with libreoffice- on the image [14:25] seb128: I'm happy to make a merge proposal this evening for the Ubuntu seeds. [14:25] flexiondotorg, that would be good, thanks [14:26] We've only just finished the QA for minimal, so I wasn't able to say for sure if everything was correct until this morning. [14:26] Laney, nice to see that, https://bugzilla.gnome.org/show_bug.cgi?id=783789 is still open and had cheese/g-c-c changes [14:26] Gnome bug 783789 in User Accounts "User accounts panel is slow to open" [Normal,Reopened] [14:26] but maybe new gstreamer helps [14:27] didrocks, from his diff it seems they remove uno-libs3 and ure, so more "disk noise" than anything user visible [14:27] maybe it never happened on my laptop [14:28] didrocks, same for others packages, removing libs or -common in extra [14:28] it stopped happening on this machine for me, but I can test on my other one [14:29] seb128: yeah, some leftover are for sure around, but there wasn't any libreoffice-* for sure [14:29] didrocks, right, we had -math in our list which they didn't for some reason, so NOTOURBUG :) [14:30] ahhhh [14:30] so the premise was wrong, sorry ;) [14:30] but yeah, leftover would have to be cleaned up manually, as we discussed, until we have the new way of doing things (and no request at the last minute ;)) [14:30] * didrocks is sad that deps aren't marked as auto-installed… [14:31] otherwise, minimal install, apt autoremove -> update the list [14:31] would have be a nice workaround [14:38] yeah [15:15] Laney: someone on the forums mentioned that gnome-todo-common was still installed for minimal. Not sure how much we care about extra libraries & data packages like that with our current implementation [15:16] that's what didrocks was just talking about [15:19] jbicha, see backlog and what flexiondotorg wrote, it's true for a couple of libs and -common (like shotwell) [15:19] we don't care much imho but we can as well clean those [15:20] it's going to be a moving target [15:29] is it known that gnome-characters doesn't appear in the shell when searching for it? [15:29] I just verified the package ships a desktop file, and it's not hidden by default [15:31] I'm still on artful (upgrading on Monday). I have it under "utilities" [15:31] oSoMoN, deb or snap? [15:32] deb [15:32] d'oh [15:32] I was searching for "char" [15:32] if I search for "car" then I get it [15:32] ;) [15:32] French localization [15:32] you are french, remember :p [15:33] a little less every day, it seems :) [15:34] that should be a feature request for it to show results for other languages [15:34] English might make sense worldwide. Some countries also have other second languages [15:35] yep, typically here people might want to search both in Spanish and Catalan, as they use both languages (sometimes even in the same sentence) [15:36] oSoMoN: are you in Catalonia then? [15:36] yes [15:36] actually, today marks 10 years of my arrival here :) [15:40] jbicha, did you see bug #1752535? [15:40] bug 1752535 in ubiquity (Ubuntu) "ubiquity crashes in start_slideshow - Could not locate webkit_web_context_get_default: 'webkit_web_context_get_default': /usr/lib/x86_64-linux-gnu/libjavascriptcoregtk-4.0.so.18: undefined symbol: webkit_web_context_get_default" [High,Confirmed] https://launchpad.net/bugs/1752535 [15:45] jibel, ^ is that an error you saw yourself? [15:53] seb128: not before you mentioned it. Strange bug. Didn't have many reports before bionic [15:53] yeah, it's a bit weird [15:53] a missing symbol should impact everyone or nobody [16:02] seb128, no, I tried several times in several ways but didn't reproduce it myself [16:03] k [16:03] thanks [16:52] seb128, bug 1752579 updated. [16:52] bug 1752579 in brlaser (Ubuntu) "Needs sponsoring: Upload brlaser 4" [High,New] https://launchpad.net/bugs/1752579 [17:05] tkamppeter, thx [17:11] woo, thanks jbicha https://gitlab.gnome.org/GNOME/gnome-control-center/merge_requests/2 [17:13] willcooke: the gnome-getting-started videos thing is still on your list somewhere, right? :) [17:18] jbicha, I thought it had moved on since then, and they just needed someone to run the scripts? [17:18] I said I'd do that, but I needed help getting the env set up, and then... things get hazy but I think you said someone else was doing it? [17:26] Hey, anyone know why my laptop would suddenly stop supporting Wayland on bionic? (I was running on Wayland, locked it up and on reboot cannot select a Wayland session) [17:28] alan_g: you're probably using swrast [17:28] for whatever the reason [17:29] meaning the native driver failed [17:29] Never heard of it [17:29] software fallback [17:29] Oh [17:29] Hardware seems fine: Wayland works on Fedora 27 [17:30] Any idea where to look? [17:30] dmesg [17:30] then xorg log [17:33] What am I looking for? [17:34] which gpu? [17:35] intel [17:35] check if i915 is loaded [17:36] Some happy i915 dmesg lines, so I guess it is [17:36] are you logged in on X? [17:37] yes [17:37] check glxinfo [17:38] It spews a load [17:38] What am I looking for? [17:38] never looked at it before? [17:38] No [17:39] glxinfo | grep renderer [17:39] does it say Intel [17:39] llvmpipe [17:40] right [17:40] what about 'apt-cache policy libgl1' [17:40] installed or not [17:42] Installed: (none) [17:42] good [17:42] pastebin xorg log [17:45] Where's that? I don't see it under /var/log [17:45] pastebinit $HOME/.local/share/xorg/Xorg.0.log [17:48] http://paste.ubuntu.com/p/bRDgtxdW38/ [17:49] check the timestamp of that one [17:50] it shows a normal session [17:51] Yeah, that's 3 hours ago [17:51] so it's not your current session then [17:51] http://paste.ubuntu.com/p/NYMY2yc7XJ/ [17:52] apt-cache policy libegl1 [17:52] if you have that installed, remove it [17:52] It is [17:54] libegl1 has a lot of Qt deps [17:55] wonderful [17:55] so that's why kubuntu users got it [17:57] and others too, but I thought it was kubuntu specific after seeing a bug about this [17:57] So, if I install Qt apps it breaks Wayland? [17:58] install libegl1-mesa [17:58] That sort of fits - I installed qterminal [17:58] ack [17:58] it should satisfy the dep [17:58] iirc [17:58] hmm no [17:59] Hm not qterminal, that seems to be OK [18:01] Anyway, thanks tjaalton! I'm back in action [18:01] yw [18:03] Ah! it was the qtwayland5 from proposed. [18:11] oh === alan_g is now known as alan_g|EOD [18:15] and also plasma-framework depends on libegl1 [18:15] willcooke: no one else has volunteered to generate the videos [18:21] jbicha, if you can talk me through setting up the env I'll do it [18:21] jbicha, ah, you know what - lemme try and do it in the morning. HHCIB [18:21] I'm ready for EOD now [18:23] willcooke: I've never done it before. talk to jimmac or pmkover or ask for help on the bug [18:24] jbicha, will do [18:24] pmkovar [18:30] Hi everyone, i need working 18.04 build can someone please share with me? [18:31] Pharma_, http://cdimage.ubuntu.com/daily-live/current/ [18:31] Is not working [18:31] both 1st of march and 28 of feb [18:31] error on installation and i think there is bug report on launchpad [18:34] I had image from ~26 of feb that one worked but i overwrote it, and my current setup died when installed nvidia driver [18:34] hmm, 20th of February worked... [18:35] I've got a 15th Feb version here [18:35]