[00:06] robert_ancell, hey, when running gnome-software from code, is there any special one needs to do to get polkit permissions to work with the snapd plugin ? I get things like "Requires authentication with @snapd" [00:28] ahayzen, no, it should be automatic. [00:29] ahayzen, what versions of g-s/snapd etc? [00:31] robert_ancell, gnome-software master and snapd 2.42.1, this in on fedora 31, as i couldn't get gnome-software master to build on ubuntu/debian before :-) (maybe it can now though) [00:34] ahayzen, can you try: [00:34] $ curl --unix-socket /run/snapd.socket http://localhost/v2/logs [00:34] $ curl --unix-socket /run/snapd.socket -H 'X-Allow-Interaction: true' http://localhost/v2/logs [00:34] The former should fail with an auth error, the latter should prompt for a Polkit auth [00:34] The command is otherwise safe and will just return some system logs [00:35] yup those work :-) [00:35] All G-S does is set the HTTP headers, and snapd does the Polkit auth. Perhaps it's not working in Fedora for some reason [00:36] yeah, must be broken somehow. I'll submit my code change i have anyway as WIP as most of it i can test, and you can tell me if it's crazy :-) [00:37] sure, @ me on the MR and I'll have a look. [00:38] thanks! :-) === Fudge is now known as Guest75415 [05:51] morning everyone [05:52] Morning jibel [05:53] * duflu just noticed his drink coaster is a Red Hat CD from 1997 and smiles === duflu changed the topic of #ubuntu-desktop to: Home of the Desktop Team, https://wiki.ubuntu.com/DesktopTeam | For help or questions, try #ubuntu | Work (read-only for non-developers): https://trello.com/b/mxaCZTVc [06:02] Hey duflu [09:02] moin [09:03] feel like it's going to be a quiet one :> [09:05] Pin drops [09:05] Morning Laney [09:10] good morning Laney [09:10] glad you bring some animation [09:14] guten morgen duflu und jibel [09:14] wie gehts? [09:15] es geht mir gut, danke [09:19] auch gut. ich freue mich auf die weichnachtspause! [09:19] und dir? [09:19] * Laney drags things out of ancient brain compartments [09:20] umm, same [09:21] my German is too rusty [09:35] mine too. I studied german for 7 years and never used it [09:38] same [09:38] will get the chance in 2020 though! [10:16] I even worked for a german company and never spoke a single word of german [10:18] no wonder tests were passing, I was testing the wrong code :((( [10:18] I need holidays [11:36] I will be in Monday and Tuesday. For anyone I don't see then, happy holidays === hggdh is now known as QuosqueTandem [15:16] getting all caught up on USN refreshes before the EOY [15:18] hey seb128 [15:18] is seb128 here today? [15:18] he just joined :) [15:19] he's not supposed to be here [15:19] hey [15:19] * kenvandine glares at seb128 [15:19] I don't see joiners [15:19] I travelled on wednesday and said I wouls swap some hours [15:19] salut seb128 [15:19] lut jibel [15:19] hey kenvandine [15:19] what's up? [15:19] getting caught up on all these USN refreshes :) [15:20] not much, very quiet day, fixing the testsuite of ubuntu-release-upgrader before holidays [15:20] holiday coming \o/ [15:20] seb128: i don't think i can get to grilo [15:20] man need it so much [15:20] moin seb128 kenvandine [15:20] +I [15:20] gotta finish these refreshes and a couple other bits of and bobs before the new year [15:20] kenvandine, no worry, as I said it can wait for after holidays [15:20] hey Laney [15:21] sorry... last minute push :) [15:21] kenvandine, I noticed the trello spam, that's still not fixed :p [15:21] lol [15:48] good morning folks! === ginggs_ is now known as ginggs [17:58] how do i restart multipassd again? [17:59] answer: sudo snap restart multipass [18:01] grumble grumble multipass is still taking forever to start a vm and then fails saying it can't find snap/snapcraft.yaml :( [18:02] ok there it goes [18:02] finally [18:20] finishing the year on an upload that doesn't build anywhere /o\ [18:20] https://launchpad.net/ubuntu/+source/mutter/3.34.2-1ubuntu1 [18:20] sounds about right [18:21] works on my system [18:21] fml [18:22] will have to hope I get a chance to look into it [18:22] BYEEEEEEEEEEEEE [18:22] cheers!! [18:22] happy holidays :) [18:25] Laney, enjoy the holidays! [18:35] tjaalton, the liglvnd update seems to make things angry, looks like the -dev doesn't bring libGL anymore which fails random autopkgtests and builds [18:38] seb128: things moved to libgl-dev etc [18:40] could be they were tested against new mesa + old glvnd [18:40] glvnd seems to be out of new now [18:41] tjaalton, see the libglvnd section on https://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses_by_team.html#desktop-package [18:41] yes, I newed it today [18:41] tjaalton, that sounds weird, new mesa was waiting on glvnd to be newed to build [18:41] but yeah, you are right, autopkgtest only pick the one package it tests from proposed [18:42] should have both [18:42] if they need to go in lock step then you miss a Breaks? [18:42] maybe [18:42] or versioned depends [18:44] libgl-dev has breaks on old mesa [18:44] etc [18:44] and replaces [18:45] so it's new glvnd and old mesa [18:48] tjaalton, unsure why it manages to try those autopkgtests with the old mesa then [18:48] tjaalton, anyway either we need to fix the dependencies for that to not happen [18:48] or we need to manually retry those with a trigger on the new mesa version [18:49] I don't see what to change without making circular deps [18:54] tjaalton, well, https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-focal/focal/amd64/d/ddnet/20191220_164053_023ce@/log.gz [18:54] tjaalton, that installs libglvnd-dev and no libgl-dev so maybe that one should also have the Breaks to force the mesa upgradE? [18:55] that's just a broken package, libglvnd-dev never provided libGL.so [18:56] nah, it installs libgl1-mesa-dev which used to provide it [18:57] new version of it depends on libgl-dev now, this test installed the old version [18:58] ah actually [18:58] libglvnd-dev did provide the .so, but not headers [19:00] so adding the lib{egl,gl,gles}-dev pkgs as libglvnd-dev depends should fix it [19:00] new mesa isn't actually needed for this [19:00] sounds like tyhe right way to fix it [19:00] it's just that something should pull in libgl which doesn't atm [19:02] right [19:03] didn't realize so many pkgs had actually started to build-depend on libglvnd-dev instead of mesa-common-dev etc [19:03] or any at all.. [19:04] anyway, I'll push a new one to debian [19:27] this is probably something to do with why mutter fails too [19:27] so I hope for it to magically fix itself :> [19:39] tomorrow-ish [23:22] ok i'm off. see yall next year!!