[05:06] goood morning :) [05:07] * zyga-mbp spent all of yesterday deep inside the cloud but is back now [05:47] zyga-mbp: 'morning! [05:47] o/ :) [05:48] morning [05:48] mardy: zyga-mbp: hey [05:59] hm opensuse.org is down [06:03] yeah, it looks down from here as well [06:14] PR snapd#10837 opened: cmd/libsnap-confine-private: fix set but unsed variable in the unit tests [06:14] trivial PR ^^^ [06:15] and should make clang 13 happy, apparently it's more picky now [06:18] mborzecki +1 [06:35] ehh so https://forum.snapcraft.io/t/systemd-services-running-snap-apps-fail-to-associate-journal-logs/26714 someone tries to use a snap application as a service (defined separately, outside of snapd's knowledge), then tracking gets in the way and moves the app (since it's an app) to a separate sope [06:55] woah, opensuse landed a rebuild with new glibc too [07:01] mborzecki: hi! [07:01] morning [07:03] good morning pstolowski and mardy [07:03] pstolowski: mvo: hey [07:04] hey mborzecki [07:08] mborzecki ouch, interesting [07:08] mborzecki there's a gnome protocol for app tracking, also based on scopes [07:08] I guess it just doesn't matter there because it's a part of a desktop app [07:09] perhaps app tracking should bail-out if invoked in a systemd service context? [07:10] that would break the user, perhaps we shoudl just show a warning and continue [07:21] mvo: hi! [07:21] I need to leave for about 1 hour [07:24] PR snapd#10838 opened: Component tests [07:29] PR snapd#10831 closed: osutil/disks: add new `CreateLinearMapperDevice` helper [07:34] PR snapd#10834 closed: tests: fix netplan test on i386 architecture [07:34] PR snapd#10835 closed: tests: fix lxd-mount-units test which is based on core20 in ubuntu focal system [07:51] mvo: hi, can you merge this https://github.com/snapcore/snapd/pull/10812 please? [07:51] PR #10812: o/ifacestate: don't fail remove if disconnect hook fails [07:56] mardy: hey, can you take a look at https://github.com/snapcore/snapd/pull/10737 again? [07:56] PR #10737: o/snapstate: validation sets enforcing on update

[08:03] quick errand, back in 30 [08:35] pstolowski: approved -- I also have a question, but it's just to make sure the branches are intentional [08:40] mardy: can you take another look at https://github.com/snapcore/snapd/pull/10826 ? [08:40] PR #10826: cmd/libsnap-confine-private: workaround BPF memory accounting, update apparmor profile

[08:42] heh native mattermost app is kinda broken, whenever i click on a link it tries to open it in firefox, but in such a way that firefox shows a dialog that profile is in use [08:46] mborzecki: approved [08:50] had to restart the CI job in #10837, seems like there were network problems [08:50] Bug #10837: Segfault when try to print using cups. [08:50] PR #10837: cmd/libsnap-confine-private: fix set but unsed variable in the unit tests [08:50] mardy: thanks! [08:58] damn, this new github ui is confusing indeed [08:59] hahah [09:09] PR snapcraft#3584 opened: extension/desktop: Prevent keeping updating when there is no user use… [10:28] mborzecki nice work on that .1 [10:29] I'll happily review, I've been following that thread [10:30] hm? [10:40] PR snapd#10839 opened: cmd/snap-confine: update s-c apparmor profile to allow versioned ld.so [10:41] yes, this :) [10:43] done [10:50] ah, haha [11:10] PR snapd#10837 closed: cmd/libsnap-confine-private: fix set but unused variable in the unit tests [11:16] mvo: would you be able to merge https://github.com/snapcore/snapd/pull/10571? It has received 2+ reviews, but I'm getting some unrelated CI failures that prevent me from landing it myself. [11:16] PR #10571: daemon: implement access checkers for themes API [11:16] jamesh: sure [11:17] mvo: most recent CI run is from today after merging master [11:58] hello, is it possible to disable ssl_verify for a `snapcraftctl pull` in a snapcraft.yaml ? [11:59] ijohnson: thanks for looking at kdenlive for me. looking in the journald output I see the supporting binary kdenlive_renderer is crashing after it fails to load /snap/core20/.../libacl etc and I've no idea why it would be loading it from there instead of /usr/lib/libacl etc there's various scripts in the command chain that alter lots of environment variables so meh maybe I'll have to play with lots of those and see what happens [12:01] https://invent.kde.org/packaging/snapcraft-kde-applications/-/blob/Neon/release/kdenlive/kdenlive-launch/kdenlive-launch is the only one specific to kdenlive but I don't think setting those would cause the issue [12:10] PR snapd#10788 closed: many: rename systemd.Kind to Backend for a bit more clarity [12:12] hm did we land a change in spread to run with a single worker? [12:19] does anyone remember the trick for time.Time mocking in the tests to make DeepEquals happy about time fields and not choke about its internal bits that differ? [12:21] pstolowski, I only did this once but IIRC you can round the nanonseconds/internal noise [12:21] and then perform the comparison [12:24] zyga: will try, thanks... i've a vague memory there was something else but completely forgot [12:50] PR snapd#10840 opened: tests/main/security-device-cgroups: fix when both variants run on the same host [13:13] hmm [13:14] mborzecki, not signed into github and away from token but https://github.com/snapcore/snapd/blob/f33c0179a4bef62c224160d77c996e5f9307d6ca/tests/main/cgroup-tracking/task.yaml#L77 [13:14] mborzecki, nitpicky a bit but we compare systemd version in one more place [13:15] and I wanted to make sure you were at least aware of the way [13:41] mardy: let's sync monday morning about the app tracking stuff [13:41] maybe it's actually one problem [13:54] mborzecki: OK! [13:55] i suspect it's ok for normal user, but maybe not so ok for root [14:06] PR snapd#10841 opened: interfaces/dsp: add a usb rule to the ambarella flavor <⚠ Critical> [14:09] mborzecki root doesn't have app tracking, we equate root to services, no? [14:10] zyga: not really iirc [14:10] hmm, perhaps I remember wrong then [14:11] PR snapd#10842 opened: interfaces/apparmor/template.go: allow inspection of dbus mediation level [14:11] ah right https://github.com/snapcore/snapd/blob/master/cmd/snap/cmd_run.go#L1158 [14:16] PR snapd#10839 closed: cmd/snap-confine: update s-c apparmor profile to allow versioned ld.so [14:16] yay [14:18] hm we just continue if tracking could not be established, which probably is ok [14:18] i should make thhe devie cgroup code verify that we're in a scope/service of a snap, i.e. there's a `snap.` prefix and `.scope` || `.service` suffix [14:18] Jonathan Esk-Riddell: maybe something to discuss in the snapcraft clinic today [14:19] mborzecki, mmm [14:19] yeah [14:19] that is a good idea [14:19] those are not synchronized today [20:32] PR snapd#10840 closed: tests/main/security-device-cgroups: fix when both variants run on the same host [20:54] diddledani: hello :) is there a way to go from https://snapstats.org/snaps/multipass to a list of releases for this snap?