[04:59] morning [05:04] mborzecki: 'morning! [05:04] mardy: hey [06:19] morning [06:45] pstolowski: hi! [06:50] pstolowski: heya [06:51] PR snapd#10640 opened: cmd, packaging: import BPF headers from kernel, detect whether host headers are usable [06:53] Just saw this post on Twitter about a snapd thread waking every 20 microseconds: https://twitter.com/rbtcollins/status/1428237490171912197 -- is there anything obvious on our side that could be causing that? [06:58] jamesh: i don't think there is anything obvious, don't think we do anything with such high frequency [06:59] pstolowski: yeah. The only ticker I could see using a timeout measured in microseconds was the systemd watchdog, which should only be firing every 2.5 minutes [07:01] PR snapd#10641 opened: packaging: remove TEST_GITHUB_AUTOPKGTEST support [07:06] mardy: pstolowski: 10640 i quite simple, mostly code imports and a bit of autotools, maybe you'd like to take a look? [07:09] jamesh: i don't see it on my box (but i booted my box 1h ago, maybe it needs time) [07:17] jamesh: i replied to the tweet, maybe snap debug stacktraces will shed some light [07:20] pstolowski: jamesh: and as expected snapd and its threads are doing literally nothing here, all threads are nicely asleep [07:20] yeah i didn't see snapd at all in powertop [07:21] something went astray there [07:21] maybe an ongoing change? [07:22] or something about snapd being reexeced and the actual binay being produced by an older go (mine was built with go 1.16.7 [07:22] mborzecki: he said no FDs, i guess we would see some? [07:23] pstolowski: hm I see 11 fds open here (incl std* ones) [07:23] yeah weird [07:41] PR snapd#10642 opened: interfaces/u2f-devices: add Nitrokey FIDO2 [08:02] mborzecki: approved [08:11] thanks [08:11] PR snapd#10643 opened: testutil: add DeepContainerMatches Checker [08:28] good morning :) [08:28] zyga: hi! [08:29] 2nd? :D [08:29] more IRC on extra computers? [08:35] zyga: yes, now in the park with my laptop [08:36] too lazy to logoff from my desktop pc at home :-) [08:52] mardy_2nd, nice :) [10:30] what is `snapst.Sequence` in this context? https://github.com/snapcore/snapd/pull/10473/files#r691178628 [10:30] PR #10473: interfaces/builtin: add mount-control interface [11:12] PR snapd#10644 opened: [RFC] o/snapstate: don't hold some snaps if not all snaps can be held by the given gating snap [11:12] mardy: snapst.Sequence is an array of side infos corresponding to all installed revisions of the given snap (and current index points to one of them) [11:12] mardy: we sometimes examine its length e.g. when removing the snap to know if we are dropping the snap completely [11:12] mardy: is this what you were asking for? [11:13] pstolowski: exactly this, thanks a lot :-) [11:13] now it's clear [11:28] meh, found a weird bug (problem?) where snap run cannot create a tracking scope for the snap application when running as root on fedora [11:28] idk why [11:29] but the side effect is that the process stays in the user's scope (user-0) and then s-c proceeds to attach a program to that cgroup which really affects the user's cgroup :/ [11:29] what a mess [11:32] pffff [11:32] i think i know what's wrong [11:32] still, a mess [11:47] PR snapd#10645 opened: secboot: switch main key KDF memory cost to 32KB <:rocket: bootspeed> [11:56] mvo: can you land https://github.com/snapcore/snapd/pull/10546 (unrelated failures)? [11:56] PR #10546: daemon, o/snapstate: handle IgnoreValidation flag on install (2/3) [12:00] https://bugs.launchpad.net/snappy/+bug/1933699 looks like a problem with translations, those are stored in lp too right? [12:00] Bug #1933699: main.go:176: description of prepare-image's "" [12:02] since it's about a pt_BR translation maybe miguelpires you could take a look at ^^ whether it even makes any sense? [12:11] Bug #1930175 changed: Translation error in pt_BR [12:11] Bug #1938921 changed: Snap binaries are only available for user account and not root account on Raspbian [12:28] mborzecki: Sure :) seems like there's description starting with lowercase but it's otherwise correct. Where do we keep the translations? I see some under po/ but it doesn't contain this [12:28] miguelpires: i belive they are here: https://translations.launchpad.net/snappy [12:47] Bug #1932129 changed: `snap info snapd` refers to freenode [12:47] Bug #1933394 changed: PANIC description of prepare-image's "" is lowercase in locale "pt_BR" [12:47] Bug #1938469 changed: snapd + snap core stopped working on Fedora [12:50] Bug #1932129 opened: `snap info snapd` refers to freenode [12:50] Bug #1933394 opened: PANIC description of prepare-image's "" is lowercase in locale "pt_BR" [12:50] Bug #1938469 opened: snapd + snap core stopped working on Fedora [12:53] Bug #1932129 changed: `snap info snapd` refers to freenode [12:53] Bug #1933394 changed: PANIC description of prepare-image's "" is lowercase in locale "pt_BR" [12:53] Bug #1938469 changed: snapd + snap core stopped working on Fedora [13:12] PR snapd#10546 closed: daemon, o/snapstate: handle IgnoreValidation flag on install (2/3) [13:13] mvo: can we land https://github.com/snapcore/snapd/pull/10632 ? [13:13] PR #10632: spread: temporarily fix the ownership of /home/ubuntu/.ssh on 21.10 [13:15] thanks! [13:17] PR snapd#10632 closed: spread: temporarily fix the ownership of /home/ubuntu/.ssh on 21.10 [14:02] PR snapd#10646 opened: tests: fix services-refresh-mode test [14:19] mborzecki, I see in the test security-dev-input-event-denied that after snap connect test-snapd-event:joystick and udevadm settle [14:20] sometimes the execution of test-snapd-event "-event-kbd" does not give an error [14:20] perhaps it is a timing issue [14:23] I'll try to reproduce the error but adding a sleep after settle [14:43] PR core#122 opened: hooks: replaces symlinks in /usr/share/doc with the actual files [14:55] cjp256, hi there! I've seen this weird snapcraft crash: https://paste.ubuntu.com/p/YbZSKV99wT/ - any hint on what might be happening there? [14:56] * abeato moving the question to the snapcraft channel... [14:56] abeato: looks like a non-utf8 character in teh yaml? [14:58] err nvm, that's not from the yaml :thinking: [15:32] PR snapd#10637 closed: tests: remove the test user just when it was installed on create-user-2 test === King_InuYasha is now known as Conan_Kudo === Conan_Kudo is now known as King_InuYasha [16:19] PR core20#109 opened: hooks: add bootchart configuration [16:38] PR core#123 opened: tests: move from travis.yaml to github actions [17:18] PR snapd#10645 closed: secboot: switch main key KDF memory cost to 32KB <:rocket: bootspeed> [17:58] PR core#123 closed: tests: move from travis.yaml to github actions [18:03] PR core#124 opened: workflow: add basic smoke test [18:18] PR snapd#10647 opened: o/snapstate: fail remove if snap can't be found [19:48] PR snapd#10648 opened: .github/workflows: add codedov again [20:23] PR snapd#10619 closed: secboot: use half the mem for KDF in AddRecoveryKey <:rocket: bootspeed> [20:54] PR snapd#10649 opened: release: 2.51.6 [22:04] PR snapd#10649 closed: release: 2.51.6 [22:09] PR snapd#10650 opened: packaging: merge 2.51.6 changelog back to master [22:14] PR snapd#10651 opened: packaging: merge 2.51.6 changelog back to 2.52 [23:29] PR snapd#10652 opened: tests: fix timing issue on security-dev-input-event-denied test