[03:58] morning [04:31] hmmm `Skipping non-snapd systemd unit snap.lxd.workaround.service` in postrm-purge breaks the test, however where does the unit come from? [06:06] pstolowski: hey [06:06] hey mborzecki [06:07] pstolowski: tests are failing 😕 there's a weird thing in postrm-purge and user session socket is flaky but sergio opened https://github.com/snapcore/snapd/pull/10556 for this [06:10] mborzecki: oh.. the revert of services change landed, related? [06:10] pstolowski: would that create snap.lxd.workaround.service file? [06:12] mborzecki: no [06:29] mborzecki: google:ubuntu-20.04-64:tests/main/snap-services passed for me [06:29] or do i need to run both in sequence? [06:29] pstolowski: nah, i don' think there's anything related to snap services [08:22] pstolowski: looks like there is one comment for 10551, are you on it or should I add the todo? I'm keen to land this so that I can branch 2.52 :) [08:23] mvo: the TODO from Samuele? i'll add in a sec [08:28] mvo: updated [08:37] pstolowski: ta! [09:46] meh, i can't get batch delete from bpf map to work, funny that deleting elements one by one works though [11:30] morning folks [11:30] morning ijohnson[m] [11:31] hey pstolowski [11:31] ijohnson: hey [11:31] I am hopeful with slightly less meetings today I can get my review finished for the auto-refresh gating hook thing [11:31] hey bboozzoo [11:31] how has this week been for y'all [11:37] ijohnson: rather silent, some tests are failing again, the postrm-purge is in particular interesting, as there's a transient systemd unit for which it isn't clear where it comes from [11:38] bboozzoo: is it one our transiennt units we create to run snap apps/hooks ? [11:38] I saw an update yesterday to this bug where it was fixed by a systemd patch someone has [11:38] ijohnson: it's called snap.lxd.workaround.service and it executes /bin/true ;) [11:38] ijohnson: https://paste.ubuntu.com/p/z32jQzcDn6/ [11:39] this one [11:39] https://bugs.launchpad.net/snapd/+bug/1928806 [11:39] bboozzoo: ah yeah that does look just like the sort of transient scope that is left around in the bug ^ [11:41] ijohnson: but those are clearly scopes, probably from runs of kubectl [11:41] I assume lxd runs this somehow as well [11:41] this one looks like it was added on purpose, with Deletegate=yes and RemainAfterExit [11:42] oh I missed the `RemainAfterExit=yes` [11:42] hmm [11:43] i'm slightly worried by snap-services test failures, a bit odd it started around landing the changes around this area, but then it doesn't fail everywhere [11:44] pstolowski: do you have logs for an example failure ? [11:45] ijohnson[m]: yes, e.g. https://github.com/snapcore/snapd/pull/10551/checks?check_run_id=3142648936 (it was failing on master already, so it's not specific to this PR) [11:45] hm i think i know the culprit now [11:45] ijohnson[m]: there is a PR from cachio for this: https://github.com/snapcore/snapd/pull/10556 [11:46] ah right that issue [11:46] yeah I think that's been around for a few weeks [11:47] ah ok, i didn't see it.. good (sort of) [11:55] ok, i've updaed my standup notes and i'm wrapping it up for the week, got some driving to do [11:55] bboozzoo: have a nice weekend, you're back on Tuesday ? [11:56] ijohnson: thanks, yes back on tue [11:56] nice, see you then [11:57] mborzecki: have fun! === dob1_ is now known as dob1 [17:51] * cachio_ afk === zyga__ is now known as zyg === zyg is now known as zyga === clobrano1 is now known as clobrano