[00:11] PR snapcraft#4079 closed: update craft-archives and support apt pinning [06:34] PR snapd#12671 opened: cmd/snap-bootstrap: Add support for essential snaps with integrity data (run mode) [11:15] PR snapd#12668 closed: image: fix odd output of 'Fetching ...' text [15:08] Hey folks. ogra you about, by any chance? Still looking to learn a bit about `refresh-mode: endure` [15:09] kyrofa, well, that is (to my knowlede) only for services where you knoe it will not do any harm ... i.e. someting tht doesnt use SNAP_DATA for important things but only SNAP_COMMON etc etc [15:09] *know [15:09] Okay so no magic as far as you're aware [15:09] the confinement should not get in your way here since your old namespace should stay around [15:10] Is that a recent development? The entire reason for app refresh awareness was because that WASN'T the case [15:10] yeah, no magic ... you should really know what you are ding when using it [15:10] *doing [15:11] the app refreh awareness was mainly because your dirs change underneath you as i understood ... so SNAP_DATA becomes un-writable [15:11] which essentially means your browser history/cache that lives in there can not be updated [15:12] the rest of confinement suld not be affected as long as the app is running [15:12] *should [15:12] * ogra needs a new kbd ... [15:12] (or new fingers) [15:14] My understanding was not that the dirs changed, but that your access to the old dirs was no longer possible due to the confinement update [15:15] Indeed, the dirs don't change at all. A new one is created and the current symlinks are updated it all [15:16] right ... [15:17] If rev 1 updates to rev 2, rev 1's data is still there, but the confinement now only allows access to rev 2, so rev 1's app explodes [15:17] you lose access ... thats the point [15:17] Right. Maybe we're talking past each other-- I was trying to say your old namespace doesn't stay around, but maybe you weren't saying what I thought you were saying there [15:17] but only on a filesystem level, the runtime stuff in ram persists just fine [15:17] Ah, indeed [15:18] if you have a service that doesnt use the disk at all or only uses _COMMON, endure will actually wrok well [15:18] *work [15:20] Any chance you know the history? That seems highly specific [15:20] I'm curious in the use-case that caused this development [15:22] Ah ha, I found https://forum.snapcraft.io/t/process-lifecycle-on-snap-refresh/140/20 [15:23] IIRC chromium [15:24] oh, you mean endure ... not sure, but IIRC a customer asked for it [15:26] Thanks ogra, I appreciate the help. Always a pleasure :) . I hope you're well? [15:27] yeah, i am 🙂 [15:27] you too ? [15:27] Yeah, things are good! Baby #6 is due here pretty soon-- our second girl! [15:28] OMG ! [15:28] half a football team ! [15:28] Yeah I'll get there one day [15:28] :P [15:28] lol [21:58] taking a look at https://bugs.launchpad.net/snapd/+bug/1998980/comments/4 [21:58] Bug #1998980: Snap sandbox blocks kcmp, used by Mesa [21:58] I am unsure where to look for "base template for all snaps" [22:00] is it https://github.com/snapcore/snapd/blob/master/interfaces/seccomp/template.go ? [22:13] PR snapd#12672 opened: interfaces/seccomp/template,Adding kcmp to allow Mesa usecases [22:33] PR snapd#12672 closed: interfaces/seccomp/template: Adding kcmp to allow Mesa usecases [22:33] PR snapd#12673 opened: interfaces/seccomp/template: Adding kcmp to allow Mesa usecases [23:08] PR snapcraft#4081 opened: Kernel plugin improvements 1