| === popey7 is now known as popey | ||
| transhumanist | hi how do I start a snap from a terminal if I have multiple isolations of the same snap? https://snapcraft.io/docs/parallel-installs ? | 02:38 |
|---|---|---|
| transhumanist | or create shortcuts to them on the desktop for that matter? | 02:39 |
| mborzecki | morning | 05:51 |
| mup | PR snapd#11695 closed: libsnap-confine-private: show proper error when aa_change_onexec() fails <Simple 😃> <Created by mardy> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/11695> | 06:45 |
| mardy | hi all | 06:51 |
| pstolowski | morning | 07:00 |
| mup | PR snapd#11566 closed: features: enable refresh-app-awareness by default <Squash-merge> <Run nested> <cherry-picked> <refresh app awareness> <Created by stolowski> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/11566> | 07:00 |
| mup | PR snapd#11690 closed: o/snapstate: exclude TypeSnapd and TypeOS snaps from refresh-app-awareness <Needs Samuele review> <cherry-picked> <refresh app awareness> <Created by stolowski> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/11690> | 07:05 |
| mup | PR snapd#11097 closed: interfaces: add ACRN hypervisor support <Squash-merge> <Needs Samuele review> <Created by igorljubuncic> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/11097> | 07:15 |
| zyga[m] | good morning | 07:20 |
| mardy | hi all | 09:02 |
| pstolowski | hey mardy | 09:12 |
| mup | PR snapd#11698 opened: o/snapstate: fix TestSnapdRefreshTasks test after two r-a-a PRs <⚠ Critical> <Simple 😃> <Skip spread> <Created by stolowski> <https://github.com/snapcore/snapd/pull/11698> | 09:16 |
| mup | PR snapd#11699 opened: interfaces/builtin/system-packages: use a broad AppArmor snippet for mounting /usr/share/*-docs <Simple 😃> <Needs security review> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/11699> | 09:31 |
| mup | PR snapd#11698 closed: o/snapstate: fix TestSnapdRefreshTasks test after two r-a-a PRs <⚠ Critical> <Simple 😃> <Skip spread> <Created by stolowski> <Merged by stolowski> <https://github.com/snapcore/snapd/pull/11698> | 09:46 |
| bdmurray | mardy: I added you to the group of people who can see crash reports in the Error Tracker | 09:58 |
| mardy | bdmurray: saw it, thanks! | 10:17 |
| mup | PR snapd#11700 opened: sandbox: move profile load/unload to sandbox/apparmor <Simple 😃> <Created by mardy> <https://github.com/snapcore/snapd/pull/11700> | 10:31 |
| mup | PR snapd#11701 opened: tests/nested: fix custom-device test <Simple 😃> <Run nested> <Created by mardy> <https://github.com/snapcore/snapd/pull/11701> | 12:02 |
| mup | PR snapd#11456 closed: tests: Initial changes to run nested tests on uc22 <Run nested> <Created by sergiocazzolato> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/11456> | 12:37 |
| mup | PR snapd#11622 closed: image/preseed, cmd/snap-preseed: create and sign preseed assertion <Preseeding 🍞> <Run nested> <Created by stolowski> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/11622> | 12:37 |
| jdstrand | ogra: hey, fyi, zoom-client 5.10.3.2778 fails to launch for me, no sandbox denials. I had to snap revert. unfortunately I don't have time now to debug but did strace and found: write(2, "[0421/074431.324810:FATAL:platform_shared_memory_region_posix.cc(255)] This is frequently caused by incorrect permissions on /dev/shm. Try 'sudo chmod 1777 /dev/shm' to fix.\n", 175 <unfinished ...> | 12:46 |
| ogra | jdstrand, reboot ... | 12:47 |
| jdstrand | ogra: I thought I read something about new snapd /dev/shm handling. I have snap 2.55.3 | 12:47 |
| ogra | that seems to be a snapd issue that solves itself after rebooting | 12:47 |
| ogra | i have seen it on one out of five machines yesterday | 12:47 |
| jdstrand | oh, that is super weird. I hope they're looking at that... | 12:48 |
| * jdstrand wonders if snap-discard-ns is enough | 12:48 | |
| * jdstrand tries | 12:48 | |
| jdstrand | (I have a lot of session state atm) | 12:48 |
| ogra | i havent filed it et | 12:49 |
| ogra | *Yet | 12:49 |
| jdstrand | ok, if I 'sudo /snap/snapd/current/usr/lib/snapd/snap-discard-ns zoom-client' then do 'sudo rmdir /dev/shm/snap.zoom-client/' then I don't have to reboot. I can say that the snap-discard-ns by itself is not enough to fix it. I don't know if the rmdir alone would fix it | 12:51 |
| jdstrand | ogra: thanks for the pointer! :) | 12:51 |
| ogra | thanks for te test ! | 12:52 |
| ogra | the rmdir alone didnt help for me | 12:52 |
| ogra | i tried that | 12:52 |
| jdstrand | boy, this bug makes me feel all those browser apps might be affected since I know they use /dev/shm | 12:53 |
| ogra | well, i think it is unusual that apps and snapd switch at the same time to it ... but yeah, if you hit that corner case ... | 12:58 |
| jdstrand | maybe, but if people refresh, say, only on the weekends, etc... | 12:59 |
| ogra | well, i mean devs ... i dont think any of our browsers use the shared-memory interface yet | 13:00 |
| jdstrand | oh I see | 13:00 |
| ogra | i was ahead with zoom since it actually completely switched its backend and hard-required a private /dev/shm | 13:01 |
| jdstrand | this makes things a little clearer on why things might fail if you go from no shared-memory to using shared-memory | 13:04 |
| ogra | right and both snaps upgraded at the same time | 13:04 |
| ogra | i think there is some hidden race somewhere | 13:04 |
| jdstrand | mardy: fyi ^. also, a snap using shared-memory (eg, the new zoom-client revision), the /dev/shm dir has world-write permissions: drwxrwxrwx 2 root root 40 Apr 21 07:53 /dev/shm/snap.zoom-client | 13:05 |
| mup | PR #21: Drop the NotFoundHandler check that relied on undefined behaviour <Created by chipaca> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/21> | 13:05 |
| jdstrand | mardy, ogra (cc amurray and mvo): here is the security bug https://bugs.launchpad.net/snapd/+bug/1969777 | 13:15 |
| mup | Bug #1969777: new shared-memory interface causes /dev/shm/<snap> dir to have world-write permissions <snapd:New> <https://launchpad.net/bugs/1969777> | 13:15 |
| jdstrand | ogra: I referenced the problem we discussed in it | 13:16 |
| ogra | jdstrand, well, the point is there was no "before" zoom has not used /dev/shm in former versions IIRC | 13:18 |
| jdstrand | ogra: I can rephrase. that directory was present with 170 though | 13:36 |
| ogra | hmm | 13:36 |
| jdstrand | that's just part of the launcher sequence | 13:37 |
| jdstrand | iirc | 13:37 |
| jdstrand | well, now I don't remember how that all works otoh, *but* that directory was present, was 0755 and I had to rmdir it | 13:38 |
| pikapika | Does snap follow apt's orders regarding when to update, apt freeezes etc or does it have its own mind? | 15:37 |
| Wimpy | pikapika: snap doesn't interpret apt directives. It is a completely different system. | 15:40 |
| pikapika | Wimpy, how do we control a given snap | 15:40 |
| pikapika | ie can we tell it to only upgrade manually | 15:40 |
| Wimpy | What do you mean by control? | 15:40 |
| pikapika | and where is it's configuration stored | 15:40 |
| Wimpy | No, snaps always auto update. | 15:40 |
| pikapika | How can that be disabled? | 15:41 |
| Wimpy | You can defer updates for a while, but after 30 days they will update. | 15:41 |
| pikapika | I see | 15:41 |
| Wimpy | It can not be completely disabled. | 15:41 |
| ogra | you can delay updates by 30 days as Wimpy said ... but the new desktoip will have additional integration | 15:42 |
| pikapika | ogra, where is a snaps .config folder | 15:43 |
| ogra | you get a notification about upgrades ... and eventually you will also be able to manage them | 15:43 |
| ogra | pikapika, ~/snap/<snapname>/current/ | 15:43 |
| ogra | pikapika, or if you mean system wide you use "snap set system ..." https://snapcraft.io/docs/system-options | 15:45 |
| pikapika | ogra, is snap set system analogous to etc/ ? | 15:46 |
| ogra | it is for global snap serttings | 15:47 |
| pikapika | well | 15:47 |
| pikapika | it doesn't seem to be app specific | 15:48 |
| pikapika | so its not analogous | 15:48 |
| ogra | no, thats up to the app packagers | 15:48 |
| mardy | jdstrand: thanks, indeed that's not right | 16:00 |
| jdstrand | ogra: fyi, the 'Launch meeting' functionality from the browser doesn't work right in zoom-client (separate from the other issue). (This is where you have a zoom url in say your google calendar, click it, takes you to a page that has a button to click that opens zoom) | 16:01 |
| jdstrand | ogra: you may need browser-support looking at the denials | 16:01 |
| jdstrand | (if you don't already) | 16:01 |
| jdstrand | I reverted again | 16:02 |
| ogra | i dont | 16:02 |
| ogra | will add ... | 16:02 |
| ogra | thanks for the report ! | 16:02 |
| jdstrand | (in a meeting; a zoom meeting ;) ) | 16:02 |
| jdstrand | thanks! | 16:02 |
| jdstrand | ogra: ping me and I can test the new version | 16:03 |
| ogra | will do | 16:03 |
| mup | PR snapd#11702 opened: tests: add spread test to test upgrade from release snapd to current <Created by mvo5> <https://github.com/snapcore/snapd/pull/11702> | 17:58 |
| mup | PR snapd#11703 opened: tests: initial set of tests to uc22 nested execution <Run nested> <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/11703> | 21:29 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!