| 11 Feb 2025 |
 Arian | Systemd folks told me vmspawn is basically untested and they never tried it on aarch64 | 20:15:27 |
| Arian | Mkosi qemu wrapper is also broken on aarch64 | 20:15:38 |
 raitobezarius | in true systemd fashion | 20:27:04 |
 antifuchs | oooh, is it basic.target that gets waited on there? I guess that's my culprit | 20:39:46 |
| Arian |
in true systemd fashion
I made the mistake at looking all issues tagged with journal in the systemd issue tracker today
| 21:43:25 |
| Arian | *
in true systemd fashion
I made the mistake at looking all issues tagged with journal in the systemd issue tracker today
| 21:43:32 |
| Arian | I’m convinced it’s physically impossible to logship journal logs without occasional corruption | 21:43:55 |
| 12 Feb 2025 |
|  arcayr joined the room. | 02:50:36 |
 magic_rb | so ive got a weird setup where i've got some disks on my server which are unlocked by me after the rest of the system boots up. therefore for example /mnt/disk1 wont be available until i unlock it. But when I do need the equivalent of machinectl bind uk3s /mnt/disk1/infrastructure/buildbot /data/buildbotto be ran. I'm not sure whats the best way to automate that in systemd. Maybe a service unit depending onmnt-disk1-infrastructure-buildbot.mount`? | 07:49:51 |
 gdamjan | automate what? | 15:58:25 |
| gdamjan | do you want to automate the unlocking?
there are some options here https://www.freedesktop.org/software/systemd/man/systemd-cryptenroll.html
with tpm2 or fido2 keys | 15:59:05 |
 aloisw | In reply to @arianvp:matrix.org
Yeh seems qemu only does acpi for intel I just tried it with TCG (on an x86_64 host) and the aarch64 guest does seem to have ACPI. | 17:14:08 |
|  @kira:jakira.space changed their display name from kira to Kira. | 19:59:45 |
| 15 Feb 2025 |
|  BenjB83 joined the room. | 10:18:44 |
| BenjB83 changed their display name from Benjamín Buske to BenjB83. | 10:43:01 |
| 16 Feb 2025 |
| aloisw changed their profile picture. | 10:14:36 |
 K900 | @ElvishJerricco @infinisil FYI https://github.com/NixOS/nixpkgs/commit/daadcbdaacc352a2031d53eaf6f8f1eedc263bcb | 16:42:55 |
 ElvishJerricco | I could swear we already had that... | 16:43:35 |
| K900 | We had pkgs.bash | 16:43:45 |
| K900 | And on staging-next pkgs.bash == pkgs.bashInteractive | 16:43:51 |
| K900 | And pkgs.runtimeShell = ${pkgs.bashNonInteractive}/bin/bash | 16:43:59 |
 infinisil | In reply to @k900:0upti.me
@ElvishJerricco @infinisil FYI https://github.com/NixOS/nixpkgs/commit/daadcbdaacc352a2031d53eaf6f8f1eedc263bcb Nice, thanks | 16:46:47 |
| ElvishJerricco | K900: oh, like pkgs.bash was changed to bashInteractive? Why? | 16:48:52 |
| K900 | Convenience, primarily | 16:49:10 |
| K900 | @infinisil's PR has detailed motivation | 16:49:19 |
| K900 | That I generally agree with | 16:49:24 |
| ElvishJerricco | link? | 16:49:40 |
| K900 | Hold on need to find it | 16:50:11 |
| K900 | Currently rolling back all the shit I just made not boot | 16:50:24 |
| infinisil | ElvishJerricco: https://github.com/NixOS/nixpkgs/pull/379368 | 16:50:36 |
