| 21 May 2021 |
|  Rosario Pulella joined the room. | 19:43:13 |
| 4 Apr 2025 |
 ElvishJerricco | all in all, I think the fix is to have the test's shutdown script pipe its output to /dev/kmsg or something like that, which would have been the dumb workaround I would have used if I hadn't bothered to figure all this out | 22:06:52 |
| 21 May 2021 |
|  hax404 joined the room. | 20:42:16 |
| 4 Apr 2025 |
| ElvishJerricco | but now I feel better about it actually being a good fix :P | 22:07:00 |
| 21 May 2021 |
|  Emantor changed their profile picture. | 20:42:52 |
| 4 Apr 2025 |
| ElvishJerricco | * ok but I think it's not actually using that function. Based on main() and log_open(), I think it's defaulting to setting the log target to kmsg for PID1 (which means this output goes to each type of console), until the journal starts and it switches to that. At shutdown, based on become_shutdown(), PID1 execs systemd-shutdown with --log-target=kmsg. At no point is stdout or stderr redirected, meaning I think that PID1 and subprocesses like systemd generators or systemd-shutdown scripts will output to /dev/console, which only represents one tty, which is now tty0 in our VM tests | 22:09:21 |
| 21 May 2021 |
|  Roos joined the room. | 22:39:27 |
| 4 Apr 2025 |
|  bdd left the room. | 23:18:56 |
| 5 Apr 2025 |
|  @tinybronca:sibnsk.net left the room. | 15:48:46 |
| 22 May 2021 |
|  globin joined the room. | 00:06:24 |
|  Server Stats Discoverer (traveler bot) joined the room. | 00:17:52 |
|  Matrix Traveler (bot) joined the room. | 00:29:39 |
 @grahamc:nixos.org | I've experimentally placed this in a Nix Teams subspace, let me know if this doesn't feel like a good fit. | 00:39:55 |
|  casey © joined the room. | 03:04:00 |
| 7 Apr 2025 |
| ElvishJerricco | Arian: btw I think I've realized that the reason nixosTests.systemd-repart.create-root fails only intermittently is because of the thing I found | 04:17:25 |
| 22 May 2021 |
|  mvnetbiz joined the room. | 10:51:30 |
| 7 Apr 2025 |
 Arian | Oh no | 08:25:13 |
| 22 May 2021 |
 Gytis Ivaskevicius | In reply to @andi:kack.it
What is our current stance on DynamicUser=true with state directories on different disks/partitions? Mounting fileSystems to /var/lib/private/... has never been a great experience. You can bind mount something in /srv and go on from there | 11:23:16 |
| 7 Apr 2025 |
|  dblsaiko ⚧︎ 🔪 left the room. | 19:34:30 |
| dblsaiko ⚧︎ 🔪 joined the room. | 19:34:49 |
| 22 May 2021 |
 andi- | In reply to @gytis-ivaskevicius:matrix.org
You can bind mount something in /srv and go on from there That isn't how DynamicUser with StateDirectory is going to work. You can't specify absolut paths with StateDirectory (anymore). You can only give it a "namespace" (and subdirs IIRC of that same). But given that /var/lib/private is an implementation detail of systemd I'd rather not mount into that folder.
To have things on /srv would require that the dynamic user id that I spawn with obtains ownership of everything in that subdir there. That is something I would have to implement manually (with preStart or whatever).
| 11:56:11 |
| Arian | grahamc (he/him) could you make me a moderator/admin ? | 11:59:43 |
| 9 Apr 2025 |
 K900 | Is there a systemd knob to mount debugfs by default? | 07:52:46 |
| 22 May 2021 |
| @grahamc:nixos.orgchanged room power levels. | 12:00:16 |
| 9 Apr 2025 |
| K900 | I feel like I remember seeing something | 07:52:57 |
| K900 | But not what or where | 07:52:59 |
| 22 May 2021 |
| andi- removed their profile picture. | 13:25:58 |
| 9 Apr 2025 |
| ElvishJerricco | K900: we did a PR recently. That should be default right now | 08:13:45 |
| 22 May 2021 |
| andi- set a profile picture. | 13:34:08 |
| 9 Apr 2025 |
| K900 | Hm | 08:14:28 |
