| 20 May 2026 |
 K900 | Yeah but we don't know what a good fix is | 11:54:41 |
| K900 | And we probably don't want to ship this staging cycle out before we have that | 11:54:55 |
| K900 | I have a fix but it's horrible and I don't trust myself enough to send it | 11:55:12 |
 Vladimír Čunát | So should I prevent unstable-small from getting this? | 11:59:11 |
| K900 | I think so | 11:59:26 |
| Vladimír Čunát | I wouldn't expect it very relevant to desktop usage, but 🤷 | 11:59:39 |
| K900 | Well -small is maybe fine, large channels definitely not | 12:00:09 |
 hexa | how big is your xdg shit in rebuilds | 12:00:19 |
| hexa | because we need to eat the staging-next anyhow | 12:00:26 |
| hexa | so nixos-unstable would run for a day or two anyhow | 12:00:40 |
| K900 | systemd.package big at the very least | 12:01:00 |
| hexa | a systemd rebuild is around 25k rebuilds per platform | 12:02:36 |
| K900 | systemd.package NixOS option | 12:03:16 |
| K900 | So one systemd and all the nixos tests | 12:03:24 |
 André Lima | Are autostart entries not running, is that it? And is this also a problem for upstream or just for NixOS?
Also, just to make sure I understand it, iirc GNOME hasn't supported autostart desktop entries in a while, is it also affected by this regression? | 12:18:57 |
| K900 | Autostart entries are not running | 12:19:30 |
| K900 | Problem kinda for upstream in weird cases, NixOS always | 12:19:38 |
| K900 | GNOME presumably uses the same standard systemd generator as everything else | 12:19:50 |
| André Lima | Right. I just checked and GNOME didn't stop supporting autostart, it just removed it's built-in service manager. Autostart entries are handled via systemd, like you said | 12:24:09 |
| Vladimír Čunát | I see context in the systemd room, which is maybe https://matrix.to/#/!SsuQHVzf59iBgk6YIm:0upti.me?via=nixos.org&via=matrix.org&via=tchncs.de (at least my Element thinks so; behaves weird) | 12:45:58 |
| hexa | firefox on nixos-unstable contains a known rce | 22:08:02 |
| hexa | 150.0.2 | 22:08:05 |
| hexa | I updated 150.0.2 -> 150.03 on May 12th fwiw | 22:08:42 |
| hexa | and -> 151.0 on may 18th | 22:09:03 |
| hexa | those delays are bonkers | 22:09:16 |
 ghpzin | Isn't it 150.0.3 on both channels now ?
Unless "known rce" part is about 150.0.3 and not 150.0.2 | 22:22:59 |
| hexa | it is about 150.0.3 | 22:29:54 |
| hexa | nixos-unstable indeed has 150.0.3 | 22:30:34 |
| hexa | through a half-cancelled eval that advanced | 22:30:47 |
| hexa | god, these are confusing | 22:30:51 |
