17 Jul 2025 |
| @imadalin:matrix.org left the room. | 16:14:55 |
18 Jul 2025 |
| @dmjio:matrix.org set a profile picture. | 00:21:19 |
| @nyxvectar:matrix.org changed their display name from Nyxvectar to Nyxvectar Yan. | 09:55:16 |
| Sandro 🐧 joined the room. | 12:29:06 |
| @haauler:matrix.org joined the room. | 14:22:45 |
| @jonhermansen:matrix.org left the room. | 15:01:35 |
| @dakammy:matrix.org joined the room. | 17:01:13 |
| @dakammy:matrix.org left the room. | 17:03:11 |
19 Jul 2025 |
| mrdev023 joined the room. | 10:29:51 |
mrdev023 | Hi everyone,
I want continue this package https://github.com/NixOS/nixpkgs/pull/170443 but i newbie about FHS management.
Here https://github.com/mrdev023/nixpkgs/commit/f4fc41b9498c455b5ebc3a50163f29b38a27acf4#diff-091e9b8c02fccb9fc7059db6a16c5f22a00e88427b29749bd917bdeadc785ac3R10 i got this error error: path '/nix/store/h8ya86bfricd2vkm2mr7gfbyig0yf727-source/pkgs/build-support/build-fhs-userenv/chrootenv' does not exist
What's the recommended solution to replace this ?
| 10:34:21 |
[0x4A6F] | Hey, better look into the #gaming:nixos.org channel or the general channel. This is for deployment tools. | 10:51:22 |
mrdev023 | Okay thx | 10:52:15 |
20 Jul 2025 |
| @dmjio:matrix.org left the room. | 09:57:48 |
21 Jul 2025 |
| @nyxvectar:matrix.org left the room. | 16:05:29 |
| @jovalie:matrix.org joined the room. | 23:47:56 |
22 Jul 2025 |
| @federicodschonborn:matrix.org changed their display name from The Worm 🏳️🌈 (he/they) to Wormy McWormface 🏳️🌈 (he/they). | 18:28:33 |
| Max joined the room. | 18:56:52 |
23 Jul 2025 |
| AssertInequality joined the room. | 16:39:57 |
AssertInequality | Hi Folks Does disko accept regex as device ? docs doesn't necessarily have a "spec"
On Hetzner, an attached volume is showing up as /dev/sda for some weird reason, with the VPS's built-in disk showing up as /dev/sdb
I noticed that /dev/disk/by-id has consistent naming with built-in storage having QEMU in the name and attached volume storage having HC
Since I'm trying to use the nixos-anywhere terraform module, I realistically won't be able to rely on the inconsistent /dev/sdX device names.
| 16:45:49 |
24 Jul 2025 |
uep | can you find something more consistent in one of the other subdirs, maybe one of /dev/disk/by-{path,id} | 07:11:26 |
uep | id probably has some "random" elements like a wwn, but path might be stable for a given vps instance shape at least? | 07:12:14 |
AssertInequality | I think path should work, I see no reason it should change
Here's how it's structured:
pci-0000:06:00.0-scsi-0:0:0:0 -> ../../sr0
pci-0000:06:00.0-scsi-0:0:0:1 -> ../../sdb
pci-0000:06:00.0-scsi-0:0:0:1-part/
pci-0000:06:00.0-scsi-0:0:0:1-part1 -> ../../sdb1
pci-0000:06:00.0-scsi-0:0:0:1-part2 -> ../../sdb2
pci-0000:06:00.0-scsi-0:0:0:2 -> ../../sda
pci-0000:06:00.0-scsi-0:0:0:2-part/
pci-0000:06:00.0-scsi-0:0:0:2-part1 -> ../../sda1
pci-0000:06:00.0-scsi-0:0:0:2-part9 -> ../../sda9
So I think I can safely assume that :1 would always be the built-in storage
| 12:10:45 |
uep | It looks that way, and the a/b ordering is just timing related for the probe order. That's what I'd expect, this looks promising but if course make sure | 22:58:40 |
[0x4A6F] | I think you should stick to /dev/disk/by-id/scsi-0HC_Volume_XXXXXXXXX for the Hetzner Cloud Volume. | 23:05:39 |
25 Jul 2025 |
uep | i think that's where the regex question started; the pattern is easy to see but unless disko supports matching against it, you can't use it without knowing the id in advance for any given node | 00:32:07 |