| 15 Jan 2025 |
 Mic92 | Maybe we can discuss this here. | 15:56:50 |
| Mic92 | Will vouch for a Nix release today. But maybe we can test a backport of this patch first. | 15:57:34 |
| Mic92 | Started the backport process | 15:58:25 |
| Mic92 | Also this just improves logging and doesn't explain how the error was created. | 15:59:10 |
 abathur | In reply to @joerg:thalheim.io
Maybe we can discuss this here. Either's fine by me, but I responded in the issue for now since we just discussed this in the installer WG meeting an hour ago and some others may be out there duplicating effort | 16:27:21 |
| Mic92 | abathur: answered. | 17:30:34 |
| Mic92 | abathur: Unless something gets in the way, I would take a stab at it, this weekend. | 17:35:55 |
| Mic92 | abathur: to speed up things, it would help if I add access to this repository | 17:45:55 |
| Mic92 | https://github.com/NixOS/experimental-nix-installer/ | 17:46:02 |
| Mic92 | I might need to change repository settings | 17:46:08 |
| abathur | that's fine by me | 17:49:16 |
| abathur | I don't have the keys to actually add you if you can't, had to ask around before to get matthew added | 17:50:19 |
 emily | #org_owners:nixos.org | 17:51:31 |
| abathur | Mic92: regarding s3, we re-did the release process to use GH releases as zimbatm suggested with the idea of just simple redirects in mind, but the upstream workflows @ detsys upload to S3 | 17:54:41 |
| abathur | * Mic92: regarding s3, we re-did the release workflow to use GH releases as zimbatm suggested with the idea of just simple redirects in mind, but the upstream workflows @ detsys upload to S3 | 17:54:55 |
| Mic92 | abathur: I think s3 is fine. Github also still doesn't do ipv6 anyway | 17:56:18 |
| emily | does v6 matter if it's behind fastly? | 17:57:13 |
| abathur | In reply to @joerg:thalheim.io
abathur: I think s3 is fine. Github also still doesn't do ipv6 anyway just making sure you know there's prior art for the workflows if you go that way | 17:57:40 |
| Mic92 | abathur: it's fairly similar to our normal nix release workflow. | 17:58:11 |
| Mic92 | No. Could be also in front of github releases, I suppose. | 18:00:56 |
|  SomeoneSerge (back on matrix) changed their display name from SomeoneSerge (utc+3) to SomeoneSerge. | 19:02:53 |
| 16 Jan 2025 |
 p14 | I am surprised to see /realisations in a brand new nix binary cache only used for nixpkgs. Assumption: the derivations I am looking at with doi files are not CA. (E.g gnu hello), since CA is not in use in nixpkgs. Assumption: realisations are a CA feature and not necessary.
The flip side is that my builder is not set up to copy realisations into the cache. Is that broken, if CA is not in use? | 10:15:55 |
|  j4m3s set their display name to j4m3s. | 10:28:00 |
| 17 Jan 2025 |
|  @550mg:matrix.org joined the room. | 07:16:39 |
| @550mg:matrix.org set a profile picture. | 07:16:54 |
| @550mg:matrix.org changed their profile picture. | 07:17:03 |
| @550mg:matrix.org changed their profile picture. | 07:18:44 |
 Las | Does the Nix sandboxing code sandbox time too? Does it prevent someone from using __vdso_clock_gettime, or the equivalent of that, somehow? | 15:57:24 |
 K900 | No | 15:57:33 |
| K900 | It does not attempt to | 15:57:37 |
