| 21 May 2026 |
 Vladimír Čunát | * You don't need to push it into that branch. It suffices to create a commit on that branch (to avoid too many changes) and we can switch to it easily. | 12:17:45 |
 Sergei Zimmerman (xokdvium) | Ok, I’ll try to get to that this/next week | 12:19:20 |
| Vladimír Čunát | Thanks. Ping us e.g. in this chat 🙂 | 12:20:10 |
| Sergei Zimmerman (xokdvium) | In the meantime it might be possible to cleanly cherry-pick https://github.com/NixOS/nix/pull/15855, but I’m not sure how robust it is | 12:21:22 |
| Vladimír Čunát | It does not apply cleanly. | 12:30:01 |
| Vladimír Čunát | (and I don't know the codebase, so I don't dare trying to resolve this) | 12:30:19 |
 hexa | that is the plan already | 13:15:23 |
|  Fovir joined the room. | 15:42:14 |
 mynacol | Did you check if going over IPv6 would help? (You need a non-default dualstack hostname). From the error messages, I don't think so, but maybe worth a try. | 18:37:06 |
| 22 May 2026 |
| Vladimír Čunát | I don't think we have tried that. | 03:38:00 |
 Vinceff | Hello, just a quick question, is there any news about GNOME 50?
Nixos 26.05 is supposed to be released next week or the week after at the latest, and there's no news about GNOME in unstable. 🥺 | 08:14:58 |
 K900 | It has been merged with staging-next AFAIK? | 08:15:24 |
| Vladimír Čunát | Yes. | 08:15:30 |
| Vladimír Čunát | So it's just about channel updates which makes the question on-topic actually 😁 | 08:16:01 |
| Vladimír Čunát | nixos-unstable will update to f83fc3c307e7 - but we're missing NixOS tests. | 08:16:48 |
| Vladimír Čunát | (nixpkgs-unstable has already updated to include these big changes) | 08:17:12 |
| Vladimír Čunát | * nixos-unstable will update to f83fc3c307e7 - but Hydra hasn't built many of the NixOS tests so far. | 08:17:55 |
| Vladimír Čunát | * nixos-unstable will update to f83fc3c307e7, as channel-blockers have passed already - but Hydra hasn't built many of the NixOS tests so far. | 08:18:15 |
| 23 May 2026 |
 tbutter | Not sure if this is really 'infrastructure', but i saw multiple checks on PRs today with "RequestError [HttpError]: Bad credentials". Is this a known error? | 14:52:52 |
| tbutter | example: https://github.com/NixOS/nixpkgs/actions/runs/26335565307/job/77528609860 | 14:53:12 |
| K900 | Github problem | 14:53:28 |
| tbutter | * Not sure if this is really 'infrastructure', but i saw multiple checks on PRs today with "RequestError [HttpError]: Bad credentials". Is this a known issue? | 14:53:32 |
| tbutter | the correct way to let it retry is to close / reopen the PR? | 14:54:18 |
| K900 | Yes | 15:00:00 |
| tbutter | thank you | 15:00:22 |
 emily | do we just need to https://github.com/NixOS/infra/pull/1030 harder? | 16:04:49 |
| Sergei Zimmerman (xokdvium) | In reply to @emilazy:matrix.org
do we just need to https://github.com/NixOS/infra/pull/1030 harder? No sadly, the error in question is not currently retryable because for whatever reason AWS does 400 for stuff that should be retried… | 19:06:47 |
| Sergei Zimmerman (xokdvium) | I don’t know if the new queue runner handles this at all btw | 19:07:37 |
| Sergei Zimmerman (xokdvium) | I don’t see any handling of this in the object_store crate, so our mileage may vary | 19:17:36 |
| 24 May 2026 |
 Arian | For rust I'd just use the AWS SDK to be honest | 10:18:32 |
