| 6 Oct 2025 |
 SomeoneSerge (back on matrix) | One shared castore, multiple snix-stores | 09:59:44 |
 Matt Sturgeon | In reply to @joerg:thalheim.io
@mattsturg:matrix.org: if you go to the eval build step, there is a restart button after github login When I tried that for a merge group, I got:
fatal: couldn't find remote ref gh-readonly-queue/nixos-25.05/pr-3724-a30decbd5fc231e84dfefeb75bc7f57d8167726c
Because GitHub cleans up the merge group ref when it is dequeued, which happens as soon as it receives the failed status check. | 10:45:20 |
| Matt Sturgeon | (maybe buildbot should fetch the actual git object instead of fetching by reference to avoid this failure mode?) | 10:46:22 |
 Mic92 | In reply to @mattsturg:matrix.org
(maybe buildbot should fetch the actual git object instead of fetching by reference to avoid this failure mode?) Fetching by git commit is not a supported feature of the protocol | 11:02:44 |
| Mic92 | However we should not cache fetch errors in the first place | 11:03:21 |
| Mic92 | Can you open an issue | 11:03:30 |
| Mic92 | ? | 11:03:33 |
| Matt Sturgeon | In reply to @joerg:thalheim.io
Fetching by git commit is not a supported feature of the protocol You can do it via the git CLI git fetch <remote> <commit> (so I assume the git protocol supports it) and also via flakeref (if you know the commit id, at least: nix build github:NixOS/nixpkgs/<commit>). Which protocol does buildbot use to fetch? | 11:07:06 |
| Matt Sturgeon | In reply to @joerg:thalheim.io
However we should not cache fetch errors in the first place I haven't checked yet if that error caused cached failures; I saw it in a rebuild attempt | 11:07:40 |
| Mic92 | Redacted or Malformed Event | 11:08:25 |
| Mic92 | Will check if this possible | 11:09:32 |
| Mic92 | In my memory this was not possible | 11:09:42 |
| Matt Sturgeon | Gotcha. I'll open an issue in a bit then. Thanks | 11:10:21 |
| Mic92 | I would expect that the commit in this case is also garbage collected | 11:15:47 |
| Matt Sturgeon | Maybe, although anecdotally I've noticed GitHub almost never runs gc; you can almost always get commits from old rebases, for example, even if you've done several subsequent force-pushes. | 11:20:18 |
| 11 Oct 2025 |
|  W47NUT joined the room. | 05:20:53 |
| 6 Oct 2025 |
|  Manuel A joined the room. | 12:05:27 |
| 17 Oct 2025 |
|  @aelita:tchncs.de joined the room. | 16:03:33 |
| 6 Oct 2025 |
 dramforever | fetching a commit by full id is specified but is not necessarily allowed on any git server | 12:31:17 |
| 19 Oct 2025 |
| @aelita:tchncs.de removed their profile picture. | 01:48:22 |
| 6 Oct 2025 |
| dramforever | i guess github allows it then, but i don't know to what extent (e.g. does it have to be reachable?) | 12:31:37 |
| 19 Oct 2025 |
| @aelita:tchncs.de removed their display name Aelita (they/she). | 01:48:24 |
| 6 Oct 2025 |
 Grimmauld (any/all) | In reply to @mattsturg:matrix.org
Maybe, although anecdotally I've noticed GitHub almost never runs gc; you can almost always get commits from old rebases, for example, even if you've done several subsequent force-pushes. they do run GC, but often only on very busy repos | 12:32:24 |
| 19 Oct 2025 |
| @aelita:tchncs.de left the room. | 01:48:25 |
| 7 Oct 2025 |
|  martijn removed their profile picture. | 10:54:48 |
| 24 Oct 2025 |
|  saygo.2 joined the room. | 21:12:03 |
| 7 Oct 2025 |
| martijn set a profile picture. | 10:56:12 |
| 9 Oct 2025 |
|  srhb set a profile picture. | 07:08:12 |
|  @nullcube:matrix.org left the room. | 10:38:04 |
| Matt Sturgeon | AFAICT, buildbot isn't picking up github merge queue groups anymore.
E.g. #3762 is queued, but isn't showing up on the nixvim/nix-eval page.
The last time we saw buildbot trigger on a queue event was yesterday: build 7028.
| 15:34:58 |
