17 Oct 2024 |
hexa (signing key rotation when) |
Out of memory: Killed process 817379 (python3.12)
| 11:08:52 |
Gaétan Lepage | Ok :/ Seems to be back now | 11:13:09 |
| Mic92 changed their display name from Mic3000 🌋 to Mic92. | 12:22:31 |
Mic92 | In reply to @arianvp:matrix.org Wondering if we can instruct people to point their flakes to the nixpkgs tarball instead? On our CDN? Git shallow clones are also faster | 13:13:55 |
Arian | Okay seems to me we should kill the GitHub API support stuff then? Like what is the point | 13:56:58 |
emily | IIRC GitHub don't like you doing a bunch of shallow clones? | 14:00:54 |
emily | I think they complained about Homebrew or CocoaPods or something doing that | 14:01:01 |
Mic92 | In reply to @dgrig:erethon.com Yes, I've blocked all the nixos infra meetings in my calendar so I can attend them. So you come to the meeting today? | 15:02:46 |
dgrig | In reply to @joerg:thalheim.io So you come to the meeting today? Yes, it's in ~30', right? | 15:28:47 |
Mic92 | @dgrig:erethon.com: yes 15 minutes | 15:42:59 |
Mic92 | https://jitsi.lassul.us/nixos-infra | 16:00:00 |
| bendlas changed their profile picture. | 17:37:40 |
18 Oct 2024 |
Artturin | Can someone check if https://github.com/NixOS/infra/blob/master/modules/tarball-mirror.nix is working?
Locally I'm getting
$ NIX_PATH=nixpkgs=. ./maintainers/scripts/copy-tarballs.pl --dry-run --debug \
--expr 'import <nixpkgs/maintainers/scripts/all-tarballs.nix>' \
--exclude 'registry.npmjs.org|mirror://kde|mirror://xorg|mirror://kernel|mirror://hackage|mirror://gnome|mirror://apache|mirror://mozilla|pypi.python.org'
Undefined subroutine &main::isValidPath called at ./maintainers/scripts/copy-tarballs.pl line 25.
fileparse(): need a valid pathname at ./maintainers/scripts/copy-tarballs.pl line 77.
END failed--call queue aborted.
| 19:56:36 |
Mindavi | isValidPath needs to be prefixed these days with the store I think | 20:15:48 |
Mindavi | Maybe the script broke with a nix update? | 20:15:57 |
vcunat | The service seems to be working as usual, at a glance.
Oct 18 05:43:17 pluto mirror-tarballs-start[3020740]: mirrored 19 files, already have 18597 files
| 20:35:59 |
vcunat | I'd also expect that nix(pkgs) update is the main difference here. | 20:38:15 |
19 Oct 2024 |
K900 | The keymap tests are flaking again and blocking nixos-unstable | 06:00:44 |
K900 | I'd appreciate a review on https://github.com/NixOS/nixpkgs/pull/349701 so maybe we can actually figure out why | 06:00:57 |
K900 | In reply to@k900:0upti.me I'd appreciate a review on https://github.com/NixOS/nixpkgs/pull/349701 so maybe we can actually figure out why OK I'm pretty sure I reproduced and fixed it | 06:34:27 |
K900 | (same PR) | 06:34:31 |
K900 | OK I'm just going to merge it | 06:37:07 |
K900 | And start another unstable-large | 06:37:12 |
K900 | Because I've restarted the colemak test like 10 times | 06:37:20 |
K900 | And it's still stuck | 06:37:22 |
vcunat | Hydra/rhea stuck because of full drive, trying to fix | 13:12:38 |
hexa (signing key rotation when) | Download image.png | 13:14:43 |
hexa (signing key rotation when) | a clear cut case for predict_linear 😄 | 13:15:35 |
hexa (signing key rotation when) | nix.gc.options = ''--max-freed "$((400 * 1024**3 - 1024 * $(df -P -k /nix/store | tail -n 1 | ${pkgs.gawk}/bin/awk '{ print $4 }')))"'';
| 13:18:48 |
hexa (signing key rotation when) | gross | 13:18:50 |