| 31 Oct 2025 |
 eveeifyeve | * Huh is it failing to download? Because I don't see the log file for the fail. | 19:32:42 |
| eveeifyeve | trying https://github.com/fumiama/base16384/archive/v2.3.2.tar.gz
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
100 36802 0 36802 0 0 96638 0 --:--:-- --:--:-- --:--:-- 96638
unpacking source archive /private/tmp/nix-build-source.drv-0/download.tar.gz
| 19:37:09 |
| eveeifyeve | * Because all I see is this as log:
trying https://github.com/fumiama/base16384/archive/v2.3.2.tar.gz
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
100 36802 0 36802 0 0 96638 0 --:--:-- --:--:-- --:--:-- 96638
unpacking source archive /private/tmp/nix-build-source.drv-0/download.tar.gz
| 19:37:23 |
 vcunat | The hash is wrong. | 19:47:59 |
| vcunat | You'd see that if you tried it locally. | 19:48:26 |
| eveeifyeve | No I am about to. | 19:48:50 |
| eveeifyeve | * No I am about to.
Also when I do this change make the branch staging right? | 19:49:27 |
| eveeifyeve | * No I am about to.
Also when I do this change, make the branch staging right? | 19:49:33 |
| eveeifyeve | * No I am about to.
Also when I do this change, make the branch staging right? It's going to produce 36 rebuilds. | 19:51:44 |
 leona | no, 36 rebuilds is totally fine for master | 20:19:01 |
| 1 Nov 2025 |
| eveeifyeve | https://github.com/NixOS/nixpkgs/pull/457444 | 04:14:07 |
| eveeifyeve | * base16384: fix hash pr: https://github.com/NixOS/nixpkgs/pull/457444 | 04:15:53 |
| eveeifyeve | I am working on fixing the jdk17 on x86_64-linux idk if anyone is working on fixing that as we speak. | 04:28:31 |
| eveeifyeve | Huh It looks like it built and builds locally for x86_64-linux openjdk 17, but hydra is not registering that it built properly. | 04:36:13 |
| eveeifyeve | Moved on to pytorch as that is really important. | 04:52:00 |
| eveeifyeve | * Moved on to pytorch as that is really important. Seems like it's failing on thrust/context.h not being included: https://matrix.to/#/!eWOErHSaiddIbsUNsJ:nixos.org/$cNdgxQAI-1aVCYJecQVVzFcOp9jajxUtmlre3Xa-LmE?via=nixos.org&via=matrix.org&via=nixos.dev | 04:52:30 |
| eveeifyeve | I think I might have a blocker for zero failure builds. https://matrix.to/#/!eWOErHSaiddIbsUNsJ:nixos.org/$T7saEwa5A0hPvY_5Jc51IqSxMEoDeNWnu4ME7kgMQqk?via=nixos.org&via=matrix.org&via=nixos.dev | 05:46:41 |
| eveeifyeve | * I think I might have a blocker for zero failure builds. pytorch is expecting thrust/context.h but it no longer exists. https://matrix.to/#/!eWOErHSaiddIbsUNsJ:nixos.org/$T7saEwa5A0hPvY_5Jc51IqSxMEoDeNWnu4ME7kgMQqk?via=nixos.org&via=matrix.org&via=nixos.dev | 05:47:16 |
| leona | there is no concept like “blocker for ZHF”. We won’t reach zero hydra failures, we strive for it tho. It’s only an effort to fix as much as we can. Maybe it’s a release blocker, but I would say pytorch not building is not even that | 10:34:37 |
|  Matt H changed their display name from Matt H to Matt H (Back Monday). | 11:01:50 |
| 2 Nov 2025 |
|  tom joined the room. | 08:36:54 |
| 3 Nov 2025 |
| Matt H changed their display name from Matt H (Back Monday) to Matt H. | 07:51:46 |
|  winston joined the room. | 13:57:42 |
| 5 Nov 2025 |
|  JacoMalan1 joined the room. | 07:42:50 |
| JacoMalan1 | Hey guys, looking at the release schedule, no new modules will be added to 25.11 right? | 07:44:06 |
| leona | You can add new modules and packages, as they're not breaking changes | 07:45:26 |
| JacoMalan1 | Thank you! | 07:45:55 |
| vcunat | Even to 25.05 I guess, still. | 07:46:06 |
| 6 Nov 2025 |
|  liminor joined the room. | 08:57:30 |
|  krauterbaquette joined the room. | 20:49:28 |
