| 8 Jan 2026 |
 whispers (it/fae) | maybe related: some builds i ran on master were trying to build cargo for aarch64-linux, which absolutely should not have needed a rebuild without staging cycle | 04:53:40 |
| whispers (it/fae) | * maybe related: some builds i ran on master a few hours ago were trying to build cargo for aarch64-linux, which absolutely should not have needed a rebuild without staging cycle | 04:54:37 |
| whispers (it/fae) | * maybe related: some builds i ran on master a few hours ago were trying to build cargo for aarch64-linux, which absolutely should not have needed a rebuild without a staging cycle. so something deep in the dependency tree changed somehow | 04:55:36 |
| whispers (it/fae) | looking at hydra, it seems like cargo (probably some dependent?) had a rebuild on every platform besides x86_64-linux | 05:05:00 |
| whispers (it/fae) | https://github.com/NixOS/nixpkgs/pull/472376 probably this PR | 05:06:15 |
 Fabián Heredia | weird that x86_64-linux would be 0 rebuilds and ofborg/eval also indicates 0 rebuilds 😱 | 05:07:35 |
| Fabián Heredia | 
Download image.png | 05:08:23 |
| Fabián Heredia | Yeah currently x86_64-linux chilling | 05:08:31 |
| whispers (it/fae) | * https://github.com/NixOS/nixpkgs/pull/472376 maybe this PR? | 05:10:35 |
| whispers (it/fae) | Redacted or Malformed Event | 05:17:28 |
| whispers (it/fae) | * if i'm right and it is that pr, i have a guess at how this happened. i wonder if https://github.com/NixOS/nixpkgs/pull/464707 made this PR become a mass-rebuild (once it reached master with the recent staging-next merge) between the time of the initial CI rebuild calculation/label application and merge time | 05:18:23 |
| whispers (it/fae) | 
Download image.png | 05:40:03 |
| whispers (it/fae) | Redacted or Malformed Event | 05:40:12 |
| whispers (it/fae) | * i don't fully understand this interaction with RUSTFLAGS but it may also have something to do with https://github.com/NixOS/nixpkgs/pull/464707 | 05:42:25 |
| whispers (it/fae) | * https://github.com/NixOS/nixpkgs/pull/472376 maybe this PR?
EDIT: yep, seems like it
$ nixpkgs-review pr --no-shell --print-result 472376 --systems aarch64-linux -c merge --eval local
-- snip
--------- Impacted packages on 'aarch64-linux' ---------
26438 packages updated:
| 06:15:07 |
| whispers (it/fae) | * https://github.com/NixOS/nixpkgs/pull/472376 maybe this PR?
EDIT: yep, seems like it
$ nixpkgs-review pr --no-shell --print-result 472376 --systems aarch64-linux --checkout merge --eval local
-- snip
--------- Impacted packages on 'aarch64-linux' ---------
26438 packages updated:
| 06:17:07 |
| whispers (it/fae) | * https://github.com/NixOS/nixpkgs/pull/472376 maybe this PR?
EDIT: yep, seems like it
$ nixpkgs-review pr --no-shell --print-result 472376 --systems aarch64-linux --checkout merge --eval local
(output elided)
--------- Impacted packages on 'aarch64-linux' ---------
26438 packages updated:
(output elided)
| 06:17:27 |
 Vladimír Čunát | I'm cancelling the jobs and notified that PR. | 06:29:23 |
| Vladimír Čunát | Thanks for finding the difference. | 06:29:40 |
| Vladimír Čunát | Hydra probably will have spare time for *-linux, by the way. (but the cargo also rebuilds on *-darwin) | 06:39:41 |
| Vladimír Čunát | I might get crazy from all the darwin and pythonPackages fallout on staging-next-25.11. | 14:30:23 |
| Vladimír Čunát | nixpkgs-unstable is also blocked for a week now, because of some darwin-only issue, BTW. | 14:32:23 |
 hexa | twisted fails differently than on staging-next 🤔 restarting once more | 14:34:49 |
| Fabián Heredia | I only have x86_64-linux, can someone on one of the 3 affected platforms (aarch64-linux, aarch64-darwin, x86_64-darwin) review https://github.com/NixOS/nixpkgs/pull/478053 ? | 15:37:22 |
 Ben Sparks | not sure what there is to do here :/ nixpkgs-review states 0 rebuilds, as does the PR | 15:51:00 |
| Ben Sparks | * not sure what there is to do here? nixpkgs-review states 0 rebuilds, as does the PR | 15:51:07 |
| Ben Sparks | * not sure what there is to do here? nixpkgs-review states 0 rebuilds, as does the PR (posted from an aarch64-darwin machine) | 15:52:04 |
| Fabián Heredia | It is reapplying changes from a PR that was labeled 0 rebuilds but ended up being about 50k on hydra: https://matrix.to/#/!UNVBThoJtlIiVwiDjU:nixos.org/$CchYF6HrOphocWTwXpEGkBlcRn_Jm1EU9LG8YztjibU?via=nixos.org&via=matrix.org&via=tchncs.de | 15:52:19 |
 emily | did we figure out why the rebuilds weren't reported? | 16:08:36 |
| emily | oh | 16:09:03 |
