| 9 Apr 2026 |
Alyssa Ross | That's a very small rebuild count. | 11:55:51 |
Vladimír Čunát | Yes, especially as we don't have a large fraction finished yet (~53k successes / ~280k). | 11:58:49 |
K900 | Uhh it's 58818 rebuilds according to CI | 12:50:33 |
Vladimír Čunát | Why do you even need a rebuild on isGNU? | 12:55:10 |
Vladimír Čunát | Besides, I believe you paste the strings incorrectly (without a space). | 12:56:01 |
Sapii | Redacted or Malformed Event | 23:29:14 |
| 10 Apr 2026 |
Randy Eckenrode | That was pointed out in the PR. I fixed it not to cause rebuilds (and fixed the spacing issue). | 00:19:32 |
Randy Eckenrode | Apple appears to have fixed the issue in libiconv-113. I’m working on an update to the source releases and 26.x SDK. If it is indeed fixed, I’ll drop the workarounds in the update PR. | 02:07:05 |
K900 | Nice | 05:44:57 |
Yureka (she/her) | python3.pkgs.scipy is broken on staging-next | 06:08:26 |
Yureka (she/her) | (aarch64-linux) | 06:10:35 |
K900 | It's not | 06:28:54 |
K900 | It's flaky as fuck | 06:28:56 |
K900 | It built for me after like five tries | 06:29:02 |
Yureka (she/her) | I have 104 failing tests | 06:38:40 |
Yureka (she/her) | on aarch64? | 06:38:55 |
K900 | Yep | 06:39:43 |
K900 | /nix/store/r6sngalyggia24wa2vq2afnh3k71vf2s-python3.13-scipy-1.17.1 right? | 06:41:10 |
Yureka (she/her) | /nix/store/bs4xh6nw6fhiz2lpxgy3rq703xr5wfqb-python3.13-scipy-1.17.1.drv | 06:42:41 |
Yureka (she/her) | ah yes | 06:42:54 |
Yureka (she/her) | that's the outPath | 06:42:56 |
Yureka (she/her) | hmm, I will try a few more times | 06:43:07 |
K900 | https://static.0upti.me/scipy.nar.zst | 06:46:35 |
K900 | If you want to just be done with it | 06:46:42 |
Yureka (she/her) | I'm trying once more on my one slow 4K pagesize machine | 07:05:54 |
Yureka (she/her) | openblas already didn't want to build locally but did on that machine | 07:06:06 |
Vladimír Čunát | Retrying isn't a real solution. Such tests should be either fixed or skipped. | 07:11:42 |
Vladimír Čunát | Otherwise we'll be dealing with the issue every rebuild again. | 07:12:06 |
Vladimír Čunát | * Otherwise we'll be dealing with the issue on every rebuild again. | 07:12:10 |
Yureka (she/her) | The failing tests I'm seeing also look somewhat serious | 07:16:13 |