| 25 Jun 2026 |
 hexa | doing | 10:08:06 |
| hexa | deployed | 10:20:31 |
| hexa | Mic92 I suppose you want a queue-runner restart? | 10:20:40 |
| hexa | but please do confirm | 10:23:45 |
 Mic92 | might be not actually required. | 10:30:37 |
 Vladimír Čunát | Now I still get
Failed: error: path '/nix/store/d3dhzz7k3yvpg51m3pb6cnkhza8mp6sr-source' is not valid
| 11:28:48 |
| Vladimír Čunát | (less than 20 minutes ago) | 11:29:04 |
| Vladimír Čunát | * (less than 20 minutes ago, step 4 of this build) | 11:29:30 |
| Vladimír Čunát | Though I do respect that these kind of bugs aren't easy to fix. | 11:30:32 |
| Vladimír Čunát | * Though I do respect that these kinds of bugs aren't easy to fix. | 11:30:39 |
| hexa | why do we have three places where nixexprs tarballs are created? | 11:54:54 |
| hexa | I understand the one in nixos/lib/make-channel.nix and the one in pkgs/top-level/make-tarball.nix | 11:55:21 |
| hexa | but not the one in pkgs/build-support/release/default.nix | 11:55:34 |
| hexa | Redacted or Malformed Event | 11:55:47 |
| hexa | can't find any reference to releaseTools.channel | 11:59:23 |
| hexa | is it ded? | 11:59:28 |
| hexa | 8f98a6d39bc9bfd4bde8d9cd692ca7ef92111809 states
It turned out that my assumption was wrong and we do use the attribute in Nixpkgs, namely via releaseTools.channel, which is similar to what we're doing in Vuizvui.
| 12:09:04 |
| hexa | no idea where | 12:09:36 |
| hexa | do we have any perl enjoyers here? https://github.com/NixOS/infra/pull/1103 | 13:35:50 |
| Mic92 | DROCM_VERSION=70203 -Wall -Wno-unused-function -O3 -DNDEBUG -x hip -MD -MT CMakeFiles/testing_dpotri.dir/testing/testing_dpotri.cpp.o -MF CMakeFiles/testing_dpotri.dir/testing/testing_dpotri.cpp.o.d -o CMakeFiles/testing_dpotri.dir/testing/testing_dpotri.cpp.o -c /build/source/testing/testing_dpotri.cpp > error: unable to open output file '/build/testing_dpotri-gfx906-68b29b.o': 'No space left on device' > 1 error generated when compiling for gfx906. > ninja: build stopped: subcommand failed. For full logs, run: nix log /nix/store/j47xkc043h5ky66z93zlwj2wp86m53b7-magma-2.9.0.drv | 14:28:08 |
| Mic92 | out of disk | 14:28:11 |
| Mic92 | not a bug I'd say | 14:28:26 |
 dgrig | Not a perl enjoyer (the opposite tbh), but the .diff view in github makes things pop up. I left a quick review without running/testing it | 14:40:40 |
| dgrig | * Not a perl enjoyer (the opposite tbh), but the .diff view in github makes things pop up. I left a quick non-blocking review without running/testing it | 14:41:00 |
| hexa | Thanks, I had hoped perlnavigator and perlcritic would be more attentive 🫠 | 14:41:19 |
| Vladimír Čunát | No, that's log from a different attempt. | 15:26:46 |
| Vladimír Čunát | This one didn't show any log. | 15:26:50 |
| Mic92 | ok than I am missing context here. | 15:27:07 |
| Vladimír Čunát | That's why I was posting the step number. | 15:27:40 |
| Mic92 | Okay, do you already know if this is a flakiness issue or it will always reproduce? | 15:31:39 |
