| 14 Jan 2026 |
 Sofie 🏳️⚧️ (she/her) | what is their tld 😭 | 13:44:56 |
 K900 | Over 12800 locations according to wikipedia | 13:45:53 |
| K900 | At that scale you basically have to run your own infra | 13:46:08 |
 just1602 | For a fancy homelab that ressemble production like env with UI and everything I'd look into incus + garage or something like that | 13:47:08 |
 ginkogruen | I’m using the git version of lix from nixpkgs on my mac.
But I somehow don’t get a cached build (which I think I did previously) and it fails during building when running tests.
> FAILED ../tests/functional2/commands/test_build/test_reference_checks.py::test_regression_partial_build[files0]
> FAILED ../tests/functional2/commands/test_build/test_reference_checks.py::test_structured_attrs_discard[files0]
Is this something known because I couldn’t find issues (on nixpkgs & lix) regarding that so far? | 13:48:43 |
 Marie | it's the name of the company | 13:52:21 |
| Marie | they have their own tld because they have too much money | 13:52:30 |
 truh | claiming a color for your company tld is almost as annoying as .fox | 13:53:44 |
 raitobezarius | if you are private cloud shaped, OpenStack makes sense | 14:11:47 |
| raitobezarius | ceph is a filesystem technology | 14:11:53 |
| raitobezarius | containers is one way to offer compute, it's orthogonal to openstack | 14:12:04 |
| raitobezarius | you compare OpenStack to like commercial products that VMware offers | 14:12:19 |
| raitobezarius | on mac, we know there are issues with macOS | 14:13:20 |
| raitobezarius | especially on Tahoe | 14:13:27 |
| raitobezarius | https://gerrit.lix.systems/c/lix/+/4910 related | 14:13:38 |
| ginkogruen | In reply to @raitobezarius:matrix.org
https://gerrit.lix.systems/c/lix/+/4910 related Thanks for pointing me to the issue :) This will arrive in nixpkgs relatively soon right? | 14:17:13 |
| raitobezarius | there's no true fix for now | 14:17:21 |
| raitobezarius | the workaround is to do doCheck = false; or retry frequently enough | 14:17:30 |
| raitobezarius | we are a bit blocked because introducing the shebang breaks things and we need to introduce the shebang i feel like | 14:18:00 |
| raitobezarius | i have no idea yet how we are going to overcome this | 14:18:05 |
| ginkogruen | Maybe I’ll pin lix then until it’s fixed. | 14:18:17 |
| raitobezarius | it's not a lix regression alas | 14:18:27 |
| raitobezarius | what Lix (and Nix) have been doing has always been considered unsafe to some extent by macOS I think | 14:18:44 |
| raitobezarius | recent macOS upgrades started to exhibit breakage harder | 14:18:52 |
| raitobezarius | sufficiently so that everyone runs into it from time to time | 14:18:58 |
| raitobezarius | it unfortunately also extend to nix-shell -p itself | 14:19:14 |
| ginkogruen | Oh then I’ve misunderstood what this build failure is about. | 14:19:20 |
| raitobezarius | (or any runCommand primitive) | 14:19:23 |
| raitobezarius | I'm not sure neither what it is, but there's an ongoing major problem with macOS related to the execution of anything without a shebang, we fixed many of those but nix-shell-based workloads are broken and we test many things in the testsuite | 14:19:59 |
| raitobezarius | (are broken non-deterministically, that is, they work N % of the time) | 14:20:16 |
