My unified nix flake for all configuration management.
Find a file
Daniel Flanagan 014c690336
All checks were successful
/ build-host (map[host:beefcake]) (push) Successful in 54s
/ build-host (map[host:dragon]) (push) Successful in 1m17s
/ build-host (map[host:flipflop]) (push) Successful in 1m8s
/ build-host (map[host:foxtrot]) (push) Successful in 1m14s
/ build-host (map[host:router]) (push) Successful in 45s
/ build-devshell (push) Successful in 18s
/ flake-check (push) Successful in 4m58s
feat: integrate deploy-rs
2025-03-20 22:22:27 -05:00
.forgejo/workflows fix: add echo command at end of jobs for clarity 2025-03-20 12:38:34 -05:00
.helix Format 2025-02-14 13:31:18 -06:00
lib feat: integrate deploy-rs 2025-03-20 22:22:27 -05:00
packages feat: integrate deploy-rs 2025-03-20 22:22:27 -05:00
secrets Uwu 2025-02-11 01:27:43 -06:00
.envrc Fprintd fixes? 2024-03-24 14:34:44 -05:00
.gitignore Implement pre-commit using the pre-commit-hooks flake 2024-02-23 10:35:41 -06:00
.sops.yaml Roll dragon secrets again for second re-install 2025-01-06 11:04:37 -06:00
flake.lock feat: integrate deploy-rs 2025-03-20 22:22:27 -05:00
flake.nix feat: integrate deploy-rs 2025-03-20 22:22:27 -05:00
readme.md fix: move links up top 2025-03-20 14:08:24 -05:00

Details

I aim for declaring everything where possible, but I definitely break the rules if it's convenient enough and try to write it down when I do so. I also don't like repeating myself in configuration. I am mostly focused on being pragmatic and getting things done and doing so in a way that makes it easy to keep doing so (maintainability), but I also occasionally fall into experimenting with big refactors of the code for no real gain.

Layers

  • Common defaults (machines that I might not interact with at all or only rarely, such as backup targets)
  • Machines that I interact with only remotely (shell class)
  • Machines that I interact with directly, such as a laptop or my desktop workstation (desktop class)