[–] 52 points 2 months ago (2 children)

Love how they're trotting out the piracy argument while AI companies have done the most widespread pirating of information and data to date yet operate openly and unscathed by accusations by this group.

  • source
  • parent
  • context
  • [–] 96 points 2 months ago (11 children)

    The article is a short read, anyone who usually skips over the headline is welcome to read it.

    the trade group speaking for corporate giants like Microsoft, Sony, and Electronic Arts, [...] stood up in front of lawmakers and declared that community-operated, private game servers are essentially a form of illegal piracy. To back that up, they actually pointed at community-run Minecraft servers, arguing that running your own multiplayer instance outside of corporate control is a black market threat that lacks proper safety standards [...] They are trying to shape the law and public narrative to brand independent, user-controlled hardware as suspicious, unsafe, and legally shady.

  • source
  • [–] 3 points 2 months ago

    I have had issues with Immich updates breaking my setup at least 3 times over the course of ~2 years hosting it, sometimes it's fixable but other times it's been easier to rebuild from scratch and restore the last backup

  • source
  • [–] 4 points 2 months ago (1 child)

    I would need logs to attempt to troubleshoot that. Might have to look at a Zima OS community to learn the troubleshooting steps. Otherwise, try reading the latest files in /var/log.

  • source
  • [–] 2 points 2 months ago* (last edited 2 months ago)

    No. There’s no way to force it.

    And the understanding is off in one spot: nothing gets “automatically pulled without verification.” Every source fetch has a pinned sha256 and fails loudly if it doesn’t match. It’s just that a hash proves the bytes didn’t change, not that upstream authored them.

    Signatures do exist in Nix, but only on binary cache paths. Fixed-output derivations (every fetchurl, every fetchFromGitHub) are exempt by design, because the hash is the identity. That’s not a setting, it’s the model.

    And there’s no signature to check even if you wanted one. Almost no nixpkgs package records an upstream key. So a hypothetical “signed only” mode would reject basically all of nixpkgs.

    Guix does what you’re asking with guix git authenticate. Nix has no equivalent.….yet.

  • source
  • parent
  • context
  • [–] [S] 1 point 2 months ago (1 child)

    Is there any way you can force it to use only software that's crypographically signed (not hashes)?

    My understanding is that nix uses a mix of source-based packages that are automatically pulled without verifying the sources crypgographically.

  • source
  • parent
  • context
  • [–] 1 point 2 months ago* (last edited 2 months ago) (2 children)

    Yes. Thats how I do it actually (not with apt). You can even point the runner at the nix community cache to prevent overly long build times. I have never trusted the Docker style of “I crammed everything I needed into this amorphous, way-too-capable, badly sandboxed, portable Alpine Linux box. Trust me bro, it’s secure” and prefer “here’s code showing exactly how I build this software with pinned, hashed versions that pass checks and you can use them too” with a simple flake file.

    Here’s some pseudo code demonstrating it:

    flake.nix

    {
      description = "hello-cli: pinned end to end by a lockfile, built in a sandbox";
    
      inputs = {
        nixpkgs.url = "github:NixOS/nixpkgs/nixos-25.05";
        flake-utils.url = "github:numtide/flake-utils";
      };
    
      outputs = { self, nixpkgs, flake-utils }:
        flake-utils.lib.eachDefaultSystem (system:
          let
            pkgs = import nixpkgs { inherit system; };
          in {
            packages.default = pkgs.rustPlatform.buildRustPackage {
              pname = "hello-cli";
              version = "0.1.0";
              src = ./.;
    
              # No magic hash to paste. The committed Cargo.lock IS the pin:
              # Nix reads it and fetches every crate as a fixed-output
              # derivation. Change a dep, the lock changes, the build changes.
              cargoLock.lockFile = ./Cargo.lock;
    
              meta.mainProgram = "hello-cli";
            };
    
            devShells.default = pkgs.mkShell {
              packages = with pkgs; [ cargo rustc rust-analyzer clippy ];
            };
          });
    }
    

    And a pertinent example runner.yml file:

    
    when:
      - event: [push, pull_request, tag]
    
    steps:
      - name: build
        image: nixos/nix:2.28.3
        environment:
          NIX_CONFIG: |
            experimental-features = nix-command flakes
            accept-flake-config = false
            sandbox = true
            substituters = https://cache.nixos.org/ https://nix-community.cachix.org/
            trusted-public-keys = cache.nixos.org-1:6NCHdD59X431o0gWypbMrAURkbJ16ZPMQFGspcDShjY= nix-community.cachix.org-1:mB9FSh9qf2dCimDSUo8Zy7bkq5CX+/rkCWyvRCYg3Fs=
        commands:
          # stale lock is a hard failure, so CI can never silently float
          - nix build .#default --no-update-lock-file --print-build-logs
          - ./result/bin/hello-cli nix
          # the money shot
          - nix path-info -r --closure-size --human-readable ./result
          # prove it again from scratch with zero cache and zero network
          - nix build .#default --rebuild --option substituters "" --print-build-logs
    
    
    
  • source
  • parent
  • context
  • [–] 1 point 2 months ago

    If you are on Windows then there is no reason to use it. Windows Defender is excellent and MS put A LOT of effort in making sure it’s that way.

    If you’e on Unix/Linux it’s pointless other than to scan stuff thats come from a Windows network.

  • source
  • [–] 1 point 2 months ago

    Clamav is a full antivirus, but is missing on access scanning. It's used a lot as a plugin to server software as you can feed it files and get a yes or no back.

    I would not recommend using it as endpoint protection, for that I would look at bitdefender. I use at work and it does the job

  • source
  • [–] 6 points 2 months ago (3 children)

    Not really. I mean it does the job, it's a real AV that will catch malware, but it's not a modern product. Its primary use is for corporations needing to meet a "you must have antivirus" compliance requirement.

    The best defense against Linux malware is reviewing stuff before you run it. And not running random stuff as root. And not piping curl/wget to bash, especially not sudo bash.

    And of course don't expose stuff to the Internet unless you have to. And if you do, isolate each service, so that an attacker can't pivot once they're in your network.

  • source
  • [–] [S] 1 point 2 months ago

    Thanks. I haven't looked into podman much. It came up in our research, but I saw cosign in the docs and assumed it had the same vulnerabilities. I'd be curious of how to use podman with GPG (and assume you're talking about the publishers signing the release, not me signing it?)

    Do you have any examples of popular, public images that you can use with podman, and verify the signature with GPG? Do you have any link to any guide that shows how to do this?

  • source
  • parent
  • context
  • [–] 1 point 3 months ago*

    The one situation I can think of that fits this write-up is someone selfhosting for themself only - they would probably not use more than one or two services at a time and so could be interested in knowing how many they can stack/stand up idle on the same box or vps. Even then, a single person hosting an -arr stack and a code forge with ci/cd can easily find themselves triggering simultaneous active compute on several of these services.

  • source
  • parent
  • context
  • [–] 2 points 3 months ago

    When was this measured, and how were the apps installed?

    The current latest release of forgejo is 16.0.2, which released almost two weeks ago. How did you get 7.0.9 running, compiled from source? I mainly ask because you state

    No docker overhead in the numbers

    Which I can't tell if that means you've found a way to measure only the footprint of the in-container process(es) or that you installed these without docker entirely.

    And your smeltworks website is giving me a 502 error /cloudflare status page, so I can't check the supposed commands used.

  • source
  • view more: ‹ prev next ›