Example in docs/disko-images.md generates broken result script

[RadxaYuntian]

Step to reproduce

Running following script based on `docs/disko-images.md` example.

#!/usr/bin/env bash

set -euo pipefail

(
    rm -rf .git flake.lock flake.nix simple-efi.nix
)

cat << EOF > flake.nix
# save this as flake.nix
{
  description = "A disko images example";

  inputs = {
    nixpkgs.url = "github:NixOS/nixpkgs/nixos-unstable";
    disko.url = "github:nix-community/disko/latest";
    disko.inputs.nixpkgs.follows = "nixpkgs";
  };

  outputs = { self, disko, nixpkgs }: {
    nixosConfigurations.myhost = nixpkgs.lib.nixosSystem {
      system = "x86_64-linux";
      modules = [
        # You can get this file from here: https://github.com/nix-community/disko/blob/master/example/simple-efi.nix
        ./simple-efi.nix
        disko.nixosModules.disko
        ({ config, ... }: {
          # shut up state version warning
          system.stateVersion = config.system.nixos.version;
          # Adjust this to your liking.
          # WARNING: if you set a too low value the image might be not big enough to contain the nixos installation
          disko.devices.disk.main.imageSize = "10G";
        })
      ];
    };
  };
}
EOF

wget https://github.com/nix-community/disko/raw/refs/heads/master/example/simple-efi.nix

git init
git add .

nix build .#nixosConfigurations.myhost.config.system.build.diskoImagesScript
sudo ./result --build-memory 2048

Execution output

[excalibur@nixos:~/disko-test]$ ./test.sh 
--2024-11-27 09:56:55--  https://github.com/nix-community/disko/raw/refs/heads/master/example/simple-efi.nix
Resolving github.com (github.com)... 20.205.243.166
Connecting to github.com (github.com)|20.205.243.166|:443... connected.
HTTP request sent, awaiting response... 302 Found
Location: https://raw.githubusercontent.com/nix-community/disko/refs/heads/master/example/simple-efi.nix [following]
--2024-11-27 09:56:55--  https://raw.githubusercontent.com/nix-community/disko/refs/heads/master/example/simple-efi.nix
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 2606:50c0:8003::154, 2606:50c0:8000::154, 2606:50c0:8001::154, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|2606:50c0:8003::154|:443... failed: Connection refused.
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|2606:50c0:8000::154|:443... failed: Connection refused.
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|2606:50c0:8001::154|:443... failed: Connection refused.
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|2606:50c0:8002::154|:443... failed: Connection refused.
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.110.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 743 [text/plain]
Saving to: ‘simple-efi.nix’

simple-efi.nix                                           100%[================================================================================================================================>]     743  --.-KB/s    in 0s      

2024-11-27 09:56:55 (86.1 MB/s) - ‘simple-efi.nix’ saved [743/743]

Initialized empty Git repository in /home/excalibur/disko-test/.git/
warning: Git tree '/home/excalibur/disko-test' is dirty
warning: creating lock file '/home/excalibur/disko-test/flake.lock'
warning: Git tree '/home/excalibur/disko-test' is dirty
warning: error: unable to download 'https://mirrors.ustc.edu.cn/nix-channels/store/1zw7cxaz47bzhklwc5hdbxkxwh9plfd7.narinfo': Couldn't connect to server (7); retrying in 320 ms
[sudo] password for excalibur: 
/nix/store/skfbvi579f1wa30wa48c6qap28pswghn-vm-run: line 4: /setup: No such file or directory

[excalibur@nixos:~/disko-test]$ head /nix/store/skfbvi579f1wa30wa48c6qap28pswghn-vm-run
if [ -f "${NIX_ATTRS_SH_FILE-}" ]; then
  source "$NIX_ATTRS_SH_FILE"
fi
source $stdenv/setup
export > saved-env

PATH=/nix/store/k48bha2fjqzarg52picsdfwlqx75aqbb-coreutils-9.5/bin
mkdir xchg
mv saved-env xchg/

vm-run script is called here. Maybe the override caused some issues?

[RadxaYuntian]

Seems to be caused by this commit.

Test the flake with nixpkgs.url = "github:NixOS/nixpkgs/26eba25577388072ba13b1682860fe872ba39942";: issue mentioned above.

Test the flake with ecdda28a3a3a5bfa4d2b761a390f8b6ca3bdd6b3 (the parent commit):

at 15:49:12 ❯ ./result
Formatting 'main.raw', fmt=raw size=10737418240
qemu-kvm: -virtfs local,path=/build,security_model=none,mount_tag=sa: cannot initialize fsdev 'sa': failed to open '/build': No such file or directory

Test the flake with 23e89b7da85c3640bbc2173fe04f4bd114342367 (a random commit I have on another project with working diskoImagesScript): appears to be fine.

[RadxaYuntian]

Seems only affects diskoImagesScript. When I build for diskoImages VM can start.

[iFreilicht]

Thank you for the detailed report! I was able to reproduce the issue, I'm looking into it now.

[iFreilicht]

It seems this is actually caused by NixOS/nixpkgs#354535

I tried to fix this by making the build dir configurable in vmTools (see NixOS/nixpkgs@8812f20) and then overriding that setting in make-disk-image.nix.

This gets me further, but I end up with this:

EFI stub: Loaded initrd from LINUX_EFI_INITRD_MEDIA_GUID device path
loading kernel modules...
mounting Nix store...
mounting host's build directory...
starting stage 2 (/nix/store/6nz6pk6kf7w83lqq4a29h368f97vhyli-vm-run-stage2)
/nix/store/6nz6pk6kf7w83lqq4a29h368f97vhyli-vm-run-stage2: line 3: empty_build_dir/xchg/saved-env: No such file or directory
[    0.445652] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000100
[    0.446310] CPU: 0 PID: 1 Comm: 6nz6pk6kf7w83lq Not tainted 6.6.63 #1-NixOS
[    0.446956] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS unknown 02/02/2022
[    0.447613] Call Trace:
[    0.447816]  <TASK>
[    0.447999]  dump_stack_lvl+0x47/0x70
[    0.448289]  panic+0x180/0x340
[    0.448561]  do_exit+0x956/0xad0
[    0.448851]  do_group_exit+0x31/0x80
[    0.449178]  __x64_sys_exit_group+0x18/0x20
[    0.449581]  do_syscall_64+0x39/0x90
[    0.449897]  entry_SYSCALL_64_after_hwframe+0x78/0xe2
[    0.450366] RIP: 0033:0x7f1f3ba5dd1d
[    0.450808] Code: 45 31 c0 45 31 d2 45 31 db c3 0f 1f 00 f3 0f 1e fa 48 8b 35 e5 e0 10 00 ba e7 00 00 00 eb 07 66 0f 1f 44 00 00 f4 89 d0 0f 05 <48> 3d 00 f0 ff ff 76 f3 f7 d8 64 89 06 eb ec 0f 1f 40 00 f3 0f 1e
[    0.452877] RSP: 002b:00007ffdd4a23238 EFLAGS: 00000246 ORIG_RAX: 00000000000000e7
[    0.453682] RAX: ffffffffffffffda RBX: 00007f1f3bb6dfa8 RCX: 00007f1f3ba5dd1d
[    0.454333] RDX: 00000000000000e7 RSI: ffffffffffffff88 RDI: 0000000000000001
[    0.455080] RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000
[    0.455843] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
[    0.456475] R13: 0000000000000001 R14: 00007f1f3bb6c680 R15: 00007f1f3bb6dfc0
[    0.457136]  </TASK>
[    0.457362] Kernel Offset: disabled
[    0.457676] Rebooting in 1 seconds..
Virtual machine didn't produce an exit code.

And frankly, I'm not sure how to continue at this point.

[iFreilicht]

@RadxaYuntian please be aware that this will only be fixed once NixOS/nixpkgs#360413 is merged. However, we can't do anything else from our side now, so the issue being closed is correct.