hey everyone, im very new to disko and disk partitioning/management in general, but i had a few questions as i prep to use nixos-anywhere to setup/expand my homelab.

i have 5 Intel NUC computers, each with a SSD and a 1.8T HDD. The issue is that some of them report /dev/sda to be the SSD and /dev/sdb to be the HDD, while some do reverse. is this a problem when working with disko? aka would i be able to reuse the same disko config for each one? or do i necessarily need to tailor each one specifically to the output of lsblk?

hey everyone, im very new to disko and disk partitioning/management in general, but i had a few questions as i prep to use nixos-anywhere to setup/expand my homelab.

i have 5 Intel NUC computers, each with a SSD and a 1.8T HDD. The issue is that some of them report /dev/sda to be the SSD and /dev/sdb to be the HDD, while some do reverse. is this a problem when working with disko? aka would i be able to reuse the same disko config for each of the 5 computers? or do i necessarily need to tailor each one specifically to the output of lsblk?

hey everyone, im very new to disko and disk partitioning/management in general, but i had a few questions as i prep to use nixos-anywhere to setup/expand my homelab.

i have 5 Intel NUC computers, each with a SSD and a 1.8T HDD. The issue is that some of them report /dev/sda to be the SSD and /dev/sdb to be the HDD, while some do reverse. is this a problem when working with disko? aka would i be able to reuse the same disko config for each of the 5 computers? or do i necessarily need to tailor each one specifically to the output of its respective host's lsblk?

im especially confused here:

{
  disko.devices.disk.example = {
    type = "disk";
    device = "/dev/nvme0n1";
    content = {
      type = "gpt";
      partitions = {
        ESP = {
          size = "512MiB";
          type = "EF00";
          priority = 1;
          content = {
            type = "filesystem";
            format = "vfat";
            mountpoint = "/boot";
            mountOptions = [ "umask=0077" ];
          };
        };
        root = {
          size = "100%";
          priority = 2;
          content.format = "ext4";
        };
      };
    };
  };
}

how do i know what the Type of the partition is? and also the file system format (i.e. "vfat" vs "ext4")? are these things i just need to know about the hardware beforehand?

so when adding a new machine to my network, i should aim for a workflow that looks like:

1. ssh into the target and run ls -l /dev/disk/by-id/
2. using that information, write a disko configuration at ./modules/node-specific/machine-X/disko.nix
3. Run nixos-anywhere using that disko.nix
4. Add a node to my deploy-rs setup, which reuses that same disko.nix (and anything else in node-specific/machine-X/)
5. Future rebuilds are done via deploy-rs

In reply to @caraiiwala:beeper.com
https://github.com/nix-community/disko/issues/678 is what I'm referring to

/nix/store/4ps8pa311972zw8k9c7z21w04qhwpr8j-grub-2.12/sbin/grub-install: warning: this GPT partition label contains no BIOS Boot Partition; embedding won't be possible.
/nix/store/4ps8pa311972zw8k9c7z21w04qhwpr8j-grub-2.12/sbin/grub-install: warning: Embedding is not possible.  GRUB can only be installed in this setup by using blocklists.  However, blocklists are UNRELIABLE and their use is discouraged..
/nix/store/4ps8pa311972zw8k9c7z21w04qhwpr8j-grub-2.12/sbin/grub-install: error: will not proceed with blocklists.
/nix/store/dhn57d6hqajp0j77l33alsqwvv54kv5a-install-grub.pl: installation of GRUB on /dev/sda failed: No such file or directory
Failed to install bootloader

# sgdisk -c 1:disk-nvme-ESP /dev/sda1

***************************************************************
Found invalid GPT and valid MBR; converting MBR to GPT format
in memory. 
***************************************************************

Unable to set partition 1's name to 'disk-nvme-ESP'!
Error encountered; not saving changes.