so... does disko offer some configuration mode where I could:

1. update configuration of my zpool without erasing anything
2. don't touch the other zpool that is there in the NixOS configuration

Hi, I’m installing nixos in a vm using disko. The installation finishes without errors, but when booting into the system, it drops into emergency mode because /dev/disk/by-partlabel/ROOT cannot be found.

I’ve verified that /dev/disk/by-partlabel/ROOT does exist and can be mounted manually from the iso.

This is my disko config: https://github.com/god464/flake/blob/master/hosts/server/disk.nix

Is there anything I’m missing?

Hope someone can help: Have tiny home server/router with two USB SSDs forming a ZFS mirror. Also each disk has swap and ESP partition. One disk seems to die now (dmesg says usb 3-1: device not accepting address). Even though ZFS should give redundancy, the system becomes unresponsive. Boot loader is on the dying disk. Adding the mountpoint on the second disk does not have effect after rebuild and reboot:

          ESP = {
            size = "512M";
            type = "EF00";
            content = {
              type = "filesystem";
              format = "vfat";
              mountpoint = "/boot-fallback";         # added
              mountOptions = [ "umask=0077" ]; # added
            };
          };

Do I need to do something extra after changing the disko.nix file (adding the two lines above)?
I want to get redundancy for the bootloader. Do I need grub for this (used systemd-boot so far)?