Introduction

ZFSBootMenu is a Dracut module that intends to provide Linux distributions with an experience similar to FreeBSD's bootloader. By taking advantage of ZFS features, it allows a user to have multiple "boot environments" (with different distros, for example), manipulate snapshots before booting, and, for the adventurous user, even bootstrap a system installation via zfs recv.

In broad strokes, it works as follows:

• Via direct EFI booting, rEFInd, syslinux, etc, boot a Linux kernel along with an initramfs containing ZFSBootMenu.
• Look for zfsbootmenu in the kernel command line.
  • Optionally specify a default pool (if multiple are present).
• Find all healthy ZFS pools and import them.
• If a specific pool was set, look for the bootfs pool value. Prefer this boot environment.
  • If no pool was defined in the command line, use the bootfs value on the first-found pool.
  • If a bootfs value is defined, start a 10 second (by default) countdown to boot that environment with the highest versioned kernel found in /boot.
  • If no bootfs value is defined, find every filesystem that mounts to / with a /boot directory, and find every kernel image. Prompt for boot environment selection via a fuzzy finder.
    • If needed, prompt for encryption passphrases.
• Once the countdown has been reached for the bootfs-selected environment, prompt for the encryption passphrase if needed.
• Mount the filesystem and find the highest versioned kernel in /boot in the selected boot environment.
• Load the selected kernel and initramfs with the kernel command line defined in the org.zfsbootmenu:commandline property (or, as a fallback, /etc/default/grub) into memory with kexec.
• Unmount all ZFS filesystems.
• Boot the final kernel and initramfs.

At this point, you'll be booting into your usual OS-managed kernel and initramfs, along with any arguments needed to correctly boot your system.

This tool makes uses of the following additional software:

• fzf or skim
• kexec-tools
• mbuffer
• Linux Kernel
• ZFS on Linux
• dracut

ZFSBootMenu has been tested successfully with Kernel 5.8.14, Dracut 050 and OpenZFS 2.0.0-rc4.

System prereqs

To ensure the boot menu can find your kernels, you'll need to ensure /boot resides on your ZFS filesystem. An example filesystem layout is as follows:

NAME                           USED  AVAIL     REFER  MOUNTPOINT
zroot                          278G   582G       96K  none
zroot/ROOT                    10.9G   582G       96K  none
zroot/ROOT/void.2019.10.04    1.20M   582G     7.17G  /
zroot/ROOT/void.2019.11.01    10.9G   582G     7.17G  /
zroot/home                     120G   582G     11.8G  /home

There are two boot environments created, identified by mounting to /. The environment that this system will boot into is defined by the bootfs value set on the zroot zpool.

NAME   PROPERTY  VALUE                       SOURCE
zroot  bootfs    zroot/ROOT/void.2019.11.01  local

On start, ZFSBootMenu will find the highest versioned kernel in zroot/ROOT/void.2019.11.01/boot, confirm that a matching initramfs is present, and default to booting the OS with that.

Installation

Kernel command line arguments should be configured by setting the org.zfsbootmenu:commandline ZFS property for each boot environment. If the property is not defined for a boot environment or its parents, command line arguments will be taken from the GRUB_CMDLINE_LINUX_DEFAULT variable defined in the boot environment's /etc/default/grub file, if it exists and the variable is set. Do not set any root= or any other pool-related options in the kernel command line; these will be filled in automatically when a boot environment is selected.

For example, I have the following command line arguments set for my boot environment:

zfs.zfs_arc_max=8589934592 elevator=noop

Because ZFS properties are inherited by default, it is possible to set the org.zfsbootmenu:commandline property on a common parent to apply the same arguments to multiple environments. Setting the property locally on individual boot environments will override the common defaults.

EFI

ZFSBootMenu integrates nicely with an EFI system. There will be two key things to configure here.

• The mountpoint of the EFI partition and its contents.
• The mountpoint of the boot environment /boot and its contents.

Each boot environment should have its /boot directory in the ZFS filesystem. Using the above example, zroot/ROOT/void.2019.11.01 would contain /boot with kernel/initramfs pairs:

# ls /boot
config-5.3.18_1
config-5.4.6_1
efi
initramfs-5.3.18_1.img
initramfs-5.4.6_1.img
System.map-5.3.18_1
System.map-5.4.6_1
vmlinuz-5.3.18_1
vmlinuz-5.4.6_1

Once /boot is contained in a boot environment, it is necessary to install the boot menu files. Typically, EFI partitions (ESP) are mounted to /boot/efi, and contain a number of sub-directories. In this example, /boot/efi/EFI/void holds the ZFSBootMenu kernel and initramfs.

# lsblk -f /dev/sda
NAME   FSTYPE LABEL UUID                                 FSAVAIL FSUSE% MOUNTPOINT
sdg
├─sda1 vfat         AFC2-35EE                               7.9G     1% /boot/efi
└─sda2 swap         412401b6-4aec-4452-a6bd-6fc20fbdc2a5                [SWAP]

# ls /boot/efi/EFI/void/
initramfs-0.7.4.img
initramfs-0.7.5.img
vmlinuz-0.7.4
vmlinuz-0.7.5

With this layout, you'll now need a way to boot the kernel and initramfs via EFI. This can be done via a manual entry set via efibootmgr, or it can be done with rEFInd.

If you do not generate the ZFSBootMenu initramfs locally, you'll need to identify the following additional details:

• Your system's hostid (hostid). It's important that this command be executed as root, to ensure that it returns the correct value.
• Your boot pool name, if you have multiple.
• The disk path and partition index of your EFI partition. (/dev/sda, part 1)

efibootmgr

efibootmgr --disk /dev/sda \
  --part 1 \
  --create \
  --label "ZFSBootMenu" \
  --loader /vmlinuz-0.7.5 \
  --unicode 'root=zfsbootmenu:POOL=zroot ro initrd=\EFI\void\initramfs-0.7.5.img quiet spl_hostid=a8c0a2a8' \
  --verbose

Take note to adjust root=zfsbootmenu:POOL=, spl_hostid=, --disk and --part to match your system configuration.

Each time ZFSBootMenu is updated, a new EFI entry will need to be manually added, unless you disable versioning in the ZFSBootMenu configuration.

rEFInd

rEFInd is considerably easier to install and manage. Refer to your distribution's packages for installation. Once rEFInd has been installed, you can create refind_linux.conf in the directory holding the ZFSBootMenu files (/boot/efi/EFI/void in our example):

"Boot Default BE" "ro quiet loglevel=0 timeout=0 root=zfsbootmenu:POOL= spl_hostid="
"Select BE" "ro quiet loglevel=0 timeout=-1 root=zfsbootmenu:POOL= spl_hostid="

As with the efibootmgr section, the root=zfsbootmenu:POOL= and spl_hostid= options need to be configured to match your environment.

This file will configure rEFInd to create two entries for each kernel and initrams pair it finds. The first will directly boot into the environment set via the bootfs pool property. The second will force ZFSBootMenu to display an environment / kernel / snapshot selection menu, allowing you to boot alternate environments, kernels and snapshots.

Kernel command line options

The zfsbootmenu(7) manual page describes command line options for ZFSBootMenu kernels in detail.

ZFS properties

The zfsbootmenu(7) manual page describes ZFS properties interpreted by ZFSBootMenu.

initramfs creation

bin/generate-zbm can be used to create an initramfs on your system. It ships with Void-specific defaults in etc/zfsbootmenu/config.yaml. To create an initramfs, the following additional tools/libraries will need to be available on your system:

• For inclusion in the initramfs:
  • fzf or skim
  • kexec-tools
  • mbuffer
• For running bin/generate-zbm:
  • perl Sort::Versions
  • perl Config::IniFiles
  • perl YAML::PP
  • perl boolean

If you want to create a unified EFI executable (which bundles the kernel, initramfs and command line), you will also need:

• linuxx64.efi.stub (typically packaged with gummiboot or systemd-boot)

Your distribution should have packages for these already.

config.yaml

config.yaml is used to control the operation of generate-zbm.

Conversion of legacy configurations

In prior versions of ZFSBootMenu, an INI format was used for configuration. In general, migration to the new format is not automatic, but generate-zbm can perform the migration if your distribution package has not done it for you. To migrate an existing configuration, just run

generate-zbm --migrate [ini-config] [--config yaml-config]

By default, the output YAML will be written to /etc/zfsbootmenu/config.yaml; use the --config argument to customize the output location.

The argument [ini-config] to --migrate is optional. When it is not provided, generate-zbm will derive an input file by replacing the .yaml extension from the output file with a .ini extension.

If (and only if) generate-zbm is run without a --config option (i.e., it attempts to load the default /etc/zfsbootmenu/config.yaml) and the default configuration does not exist, generate-zbm will behave as if it had been passed the --migrate /etc/zfsbootmenu/config.ini option.

Whenever generate-zbm attempts to migrate configuraton files, it will exit with a zero exit code on successful conversion and a nonzero exit code if problems were encountered during the conversion. No boot images will be produced in the same invocation as a migration attempt.

Dealing with driver conflicts

For some combination of hardware and kernel modules, the ZFSBootMenu kernel may leave hardware in an unexpected state and prevent the boot environment from properly initializing and attaching drivers. The simplest way to avoid this issue is to disable the affected kernel modules in ZFSBootMenu, leaving all hardware initialization to the final kernel. For example, if Nvidia graphics hardware does not function as expected, a dracut configuration file can be added to /etc/zfsbootmenu/dracut.conf.d to exclude the nouveau and nvidia drivers from ZFSBootMenu. Adding the line

omit_drivers+=" nouveau nvidia "

to a file called, e.g., /etc/zfsbootmenu/dracut.conf.d/nvidia.conf should restore expected functionality to your boot environment after recreating your ZFSBootMenu image with generate-zbm.

In other cases, it is not possible to exclude drivers without depriving ZFSBootMenu of critical hardware support. For example, some XHCI USB controllers may not be properly initialized after a kexec, leaving a boot environment without USB devices like a keyboard. However, excluding XHCI drivers from ZFSBootMenu would make the same keyboard inoperable in the boot menu, making it impossible to interact with the menus. ZFSBootMenu provides "teardown hooks" that can sometimes be used to address these situations. Teardown hooks are invoked immediately before a target kernel is booted via kexec and provide an opportunity to run last-minute commands to prepare the system for the boot. Scripts may be registered as teardown hooks by adding text of the form

zfsbootmenu_teardown+=" <path to script> "

where <path to script> points to an executable script or program. A sample XHCI teardown script demonstrates the use of teardown hooks to unbind the XHCI driver from the USB controllers in the ZFSBootMenu kernel before launching the selected boot environment, allowing the next kernel to properly initialize the controller.

Native encryption

ZFSBootMenu can import pools or filesystems with native encryption enabled. If your boot environments are not encrypted but, for example, /home is, you will not receive a decryption prompt during boot. To ensure that you can decrypt your pool to load the kernel and initramfs, you'll need to you have the filesystem parameters configured correctly.

zfs get all zroot | egrep '(encryption|keylocation|keyformat)'
zroot  encryption            aes-256-gcm                -
zroot  keylocation           file:///etc/zfs/zroot.key  local
zroot  keyformat             passphrase                 -
zroot  encryptionroot        zroot                      -

It's critical that keyformat is set to passphrase, otherwise you'll be unable to enter the correct value in the boot loader. ZFS on Linux currently supports only one key, but in a way which we can exploit: if you configure the keylocation value to a file on disk, put your passphrase in that, and then include that file into the FINAL initramfs (the OS-managed one), you won't receive a second password prompt on boot. You'll still receive a password prompt in the boot loader, since we can force a prompt for passphrase input.

For Dracut-based systems, this can be done by creating a /etc/dracut.conf.d/zol.conf file with the following contents:

install_items+=" /etc/zfs/zroot.key "

It's critical that you do not include this key file into the ZFSBootMenu initramfs, since that file exists on an unencrypted volume - leaving your pool essentially wide-open.