Installing Arch Linux on ZFS
Related articles
This article details the steps required to install Arch Linux onto a ZFS root filesystem.
Contents
Installation
See ZFS#Installation for installing the ZFS packages. If installing Arch Linux onto ZFS from the archiso, it would be easier to use the archzfs repository.
Embedding archzfs into archiso
See ZFS article.
Arch ZFS installation scripts
Manually installing Arch using ZFS is quite an involved undertaking but thankfully there are scripts to simplify the process such as ALEZ and install-raidz.
Partition the destination drive
Review Partitioning for information on determining the partition table type to use for ZFS. ZFS supports GPT and MBR partition tables.
ZFS manages its own partitions, so only a basic partition table scheme is required. The partition that will contain the ZFS filesystem should be of the type bf00
, or "Solaris Root".
When using GRUB as your bootloader with an MBR partition table there is no need for a BIOS boot partition. Drives larger than 2TB require a GPT partition table and you should use parted to create the partitions for GPT. BIOS/GPT and UEFI/GPT configurations require a small (1/2MB) BIOS boot partition to store the bootloader. If you are using a UEFI-only bootloader you should use GPT.
Depending upon your choice of bootloader you may or may not require an EFI partition. GRUB, when installed on a BIOS machine (or a UEFI machine booting in legacy mode) using either MBR or GPT doesn't require an EFI partition. Consult Boot loaders for more info.
Partition scheme
Here is an example of a basic partition scheme that could be employed for your ZFS root install on a BIOS/MBR installation using GRUB:
Part Size Type ---- ---- ------------------------- 1 XXXG Solaris Root (bf00)
Using GRUB on a BIOS (or UEFI machine in legacy boot mode) machine but using a GPT partition table:
Part Size Type ---- ---- ------------------------- 1 2M BIOS boot partition (ef02) 2 XXXG Solaris Root (bf00)
Another example, this time using a UEFI-specific bootloader (such as rEFInd) and GPT:
Part Size Type ---- ---- ------------------------- 1 2M BIOS boot partition (ef02) 2 100M EFI boot partition (ef00) 3 XXXG Solaris Root (bf00)
ZFS does not support swap files. If you require a swap partition, see ZFS#Swap volume for creating a swap ZVOL.
Example parted commands
Here are some example commands to partition a drive for the second scenario above ie using BIOS/legacy boot mode with a GPT partition table and a (slighty more than) 1MB BIOS boot partition for GRUB:
# parted /dev/sdx (parted)mklabel gpt (parted)mkpart non-fs 0% 2 (parted)mkpart primary 2 100% (parted)set 1 bios_grub on (parted)set 2 boot on (parted)quit
You can achieve the above in a single command like so:
parted --script /dev/sdx mklabel gpt mkpart non-fs 0% 2 mkpart primary 2 100% set 1 bios_grub on set 2 boot on
If you are creating an EFI partition then that should have the boot flag set instead of the root partition.
Format the destination disk
If you have opted for a boot partition as well as any other non-ZFS system partitions then format them. Do not do anything to the Solaris partition nor to the BIOS boot partition. ZFS will manage the first, and your bootloader the second.
Setup the ZFS filesystem
First, make sure the ZFS modules are loaded,
# modprobe zfs
Create the root zpool
# zpool create -f zroot /dev/disk/by-id/id-to-partition-partx
Create your datasets
Instead of using conventional disk partitions, ZFS has the concept of datasets to manage your storage. Unlike disk partitions, datasets have no fixed size and allow for different attributes, such as compression, to be applied per dataset. Normal ZFS datasets are mounted automatically by ZFS whilst legacy datasets are required to be mounted using fstab or with the traditional mount command.
One of the most useful features of ZFS is boot environments. Boot environments allow you to create a bootable snapshot of your system that you can revert to at any time instantly by simply rebooting and booting from that boot environment. This can make doing system updates much safer and is also incredibly useful for developing and testing software. In order to be able to use beadm to manage boot environments your datasets must be configured properly. Key to this are that you split your data directories (such as /home
) into datasets that are distinct from your system datasets and that you do not place data in the root of the pool as this cannot be moved afterwards.
You should always create a dataset for at least your root filesystem and in nearly all cases you will also want /home
to be in a separate dataset. You may decide you want your logs to persist over boot environments. If you are a running any software that stores data outside of /home
(such as is the case for database servers) you should structure your datasets so that the data directories of the software you want to run are separated out from the root dataset.
With these example commands, we will create a basic boot environment compatible configuration comprising of just root and /home
datasets with lz4 compression to save space and improve IO performance:
# zfs create -o mountpoint=none zroot/data # zfs create -o mountpoint=none zroot/ROOT # zfs create -o compression=lz4 -o mountpoint=/ zroot/ROOT/default # zfs create -o compression=lz4 -o mountpoint=/home zroot/data/home
Configure the root filesystem
If you have just created your zpool, it will be mounted in a dir at the root of your tree named after the pool (ie /zroot). If the following set commands fail, you may need to unmount any ZFS filesystems first:
# zfs umount -a
Now set the mount points of the datasets:
# zfs set mountpoint=/ zroot/ROOT/default # zfs set mountpoint=legacy zroot/data/home
and put them in /etc/fstab
/etc/fstab
# <file system> <dir> <type> <options> <dump> <pass> zroot/ROOT/default / zfs defaults,noatime 0 0 zroot/data/home /home zfs defaults,noatime 0 0
All legacy datasets must be listed in /etc/fstab
or they will not be mounted at boot.
Set the bootfs property on the descendant root filesystem so the boot loader knows where to find the operating system.
# zpool set bootfs=zroot/ROOT/default zroot
Export the pool,
# zpool export zroot
Finally, re-import the pool,
# zpool import -d /dev/disk/by-id -R /mnt zroot
If there is an error in this step, you can export the pool to redo the command. The ZFS filesystem is now ready to use.
Be sure to bring the zpool.cache
file into your new system. This is required later for the ZFS daemon to start.
# cp /etc/zfs/zpool.cache /mnt/etc/zfs/zpool.cache
if you do not have /etc/zfs/zpool.cache
, create it:
# zpool set cachefile=/etc/zfs/zpool.cache zroot
Install and configure Arch Linux
Follow the following steps using the Installation guide. It will be noted where special consideration must be taken for ZFSonLinux.
- First mount any legacy or non-ZFS boot or system partitions using the mount command.
- Install the base system.
- The procedure described in Installation guide#Fstab is usually overkill for ZFS. ZFS usually auto mounts its own partitions, so we do not need ZFS partitions in
fstab
file, unless the user made legacy datasets of system directories. To generate thefstab
for filesystems, use:
# genfstab -U -p /mnt >> /mnt/etc/fstab
- Edit the
/etc/fstab
:
- You need to add the Arch ZFS repository to
/etc/pacman.conf
, sign its key and install zfs-linux (or zfs-linux-lts if you are running the LTS kernel) within the arch-chroot before you can update the ramdisk with ZFS support.
- When creating the initial ramdisk, first edit
/etc/mkinitcpio.conf
and addzfs
before filesystems. Also, movekeyboard
hook beforezfs
so you can type in console if something goes wrong. You may also remove fsck (if you are not using Ext3 or Ext4). YourHOOKS
line should look something like this:
HOOKS="base udev autodetect modconf block keyboard zfs filesystems"
Install and configure the bootloader
For BIOS motherboards
Follow GRUB#BIOS systems to install GRUB onto your disk. grub-mkconfig
does not properly detect the ZFS filesystem, so it is necessary to edit grub.cfg
manually:
/boot/grub/grub.cfg
set timeout=2 set default=0 # (0) Arch Linux menuentry "Arch Linux" { linux /vmlinuz-linux zfs=zroot rw initrd /initramfs-linux.img }
if you did not create a separate /boot partition, kernel and initrd paths have to be in the following format:
/dataset/@/actual/path
Example with Arch installed on the main dataset (not recommended - this will not allow for boot environments):
linux /@/boot/vmlinuz-linux zfs=zroot rw initrd /@/boot/initramfs-linux.img
Example with Arch installed on a separate dataset zroot/ROOT/default:
linux /ROOT/default/@/boot/vmlinuz-linux zfs=zroot/ROOT/default rw initrd /ROOT/default/@/boot/initramfs-linux.img
When you come to installing GRUB, you are likely to get an error like:
Failed to get canonical path of /dev/ata-yourdriveid-partx
Until this gets fixed, the easiest workaround is to create a symbolic link from the regular Linux device name of the partition to the device name GRUB is looking for:
# ln -s /dev/sdax /dev/ata-yourdriveid-partx
For UEFI motherboards
Use EFISTUB
and rEFInd
for the UEFI boot loader. The kernel parameters in refind_linux.conf
for ZFS should include zfs=bootfs
or zfs=zroot
so the system can boot from ZFS. The root
and rootfstype
parameters are not needed.
Unmount and restart
We are almost done!
# exit # umount /mnt/boot (if you have a legacy boot partition) # zfs umount -a # zpool export zroot
Now reboot.
After the first boot
If everything went fine up to this point, your system will boot. Once.
For your system to be able to reboot without issues, you need to enable the zfs.target
to auto mount the pools and set the hostid.
For each pool you want automatically mounted execute:
# zpool set cachefile=/etc/zfs/zpool.cache <pool>
Enable the target with systemd:
# systemctl enable zfs.target
When running ZFS on root, the machine's hostid will not be available at the time of mounting the root filesystem. There are two solutions to this. You can either place your spl hostid in the kernel parameters in your boot loader. For example, adding spl.spl_hostid=0x00bab10c
, to get your number use the hostid
command.
The other, and suggested, solution is to make sure that there is a hostid in /etc/hostid
, and then regenerate the initramfs image which will copy the hostid into the initramfs image. To write the hostid file safely you need to use a small C program:
#include <stdio.h> #include <errno.h> #include <unistd.h> int main() { int res; res = sethostid(gethostid()); if (resĀ != 0) { switch (errno) { case EACCES: fprintf(stderr, "Error! No permission to write the" " file used to store the host ID.\n" "Are you root?\n"); break; case EPERM: fprintf(stderr, "Error! The calling process's effective" " user or group ID is not the same as" " its corresponding real ID.\n"); break; default: fprintf(stderr, "Unknown error.\n"); } return 1; } return 0; }
Copy it, save it as writehostid.c
and compile it with gcc -o writehostid writehostid.c
, finally execute it and regenerate the initramfs image:
# ./writehostid # mkinitcpio -p linux
You can now delete the two files writehostid.c
and writehostid
. Your system should work and reboot properly now.