ZFS as Root Filesystem

Introduction

This tutorial will show you how to install Funtoo on ZFS (rootfs). This tutorial is meant to be an "overlay" over the Regular Funtoo Installation. Follow the normal installation and only use this guide for steps 2, 3, and 8.

Introduction to ZFS

Since ZFS is a new technology for Linux, it can be helpful to understand some of its benefits, particularly in comparison to BTRFS, another popular next-generation Linux filesystem:

• On Linux, the ZFS code can be updated independently of the kernel to obtain the latest fixes. btrfs is exclusive to Linux and you need to build the latest kernel sources to get the latest fixes.

• ZFS is supported on multiple platforms. The platforms with the best support are Solaris, FreeBSD and Linux. Other platforms with varying degrees of support are NetBSD, Mac OS X and Windows. btrfs is exclusive to Linux.

• ZFS has the Adaptive Replacement Cache replacement algorithm while btrfs uses the Linux kernel's Last Recently Used replacement algorithm. The former often has an overwhelmingly superior hit rate, which means fewer disk accesses.

• ZFS has the ZFS Intent Log and SLOG devices, which accelerates small synchronous write performance.

• ZFS handles internal fragmentation gracefully, such that you can fill it until 100%. Internal fragmentation in btrfs can make btrfs think it is full at 10%. Btrfs has no automatic rebalancing code, so it requires a manual rebalance to correct it.

• ZFS has raidz, which is like RAID 5/6 (or a hypothetical RAID 7 that supports 3 parity disks), except it does not suffer from the RAID write hole issue thanks to its use of CoW and a variable stripe size. btrfs gained integrated RAID 5/6 functionality in Linux 3.9. However, its implementation uses a stripe cache that can only partially mitigate the effect of the RAID write hole.

• ZFS send/receive implementation supports incremental update when doing backups. btrfs' send/receive implementation requires sending the entire snapshot.

• ZFS supports data deduplication, which is a memory hog and only works well for specialized workloads. btrfs has no equivalent.

• ZFS datasets have a hierarchical namespace while btrfs subvolumes have a flat namespace.

• ZFS has the ability to create virtual block devices called zvols in its namespace. btrfs has no equivalent and must rely on the loop device for this functionality, which is cumbersome.

The only area where btrfs is ahead of ZFS is in the area of small file efficiency. btrfs supports a feature called block suballocation, which enables it to store small files far more efficiently than ZFS. It is possible to use another filesystem (e.g. reiserfs) on top of a ZFS zvol to obtain similar benefits (with arguably better data integrity) when dealing with many small files (e.g. the portage tree).

For a quick tour of ZFS and have a big picture of its common operations you can consult the page ZFS Fun.

Disclaimers

   Warning

This guide is a work in progress. Expect some quirks.

Today is 2015-05-12. ZFS has undertaken an upgrade - from 0.6.3 to 0.6.4. Please ensure that you use a RescueCD with ZFS 0.6.3. At present date grub 2.02 is not able to deal with those new ZFS parameters. If you want to use ZFS 0.6.4 for pool creation, you should use the compatability mode.

You should upgrade an existing pool only when grub is able to deal with - in a future version ... If not, you will not be able to boot into your system, and no rollback will help!

Please inform yourself!

   Important

Since ZFS was really designed for 64 bit systems, we are only recommending and supporting 64 bit platforms and installations. We will not be supporting 32 bit platforms!

Downloading the ISO (With ZFS)

In order for us to install Funtoo on ZFS, you will need an environment that already provides the ZFS tools. Therefore we will download a customized version of System Rescue CD with ZFS included.

Name: sysresccd-4.2.0_zfs_0.6.2.iso  (545 MB)
Release Date: 2014-02-25
md5sum 01f4e6929247d54db77ab7be4d156d85

Download System Rescue CD with ZFS

Creating a bootable USB from ISO (From a Linux Environment)

After you download the iso, you can do the following steps to create a bootable USB:

Make a temporary directory
root # mkdir /tmp/loop

Mount the iso
root # mount -o ro,loop /root/sysresccd-4.2.0_zfs_0.6.2.iso /tmp/loop

Run the usb installer
root # /tmp/loop/usb_inst.sh

That should be all you need to do to get your flash drive working.

Booting the ISO

   Warning

When booting into the ISO, Make sure that you select the "Alternate 64 bit kernel (altker64)". The ZFS modules have been built specifically for this kernel rather than the standard kernel. If you select a different kernel, you will get a fail to load module stack error message.

Creating partitions

There are two ways to partition your disk: You can use your entire drive and let ZFS automatically partition it for you, or you can do it manually.

We will be showing you how to partition it manually because if you partition it manually you get to create your own layout, you get to have your own separate /boot partition (Which is nice since not every bootloader supports booting from ZFS pools), and you get to boot into RAID10, RAID5 (RAIDZ) pools and any other layouts due to you having a separate /boot partition.

gdisk (GPT Style)

A Fresh Start:

First lets make sure that the disk is completely wiped from any previous disk labels and partitions. We will also assume that /dev/sda is the target drive.

root # sgdisk -Z /dev/sda

   Warning

This is a destructive operation and the program will not ask you for confirmation! Make sure you really don't want anything on this disk.

Now that we have a clean drive, we will create the new layout.

First open up the application:

root # gdisk /dev/sda

Create Partition 1 (boot):

Command: n ↵
Partition Number: ↵
First sector: ↵
Last sector: +250M ↵
Hex Code: ↵

Create Partition 2 (BIOS Boot Partition):

Command: n ↵
Partition Number: ↵
First sector: ↵
Last sector: +32M ↵
Hex Code: EF02 ↵

Create Partition 3 (ZFS):

Command: n ↵
Partition Number: ↵
First sector: ↵
Last sector: ↵
Hex Code: bf00 ↵

Command: p ↵

Number  Start (sector)    End (sector)  Size       Code  Name
   1            2048          514047   250.0 MiB   8300  Linux filesystem
   2          514048          579583   32.0 MiB    EF02  BIOS boot partition
   3          579584      1953525134   931.2 GiB   BF00  Solaris root

Command: w ↵

Format your /boot partition

root # mkfs.ext2 -m 1 /dev/sda1

Create the zpool

We will first create the pool. The pool will be named tank. Feel free to name your pool as you want. We will use ashift=12 option which is used for a hard drives with a 4096 sector size.

root #   zpool create -f -o ashift=12 -o cachefile=/tmp/zpool.cache -O normalization=formD -m none -R /mnt/funtoo tank /dev/sda3 

Create the zfs datasets

We will now create some datasets. For this installation, we will create a small but future proof amount of datasets. We will have a dataset for the OS (/), and your swap. We will also show you how to create some optional datasets: /home, /usr/src, and /usr/portage. Notice, datasets are examples only and not strictly required.

Create some empty containers for organization purposes, and make the dataset that will hold /
root #  zfs create -p tank/funtoo
root #  zfs create -o mountpoint=/ tank/funtoo/root

Optional, Create swap
root #  zfs create tank/swap -V 2G -b 4K
root #  mkswap /dev/tank/swap
root #  swapon /dev/tank/swap

Optional, but recommended datasets: /home
root #  zfs create -o mountpoint=/home tank/funtoo/home

Optional datasets: /usr/src, /usr/portage/{distfiles,packages}
root #  zfs create -o mountpoint=/usr/src tank/funtoo/src
root #  zfs create -o mountpoint=/usr/portage -o compression=off tank/funtoo/portage
root #  zfs create -o mountpoint=/usr/portage/distfiles tank/funtoo/portage/distfiles
root #  zfs create -o mountpoint=/usr/portage/packages tank/funtoo/portage/packages

Installing Funtoo

Pre-Chroot

Go into the directory that you will chroot into
root # cd /mnt/funtoo

Make a boot folder and mount your boot drive
root # mkdir boot
root # mount /dev/sda1 boot

Now download and extract the Funtoo stage3 ...

   Note

It is trully recommended to use the current version and generic64. That reduces the risk of a broken build.

After successfull ZFS installation and successfull first boot, the kernel may be changed using the eselect profile set ... command. If you create a snapshot before, you may allways come back to your previous installation, with some simple steps ... (rollback your pool and in the worst case configure and install the bootloader again)

Once you've extracted the stage3, do a few more preparations and chroot into your new funtoo environment:

Bind the kernel related directories
root # mount -t proc none proc
root # mount --rbind /dev dev
root # mount --rbind /sys sys

Copy network settings
root # cp -f /etc/resolv.conf etc

Make the zfs folder in 'etc' and copy your zpool.cache
root # mkdir etc/zfs
root # cp /tmp/zpool.cache etc/zfs

Chroot into Funtoo
root # env -i HOME=/root TERM=$TERM chroot . bash -l

   Note

How to create zpool.cache file?

If no zpool.cache file is available, the following command will create one:

root # zpool set cachefile=/etc/zfs/zpool.cache tank

Install/PortageTree

Add filesystems to /etc/fstab

Before we continue to compile and or install our kernel in the next step, we will edit the /etc/fstab file because if we decide to install our kernel through portage, portage will need to know where our /boot is, so that it can place the files in there.

Edit /etc/fstab:

   /etc/fstab

# <fs>                  <mountpoint>    <type>          <opts>          <dump/pass>

/dev/sda1               /boot           ext2            defaults        0 2

# If you set up a swap partition, you will have to add this line as well
/dev/zvol/tank/swap     none            swap            defaults        0 0

Building kernel, initramfs and grub to work with zfs

Install genkernel and initial kernel build

We need to build a genkernel initially:

root # emerge genkernel

Build initial kernel (required for checks in sys-kernel/spl and sys-fs/zfs):
root # genkernel kernel --no-clean --no-mountboot 

Installing the ZFS userspace tools and kernel modules

Emerge No results. This package will bring in No results, and No results as its dependencies:

root # emerge zfs

Check to make sure that the zfs tools are working. The zpool.cache file that you copied before should be displayed.

root # zpool status
root # zfs list

   Note

If /etc/mtab is missing, these two commands will complaine. In that case solve that with:

root # grep -v rootfs /proc/mounts > /etc/mtab

Add the zfs tools to openrc.

root # rc-update add zfs boot

If everything worked, continue.

Install GRUB 2

Install grub2:

root # echo "sys-boot/grub libzfs -truetype" >> /etc/portage/package.use
root # emerge grub

Now install grub to the drive itself (not a partition):

root # grub-install /dev/sda

Initial kernel build

Build now kernel and initramfs with --zfs

root # genkernel all --zfs --no-clean --no-mountboot --callback="emerge @module-rebuild"

Using the debian-sources, the following command may give better results:

root # genkernel all --zfs --no-clean --no-mountboot --callback="emerge spl zfs-kmod zfs"

Configuring the Bootloader

Using the genkernel you must add 'real_root=ZFS=<root>' and 'dozfs' to your params. Edit /etc/boot.conf:

   /etc/boot.conf

"Funtoo ZFS" {
        kernel kernel[-v]
        initrd initramfs-genkernel-x86_64[-v]
        params real_root=ZFS=tank/funtoo/root
        params += dozfs=force
}

The command boot-update should take care of grub configuration:

Install boot-update (if it is missing):
root # emerge boot-update

Run boot-update to update grub.cfg
root # boot-update

   Note

If boot-updatefails, try this:

root # grub-mkconfig -o /boot/grub/grub.cfg

Now you should have a new installation of the kernel, initramfs and grub which are zfs capable. The configuration files should be updated, and the system should come up during the next reboot.

   Note

If The luks integration works basically the same way.

Final configuration

Clean up and reboot

We are almost done, we are just going to clean up, set our root password, and unmount whatever we mounted and get out.

Delete the stage3 tarball that you downloaded earlier so it doesn't take up space.
root # cd /
root # rm stage3-latest.tar.xz

Set your root password
root # passwd
>> Enter your password, you won't see what you are writing (for security reasons), but it is there!

Get out of the chroot environment
root # exit

Unmount all the kernel filesystem stuff and boot (if you have a separate /boot)
root # umount -l proc dev sys boot

Turn off the swap
root # swapoff /dev/zvol/tank/swap

Export the zpool
root # cd /
root # zpool export tank

Reboot
root # reboot

   Important

Don't forget to set your root password as stated above before exiting chroot and rebooting. If you don't set the root password, you won't be able to log into your new system.

and that should be enough to get your system to boot on ZFS.

After reboot

Forgot to reset password?

System Rescue CD

If you aren't using bliss-initramfs, then you can reboot back into your sysresccd and reset through there by mounting your drive, chrooting, and then typing passwd.

Example:

root # zpool import -f -R /mnt/funtoo tank
root # chroot /mnt/funtoo bash -l
root # passwd
root # exit
root # zpool export -f tank
root # reboot

Create initial ZFS Snapshot

Continue to set up anything you need in terms of /etc configurations. Once you have everything the way you like it, take a snapshot of your system. You will be using this snapshot to revert back to this state if anything ever happens to your system down the road. The snapshots are cheap, and almost instant.

To take the snapshot of your system, type the following:

root # zfs snapshot -r tank@install

To see if your snapshot was taken, type:

root # zfs list -t snapshot

If your machine ever fails and you need to get back to this state, just type (This will only revert your / dataset while keeping the rest of your data intact):

root # zfs rollback tank/funtoo/root@install

   Important

For a detailed overview, presentation of ZFS' capabilities, as well as usage examples, please refer to the ZFS Fun page.

Troubleshooting

Starting from scratch

If your installation has gotten screwed up for whatever reason and you need a fresh restart, you can do the following from sysresccd to start fresh:

Destroy the pool and any snapshots and datasets it has
root # zpool destroy -R -f tank

This deletes the files from /dev/sda1 so that even after we zap, recreating the drive in the exact sector
position and size will not give us access to the old files in this partition.
root # mkfs.ext2 /dev/sda1
root # sgdisk -Z /dev/sda

Now start the guide again :).

Starting again reusing the same disk partitions and the same pool

If your installation has gotten screwed up for whatever reason and you want to keep your pole named tank than you should boou into the Rescue CD / USB as done before.

import the pool reusing all existing datasets:
root # zpool import -f -R /mnt/funtoo tank

Now you should wipe the previous installation off:

let's go to our base installation directory:
root # cd /mnt/funtoo

and delete the old installation: 
root # rm -rf *

Now start the guide again, at "Pre-Chroot"