Tag Archives: ubuntu

ZFS Home Directory

I tend to hop from Linux distro to Linux distro. One of the things that makes doing so much easier is keeping my home folder on a separate disk. That way I can re-install distributions to my heart’s content without fear of losing my files and settings.

I’m also a big fan of ZFS (ZFS on Ubuntu Server). That means jumping through a few extra hoops to setup ZFS on a separate drive as well as re-importing the zpool every time I swap distributions, but I find it’s well worth it. Here’s a handy guide on how to do just that! I’ll be showing the steps for Fedora and Kubuntu, but they should generally apply to other distros as well.

Disclaimer: I’ve not a ZFS expert, but these steps have worked very well for me on multiple systems. YMMV.

One quick note: ZFS works best with plenty of RAM (it will use everything available to keep data cached). If you are on a RAM-limited system, you can do something similar with encrypted XFS or EXT4.

Pre-Step: Setup Encrypted Home Drive

I’ll be configuring ZFS to use an encryption key stored on the root drive. This is only secure if the root drive is also encrypted. Make sure when you install Linux you tell the installer to use drive encryption.

It will look like this in Fedora:

Encrypted root drive in Fedora

And like this in Ubuntu:

Encrypted root drive in Kubuntu

You’ll be asked to set a password that’s used to encrypt your root drive. You cannot change this password and you’ll be asked to enter this password every time you boot your computer, so make sure you do not forget it!

Don’t worry about configuring your 2nd drive with your home folder during the installation. I find it’s much easier to have the distribution do it’s typical install, then go back and mount your new /home. Just make sure that you create yourself as an administrator or have a root password set.

Once you’ve installed your new distro, reboot into it, but don’t log in. Your computer will get grumpy if you’re logged into a desktop environment while swapping out your home directory.

Press Control-Alt-F3 to get to a terminal window then log in as yourself if you made your account an administrator, or ‘root’ if you did not.

ZFS Installation

Fedora

Make sure Fedora up-to-date

sudo dnf -y update

If there are any updates, reboot (sudo reboot), press Control-Alt-F3, and log back in.

Install ZFS for Fedora by following the official steps below, do not use the zfs-fuse package included with Fedora: https://openzfs.github.io/openzfs-docs/Getting%20Started/Fedora/index.html

Kubuntu/Ubuntu
sudo apt install zfsutils-linux

Creating a New Home Drive

If you already have created a home drive and are re-attaching after re-installing Linux, skip to Importing an Existing Home Drive.

To make things easier, I’ll be running all of the commands as root by first running:

sudo -s

Create an encryption key that will be used to encrypt and decrypt your home drive. Make sure this is only stored on an encrypted root drive and that you have backed up this key somewhere safe. If you lose this key you will lose all access to your drive. You’ve been warned 😉

dd if=/dev/urandom of=/etc/home.key bs=32 count=1 && chmod 600 /etc/home.key

Next you’ll need to find out the name of your drive. Since easy names (e.g. sda, sdb) can change, we want to set it up by something that will not change. I’ll be using the device’s physical location.

Let’s make sure we know which drive has Linux installed on it, and which is going to be used for our home drive, by runing:

lsblk

This will list all of your drives (also called block devices), any partitions on them, and where those partitions are mounted. My output (on a virtual machine) looks like the following. On real hardware, your devices will probably be called sda and sdb (if they’re SATA), or nvme0n1 and nvme1n1 (if they’re nvme):

NAME                                          MAJ:MIN RM  SIZE RO TYPE  MOUNTPOINTS
sr0                                            11:0    1    2G  0 rom    
zram0                                         251:0    0  5.8G  0 disk  [SWAP]
vda                                           252:0    0   64G  0 disk   
├─vda1                                        252:1    0    1G  0 part  /boot
└─vda2                                        252:2    0   63G  0 part   
 └─luks-a954d91b-fda3-4c22-90a6-2b35554129b1 253:0    0   63G  0 crypt /home
                                                                       /
vdb                                           252:16   0  128G  0 disk  

I can see here that my disk with Linux installed on it is called vda, since it has multiple partitions (vda1 and vda2) that are all mounted (as /boot and /). The disk with nothing installed on it is vdb. Therefore, I’ll need to check the physical location of vdb. Please comment below if you’re having trouble figuring out which drive is which and I’ll try to give you a hand!

To list all disks by their location, run:

ls -lh /dev/disk/by-path/

The result will look something like this:

total 0
lrwxrwxrwx. 1 root root  9 Jan 28 09:42 pci-0000:00:1f.2-ata-1 -> ../../sr0
lrwxrwxrwx. 1 root root  9 Jan 28 09:42 pci-0000:00:1f.2-ata-1.0 -> ../../sr0
lrwxrwxrwx. 1 root root  9 Jan 28 09:42 pci-0000:07:00.0 -> ../../vda
lrwxrwxrwx. 1 root root 10 Jan 28 09:42 pci-0000:07:00.0-part1 -> ../../vda1
lrwxrwxrwx. 1 root root 10 Jan 28 09:42 pci-0000:07:00.0-part2 -> ../../vda2
lrwxrwxrwx. 1 root root  9 Jan 28 09:42 pci-0000:08:00.0 -> ../../vdb

This tells me that the path I’ll be using is /dev/disk/by-path/pci-0000:08:00.0, since that’s the one that’s being called vdb (see the end of the last line).

We’re finally ready to create our ZFS filesystem! First we create a zpool that encompasses all of the drives we’ll be using (we’ll just be using one, but ZFS can be mirrored or RAIDed in more advanced setups).

The command we’ll run is:

zpool create homepool -O xattr=sa -O acltype=posixacl -O atime=off -O compression=lz4 -O encryption=aes-256-gcm -O keyformat=raw -O keylocation=file:///etc/home.key -o ashift=12 /dev/disk/by-path/[your disk here]

Here’s what some of those options mean:
ashift=12 : This specifies the drive’s block size. From what I’ve cobbled together, use the number 12 for most use cases unless it’s a Samsung NVME or you know your drive uses 8K clusters. In that case, use 13.
homepool: this is the name we’ve given to the zpool. You can use something else if you’d prefer.
compression=lz4: This compresses all data, increases the performance of ZFS, and essentially costs no additional CPU resources. More information here: https://www.servethehome.com/the-case-for-using-zfs-compression/
encyption=aes-256-gcm: Use AES 256 GCM encryption which is both highly secure and hardware accelerated

Now, let’s check out that brand new zpool!

zpool status

You should see something like this:

  pool: homepool
 state: ONLINE
config:

        NAME                STATE     READ WRITE CKSUM
        homepool            ONLINE       0     0     0
          pci-0000:08:00.0  ONLINE       0     0     0

errors: No known data errors

A zpool is a container for filesystems. Now that we’ve got one, we can create a filesystem where our home drive will live. In all of the steps below, replace [user] with your username.

zfs create homepool/[user]

To see information on this new filesystem, you can run:

zfs list

Now, let’s replace our old home drive (that was created when Linux was installed) with the filesystem on our second drive:

cd /home
mv /home/[user] /home/[user].bak
mkdir /home/[user]/
zfs set mountpoint=/home/[user] homepool/[user]
zfs set mountpoint=none homepool
chmod --reference=/home/[user].bak /home/[user]
mv /home/[user].bak/* /home/[user]/
mv /home/[user].bak/.* /home/[user]/
rmdir /home/[user].bak
chown -R [user]:[user] /home/[user]
#For Fedora and other distros with selinux, run the next line too:
restorecon -vR /home

Linux doesn’t yet load keys for encrypted zfs mounts automatically. You’ll need to create a simple service to automatically load zfs encryption keys on boot.
Like most good things, this is from the Arch Linux wiki: https://wiki.archlinux.org/title/ZFS#Unlock_at_boot_time:_systemd
You MUST do this before you reboot or you will not be able to log in graphically. If you forget, press Control-Alt-F3 and log into the console.

nano /etc/systemd/system/zfs-load-key.service

Type in the following (if you’re uncomfortable typing by hand, you should be able to switch to the graphical login (Fedora: Control-Alt-F2, *buntu Control-Alt-F1) and copy paste).

[Unit]
Description=Load encryption keys
DefaultDependencies=no
After=zfs-import.target
Before=zfs-mount.service

[Service]
Type=oneshot
RemainAfterExit=yes
ExecStart=/usr/sbin/zfs load-key -a
StandardInput=tty-force

[Install]
WantedBy=zfs-mount.service

Next, tell Linux to start the new service every time it boots:

systemctl enable zfs-load-key

Finally, reboot and log in normally to make sure everything works as anticipated:

reboot

Now you should have a fully functioning install with encrypted ZFS home directory! Remember to backup /etc/home.key somewhere secure that *isn’t* in your home directory, since you’ll need to copy this key back any time you re-install Linux. I’d recommend an encrypted USB key.

If you have multiple users, you can follow those same steps to create a zfs filesystem for each of them in zpool you created.

Steam

If you use Steam and want to keep your game installations separate so they don’t get backed up with zfs snapshots, you can create a separate filesystem for it.

mkdir -p /home/[user]/.local/share/Steam
sudo zfs create homepool/[user]/steam -o mountpoint=/home/[user]/.local/share/Steam

Importing an Existing Home Drive

Only follow these steps if you’ve re-installed Linux. They aren’t necessary if you just created a new zpool above.

After you’ve re-installed Linux, make sure you complete ZFS Installation above. Once those are done, you can proceed from here.

If you are not the root user yet, run:

sudo -s

Next, you’ll need to copy your backed up key to /etc/home.key

If it’s stored on an encrypted flash drive, it may be easiest to log in graphically, restore the file, then log out and return the console with Control-Alt-F3.

Once it’s restored, make sure it still has the correct permissions

chown root:root /etc/home.key && chmod 600 /etc/home.key

Rename your existing home directory:

cd /home
mv /home/[user] /home/[user].bak
mkdir /home/[user]
chmod --reference=/home/[user].bak /home/[user]
chown [user]:[user] /home/[user]

List all zpools the system can find for import:

zpool import

You should see your homepool listed

   pool: homepool
     id: 16378698673868876678
  state: ONLINE
 action: The pool can be imported using its name or numeric identifier.
 config:

        homepool            ONLINE
          pci-0000:08:00.0  ONLINE

You can now import it by name and mount the zfs filesystems:

zpool import homepool

Before the filesystems can be mounted, we’ll need to create and enable the ZFS key loading service.

nano /etc/systemd/system/zfs-load-key.service

Type in the following:

[Unit]
Description=Load encryption keys
DefaultDependencies=no
After=zfs-import.target
Before=zfs-mount.service

[Service]
Type=oneshot
RemainAfterExit=yes
ExecStart=/usr/sbin/zfs load-key -a
StandardInput=tty-force

[Install]
WantedBy=zfs-mount.service

Next, tell Linux to start the new service every time it boots, and start it now:

systemctl enable --now zfs-load-key

Finally, we can mount all of the zfs filesystems:

zfs mount -a

You can confirm they are mounted by typing:

mount

The last line of the output should be something similar to:

homepool/adam on /home/adam type zfs (rw,noatime,seclabel,xattr,posixacl)

If you’re on Fedora, be sure to run the following to make selinux happy:

restorecon -vR /home

Reboot and log in. Since all of your personal settings are saved to your home drive, everything should be exactly how you left it!

Memory Usage

If you find ZFS is using too much memory (apps keep crashing), you can adjust how much RAM ZFS uses for its cache (how much of your drive it keeps in memory for quick access).

To test different settings, set the maximum arch size in bytes and then clear the cache. This setting is temporary, so if you run intro trouble, just reboot.

sudo echo "8589934592" > /sys/module/zfs/parameters/zfs_arc_max
sudo echo 3 > /proc/sys/vm/drop_caches

Once you’ve found a size that works for you, you can set the size permanently.

sudo echo "options zfs zfs_arc_max=8589934592" >> /etc/modprobe.d/zfs.conf

Adventures in Plasma Land

Or, can a man fall in love with KDE, 20 years later

KDE has never been able to capture my heart. I remember trying KDE 1.1 or so on Madrake Linux 6 in the late 90s. It just never clicked for me. I opted instead for Enlightenment. Ever since then, I’ve tried it every year or so to see if I could understand peoples’ love for it. I didn’t fall for KDE 3.5 that so many people remember fondly, or KDE 4, which people recall much less fondly. I’ve peeked in on KDE Plamsa 5 during it’s development, but it never was able to bring me in. But here I am, in 2019, about 20 years since I started using Linux, and I’ve giving KDE Plasma 5.16.4 a go!

Background

So, why now? Well, as I said, I try KDE every now and then. Something about it always draws me in, before turning me off again. I recently ended up down an internet rabbit hole following articles on Plasma mobile, Qt Python bindings and even Qt C# bindings for .Net Core (and I love me some C#). I wondered: “Could Plasma, Plasma Mobile, Qt, and C# be the epic combo of my dreams?” Let’s find out!

Setup

I’m running KDE Plasma 5.16.4 on KDE Neon Linux. Neon is based on Ubuntu 18.04 LTS using KDE’s own, and constantly updated, repos for Plasma itself. I figured as long as I’m giving a fair shake, I ought to go right to the source. (As I’ve written this article, I’ve upgraded across a few versions of 5.x Plasma)

I’m running it on my trusty desktop with a Core i7 920, AMD Radeon RX 560, 12 GB RAM, a 512 GB SSD, and a HiDPI monitor at 3840×2160. It’s my daily driver at home for general computing, gaming, and game development.

I’m also running the X11 version of Plasma, rather than Wayland. I did some testing and Wayland seems to be particularly fluky with AMD graphics, though remarkably stable on the Intel-graphics based laptop I tested on. Your mileage may vary.

KDE System Info

Window Dressing

At work I use a Mac, so having the close, minimize and maximize icons on the left just keeps my flow going. I really thought this was going to be one of those “sorry, can’t do with Plasma” things. Oh, how wrong I was! In face, Plasma lets you customize all the icons on title bar.

Titlebar icon placement is handled by the “Look and Feel of Windows Titles” settings menu.

Bluetooth

I was able to connect a Bluetooth Microsoft Arc Mouse and my Apple AirPods without any difficulty whatsoever. Major bonus in my book!

Scaling

One of things I love about Plasma is the decimal-based resolution scaling. Wheres the GTK-based desktops I’ve used require scaling at 1x, 2x, 3x, etc., Plasma allows you too choose, for example, 1.5x. This is a huge improvement for HiDPI displays.

The caveat is, you’re probably not going to be running all Qt apps. Invariably, you’ll also run some GTK apps as well. These will ignore your scaling. This was the case for me with Unity Editor.

Luckily, there’s an easy fix!

Open kmenueditor

Find the app you need to scale

Prefix the command with: GDK_SCALE=2 GDK_DPI_SCALE=0.5

Multiple Displays

While this isn’t an issue for me on my main computer, I did want to see how Plasma handled multiple monitors in case I’m able to get a 2nd display at home someday. Using a test laptop with Intel-based graphics I had no problem at all running Plasma with two 4K monitors daisy chained with DisplayPort.

Extra Surprises

PSD Previews

One of the things that drives me crazy about Nautilus is that it doesn’t support the preview of PSD files out of the box. Perhaps there’s a plugin or setting somewhere, but not that I could find. I was thrilled to open a folder with a whole slew of PSD files and see previews of them working by default.

Latte Dock

If you’re a lover of docks like I am, I can’t recommend Latte Dock highly enough! Latte Dock is integrated beautifully into the Plasma ecosystem, with all the fun stuff like pinning and app actions. For example, the Spotify app will give you playback options right from the dock icon.

Issues

Media over SMB

Dolphin (KDE’s file manager) does fantastic job of browsing SMB shares. It’s handy to be able to view shares without actually mounting them, but there’s quite a few drawbacks. Most frustrating was getting media to play when double clicking the file. I was eventually able to get it to work with VLC by using the snap version and dragging/dropping the file into the VLC window, but this still required me putting in my username and password for each video. My recommendation: mount the share, and everything works fine. I found smb4k recommended in a forum for this, and it does a fantastic job. Just make sure to exclude it’s default mount point, ~/smb4k, from your backup jobs.

Discover

Discover is Plasma’s app installation and system update tool. It’s gotten much better over the years (and even since I began writing this article), but can still be finicky.

[Update: the awesome Nate Graham fixed what was apparently an issue with how kmenuedit comes up in search results. Thanks so much!
https://invent.kde.org/plasma/kmenuedit/-/merge_requests/5]
For example: if I search for ‘kmenu,’ I get nothing. It’s not until I search for ‘kmenuedit’ that I get a search result. I just seems by now that I should be able to do a partial search and good results.

Wherefore art thou kmenuedit?
Oh, there you are.

I will say this about Discovery: it’s ability to handle both apt and snap versions of packages is very convenient!

KRDC

KRDC is a Qt-based remote desktop app for Plasma. It works great on a regular-resolution displays, but has some strange scaling issues for me on a HiDPI display and AMD graphics. I like to have a bunch of remote desktop sessions open at once, so I’ll typically have the remote desktop display be the current size of the client window. This works great in Remmina (the GTK equivalent of KRDC), but with KRDC I can never get it working quite right. (See below)

My old Windows VM and Plex server, before I migrated it to Ubuntu

Verdict

I’m sold! I started this article about four months ago wondering when I’d switch from Plasma back to Budgie. Now, I can say without a doubt that Plasma will remain my desktop of choice for the foreseeable future. Great job Plasma team!

ZFS on Ubuntu server

Ubuntu Home Server Setup Part II

Welcome to Part II of my Ubuntu Home Server build! In Part I, I did a very basic Ubuntu Server install. In this part, I’ll be creating a ZFS pool and volumes to store all my data on.

Other parts of this guide can be found at:

Home Server With Ubuntu

Setup

I’ll be setting up a server with 8 physical drives.

Disk 0: SSD for OS

Disk 1: SSD for ZFS Intent Log (improves write performance)
(read fantastic information about it here: http://nex7.blogspot.com/2013/04/zfs-intent-log.html)

Disk 2: SSD for L2ARC caching (improves read performance)

Disk 3 – 7: HDDs for ZFS Pool (where all my data with be stored)

Quick disclosure: I’m *far* from a ZFS expert. From what I’ve gleaned, this should suffice for home / small business use. If you’re planning something enterprise-grade, find an expert!

Install Ubuntu

Perform a regular Ubuntu server installation, or use an existing server.

SSH Into the server, rather than using the console. You’ll want to be able to copy and paste when you setup the zpool.

Install ZFS

sudo apt install zfsutils-linux

Create the ZPOOL

I’ll be using RAIDZ (which is like RAID-5) to get redundancy on my disks without losing too much usable space.

ZFS offers many other options, like RAID0, 1, 6, etc. Use whichever is appropriate for your workload.

It is very strongly recommended to not use disk names like sdb, sdc, etc. Those might change across reboots.

Many of the articles I’ve read suggest using UUIDs . However, my experience on Ubuntu Server is that these are not assigned to blank disks. Therefore, I will be using disk paths instead.

These are verbose and a bit of a pain to type, but they make sure you know exactly what disk you are referring to should you need to swap drives in the future. They will also not change on reboots.

To see your installed disks run:

ls -lh /dev/disk/by-path

My output looks like

adam@normandy:~$ ls -lh /dev/disk/by-path
 total 0
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:00:1f.2-ata-5 -> ../../sr0
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:0:0 -> ../../sda
 lrwxrwxrwx 1 root root 10 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:0:0-part1 -> ../../sda1
 lrwxrwxrwx 1 root root 10 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:0:0-part2 -> ../../sda2
 lrwxrwxrwx 1 root root 10 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:0:0-part3 -> ../../sda3
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:1:0 -> ../../sdb
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:2:0 -> ../../sdc
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:3:0 -> ../../sdd
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:4:0 -> ../../sde
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:5:0 -> ../../sdf
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:6:0 -> ../../sdg
 lrwxrwxrwx 1 root root  9 Jul  8 09:06 pci-0000:02:00.0-scsi-0:0:7:0 -> ../../sdh

I chose to install Linux on my 1st drive (sda). I’ll be using sdb for the ZIL, sdc for L2ARC, and sdd, sde, sdf, sdg, and sdh to for the data pool.

First, I’ll setup the data pool. This is where SSH is handy, since you can copy/paste your paths from above.

In my example below, I’m naming my pool “data.” You can use a different name if you’d like. If your setup is like mine, you’ll create one pool with many volumes in it.

I’m using drives with 4k physical sectors, so I’m adding the option: -o ashift=12
This should increase performance, but at the cost of total storage space. You an remove this option if you don’t think it’s a good fit for you.

sudo zpool create data -o ashift=12 raidz /dev/disk/by-path/pci-0000:02:00.0-scsi-0:0:3:0 /dev/disk/by-path/pci-0000:02:00.0-scsi-0:0:4:0 /dev/disk/by-path/pci-0000:02:00.0-scsi-0:0:5:0 /dev/disk/by-path/pci-0000:02:00.0-scsi-0:0:6:0 /dev/disk/by-path/pci-0000:02:00.0-scsi-0:0:7:0

To confirm this worked, run:

zpool list

You should have something like:

adam@normandy:~$ zpool list
NAME   SIZE  ALLOC   FREE  EXPANDSZ   FRAG    CAP  DEDUP  HEALTH  ALTROOT
data  18.1T   238K  18.1T         -     0%     0%  1.00x  ONLINE  -

Next I’ll tell ZFS to use sdb as the ZFS Intent Log

sudo zpool add data log /dev/disk/by-path/pci-0000:02:00.0-scsi-0:0:1:0

Then I’ll tell ZFS to use sdc as the L2ARCH cache

sudo zpool add data cache /dev/disk/by-path/pci-0000:02:00.0-scsi-0:0:2:0

If I run zpool status, I should see my data, ZIL, and cache drives

adam@normandy:/data/download/secure$ zpool status
  pool: data
 state: ONLINE
  scan: none requested
config:

        NAME                               STATE     READ WRITE CKSUM
        data                               ONLINE       0     0     0
          raidz1-0                         ONLINE       0     0     0
            pci-0000:02:00.0-scsi-0:0:3:0  ONLINE       0     0     0
            pci-0000:02:00.0-scsi-0:0:4:0  ONLINE       0     0     0
            pci-0000:02:00.0-scsi-0:0:5:0  ONLINE       0     0     0
            pci-0000:02:00.0-scsi-0:0:6:0  ONLINE       0     0     0
            pci-0000:02:00.0-scsi-0:0:7:0  ONLINE       0     0     0
        logs
          pci-0000:02:00.0-scsi-0:0:1:0    ONLINE       0     0     0
        cache
          pci-0000:02:00.0-scsi-0:0:2:0    ONLINE       0     0     0

errors: No known data errors

Create the Filesystem

Now that the zpool exists, we can create filesystems on top of it.
A pool can have multiple filesystems. I’ll create one for media, and one for virtual machines (because that’s what I need).

sudo zfs create data/media
sudo zfs create data/vm

To confirm it was created correctly run:

zfs list

And it should look something like this:

adam@normandy:~$ zfs list
 NAME         USED  AVAIL  REFER  MOUNTPOINT
 data         210K  14.0T  36.7K  /data
 data/media  35.1K  14.0T  35.1K  /data/media
 data/vm     35.1K  14.0T  35.1K  /data/vm

I would also suggest the following tweaks. Combined, they increased my zfs throughput 50-100%! They were recommended by https://unicolet.blogspot.com/2013/03/a-not-so-short-guide-to-zfs-on-linux.html and https://www.servethehome.com/the-case-for-using-zfs-compression/ as I searched for solutions to my less-than-stellar zfs performance.

zfs set xattr=sa data/media
zfs set atime=off data/media
zfs set compression=lz4

All of your zfs filesystems are automatically mounted.

adam@normandy:~$ mount
...
data on /data type zfs (rw,xattr,noacl)
data/media on /data/media type zfs (rw,xattr,noacl)
data/vm on /data/vm type zfs (rw,xattr,noacl

You can use them just as you would any mounted filesystem. That’s it!

Basic Ubuntu Server Installation

This is Part I (the boring part) of my Ubuntu Home Server install.

Other parts can be found at:

Home Server With Ubuntu

For anyone who’s installed Ubuntu Server before, there’s not much here for you. I’m putting this here for anyone starting out with Ubuntu and for the sake of completeness.

Also, my 1st warning: this is the setup I think will serve me best for my particular situation. It may not be the best for you, and, while it’s somewhat redundant, it certainly isn’t “enterprise-grade.” You were warned ?

In the steps below, anytime you see something in brackets, replace it with the correct value for your system, without the brackets.
For example, if you see:
ssh [username]@[ip address]
You should really enter something like:
ssh me@192.168.1.1

Why Ubuntu?

All the major Linux distros are awesome. You really can’t go wrong! For servers, I’ve typically gone with Centos in the past (and on this Ubuntu server will be many Centos virtual machines). However, there is one reason I’ve decided to go with Ubuntu in this instance: ZFS. Ubuntu has ZFS baked in, whereas Centos and Fedora require recompilation of kernel modules after major OS upgrades. Since I want this box to be as turnkey as possible (if it goes down, my internet will go down as well), Ubuntu it is!

Installation

First, download the Ubuntu Server iso from Ubuntu. I’ll be using the 18.04 LTS release, since I prefer to stick to LTS releases for critical infrastructure.

https://ubuntu.com/download/server

Next, either burn the iso to a DVD or image it to a flash drive. If you use the flash drive method, I recommend Fedora Media Writer. It’s available for Windows, MacOS, and Linux, and will image pretty much any Linux distro to USB.

https://getfedora.org/en/workstation/download/

Once you’ve got a bootable DVD or flash drive, boot from it. Most servers and workstations will tell you which key to press on the keyboard to get to your BIOS/UEFI boot menu.

After booting, choose *Install Ubuntu Server.

Choose your language.

Choose your keyboard layout.

Choose Install Ubuntu.

I’m going to use DHCP for now and set static IP later when I configure the virtualization networks for KVM. If you need to configure a static IP, you can do so here.

If you use an internet proxy, set it here.

Choose the default Ubuntu mirror.

I prefer to use LVM in case I need to resize partitions in the future.

I’ll be using one SSD as a boot volume. Choose whichever drive you’ll be booting from. I’ll be using all of the rest of the drives for ZFS, so I’ll leave them as they are for now.

By default, Ubuntu will only use 4GB of your drive for the root partition. Since all of my other data will live my ZFS volumes, I’ll expand the volume to use the whole 1TB.

To change the size of the root volume, use the down arrow to chose “ubuntu-lv,” press Enter, then choose “Edit.”

Ubuntu will helpfully tell you the max size you can set the partition to. Enter that number and choose “Save.”

Choose “Done.”

Let Ubuntu know your name, your computer’s name, the username you’d like to use, and the password you’d like to use.

You now have the option of installing a secure shell server. This will allow you to log in remotely. I’ll be installing this.

You also have the option of installing some other services. You can always install these later. I’ll be skipping them and just choosing “Done.”

When the installation has finished, choose “Reboot Now.”

Remove the bootable DVD or flash drive and press Enter.

Log In

Once the server has rebooted, you can log in to the server itself or via SSH (if you installed SSH).

If you need to find out your server’s IP address for SSH, log in via the console and run the following:

ip address

Then on the computer you are using to SSH into the server run:

ssh [username]@[ip address]

Updates

Before anything else, let’s make sure everything is up-to-date.

sudo apt upgrade

Once that has completed, you may need to reboot.

sudo reboot

KDE On a Server?

Let’s get right to it: it’s not considered security-wise to install a GUI on a server. However, I’ll be using things like Handbrake and Virtual Machine Manager, so I’ll be putting on KDE. To add a bit of security and save memory, I’ll manually start KDE when I need it.

To install just the very minimum of KDE (you can always add the other bits later), run:

sudo apt install kubuntu-desktop --no-install-recommends

I’m also going to install a couple other KDE apps to make my life easier. KDE’s Konsole terminal and the dolphin file manager:

sudo apt install konsole dolphin

If you want all of KDE, and have it start be default, you can simply run this instead:

sudo apt install kubuntu-desktop

If GNOME is more your thing, you can install it with:

sudo apt install ubuntu-gnome-desktop

If you install just the minimum KDE, your server will still boot in console mode. To start KDE, simply log in and run:

startx

Since I’ll often want to use the UI remotely, I’m also going to install a package called xrdp. This will serve a desktop over the RDP protocol so I can get a desktop remotely:

sudo apt install xrdp

This will install xrdp, configure the service to start automatically, and start the service. Once it’s finished, you can connect to your server’s IP address via any remote desktop app and use the same username and password you use to log in locally.

Home Server With Ubuntu

I finally picked up a used Dell PowerEdge R720 from the fine folks at ServerMoney to replace my current home server (a Frankenstein of workstation parts).

I thought I’d document my setup for anyone that might be interested and for my future self that wondered what exactly I did in the 1st place ?

My server needs are quite diverse, so I’ll break this guide into separate posts for each one to keep things organized. (links will be active once each part is finished)

Happy serving!

Part I: Basic Ubuntu Server Install (SSH, KDE, & xrdp)
Part II: Ubuntu ZFS Setup
Part III: Ubuntu Virtualization Server with KVM
Part IV: pfSense on KVM
Part V: Plex on Ubuntu
Part VI: SMB & NFS