Environment setup

Notes taken during the setup of the system

The System¶

BIOS/UEFI¶

To begin, once the hardware thrown in the case, we have to tune the BIOS/UEFI to enable some features in the advanced pannel :

CPU Configuration >
- CPU C States Support => Enabled
- Enhanced Halt State (C1E) => Auto
- Package C State Support => Enable
- C6DRAM => Enable
Chipset Configuration >
- PCI Express Native Control => Enabled
- PCH PCIE ASPM Support => Auto
- Restore on AC/Power Loss => Power On

OS¶

For ease of use and support I will use Ubuntu Server 24.04.2 LTS even if alternative OS like nix are tempting.
To get the smallest footprint as possible, I installed the minimal configuration of Ubuntu server without any additional package except OpenSSH for remote access right after the installation.

Power savings¶

Once the BIOS configured and the OS installed, let's take a look at the power consumption by installing powertop : sudo apt install powertop .Here we can see that the CPU can't go deeper that C2.

The main suspect is the ethernet controller, to set ASPM L1 for this controller :

echo 1 | sudo tee /sys/bus/pci/drivers/r8169/0000\:01\:00.0/link/l1_aspm

Then we can set the powersave mode for the PCI-e :

echo "powersave" | sudo tee /sys/module/pcie_aspm/parameters/policy
cat /sys/module/pcie_aspm/parameters/policy # to confirm

We can also use powertop --auto-tune to tweak other things.

And to persist all those commands, let's add them to the reboot cron :

# To persist those commands
(sudo crontab -l 2>/dev/null; echo "@reboot /usr/sbin/powertop --auto-tune") | sudo crontab -
(sudo crontab -l 2>/dev/null; echo "@reboot echo 1 | sudo tee /sys/bus/pci/drivers/r8169/0000\:01\:00.0/link/l1_aspm") | sudo crontab -
(sudo crontab -l 2>/dev/null; echo "@reboot echo "powersave" | sudo tee /sys/module/pcie_aspm/parameters/policy") | sudo crontab -

Power consumption

At this point the server sips just bellow 12 Watts at idle with HX750i, the N100DC-ITX motherboard, 32 Go of RAM and a 256 Go SATA SSD. Adding 2 12x12 fans add 1.5 Watts for a total bellow 13.5 W.

UPDATE 07/01/2026 : Disks power savings¶

During the previous days I worked on hard disks for backups and RAID. Thoses disks increase significantly the overall power consumption.

My disks are WD Red Plus, they have been made for servers and therefore better tolerate spinup/spindown.

I have 5 disks : 4 WD Red Plus for the RAID5 (sdb, sdd, sde and sdf) and 1 Seagate for backups (sdc).

Let's activate the spindown for those 4 disks by adding the following lines to /etc/hdparm.conf :

/etc/hdparm.conf

command_line {
  hdparm -S 240 /dev/sdb # 240 * 5s = 20 minutes
  hdparm -S 240 /dev/sdd
  hdparm -S 240 /dev/sde
  hdparm -S 240 /dev/sdf
}

To apply, just reboot.

We can check that this does exactly what we want :

louison@homelab:~/homelab$ date
Wed Jan  7 19:52:27 UTC 2026
louison@homelab:~/homelab$ sudo hdparm -C /dev/sdb

/dev/sdb:
 drive state is:  active/idle

# Now lets wait 20 minutes (idle time defined above)

louison@homelab:~/homelab$ sudo hdparm -C /dev/sdb
[sudo] password for louison: 

/dev/sdb:
 drive state is:  standby

 # The disk fall into standby mod (spinoff)

louison@homelab:~/homelab$ ls /media/raid/jellyfin
media

louison@homelab:~/homelab$ sudo hdparm -C /dev/sdb

/dev/sdb:
 drive state is:  active/idle

# Accessing our drive (RAID in this case) will spinup the disk

Power savings

Before this improvment, the homelab was drawing between 31 and 33 Watts.
When the drive are set on standby mod it goes down to 22 Watts.

We still have to work on the backup disk !

UPDATE 11/01/2025 : Disabling disk spinoff¶

HDD spindown could cause damage

During the pasts days it appears to me that the disks spinoff is quite not the neat idea I thought for the following reasons :

The RAID5 being "unreachable" could cause Jellyfin to crash
The disks are being spinup by some process too frequently to be usefull
The disks being spindown and up multiple times a day could be harmful to the disks

I have undone the previus update.

Networking¶

Let's change the SSH default port :

Open ssh config sudo nano /etc/ssh/sshd_config
Edit #Port 22 to something else
Adding new port to firewall sudo ufw allow xxxx/tcp
Apply change by restarting SSH sudo systemctl restart ssh.service
Test the connection ssh <user>@<server_ip> -p xxxx
Removing port 22 to firewall sudo ufw deny 22/tcp

Let's disable password connection for SSH

Open ssh config sudo nano /etc/ssh/sshd_config
Setting PasswordAuthentication no
Setting PermitRootLogin prohibit-password
Apply change by restarting SSH sudo systemctl restart ssh.service

Tips

sudo ss -tulpn to list all the open ports
Don't forget to fix the IP address of the homelab in the router

Miscellaneous¶

Don't forget to set the date properly

sudo timedatectl set-timezone Europe/Paris

The tools¶

CLI utilities¶

sudo apt-get update
sudo apt install bash-completion
sudo apt install git
sudo apt install nano
sudo apt install cron
sudo apt install ufw
sudo apt install pipx && python3 -m pipx ensurepath && pipx install mkdocs && pipx inject mkdocs mkdocs-material

Docker¶

The first tool to install and Setup is Docker.

Docker installation¶

Following the official documentation :

Setup the apt repository :

# Add Docker's official GPG key:
sudo apt-get update
sudo apt-get install ca-certificates curl
sudo install -m 0755 -d /etc/apt/keyrings
sudo curl -fsSL https://download.docker.com/linux/ubuntu/gpg -o /etc/apt/keyrings/docker.asc
sudo chmod a+r /etc/apt/keyrings/docker.asc

# Add the repository to Apt sources:
echo \
  "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/ubuntu \
  $(. /etc/os-release && echo "${UBUNTU_CODENAME:-$VERSION_CODENAME}") stable" | \
  sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
sudo apt-get update

At this step you should see an url from docker from the last command

Install Docker package :

sudo apt-get install docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin

Allowing non-root users to run docker commands (from documentation) :

sudo groupadd docker
sudo usermod -aG docker $USER
newgrp docker

It's pretty usefull to get auto-completion on Docker CLI following the official documentation :

Adding the right code snippet in the bashrc :

cat <<EOT >> ~/.bashrc
if [ -f /etc/bash_completion ]; then
    . /etc/bash_completion
fi
EOT

Reloading the shell : source ~/.bashrc
Creating folder to store bash completion scripts mkdir -p ~/.local/share/bash-completion/completions
Generating the bash completion script for Docker docker completion bash > ~/.local/share/bash-completion/completions/docker

Click here if it worked

Dancing cat

Docker swarm init¶

Let's initialize our Docker Swarm cluster :

docker swarm init

Once the swarm cluster is setup we can start deploying things with ``