Public Documentation

• Welcome to Muchtall's Bookstack
• Ubuntu Template Image Preparation and Cloning
• Installing Docker on Ubuntu
• Debian Template Image Preparation and Cloning
• A Comparison of Common ESPHome-Compatible MMWave Prescence Sensors

Welcome to Muchtall's Bookstack

Nothing much here to begin with, but eventually you will find pages for public consumption in the Public Documentation area.

Ubuntu Template Image Preparation and Cloning

About

Just some documentation I use to keep track of how to provision a new, fresh Ubuntu VM template on my homelab. Some values are going to differ in your environment, such as:

• Name
• Username
• Interface names (could vary asenX0, ens160, etc depending on your hypervisor or lack thereof)
• IP Addresses, Subnets, DNS servers, and search domains

Fresh Install of VM

Fresh Ubuntu Installation

• Select and hit enter on the Try or Install Ubuntu Server option (loads by default)
• Hit Enter, leaving the the default English option
• Hit Enter, leaving the defaults of English (US) and English (US)
• Hit Enter, leaving the default of Ubuntu Server
• Arrow up on the keyboard to enX0 eth and hit Enter
• Arrow down on the keyboard to Edit IPv4 and hit Enter
• Hit Enter on Automatic (DHCP) and then arrow down on the keyboard to select Manual, then hit Enter
• Tab between fields:
  • Subnet: 192.168.160.0/23
  • Address: 192.168.160.xxx (Choose a free temporary IP from IP Address Allocations)
  • Gateway: 192.168.160.1
  • Name Servers: 192.168.160.105, 8.8.8.8, 8.8.4.4
    
  • Search domains: (blank)
  • Save (hit Enter)
• Tab to Done and hit Enter
• Hit Enter, leaving the proxy set to blank
• Hit Enter to accept the default mirror
• Tab to Done and hit Enter to accept the defaults for the storage layout
• Arrow up to ubuntu-lv (aka "mounted at /") and hit Enter
• Select Edit and hit Enter
• Tab to Size and change the value to the same as is listed next to max (probably ~?.???G)
• Tab down to Save and hit Enter
• Tab down to Done and hit Enter
• Tab to Continue and hit Enter on the Confirm destructive action dialog
• Profile Setup
  • Your Name: John Doe
  • Your server's name: ubuntu2404 (or ubuntu24041 etc for newer revisions)
    
  • Pick a username: jdoe
  • Choose a password: (my standard)
  • Confirm your password: (same)
  • Tab to Done and hit Enter
• Hit Enter, leaving the Skip for now option selected on the Upgrade to Ubuntu Pro
• Hit Spacebar on the option to Install OpenSSH Server
• If you have a GitHub account, you can optionally add your public SSH key
• Tab to Done and hit Enter
• Tab to Done and hit Enter, leaving defaults of nothing selected on Featured Server Snaps
• Wait for the "Cancel update and reboot" option to go away, which will be replaced with Reboot now

Post-install Ubuntu Configuration

• Return to the Remote Console window and Power On the VM
• Once it comes up, SSH to your VM

• Run any available upgrades and reboot
  

  sudo apt update && sudo apt upgrade -y && sudo reboot
• Reconnect via SSH

• Set up Chrony for time sync
  

  sudo apt install -y chrony && \
  sudo systemctl restart chrony && \
  chronyc tracking | grep --color=auto -e ^ -e "Last offset.*"

VIM Tweaks

While cool, some of the newer VIM plugins that come with Ubuntu can cause unexpected weirdness. Lets tweak the room VIM config so that it's less of a pain when doing a sudo vi

Apparently `sudoedit` is a thing. That loads your personal .vimrc while executing edits as root. In which case, if you use sudoedit, you may want to add these tweaks to your personal ~/.vimrc

• Fix the auto-indentation of YAML when commenting out an existing line (for both your username and root)
  

  cat << EOF >> ~/.vimrc
  " Fix auto-indentation for YAML files
  augroup yaml_fix
      autocmd!
      autocmd FileType yaml setlocal ts=2 sts=2 sw=2 expandtab indentkeys-=0# indentkeys-=<:>
  augroup END
    
  EOF
   
  sudo bash -c 'cat << EOF >> ~/.vimrc
  " Fix auto-indentation for YAML files
  augroup yaml_fix
      autocmd!
      autocmd FileType yaml setlocal ts=2 sts=2 sw=2 expandtab indentkeys-=0# indentkeys-=<:>
  augroup END
   
  EOF'

• Set VIM as the default editor
  

  sudo update-alternatives --config editor

Template Image Prep

Set up OpenSSH Key Reconfiguration

If you simply clone an Ubuntu image without resetting the OpenSSH server host keys, an attacker can take those host keys and perform a MITM SSH attack on any system that was cloned from the same image. So we have to make sure those are reset before we make the image, and then automatically regenerated on the next boot.

• Copy/Paste/Run this entire chunk of script into your terminal (creates process that checks for missing keys at boot, and regenerates them if missing):
  

  if [ `systemctl is-enabled openssh-reconfigure.service 2> /dev/null > /dev/null || true && false` ] ; then \
    echo "OpenSSH Key Reconfiguration Service already installed." ; \
  else
    sudo bash -c 'cat << EOF > /usr/local/sbin/openssh-reconfigure
  #!/bin/bash
  test -f /etc/ssh/ssh_host_dsa_key || dpkg-reconfigure openssh-server
  EOF'
  sudo chmod 700 /usr/local/sbin/openssh-reconfigure
  sudo bash -c 'cat << EOF > /etc/systemd/system/openssh-reconfigure.service
  [Unit]
  Description=OpenSSH Key Reconfiguration Service
  Before=ssh.service
   
  [Service]
  Type=simple
  ExecStart=/usr/local/sbin/openssh-reconfigure
   
  [Install]
  WantedBy=multi-user.target
  EOF' ; \
    sudo chmod 644 /etc/systemd/system/openssh-reconfigure.service ; \
    sudo systemctl enable openssh-reconfigure.service ; \
  fi

• Delete the existing keys
  

  sudo /bin/rm -v /etc/ssh/ssh_host_*

Clear the Machine ID

• Run this:
  

  sudo bash -c "truncate -s0 /etc/machine-id ; \
  rm /var/lib/dbus/machine-id ; \
  ln -s /etc/machine-id /var/lib/dbus/machine-id"

Genericize the netplan config

• Make these alterations to /etc/netplan/50-cloud-init.yaml in the enX0 section. Leave the comments for the image/template user to understand what needs to happen to re-activate networking
  Ubuntu 24.04:
  

  link-local: [ ipv4 ]   # Post-cloning, comment this out or remove
  #link-local: [ ]       # Post-cloning, un-comment this line
  #addresses:            # Post-cloning, un-comment this line
  #- 192.168.160.xxx/23  # Post-cloning, un-comment this line and set appropriately

  Setting link-local allows the network interface to come up on boot, but without DHCP or a Static IP assigned. Additionally, setting link-local to a blank array ([ ]) after initial config disabled link-local addressing which can cause problems for default route handling in some cases.

Clear the Bash, VIM, and other history

• Run this:
  

  rm -rf ~/.viminfo ~/.Xauthority ~/.cache
  sudo bash -c 'rm -rf ~/.viminfo ~/.Xauthority ~/.cache'
  sudo bash -c 'echo -n "" > /var/log/wtmp'
  sudo bash -c 'echo -n "" > /var/log/btmp'
  sudo bash -c 'echo -n "" > /var/log/lastlog'
   
  ### These should always run last
  sudo bash -c 'truncate -s0 ~/.bash_history ; history -c'
  truncate -s0 ~/.bash_history ; history -c

Shutdown

• Run this:
  

  sudo shutdown -h now

Create the Template or Image

XCP-NG

Using XOA Xen Orchestra

• Go to Home → VMs
• Change the Filters to blank
• Click on your new VM
• Click on the name of the VM
• Change the name to: TPL_U24.04_20240709172110
  • TPL means this is a user-generated template
  • U24.04 indicates this is Ubuntu Server 24.04
  • 20240709172110 Indicates the revision date of THE DOCUMENT YOU ARE READING in UTC, ie 2024 July 9th, 17:21:10. This is used as a means of revision control.
• Click on Advanced
• Click Convert to template
• Click OK on the confirmation dialog

Cloning to a New VM

XCP-NG

Using XOA Xen Orchestra

• Go to Home → VMs
• Click New VM
• Select your pool
• Set...
  • Template: (Type TPL_U to start auto-filling the options, and select your template)
  • Name: (Name of your choosing)
  • Description: (Something descriptive like: Docker server on Ubuntu 24.04)
  • VCPU: 16 (or whatever the number of cores is on the system)
  • RAM: 16 GiB (or whatever you need)
  • Topology: (Default behavior)
  • Click Show advanced settings
  • Uncheck Boot VM after creation
  •  If you wish to have this system auto-power-on when the host system/hypervisor reboots...
    
    • Check Auto power on
• Click Create
• Click Disks
• Click the disk size in the Size column
• Increase the disk size to whatever your needs are
• You may optionally migrate the VDI to a faster SR or bulk data SR if desired by using the arrow button in the last column
• When you are finished, click the start button () at the top of the page
• Click the Console to watch the startup and proceed with initial configuration

Post-Cloning Configuration

• Log into the VM via the console
• Edit /etc/netplan/50-cloud-init.yaml and configure as appropriate to your use (reserve an IP from your firewall's DHCP pool).
  • Follow the instructions in the comments we made in the template

• Apply the new netplan config
  

  sudo netplan apply

• SSH to the VM via the assigned IP using the account you created during initial installation and templating

• Configure the hostname
  

  sudo hostnamectl set-hostname myhostname

• If you resized the disk, resize the filesystem
  

  # Check the current partition configuration
  lsblk /dev/xvda | grep -e ^ -e ^xvda.* -e ".*─xvda3.*"
  
  # Resize xvda partition 3 to consume the available space
  sudo growpart /dev/xvda 3
  
  # Re-check the current partition configuration
  lsblk /dev/xvda | grep -e ^ -e ^xvda.* -e ".*─xvda3.*"
  
  # Check the current Free PE on the partition
  sudo pvdisplay | grep -e ^ -e "Free PE.*"
  
  # If you do not see any free PE's, run
  sudo pvresize /dev/xvda3
  
  # Then re-check the current Free PE on the partition
  sudo pvdisplay | grep -e ^ -e "Free PE.*"
  
  # Check the Free PE on the volume group
  sudo vgdisplay | grep -e ^ -e "Free  PE.*"
  
  # Check the LV Size on the logical volume
  sudo lvdisplay | grep -e ^ -e "LV Size.*"
  
  # Resize the LV, increasing by the number of PE's available above
  sudo lvextend -l +7680 /dev/ubuntu-vg/ubuntu-lv
  
  # Re-check the LV Size on the logical volume
  sudo lvdisplay | grep -e ^ -e "LV Size.*"
  
  # Check the Free PE on the volume group, confirming we used all available PE's
  sudo vgdisplay | grep -e ^ -e "Free  PE.*"
  
  # Check the filesystem Size
  df -h /
  
  # Resize the filesystem
  sudo resize2fs /dev/ubuntu-vg/ubuntu-lv
  
  # Re-check the filesystem to confirm larger Size
  df -h /

• Optionally Install Docker

Installing Docker on Ubuntu

Install

• Run this:
  

  ### Everything below can be copy/pasted as a one-liner
  . /etc/os-release && \
  sudo apt update && \
  sudo apt install -y apt-transport-https ca-certificates curl software-properties-common && \
  curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add - && \
  sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $VERSION_CODENAME stable" && \
  sudo apt update && \
  sudo apt install -y docker-ce docker-compose && \
   
  # Verify it's running
  sudo systemctl status docker -n 0
   
  # Update (create if need be) the docker configuration to use the CGNAT CIDR range for container networking
  sudo apt install -y moreutils jq && \
  sudo bash -c 'if [ ! -f "/etc/docker/daemon.json" ] ; then echo "{}" > /etc/docker/daemon.json ; fi ; jq '\''."default-address-pools"[0].base = "100.64.0.0/16" | ."default-address-pools"[0].size = 24'\'' /etc/docker/daemon.json | sponge /etc/docker/daemon.json' ; cat /etc/docker/daemon.json
   
  # Restart docker to pull in changes
  sudo systemctl restart docker
   
  # Set up cron task to clean up stale, unused docker images
  sudo bash -c 'cat << EOF > /etc/cron.daily/docker-image-prune
  #!/bin/bash
  ### Prune images older than 7 days
  docker image prune -af  --filter "until=$((7*24))h" 2> /dev/null > /dev/null
  EOF
  '
  sudo chmod u+x /etc/cron.daily/docker-image-prune

Per-admin first-time setup

For each local user you want to grant docker access (such as your own user)...

• Run this:
  

  # Add yourself to the Docker group
  sudo usermod -aG docker ${USER}
   
  # Disconnect and reconnect to apply permissions OR start a new privileged shell...
  sudo su - ${USER}

  The above only has to be run once. After that, you have persistent access every time you log in.

Optional Install of Portainer

Gives users an approachable GUI to manage containers

• Install via docker-compose
  

  sudo mkdir -p /opt/docker-compose/portainer
  cd /opt/docker-compose/portainer
  sudo vi docker-compose.yaml

• Paste in the following and save/exit:
  

  services:
    portainer:
      image: portainer/portainer-ce
      restart: always
      ### Generated with:
      #    docker run --rm httpd:2.4-alpine htpasswd -nbB admin 'YOUR_PASSWORD_HERE' | cut -d ":" -f 2 | sed -Ee 's/\$/\$\$/g'
      command: --admin-password '$$2y$$05$$_CHANGE_THIS_HASH_USING_THE_OUTPUT_FROM_THE_COMMAND_ABOVE'
      ports:
        - "9000:9000"
        - "9443:9443"
        - "8000:8000"
      volumes:
        - portainer_data:/data
        - /var/run/docker.sock:/var/run/docker.sock
   
  volumes:
    portainer_data:

• Set the admin password
  

  ADMIN_PW_HASH=$(docker run --rm httpd:2.4-alpine htpasswd -nbB admin "`read -p "Enter your new admin password: " -s i ; echo $i`" | cut -d ":" -f 2 | sed -Ee 's/\$/\$\$/g') ; echo "" ; \
  sudo sed -Ee 's/--admin-password '\''[^'\'']+'\''/--admin-password '\'''${ADMIN_PW_HASH}''\''/g' -i docker-compose.yaml

• Run:
  

  docker compose up -d && docker compose logs -f

  You can stop the log output with Ctrl+C

• Open the UI via https://your.hostname.or.ip:9443/
• Bypass the invalid cert warning
• Login with the admin password you set above

Debian Template Image Preparation and Cloning

ROUGH DRAFT

Fresh Install of VM

Fresh Debian Installation

To be documented

Post-install Ubuntu Configuration

• Go to the Console tab and Power On the VM
• Once it comes up, login to your VM
• Get the VM IP:
  

  ip addr

  
  
• SSH to the VM and login as your unprivileged user
• su to root (enter the root password)
  

  su -

• Edit sources.list
  

  deb http://deb.debian.org/debian bookworm main non-free-firmware
  deb-src http://deb.debian.org/debian bookworm main non-free-firmware
  
  deb http://security.debian.org/debian-security bookworm-security main non-free-firmware
  deb-src http://security.debian.org/debian-security bookworm-security main non-free-firmware
  
  deb http://deb.debian.org/debian bookworm-updates main non-free-firmware
  deb-src http://deb.debian.org/debian bookworm-updates main non-free-firmware

• Install sudo and add your user to sudo group
  

  apt-get install sudo
  usermod -aG sudo johndoe

• Logout and reconnect via SSH

• Run any available upgrades and reboot
  

  sudo apt update && sudo apt upgrade -y && sudo reboot

• Set up Chrony for time sync
  

  sudo apt install -y chrony && \
  sudo systemctl restart chrony && \
  chronyc tracking | grep --color=auto -e ^ -e "Last offset.*"
• Add /usr/sbin to default PATH

  sudo bash -c 'echo '\''PATH="/usr/sbin:$PATH"'\'' >> /etc/profile.d/usr_sbin_path.sh'

VIM Tweaks

• Install VIM basic
  

  sudo apt install vim

• Set VIM as the default editor
  

  sudo update-alternatives --config editor

Template Image Prep

Set up OpenSSH Key Reconfiguration

If you simply clone a Debian image without resetting the OpenSSH server host keys, an attacker can take those host keys and perform a MITM SSH attack on any system that was cloned from the same image. So we have to make sure those are reset before we make the image, and then automatically regenerated on the next boot.

• Copy/Paste/Run this entire chunk of script into your terminal (creates process that checks for missing keys at boot, and regenerates them if missing):
  

  if [ `systemctl is-enabled openssh-reconfigure.service 2> /dev/null > /dev/null || true && false` ] ; then \
    echo "OpenSSH Key Reconfiguration Service already installed." ; \
  else
    sudo bash -c 'cat << EOF > /usr/local/sbin/openssh-reconfigure
  #!/bin/bash
  test -f /etc/ssh/ssh_host_dsa_key || dpkg-reconfigure openssh-server
  EOF'
  sudo chmod 700 /usr/local/sbin/openssh-reconfigure
  sudo bash -c 'cat << EOF > /etc/systemd/system/openssh-reconfigure.service
  [Unit]
  Description=OpenSSH Key Reconfiguration Service
  Before=ssh.service
   
  [Service]
  Type=simple
  ExecStart=/usr/local/sbin/openssh-reconfigure
   
  [Install]
  WantedBy=multi-user.target
  EOF' ; \
    sudo chmod 644 /etc/systemd/system/openssh-reconfigure.service ; \
    sudo systemctl enable openssh-reconfigure.service ; \
  fi

• Delete the existing keys
  

  sudo /bin/rm -v /etc/ssh/ssh_host_*

Clear the Machine ID

• Run this:
  

  sudo bash -c "truncate -s0 /etc/machine-id ; \
  rm /var/lib/dbus/machine-id ; \
  ln -s /etc/machine-id /var/lib/dbus/machine-id"

Genericize the interface config

• Make these alterations to /etc/network/interfaces in the enX0 section. Leave the comments for the image/template user to understand what needs to happen to re-activate networking
  

  iface enX0 inet dhcp
  #iface enX0 inet static
  #       address 192.168.160.XXX
  #       netmask 255.255.254.0
  #       gateway 192.168.160.1
  #       dns-nameservers 192.168.160.105
  

Clear the Bash, VIM, and other history

• Run this:
  

  rm -rf ~/.viminfo ~/.Xauthority ~/.cache
  sudo bash -c 'rm -rf ~/.viminfo ~/.Xauthority ~/.cache'
  sudo bash -c 'echo -n "" > /var/log/wtmp'
  sudo bash -c 'echo -n "" > /var/log/btmp'
  sudo bash -c 'echo -n "" > /var/log/lastlog'
   
  ### These should always run last
  sudo bash -c 'truncate -s0 ~/.bash_history ; history -c'
  truncate -s0 ~/.bash_history ; history -c

Shutdown

• Run this:
  

  sudo shutdown -h now

Create the Template or Image

XCP-NG

Using XOA Xen Orchestra

• Go to Home → VMs
• Change the Filters to blank
• Click on your new VM
• Click on the name of the VM
• Change the name to: TPL_D12.7.0_20240709172110
  • TPL means this is a user-generated template
  • D12.7.0 indicates this is Debian 12.7.0
  • 20240709172110 Indicates the revision date of THE DOCUMENT YOU ARE READING in UTC, ie 2024 July 9th, 17:21:10. This is used as a means of revision control.
• Click on Advanced
• Click Convert to template
• Click OK on the confirmation dialog

A Comparison of Common ESPHome-Compatible MMWave Prescence Sensors

Models

DFRobot C4001 (SEN0609 Variant)

Not personally evaluated, but seems to be a strong contender, especially for larger spaces

• Amazon: Unavailable ("Gravity" SKU SEN0610 listed, but lacks the 8 antenna panels, using 2 instead. Limited feature set and some poor reviews)
  • DFRobot: https://www.dfrobot.com/product-2793.html
  • Digikey: https://www.digikey.com/en/products/detail/dfrobot/SEN0609/23028638?s=N4IgTCBcDaIMoFEByAGAbCgnCAugXyA
  • Mouser: https://www.mouser.com/ProductDetail/DFRobot/SEN0609?qs=HFfMDpzxxd21aKNJkzimmg%3D%3D
• Cost: $14
• Features:
  • Prescence Detection up to 16m (52ft)
    
  • Motion Detection up to 25m (82ft)
  • Horizontal Detection Angle of 100°
  • Vertical Detection Angle of 40°

LD2450

AKA: HLK-LD2450

• Amazon: https://www.amazon.com/gp/product/B0CGNM1VLM/
• Cost ($): $9 ea. @ $45 for 5
• Features
  
  • Prescence Detection up to 6m (19ft)
  • Azimuth (L/R): ±60° (120?)
  • Pitch (Up/Dn): ±35° (70?)
  • Track up to 3 targets*
  • Define up to 3 zones*
  • 2 sensors might cover the same area and more effectively than an LD1125H, for less cost

* May be an ESPHome implementation-specific limitation that can be overcome by tweaking the lambdas

LD1125H

AKA: HLK-LD1125H-24G

• Amazon: https://www.amazon.com/EC-Buying-HLK-LD1125H-24G-Millimeter-Respiratory/dp/B0C743T4H9/
• Cost ($$$): ~$25 ea. (Might be able to be sourced cheaper via AliExpress, etc)
• Features
  • Moving Body Detection up to 8m (26ft)
  • Stationary Body Detection up to 4m (13ft)
  • Ceiling Mounted Body Detection 2m (6.5ft) radius @ 3m (10ft) height
• Limitations
  • Not cost-effective when an LD2450's range will do
  • No target tracking or zones
  • Movement filtering, such as for a fan, must be done by a radius range with an implementation-specific lambdaSEN0610