Ubuntu VM Installation

Suppose you’d like to use netlab with libvirt[1] or run network devices as containers on a Windows- or MacOS-based computer. You’ll have to run the whole toolchain needed to create networking labs (netlab ⇨ Vagrant ⇨ libvirt ⇨ KVM) within a Linux virtual machine.

The easiest way to set up your lab environment is to create a Ubuntu virtual machine and use the netlab install command within that virtual machine to install the required software packages[2]. You could also run netlab on a Ubuntu instance in a public cloud.

Running Ubuntu VM on a desktop OS

Warning

Running ‌libvirt within an Ubuntu VM requires ‌nested virtualization. You don’t need nested virtualization to run Docker containers within a Ubuntu VM.

Nested virtualization was available in VMware Workstation/Fusion for years and was recently added to VirtualBox. While VMware products perform flawlessly, we experienced unacceptable performance with VirtualBox nested virtualization on some Intel CPUs.

Canonical Multipass is probably the easiest way to start a Ubuntu VM on your laptop if you don’t need nested virtualization. Create an instance with as much RAM and as many CPU cores as you can afford, and install the necessary software on it.

You can also use Vagrant or create the virtual machine yourself (using, for example, VirtualBox or VMware GUI)

Creating Ubuntu VM with Vagrant

You can use Vagrant on your computer to set up an Ubuntu VM. Vagrant will automatically:

  • Download the required virtual disk image

  • Start the virtual machine

  • Enable SSH access to the virtual machine

  • Provision the software on the virtual machine

Installation steps:

  • Install VirtualBox or VMware Fusion/Workstation

  • Install Vagrant

  • Install Vagrant VMware provider if you’re using VMware Workstation/Fusion.

  • Create an empty directory. In that directory, create a Vagrantfile with the following content. Change the memory/memsize or cpus/numvcpus settings to fit your hardware.

Vagrant.configure("2") do |config|
  config.vm.box = "generic/ubuntu2204"

  config.vm.provider "virtualbox" do |vb|
    vb.memory = "8192"
    vb.cpus = 4
    vb.customize ['modifyvm', :id, '--nested-hw-virt', 'on']
  end

  config.vm.provider "vmware_fusion" do |v|
    v.vmx["memsize"] = 8192
    v.vmx["numvcpus"] = "4"
    v.vmx["vhv.enable"] = "TRUE"
  end

  config.vm.provider "vmware_desktop" do |v|
    v.vmx["memsize"] = 8192
    v.vmx["numvcpus"] = "4"
    v.vmx["vhv.enable"] = "TRUE"
  end

  config.vm.provision "shell", privileged: false, inline: <<-SHELL
    sudo apt-get update
    sudo apt-get install -y python3-pip
    sudo pip3 install --ignore-installed networklab
    sudo pip3 install --upgrade pyopenssl cryptography
    netlab install -y ubuntu ansible libvirt containerlab
  SHELL
end

Tip

The above Vagrantfile installs Python packages as root. That differs from the recommended best practice and is used primarily because we’re setting up a single-purpose VM.

  • Execute vagrant up and wait for the installation to complete. If you’re using VMware Workstation or Fusion, specify the –provider argument in the vagrant up command when creating the VM (but not on subsequent starts).

  • Log into the virtual machine with vagrant ssh and test the installation with netlab test

Manual Virtual Machine Provisioning

  • Create an Ubuntu 22.04 or Ubuntu 24.04 virtual machine within your virtualization environment (you’ll find plenty of online tutorials). If needed, enable nested virtualization.

  • Log into the virtual machine

  • Execute these commands to download Python3 and install netlab, Ansible, vagrant, libvirt, KVM, containerlab, and Docker.

sudo apt-get update
sudo apt-get install -y python3-pip
sudo pip3 install --ignore-installed networklab
sudo pip3 install --upgrade pyopenssl cryptography
netlab install -y ubuntu ansible libvirt containerlab

Tip

  • Installing Python packages as root is not the recommended best practice. We’re using this approach because we’re setting up a single-purpose VM.

  • Ubuntu 22.04 wants you to install Python packages in a virtual environment. To stop the complaints, add the --break-system-packages option to the pip3 install command.

  • After completing the software installation, log out from the VM, log back in, and test your installation with the netlab test command. If those tests fail, you might have to use usermod to add your user to the libvirt and docker groups.

Next Steps