homelabprivacy

Own Your Hypervisor: KVM on Linux Instead of VMware

🤖 Researched and drafted automatically from the official docs, and reviewed before publishing. Commands are taken from the source projects — but always sanity-check before running anything on your own hardware.

VMware’s pricing and licensing model has tightened to the point where self-hosting a hypervisor makes hard financial sense. KVM—the kernel-based virtual machine built into Linux—gives you industrial-grade virtualization without the licensing audit risk. You get full hardware passthrough, live migration, and a rock-solid foundation for a homelab. Here’s how to set it up on bare metal.

Prerequisites

You need:

  • A modern x86-64 CPU with virtualization extensions (Intel VT-x or AMD-V). Check with:
    grep -E '(vmx|svm)' /proc/cpuinfo
    If you see output, you’re good.
  • At least 16 GB of RAM (32 GB is better for running multiple VMs).
  • A Linux distribution with kernel support. Ubuntu Server 22.04 LTS or Debian 12 work well.
  • A second NIC for bridged networking (optional but recommended to avoid trapping yourself on the host network).

Step 1: Verify Virtualization Support in BIOS

Reboot into BIOS/UEFI and enable virtualization extensions:

  • Intel systems: Look for “VT-x” or “Intel Virtualization Technology.”
  • AMD systems: Look for “SVM” or “AMD-V.”

Save and boot back into Linux. Confirm the kernel sees it:

kvm-ok

If kvm-ok is not installed, run:

sudo apt update && sudo apt install cpu-checker

You should see: “KVM acceleration can be used.”

Step 2: Install KVM and Management Tools

On Debian/Ubuntu:

sudo apt update
sudo apt install qemu-kvm libvirt-daemon-system libvirt-clients bridge-utils virt-manager

This installs:

  • qemu-kvm: The hypervisor and emulation engine.
  • libvirt: The daemon that manages VMs and networking.
  • libvirt-clients: Command-line tools (virsh, virt-install).
  • bridge-utils: For network bridging.
  • virt-manager: Optional GUI, but we’ll use CLI.

Enable and start libvirtd:

sudo systemctl enable libvirtd
sudo systemctl start libvirtd

Add your user to the libvirt group to run VMs without sudo:

sudo usermod -aG libvirt $USER
newgrp libvirt

By default, libvirt uses NAT, which works but limits VM connectivity. Bridged networking lets VMs get IPs on your LAN directly.

Identify your primary NIC:

ip link show

Let’s say it’s eth0. Create a bridge by editing /etc/netplan/01-netcfg.yaml (or your distro’s network config):

network:
  version: 2
  ethernets:
    eth0:
      dhcp4: false
  bridges:
    br0:
      dhcp4: true
      interfaces: [eth0]

Apply:

sudo netplan apply

Verify the bridge is up:

ip addr show br0

Then define the bridge in libvirt. Create /tmp/bridge.xml:

<network>
  <name>br0</name>
  <forward mode="bridge"/>
  <bridge name="br0"/>
</network>

Define and start it:

virsh net-define /tmp/bridge.xml
virsh net-start br0
virsh net-autostart br0

Step 4: Create Your First VM

Download an OS image. For Ubuntu 22.04 LTS:

cd ~/vms
wget https://cloud-images.ubuntu.com/jammy/current/jammy-server-cloudimg-amd64.img

Create a VM with 4 vCPUs, 8 GB RAM, and 50 GB disk:

virt-install \
  --name ubuntu-test \
  --vcpus 4 \
  --memory 8192 \
  --disk ~/vms/ubuntu-test.qcow2,size=50 \
  --network bridge=br0 \
  --os-type linux \
  --os-variant ubuntu22.04 \
  --import \
  --graphics none \
  --console pty,target_type=serial

This creates a VM and boots it. Once running, connect via:

virsh console ubuntu-test

Press Ctrl+] to exit the console.

Step 5: Manage VMs with virsh

List running VMs:

virsh list

Stop a VM gracefully:

virsh shutdown ubuntu-test

Force stop:

virsh destroy ubuntu-test

Start a VM:

virsh start ubuntu-test

Edit VM config (CPU, RAM, disk):

virsh edit ubuntu-test

Delete a VM and its disk:

virsh undefine ubuntu-test --remove-all-storage

Step 6: Enable Hardware Passthrough (Optional)

For GPU or NIC passthrough, you need IOMMU. Enable it in GRUB:

Edit /etc/default/grub and add to GRUB_CMDLINE_LINUX_DEFAULT:

  • Intel: intel_iommu=on
  • AMD: amd_iommu=on

Example:

GRUB_CMDLINE_LINUX_DEFAULT="quiet splash intel_iommu=on"

Update and reboot:

sudo update-grub
sudo reboot

Verify IOMMU is active:

dmesg | grep -i iommu

You should see “IOMMU enabled.” Then use virsh nodedev-list --cap pci to list PCI devices and passthrough them in VM XML.

Step 7: Keep It Secure

  • Never expose libvirtd to the network. It’s a local socket only—keep it that way.
  • Use a separate VLAN or subnet for your VMs if you’re on a shared network.
  • Run libvirtd behind a VPN or on an isolated LAN. Do not allow remote access from the internet.
  • Keep the host patched. Run sudo apt update && sudo apt upgrade regularly.
  • Use SELinux or AppArmor if your distro supports it; libvirt integrates with both.

Monitoring and Logs

Watch VM resource usage:

virt-top

Check libvirtd logs:

sudo journalctl -u libvirtd -f

View VM logs:

virsh qemu-monitor-command ubuntu-test '{"execute": "query-status"}'

Is It Worth It?

Yes. You get:

  • No licensing costs. KVM is part of the Linux kernel—free forever.
  • Full control. You own the hypervisor, the config, the VMs, everything.
  • Mature tooling. libvirt has been battle-tested in production for 15+ years.
  • Portability. Move VMs between KVM hosts via virsh dump and virsh restore.
  • No vendor lock-in. Your VM configs are standard XML; your disks are QCOW2 or raw images.

The tradeoff: you’re responsible for patching, backups, and troubleshooting. But if you’re already running a homelab, you’re used to that. KVM is simpler than VMware’s licensing maze and just as capable for homelabs and small production setups.

Related guides

← All guides