QEMU

Docker container for running virtual machines using QEMU.

Features ✨

• Create VM's which behave just like normal containers

• Manage them using all your existing tools (like Portainer)

• Configure them in a language (YAML) you are already familiar with

• Web-based viewer to control the machine directly from your browser

• Supports .iso, .img, .qcow2, .vhd, .vhdx, .vdi, .vmdk and .raw disk formats

• High-performance options (like KVM acceleration, kernel-mode networking, IO threading, etc.) to achieve near-native speed

Usage 🐳

Via Docker Compose:

services:
  qemu:
    container_name: qemu
    image: qemux/qemu-docker
    environment:
      BOOT: "https://dl-cdn.alpinelinux.org/alpine/v3.19/releases/x86_64/alpine-virt-3.19.1-x86_64.iso"
    devices:
      - /dev/kvm
    cap_add:
      - NET_ADMIN
    ports:
      - 8006:8006
    stop_grace_period: 2m

Via Docker CLI:

docker run -it --rm -e "BOOT=http://example.com/image.iso" -p 8006:8006 --device=/dev/kvm --cap-add NET_ADMIN qemux/qemu-docker

Via Kubernetes:

kubectl apply -f kubernetes.yml

FAQ 💬

• How do I use it?

  Very simple! These are the steps:

  • Set the BOOT environment variable to the URL of any disk image you want to install.

  • Start the container and connect to port 8006 using your web browser.

  • You will see the screen and can now install the OS of your choice using your keyboard and mouse.

  Enjoy your brand new machine, and don't forget to star this repo!

• How do I change the storage location?

  To change the storage location, include the following bind mount in your compose file:

  volumes:
    - /var/qemu:/storage

  Replace the example path /var/qemu with the desired storage folder.

• How do I change the size of the disk?

  To expand the default size of 16 GB, add the DISK_SIZE setting to your compose file and set it to your preferred capacity:

  environment:
    DISK_SIZE: "128G"

  This can also be used to resize the existing disk to a larger capacity without any data loss.

• How do I boot a local image?

  You can use a local image file directly, and skip the download altogether, by binding it in your compose file:

  volumes:
    - /home/user/example.iso:/boot.iso

  This way you can supply a boot.iso, boot.img or boot.qcow2 file. The URL of the BOOT variable will be ignored in this case.

• How do I boot ARM images?

  You can use qemu-arm to run ARM64-based images.

• How do I boot Windows?

  Use dockur/windows instead, as it includes all the drivers required during installation, amongst many other features.

• How do I boot macOS?

  Use dockur/macos instead, as it uses all the right settings and automaticly downloads the installation files.

• How do I boot without VirtIO drivers?

  By default, the machine makes use of virtio-scsi drives for performance reasons, and even though most Linux kernels bundle the necessary driver for this device, that may not always be the case for other operating systems.

  If your machine fails to detect the hard drive, you can modify your compose file to use virtio-blk instead:

  environment:
    DISK_TYPE: "blk"

  If it still fails to boot, you can set the value to ide to emulate a IDE drive, which is slow but requires no drivers and is compatible with almost every system.

• How do I change the amount of CPU or RAM?

  By default, the container will be allowed to use a maximum of 1 CPU core and 1 GB of RAM.

  If you want to adjust this, you can specify the desired amount using the following environment variables:

  environment:
    RAM_SIZE: "4G"
    CPU_CORES: "4"

• How do I verify if my system supports KVM?

  To verify that your system supports KVM, run the following commands:

  sudo apt install cpu-checker
  sudo kvm-ok

  If you receive an error from kvm-ok indicating that KVM acceleration can't be used, please check whether:

  • the virtualization extensions (Intel VT-x or AMD SVM) are enabled in your BIOS.

  • you are running an operating system that supports them, like Linux or Windows 11 (macOS and Windows 10 do not unfortunately).

  • you enabled "nested virtualization" if you are running the container inside a virtual machine.

  • you are not using a cloud provider, as most of them do not allow nested virtualization for their VPS's.

  If you didn't receive any error from kvm-ok at all, but the container still complains that /dev/kvm is missing, it might help to add privileged: true to your compose file (or --privileged to your run command), to rule out any permission issue.

• How do I assign an individual IP address to the container?

  By default, the container uses bridge networking, which shares the IP address with the host.

  If you want to assign an individual IP address to the container, you can create a macvlan network as follows:

  docker network create -d macvlan \
      --subnet=192.168.0.0/24 \
      --gateway=192.168.0.1 \
      --ip-range=192.168.0.100/28 \
      -o parent=eth0 vlan

  Be sure to modify these values to match your local subnet.

  Once you have created the network, change your compose file to look as follows:

  services:
    qemu:
      container_name: qemu
      ..<snip>..
      networks:
        vlan:
          ipv4_address: 192.168.0.100
  
  networks:
    vlan:
      external: true

  An added benefit of this approach is that you won't have to perform any port mapping anymore, since all ports will be exposed by default.

  Please note that this IP address won't be accessible from the Docker host due to the design of macvlan, which doesn't permit communication between the two. If this is a concern, you need to create a second macvlan as a workaround.

• How can the VM acquire an IP address from my router?

  After configuring the container for macvlan (see above), it is possible for the VM to become part of your home network by requesting an IP from your router, just like a real PC.

  To enable this mode, add the following lines to your compose file:

  environment:
    DHCP: "Y"
  devices:
    - /dev/vhost-net
  device_cgroup_rules:
    - 'c *:* rwm'

  Please note that in this mode, the container and the VM will each have their own separate IPs. The container will keep the macvlan IP, and the VM will use the DHCP IP.

• How do I add multiple disks?

  To create additional disks, modify your compose file like this:

  environment:
    DISK2_SIZE: "32G"
    DISK3_SIZE: "64G"
  volumes:
    - /home/example:/storage2
    - /mnt/data/example:/storage3

• How do I pass-through a disk?

  It is possible to pass-through disk devices directly by adding them to your compose file in this way:

  devices:
    - /dev/sdb:/disk1
    - /dev/sdc:/disk2

  Use /disk1 if you want it to become your main drive, and use /disk2 and higher to add them as secondary drives.

• How do I pass-through a USB device?

  To pass-through a USB device, first lookup its vendor and product id via the lsusb command, then add them to your compose file like this:

  environment:
    ARGUMENTS: "-device usb-host,vendorid=0x1234,productid=0x1234"
  devices:
    - /dev/bus/usb

• How can I provide custom arguments to QEMU?

  You can create the ARGUMENTS environment variable to provide additional arguments to QEMU at runtime:

  environment:
    ARGUMENTS: "-device usb-tablet"

• What image formats are supported?

  You can set the BOOT URL to any .iso, .img, .raw, .qcow2, .vhd, .vhdx, .vdi or .vmdk file.

  It will even automaticly extract compressed images, like .img.gz, .qcow2.xz, .iso.zip and many more!

Stars 🌟