Setting Up Xen Dom0 on Fedora : Xen 3.4.1 with Linux Kernel 2.6.29 on Fedora 12

Setting Up Xen Dom0 on Fedora

Xen remains a production hypervisor for enterprise deployments (AWS EC2, Citrix Hypervisor), though KVM has become the default for most Linux distributions. If you’re deploying Xen in 2026, you’ll likely be working with recent stable releases rather than the legacy versions covered in older guides.

This guide covers building a modern Xen Dom0 environment on current Fedora. Xen and kernel versions move faster than they did a decade ago, so check the official Xen release page for the latest stable build matching your hardware support requirements.

Hardware Considerations

Modern Xen deployments typically run on:

• Xeon or EPYC processors with hardware virtualization (VT-x or AMD-V)
• Server-class motherboards with IOMMU support (Intel VT-d or AMD-Vi)
• ECC memory (strongly recommended for production)
• NVMe or SSD storage for Dom0 and guest performance

Older hardware may work but lacks features like PCI passthrough and nested virtualization. Check your CPU against the Xen CPU support matrix.

Building Xen from Source

Fetch the latest stable release from xenproject.org:

mkdir -p /usr/src/xen
cd /usr/src/xen
wget https://downloads.xenproject.org/release/xen/xen-VERSION.tar.gz
tar -xf xen-VERSION.tar.gz
cd xen-VERSION

Install build dependencies on Fedora:

sudo dnf install gcc make python3-devel libuuid-devel zlib-devel \
  openssl-devel ncurses-devel bison flex perl xz-devel git

Build the hypervisor, tools, and documentation:

make -j$(nproc) dist-xen dist-tools dist-docs
cd dist
sudo sh ./install.sh

The installation script places binaries in /usr/lib/xen/, Dom0 kernel in /boot/, and tools in /usr/bin/.

Kernel Configuration

Xen requires a Dom0 kernel built with specific options. Use your distribution’s kernel source or compile from upstream. If building custom, ensure these options in your kernel .config:

CONFIG_XEN=y
CONFIG_XEN_DOM0=y
CONFIG_XEN_DEV_EVTCHN=y
CONFIG_XEN_BACKEND=y
CONFIG_XEN_BLKDEV_BACKEND=y
CONFIG_XEN_NETDEV_BACKEND=y
CONFIG_SWIOTLB=y
CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS=y

For HVM guest support:

CONFIG_KVM=y
CONFIG_HVM=y

Build and install:

make -j$(nproc)
sudo make modules_install install

If using Fedora’s standard kernel, check whether Xen support is already compiled in:

grep -i xen /boot/config-$(uname -r)

Bootloader Configuration

Update your EFI boot entries or GRUB2 configuration. For GRUB2:

sudo grubby --add-kernel=/boot/vmlinuz-VERSION --title="Fedora with Xen" \
  --initrd=/boot/initramfs-VERSION.img --args="dom0_mem=1024M,max:4096M"

Set Xen as default:

sudo grubby --set-default-index=0

Key Dom0 kernel parameters:

• dom0_mem=1024M,max:4096M — Allocate 1GB initially, cap at 4GB
• iommu=1 — Enable IOMMU for device passthrough
• console=vga — Use VGA for early boot output
• loglevel=debug — Increase verbosity for troubleshooting

Reboot into Xen:

sudo reboot

Verify you’re running Xen:

xl info

Output should show xen_version and xen_scheduler (typically credit2 in modern builds):

host                   : fedora41.example.com
release                : 6.7.0-xen
xen_major              : 4
xen_minor              : 18
xen_scheduler          : credit2
total_memory           : 16384
free_memory            : 12288

Daemon and Service Configuration

Enable Xen services for automatic startup. On modern Fedora, use systemd:

sudo systemctl enable xend
sudo systemctl start xend
sudo systemctl enable xendomains
sudo systemctl start xendomains

Check service status:

sudo systemctl status xend
sudo systemctl status xendomains

View logs in real time:

sudo journalctl -u xend -f

Disable KSM tuning to avoid conflicts:

sudo systemctl disable ksmtuned

Managing Virtual Machines

List running domains:

xl list

Create a guest configuration file (vm-ubuntu.cfg):

name = "ubuntu-guest-1"
memory = 2048
vcpus = 2
vif = [ 'mac=00:16:3e:5f:60:01,bridge=xenbr0' ]
disk = [ '/var/lib/xen/images/ubuntu.img,xvda,w' ]
kernel = "/var/lib/xen/images/vmlinuz"
ramdisk = "/var/lib/xen/images/initrd.img"
bootloader = "pygrub"

Create and start the guest:

xl create -c vm-ubuntu.cfg

The -c flag connects to the console automatically. Suspend/resume guests:

xl pause ubuntu-guest-1
xl unpause ubuntu-guest-1

Shut down cleanly:

xl shutdown ubuntu-guest-1

Force destruction if shutdown fails:

xl destroy ubuntu-guest-1

PCI Passthrough (Advanced)

For direct device access (GPU, NIC), enable IOMMU in kernel parameters, then hide devices from Dom0:

grep -i iommu /boot/config-$(uname -r)

Find your device:

lspci | grep -i nvidia

Hide from Dom0 by adding to kernel parameters:

sudo grubby --update-kernel=ALL \
  --args="pciback.hide=(0000:85:00.0)(0000:85:00.1)"

Reboot and attach to guest in config:

pci = [ '85:00.0', '85:00.1' ]

Troubleshooting

Check hypervisor logs:

dmesg | grep Xen

Verify Dom0 is using Xen kernel:

uname -r
cat /proc/xen/capabilities

If xend fails to start, check for port conflicts:

sudo lsof -i :8006

Monitor resource usage:

xl top

For nested virtualization or KVM guests under Xen, verify capabilities:

xl info | grep hvm

Xen remains stable for production workloads but requires careful planning around hardware support, kernel configuration, and ongoing maintenance. For new projects, evaluate whether KVM or containerization better fits your infrastructure needs.