Skip to content

KVM and libvirt

KVM is the Linux kernel's hardware virtualization, QEMU emulates the virtual hardware, and libvirt is the management layer that virsh and virt-install drive. Together they run virtual machines on a Linux host.

Track: Advanced · Interview weight: Low

Must-Know Facts

Fact Value Verify with
KVM Kernel module using CPU virtualization (VT-x, AMD-V) lsmod | grep kvm
CPU support vmx (Intel) or svm (AMD) flag, and /dev/kvm grep -E 'vmx|svm' /proc/cpuinfo
QEMU Emulates the virtual hardware for the guest qemu-system-x86_64 --version
libvirt Management API and daemon over KVM/QEMU systemctl status libvirtd
virsh CLI to libvirt: list, start, define domains virsh list --all
Host check Validates KVM readiness virt-host-validate
Default network NAT network default, 192.168.122.0/24 virsh net-list
Snapshot Point-in-time state of a domain virsh snapshot-create-as

The Stack

KVM turns the kernel into a hypervisor using the CPU's virtualization extensions, QEMU provides the emulated devices, and libvirt manages the domain through a stable API. virt-host-validate checks whether the host can run accelerated VMs.

virsh --version
sudo virt-host-validate qemu | head -2

Output:

11.10.0
  QEMU: Checking for hardware virtualization : FAIL (Host not compatible with KVM; HW virtualization CPU features not found. Only emulated CPUs are available; performance will be significantly limited)

This capture host is itself a lightweight VM without nested virtualization, so KVM fails and only slow emulation is possible. A bare-metal host with vmx/svm and /dev/kvm passes.

No /dev/kvm means emulation only

Without hardware virtualization, QEMU emulates the CPU in software, which is far slower. Production VMs need /dev/kvm, which requires the CPU flag and, on a VM host, nested virtualization enabled.

Managing Domains

virsh lists, starts and defines domains (libvirt's name for a VM), and virt-install creates one from an install source. A domain is defined by XML that libvirt stores, editable with virsh edit.

virsh list --all
virt-install --name web --memory 2048 --vcpus 2 \
  --disk size=20 --os-variant rocky10 --cdrom /iso/Rocky-10.iso
virsh snapshot-create-as web pre-change    # save a rollback point

The default network is NAT on 192.168.122.0/24; a bridged network instead puts guests on the physical LAN, covered in Bridges, Bonds and VLANs.

Snapshots are a rollback, not a backup

virsh snapshot-revert returns a VM to a saved state, but long-lived snapshots grow and slow the disk, and they live on the same host. Keep a real backup separately.

Common Errors

virt-host-validate reports hardware virtualization FAIL

Cause: the CPU lacks vmx/svm, or nested virtualization is off, so /dev/kvm is unavailable.

Fix: enable virtualization in firmware, or nested virtualization on the parent hypervisor; otherwise only slow emulation runs.

error: failed to connect to the hypervisor

Cause: libvirtd is not running, or the user is not in the libvirt group.

Fix: start libvirtd and add the user to libvirt, then reconnect.

Interview Checkpoints

L1: What are KVM, QEMU and libvirt, and how do they relate?

Say first: KVM is the kernel's hardware virtualization, QEMU emulates the virtual hardware, and libvirt is the management layer tools like virsh use.

Proof: lsmod | grep kvm; qemu-system-x86_64 --version; virsh list.

Follow-up: what does a VM have that a container does not? (its own guest kernel.)

L1: How do you tell if a host can run accelerated VMs?

Say first: check for the CPU virtualization flag and /dev/kvm, or run virt-host-validate.

Proof: grep -E 'vmx|svm' /proc/cpuinfo; ls /dev/kvm; virt-host-validate qemu.

Follow-up: what happens without /dev/kvm? (QEMU emulates in software, much slower.)

L2: Create and start a new VM from an ISO.

Say first: define it with virt-install, then manage it with virsh.

Proof:

virt-install --name web --memory 2048 --vcpus 2 --disk size=20 --cdrom /iso/os.iso

Follow-up: where is the VM's definition stored? (libvirt XML, edited with virsh edit.)

L2: A VM must be reachable on the physical LAN, not behind NAT. What network do you use?

Say first: a bridged network instead of the default NAT network.

Proof: attach the guest to a host bridge; the default 192.168.122.0/24 is NAT-only.

Follow-up: what does the default libvirt network give? (NAT outbound on 192.168.122.0/24.)

L3: virsh cannot connect to the hypervisor on a fresh install. How do you diagnose it?

Say first: check that libvirtd is running and the user can reach the socket.

Proof: systemctl status libvirtd; add the user to libvirt; retry virsh list; virt-host-validate for KVM support.

Follow-up: what would confirm acceleration is available? (virt-host-validate qemu passing.)

L2: Roll a VM back to a known-good state.

Say first: take a snapshot beforehand and revert to it.

Proof: virsh snapshot-create-as web pre-change; virsh snapshot-revert web pre-change.

Follow-up: why are snapshots not a backup? (they grow, slow the disk, and live on the same host.)

Captured on Rocky Linux 10.2 on an iximiuz Labs FlexBox microVM, kernel 6.1.167, 2026-09. The host has no /dev/kvm, so virsh and virt-install domain commands are shown without output; they run on a host with hardware virtualization.