Manage KVM QEMU virtual machines with virt

Prerequisite

sudo apt install qemu-kvm libvirt-clients libvirt-daemon-system bridge-utils virtinst libvirt-daemon

Add user to these groups to not have to use sudo/root for managing machines

sudo adduser $USER libvirt
sudo adduser $USER kvm

Create

Choose network type

If you want network bridge, this will make your VMs a part of your LAN and not NAT through your hosy

auto br0
iface br0 inet dhcp
    bridge_ports enp3s0  # Replace with your actual interface
    bridge_stp off
    bridge_fd 0

Or if you are using systemd network. Create these 3 files in /etc/systemd/network/

25-br0.netdev

[NetDev]
Name=br0
Kind=bridge
MACAddress=18:31:bf:25:1d:5f

30-enp10s0.network

[Match]
Name=enp10s0

[Network]
Bridge=br0

35-br0.network

[Match]
Name=br0

[Network]
DHCP=ipv4
IPv6AcceptRA=yes
Address=fda9:9699:faa:cda5::21/64

[IPv6AcceptRA]
UseDNS=yes
UseDomains=yes

then run

sudo systemctl restart systemd-networkd

Create file host-bridge.xml

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

virsh net-define host-bridge.xml
virsh net-start host-bridge
virsh net-autostart host-bridge

sudo mkdir -p /etc/qemu
echo "allow br0" | sudo tee /etc/qemu/bridge.conf
sudo chmod 0644 /etc/qemu/bridge.conf

sudo chmod u+s /usr/lib/qemu/qemu-bridge-helper

To stop/delete to start from scratch

virsh net-destroy host-bridge
virsh net-undefine host-bridge

Create

virt-install \
--name k8s-master-01 \
--ram 4096 \
--vcpus 2 \
--disk path=/mnt/ssd1/vms/kvm/k8s-master-01.qcow2,size=20 \
--os-variant debian13 \
--network network=host-bridge \
--graphics vnc \
--location 'https://deb.debian.org/debian/dists/trixie/main/installer-amd64/' \
--virt-type=kvm

then you need something to connect to the session through VNC. Cockpit for example.

After you have contact with the VM, enable console properly:

sudo systemctl enable serial-getty@ttyS0.service
sudo systemctl start serial-getty@ttyS0.service

Then connect from the host with

virsh console k8s-master-01

Also add this to GRUB

GRUB_CMDLINE_LINUX_DEFAULT="console=tty0 console=ttyS0,115200n8"

guest agents

sudo apt install -y qemu-guest-agent spice-vdagentd

maybe you need to a Enter-keypress before it launches login prompt

VM Operations

List all

virsh list --all

Start machine

virsh start debian12

Edit machine

virsh edit debian12

Power off machine

virsh shutdown debian12

Force shut power off

virsh destroy debian12

Clone machine

virt-clone --original debian12 --name debian12-copy --file /var/lib/libvirt/images/debian12-copy.qcow2

Snapshot

virsh snapshot-create-as --domain k8s-master-01 \
    --name post_ingress_stable \
    --description "Master node with working Ingress and MetalLB" \
    --atomic

Network

List

virsh net-list --all

Start network, must be done before the VM can start

virsh net-start default

Stop network

virsh net-destroy <network-name>

Delete

virsh net-undefine <network-name>

Network info

virsh net-info default

Get VMs IP addresses

virsh net-dhcp-leases default

Article on how to set up networks in KVM

Mount disk shared with host

This is in virt manager gui

• Set up shared memory
• Press "Add hardware", select "Filesystem"
• Select the host folder, and add a name for Target path like this "mount_tag_ssd1"

Typical error message: Unable to find a satisfying virtiofsd

make sure to install virtiofsd

sudo apt install virtiofsd

Add the path inside the xml right under driver tag:

  <binary path="/run/current-system/sw/bin/virtiofsd"/>

Inside the VM you mount in fstab

#virtiofs
mount_tag_ssd1 /mnt/ssd1 virtiofs rw,relatime       0       0

via virsh edit

<domain type='kvm'>
  ...
  <memoryBacking>
    <access mode='shared'/>
  </memoryBacking>
  ...
  <devices>
    <filesystem type='mount' accessmode='passthrough'>
      <driver type='virtiofs'/>
      <source dir='/mnt/raid1/your_folder'/>
      <target dir='my_raid_mount'/>
    </filesystem>
    ...
  </devices>
</domain>

Extend size of existing disk

Host:

sudo qemu-img resize debian-vm.qcow2 +20G
virsh blockresize debian-vm vda 30G

inside VM:

sudo apt install cloud-guest-utils -y
sudo growpart /dev/vda 1
sudo resize2fs /dev/sda1

NFS

Server

sudo apt install nfs-kernel-server -y

Edit /etc/exports:

/mnt/your-raid-path 192.168.1.0/24(rw,sync,no_subtree_check,no_root_squash)

/mnt/md1/bak/k8s-cluster1 192.168.1.0/24(rw,sync,no_subtree_check,no_root_squash)

change /etc/nfs.conf

[mountd]
port=32767

sudo systemctl restart nfs-kernel-server

open firewall

tcp dport { 111, 2049, 32767 } accept
udp dport { 111, 2049, 32767 } accept

Client

sudo apt install nfs-common -y

sudo showmount -e 192.168.1.21