QEMU essential

  |   Source

Start virtual machine with 8G memory and Debian Linux iso using CLI, 

qemu-img create -f qcow2 debin.img 16G
qemu-system-x86_64 -hda debian.img -boot d -cdrom "debian-testing-amd64-netinst.iso" -m 8192

Windows 10, create virtual disk "win10.qcow" and use "virt-manager" to create VM, 

qemu-img create -f qcow2 win10.qcow 30G # create a new virtual disk, windows 10 need 30G, 
qemu-img resize win10.qcow 30G # resize virtual disk 
sudo qemu-nbd -c /dev/nbd0 win10.qcow # bind virtual disk to linux device
sudo gparted /dev/nbd0 # use gparted to resize partition of virtual disk
sudo fdisk -l /dev/nbd0 # print partition info of virtual disk
sudo mount /dev/nbd0p2 /mnt/c # mount second partition of virtual disk "npd0p2" into "/mnt/c"
sudo umount /mnt/c && sudo qemu-nbd -d /dev/nbd0 # umount and disconnect virtual disk

VM can access physical USB stick on Linux,

  • Run lsusb, get below output of USB stick,
Bus 005 Device 003: ID 0930:6545 Toshiba Corp. Kingston DataTraveler 102/2.0 / HEMA Flash Drive 2 GB / PNY Attache 4GB Stick
  • Then insert below content into "<devices>" node od XML of virt-manager and start VM,
<hostdev mode='subsystem' type='usb'>
  <source startupPolicy='optional'>
  <vendor id='0x0930'/>
  <product id='0x6545'/>
  </source>
</hostdev>

If default NAT network is not active, fix with below cli, 

sudo virsh net-list --all # display network
sudo virsh net-start default  # start default network
sudo virsh net-autostart default # autostart default network

Pleae note dnsmasq on Linux host needs extra setup, 

On linux host servers, libvirtd uses dnsmasq to service the virtual networks, such as the default network. A new instance of dnsmasq is started for each virtual network, only accessible to guests in that specific network.

If you are running your own "global" dnsmasq, then this can cause your own dnsmasq to fail to start (or for libvirtd to fail to start its dnsmasq and the given virtual network). This happens because both instances of dnsmasq might try to bind to the same port number on the same network interfaces.

You have to change the global /etc/dnsmasq.conf as follows:
interface=eth0 # only handle requests from interface eth0
bind-interfaces # don't accept request from other interfaces from the beginning
Comments powered by Disqus