2

I have encountered a couple of times that a previously functioning Guest simply starts freezing on boot. (I have now reverted my VM to a previous working snapshot, so I cannot directly start up the broken guest to see what's going on.) I would like to know some general ideas of what can be wrong in this situation and how to go about debugging and recovering the guest from this state.

Steps to reproduce (I have not been able to directly reproduce this from a clean install.):

  1. Setup working guest with console via virsh and ssh access functioning.
  2. Run server over a number of reboots while everything seems functional.
  3. Eventually after some shutdown the system will be in a frozen state:
    1. virsh list shows the machine as running
    2. virsh reboot and virsh shutdown had no effect (Needed to install acpi)
    3. virsh destroy works, but then virsh start boots into the same situation where virsh console hangs at the escape character.

What would you look for? Any commong problems that would look like this? How would one fix these from the host?

For context, I am a novice with regards to KVM, QEMU, etc.

I want to setup a virtual Ubuntu server that other's can use as without any thoughts/considerations about its virtual nature.

My host machine is running Fedora 32, and I have setup the following Virtual Machine based on the standard Ubunutu live server image.

sudo virt-install \
-n ubuntu_vm \
--description "Virtual Server (Ubuntu 20.04)" \
--os-type=Linux \
--os-variant=ubuntu20.04 \
--ram=4096 \
--vcpus=4 \
--disk path=/home/libvirt/images/ubuntu_vm.img,bus=virtio,size=200 \
--graphics none \
--location /home/libvirt/images/ubuntu-20.04.1-live-server-amd64.iso \
--network bridge:br0 \
--extra-args 'console=ttyS0,115200n8 serial' \
--force --debug
user9794
  • 21
  • 2

0 Answers0