VM not working after migrating to Incus

After migrating from lxd to incus my virutal machine isn’t working, when I try: incus exec vm bash
I get:
Error: VM agent isn’t currently running

I’ve red article: Incus 0.5.1 has been released, but this doesn’t resolve my problem.

Virtual machine is debian buster, with only this services running:
lxd-agent-9p.service
lxd-agent.service

This is configuration of virtual machine instance:

incus config show netbox-D01
architecture: x86_64
config:
image.architecture: amd64
image.description: Debian buster amd64 (20200519_05:24)
image.os: Debian
image.release: buster
image.serial: “20200519_05:24”
image.type: disk-kvm.img
volatile.base_image: f0616986aaf33a7229863d355a8aa8dd5f0afa4f6f7929237f3b55c281be1f03
volatile.dev0.host_name: maca16a3dca
volatile.dev0.hwaddr: 00:16:3e:b5:7b:36
volatile.dev0.last_state.created: “true”
volatile.last_state.power: RUNNING
volatile.last_state.ready: “false”
volatile.uuid: bd14866d-48cd-44f3-b04d-d7524db0036e
volatile.uuid.generation: bd14866d-48cd-44f3-b04d-d7524db0036e
volatile.vsock_id: “27”
devices:
agent:
source: agent:config
type: disk
ephemeral: false
profiles:

  • dev
    stateful: false
    description: “”

Can you try:

mount -t 9p config /mnt
cd /mnt
./install.sh
reboot

That should refresh the units with the incus equivalent.

On host or vm?
On host I get:
mount: /mnt: special device config does not exist.
I can’t enter the vm

Would need to be in the VM, connecting either through SSH or through the VGA console using a user inside of the VM.

If you don’t have any way to get into the VM, then we’ll have to directly access the VM disk and set a root password or something so you can get into it from the console and run those commands.

This only really happens with VMs created from older images as our images for the past 6 months or so have included what’s needed to run under either LXD or Incus.

Thank you that worked.