I'm just starting to use Multipass, and I have found that some of the documentation is sparse (or I don't know how to find it properly). I am running 19.10, with QEMU and libvirt installed. I created a bridged Ethernet interface called br0 for a Windows 10 VM that I need on the same subnet as my other workstations. This part works fine. I ran the command "sudo multipass set local.driver=libvirt" and then created my first multipass instance.
The first problem I had was that I had to use the multipass connect method to log into the VM and create a user (other than the default "ubuntu" user) and make the new user a member of sudo. This was because I could not figure out how to log into the VM via the virt-manager console as the multipass sudo user ubuntu. Once that was done, I was able to log into the virt-manager console as the new user, and install openssh-server. I shut down the VM, changed the NIC in virt-manager to the bridged Ethernet interface, restarted the VM and the VM received an IP from my DHCP server and was directly accessible from any other workstation or VM on the network via SSH.
Unfortunately, doing this breaks the "multipass connect vm01" method of connecting to the VM (in fact, it breaks any multipass commands from the CLI that are directed at the instance) and the only way to connect now is via SSH (or the virt-manager console).
Is there a configuration file somewhere for instructing multipass to look for the VM on the new bridged interface? I am actually fine with it operating as a standalone VM under virt-manager (in fact, it is a fantastic way to spin up an Ubuntu VM without creating a new VM in virt-manager, attaching an ISO, etc.). But I can foresee future problems with having broken the multipass CLI link to the instance.