VMs have too low entropy, causes long delays

[twelho]

Ignite VMs currently have no (fast enough) way of gathering entropy from the hardware to fill /dev/random, which causes hangs in e.g. the boot process and SSH connections (see here).

Experiment with forwarding /dev/hwrng to the Ignite container and adding random.trust_cpu=1 to the cmdline (for host and for VM), Firecracker might pick it up then.

See also https://github.com/firecracker-microvm/firecracker/issues/1671 and https://github.com/firecracker-microvm/firecracker/issues/663.