nmstate-configuration.service failed due to wrong variable name $hostname_file
https://github.com/openshift/machine-config-operator/blob/5a6e8b81f13de2dbf606a497140ac6e9c2a00e6f/templates/common/baremetal/files/nmstate-configuration.yaml#L26

4.16.0    

always

    1. install cluster via dev-script, with node-specific network configuration
    

nmstate-configuration failed:

sh-5.1# journalctl -u nmstate-configuration
May 07 02:19:54 worker-0 systemd[1]: Starting Applies per-node NMState network configuration...
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + systemctl -q is-enabled mtu-migration
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + echo 'Cleaning up left over mtu migration configuration'
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: Cleaning up left over mtu migration configuration
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + rm -rf /etc/cno/mtu-migration
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + '[' -e /etc/nmstate/openshift/applied ']'
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + src_path=/etc/nmstate/openshift
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + dst_path=/etc/nmstate
May 07 02:19:54 worker-0 systemd[1]: nmstate-configuration.service: Main process exited, code=exited, status=1/FAILURE
May 07 02:19:54 worker-0 nmstate-configuration.sh[1565]: ++ hostname -s
May 07 02:19:54 worker-0 systemd[1]: nmstate-configuration.service: Failed with result 'exit-code'.
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + hostname=worker-0
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + host_file=worker-0.yml
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + cluster_file=cluster.yml
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + config_file=
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: + '[' -s /etc/nmstate/openshift/worker-0.yml ']'
May 07 02:19:54 worker-0 nmstate-configuration.sh[1553]: /usr/local/bin/nmstate-configuration.sh: line 22: hostname_file: unbound variable
May 07 02:19:54 worker-0 systemd[1]: Failed to start Applies per-node NMState network configuration.

cluster can be setup successfully with node-specific network configuration via new mechanism