user_ssh_mgmt step randomly fails

It sometimes works, sometimes fails with:

fatal: [localhost]: UNREACHABLE! => {"changed": false, "msg": "SSH Error: data could not be sent to remote host "xx.xx.xx.xx". Make sure this host can be reached over ssh", "unreachable": true}

I wonder if it's trying faster than Linode can create the box. Maybe a pause/retry system would help?