docker driver, start again always needs a reset why ?

[medyagh]

Unlike VM , start after start in docker invokes preparing kubernetes

right after starting a docker driver
when we start again (I added debugging statement)

we need to figure out why api server is in Sate Error when we do start again.

$ make && ./out/minikube start --vm-driver=docker
😄  minikube v1.9.2 on Darwin 10.13.6
✨  Using the docker driver based on existing profile
👍  Starting control plane node m01 in cluster minikube
🚜  Pulling base image ...
🏃  Updating the running docker "minikube" container ...
🐳  Preparing Kubernetes v1.18.0 on Docker 19.03.2 ...
    ▪ kubeadm.pod-network-cidr=10.244.0.0/16
needs reset: apiserver in state Error🌟  Enabling addons: default-storageclass, storage-provisioner
🏄  Done! kubectl is now configured to use "minikube"

I belive the root cause is that that Provosion is restarting docker that kills api server
code from fix.go

	h, err = recreateIfNeeded(api, cc, n, h)
	if err != nil {
		return h, err
	}

	// Avoid reprovisioning "none" driver because provision.Detect requires SSH
	if !driver.BareMetal(h.Driver.DriverName()) {
		e := engineOptions(cc)
		h.HostOptions.EngineOptions.Env = e.Env
		err = provisionDockerMachine(h)
		if err != nil {
			return h, errors.Wrap(err, "provision")
		}
	}

	if driver.IsMock(h.DriverName) {
		return h, nil
	}

	if err := postStartSetup(h, cc); err != nil {
		return h, errors.Wrap(err, "post-start")
	}

[medyagh]

I believe this is the root cause --> #7433

[medyagh]

I think this mystrie was due to the IP change