Change kubernetes-dashboard from RC to deployment

[kairen]

The commit changed kubernetes-dashboard from RC to deployment, the feature request from #2389. This PR has been tested on v1.9.0(b109185), more command output detail as below:

$ kubectl get node
NAME       STATUS    ROLES     AGE       VERSION
minikube   Ready     <none>    42s       v1.9.0

$ kubectl -n kube-system get po
NAME                                    READY     STATUS    RESTARTS   AGE
kube-addon-manager-minikube             1/1       Running   0          40s
kube-dns-54cccfbdf8-28cds               3/3       Running   0          36s
kubernetes-dashboard-77d8b98585-xxmxz   1/1       Running   0          36s
storage-provisioner                     1/1       Running   0          37s

$ kubectl -n kube-system get deploy
NAME                   DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
kube-dns               1         1         1            1           38s
kubernetes-dashboard   1         1         1            1           38s

$ ./out/minikube addons open dashboard
Opening kubernetes service kube-system/kubernetes-dashboard in default browser..

Dashboard's preview:

Integration testing result:

$ make integration
cp ./out/minikube-darwin-amd64 ./out/minikube
go test -v -test.timeout=30m k8s.io/minikube/test/integration --tags="integration container_image_ostree_stub containers_image_openpgp"
=== RUN   TestDocker
Environment=DOCKER_RAMDISK=yes FOO=BAR BAZ=BAT

ExecStart={ path=/usr/bin/dockerd ; argv[]=/usr/bin/dockerd -H tcp://0.0.0.0:2376 -H unix:///var/run/docker.sock --tlsverify --tlscacert /etc/docker/ca.pem --tlscert /etc/docker/server.pem --tlskey /etc/docker/server-key.pem --label provider=virtualbox --insecure-registry 10.96.0.0/12 --debug --icc=true ; ignore_errors=no ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; code=(null) ; status=0/0 }

--- PASS: TestDocker (76.73s)
=== RUN   TestFunctional
=== RUN   TestFunctional/Status
=== RUN   TestFunctional/DNS
=== RUN   TestFunctional/Logs
=== RUN   TestFunctional/Addons
=== RUN   TestFunctional/Dashboard
=== RUN   TestFunctional/ServicesList
=== RUN   TestFunctional/Provisioning
=== RUN   TestFunctional/EnvVars
=== RUN   TestFunctional/SSH
=== RUN   TestFunctional/IngressController
--- PASS: TestFunctional (1.69s)
    --- PASS: TestFunctional/Status (0.40s)
    	cluster_status_test.go:35: Checking if cluster is healthy.
    	cluster_status_test.go:45: Component: , Healthy: True.
    	cluster_status_test.go:45: Component: , Healthy: True.
    	cluster_status_test.go:45: Component: , Healthy: True.
    --- PASS: TestFunctional/SSH (0.59s)
    --- PASS: TestFunctional/ServicesList (0.78s)
    --- PASS: TestFunctional/EnvVars (0.82s)
    --- PASS: TestFunctional/Logs (1.89s)
    --- PASS: TestFunctional/Addons (9.06s)
    --- PASS: TestFunctional/Dashboard (15.40s)
    --- PASS: TestFunctional/Provisioning (35.81s)
    	util.go:329: Error: No default StorageClass yet., Retrying in 5s. 20 Retries remaining.
    	util.go:329: Error: No default StorageClass yet., Retrying in 5s. 19 Retries remaining.
    	util.go:329: Error: No default StorageClass yet., Retrying in 5s. 18 Retries remaining.
    --- PASS: TestFunctional/DNS (45.57s)
    --- PASS: TestFunctional/IngressController (54.50s)
=== RUN   TestPersistence
--- PASS: TestPersistence (70.31s)
=== RUN   TestStartStop
--- PASS: TestStartStop (112.78s)
PASS
ok  	k8s.io/minikube/test/integration	316.073s

[minikube-bot]

Can one of the admins verify this patch?

[dlorenc]

@minikube-bot ok to test

[r2d4]

@minikube-bot retest this please

[kairen]

Hi, @r2d4 could you help me to retest this? I also tried to manually run Linux-none(kubeadm bootstrapper) testing on Ubuntu 16.04, but the test can be performed successfully.

[r2d4]

I think the test issues are with the machines themselves, I'm looking into it now that I'm back from the holidays.

This LGTM though