Docker Cookbook

5.9 Creating a Single-Node Kubernetes Cluster Using Docker Compose | 151

---

That is it. You now have a one-node Kubernetes 
cluster
with all components running
as containers; 
get nodes
returns your localhost, and you can create pods, replication
controllers, and services:
$ ./kubectl get nodes
NAME LABELS STATUS
127.0.0.1 Ready
To test that you can create a new pod, you are going to run a single Nginx container:
$ ./kubectl run-container nginx --image=nginx --port=80
CONTROLLER CONTAINER(S) IMAGE(S) SELECTOR REPLICAS
nginx nginx nginx run-container=nginx 1
The 
run-container
command automatically creates a replication
controller for this container. List it with 
./kubectl get rc
.
To access this 
nginx
frontend from outside the cluster, you need to expose it as a ser‐
vice. However, when creating the service, you pass the host-only network IP of the
virtual machine to the 
kubectl
command. Otherwise, the service will be created, and
all future pods will be able to access it, but we will not be able to reach it from outside
the cluster:
$ ./kubectl expose rc nginx --port=80 --public-ip=192.168.33.10
NAME LABELS SELECTOR IP PORT
nginx run-container=nginx 10.0.0.98 80
Once the 
nginx
image is downloaded, the pod will enter running state and you will be
able to access the Nginx welcome page at 
http://192.168.33.10
:
$ ./kubectl get pods
POD IP CONTAINER(S) IMAGE(S) ...
nginx-127 controller-manager gcr.io/google_containers/...
apiserver gcr.io/google_containers/...
scheduler gcr.io/google_containers/...
nginx-461yi 172.17.0.6 nginx nginx ...
Discussion
The 
k8s.yml
Compose file shows how this was done:
etcd:
image: kubernetes/etcd:2.0.5.1
net: 
"host"
command: /usr/local/bin/etcd --addr=127.0.0.1:4001 --bind-addr=0.0.0.0:4001 \
--data-dir=/var/etcd/data
master:
image: gcr.io/google_containers/hyperkube:v0.14.1