Creating a Web Service

In this section, you will learn how to run a web server (nginx) using Kubernetes.

A Pod is the smallest deployable unit of computing that you can create and manage in Kubernetes. A Pod holds one (or more) containers. For more detailed information, refer to the Pods overview in the Kubernetes documentation.

The following steps show how to create a Pod in Kubernetes that runs a single Nginx container. We assume that you have a working cluster, and working kubectl configurations as set up in previous sections of this tutorial.

Creating a Pod

  1. Create a new Pod called webserver using a webserver container image. We are using nginx as an example here:

    $ kubectl run --generator=run-pod/v1 --image=nginx webserver
    pod/webserver created

    If you have used docker run before, this command should look familiar. It starts a Pod with an nginx image.

  2. Check that your Pod has started up and is running:

    $ kubectl get pods
    NAME        READY   STATUS              RESTARTS   AGE
    webserver   0/1     ContainerCreating   0          5s

    Initially the status will show ContainerCreating. After a few seconds you should see the status change to Running:

    $ kubectl get po
    webserver   1/1     Running   0          21s
  3. Check the details of your Pod:

    $ kubectl describe pod webserver
    Name:         webserver
    Namespace:    default
    Node:         kube-t7qvtfd34sbg-minion-0/
      Type    Reason     Age    From                                   Message
      ----    ------     ----   ----                                   -------
      Normal  Scheduled  3m10s  default-scheduler                      Successfully assigned default/webserver to kube-t7qvtfd34sbg-minion-0
      Normal  Pulling    3m8s   kubelet, kube-t7qvtfd34sbg-minion-0  Pulling image "nginx"
      Normal  Pulled     3m     kubelet, kube-t7qvtfd34sbg-minion-0  Successfully pulled image "nginx"
      Normal  Created    3m     kubelet, kube-t7qvtfd34sbg-minion-0  Created container webserver
      Normal  Started    3m     kubelet, kube-t7qvtfd34sbg-minion-0  Started container webserver
  4. Set up direct access to the Pod using the port-forward command:

    $ kubectl port-forward pod/webserver 8000
    Forwarding from -> 80
    Forwarding from [::1]:8000 -> 80

    You can now make requests to the running Pod by visiting port 8000 in a web browser on your local machine: http://localhost:8000. You should see the default “Welcome to nginx!” page.

More information

In this section we requested Kubernetes to run an nginx image. When we did this, Kubernetes performed the following (simplifed) steps:

  1. The Kubernetes master scheduled the pod to run on a free node in the cluster

  2. The node started a pod

  3. The node fetched the image from Docker Hub, based on the given image name; i.e. https://hub.docker.com/nginx

  4. The node created a container with the image and started it inside the pod

Next steps

Although the port-forward command is useful for testing out your pods, it requires access to the Kubernetes API. This is not available to the regular users who will access your service. In the next section we will set up access to your Pod from the Internet by setting up a load balancer.