Kubernetes is a container orchestration tool that builds upon 15 years of experience of running production workloads at Google, combined with best-of-breed ideas and practices from the community.

Although Kubernetes is a feature-rich project, a few key features caught our attention: namespaces, (http://kubernetes.io/docs/user-guide/namespaces/) automated rollouts and rollbacks, (http://kubernetes.io/docs/user-guide/deployments/) service discovery via DNS, (http://kubernetes.io/docs/user-guide/services/) automated container scaling based on resource usage, (http://kubernetes.io/docs/user-guide/horizontal-pod-autoscaling/) and of course, the promise of a self-healing system. (http://kubernetes.io/docs/user-guide/pod-states/#container-probes)


../../_images/k8s_01.png ../../_images/k8s_02.png ../../_images/k8s_03.png

Running Kubernetes Locally via Minikube



$ curl -Lo minikube https://storage.googleapis.com/minikube/releases/v0.18.0/minikube-linux-amd64 && chmod +x minikube && sudo mv minikube /usr/local/bin/
$ minikube get-k8s-versions
$ minikube start
$ minikube start --docker-env HTTP_PROXY=""  --docker-env HTTPS_PROXY=""
$ minikube docker-env
$ eval $(minikube docker-env)
$ docker ps
$ minikube addons list
$ minikube dashboard
$ kubectl run hello-minikube --image=gcr.io/google_containers/echoserver:1.4 --port=8080
$ minikube ssh cat /var/lib/boot2docker/profile
$ minikube stop
$ minikube delete



Minikube behind a proxy

$ minikube start  --docker-env="http_proxy=" --docker-env="https_proxy=" start
$ kubectl cluster-info
# Listing the nodes in the cluster
$ kubectl get nodes
# List cluster events
$ kubectl get events
# List services that are running in the cluster
$ kubectl get services
$ kubectl get pods
$ kubectl get pods --namespace=kube-system

To start with, we will only see one service, named kubernetes . This service is the core API server, monitoring and logging services for the pods and cluster.

Even though we have not deployed any applications on Kubernetes yet, we note that there are several containers already running. The following is a brief description of each container:

  • fluentd-gcp (fluentd-elasticsearch by Elasticsearch and Kibana)
    This container collects and sends the cluster logs file to the Google Cloud Logging service.
  • kube-ui
    This is the UI that we saw earlier.
  • kube-controller-manager
    The controller manager controls a variety of cluster functions. Ensuring accurate and up-to-date replication is one of its vital roles. Additionally, it monitors, manages, and discovers new nodes. Finally, it manages and updates service endpoints.
  • kube-apiserver
    This container runs the API server. As we explored in the Swagger interface, this RESTful API allows us to create, query, update, and remove various components of our Kubernetes cluster.
  • kube-scheduler
    The scheduler takes unscheduled pods and binds them to nodes
  • etcd
    This runs the etcd software built by CoreOS. etcd is a distributed and consistent key-value store. This is where the Kubernetes cluster state is stored, updated, and retrieved by various components of K8s.
  • pause
    The Pause container is often referred to as the pod infrastructure container and is used to set up and hold the networking namespace and resource limits for each pod.
  • kube-dns
    provides the DNS and service discovery plumbing.
  • monitoring-heapster
    This is the system used to monitor resource usage across the cluster.
  • monitoring-influx-grafana
    provides the database and user interface we saw earlier for monitoring the infrastructure.
  • skydns
    This uses DNS to provide a distributed service discovery utility that works with etcd
  • kube2Sky
    This is the connector between skydns and kubernetes . Services in the API are monitored for changes and updated in skydns appropriately.
  • heapster
    This does resource usage and monitoring.
  • exechealthz
    This performs health checks on the pods.

The environment variable


$ kube-down
$ kube-up

basic scheduling service discovery health checking pods services replication controllers labels Node (formerly minions, Note that in v1.0, minion was renamed to node,)

The pods include services for DNS, logging, and pod health checks.


Pods essentially allow you to logically group containers and pieces of our application stacks together. While pods may run one or more containers inside, the pod itself may be one of many that is running on a Kubernetes (minion) node. As we’ll see, pods give us a logical group of containers that we can then replicate, schedule, and balance service endpoints across.


apiVersion: v1
kind: Pod
    name: node-js-pod
    - name: node-js-pod
      image: bitnami/apache:latest
      - containerPort: 80
$ kubectl create -f nodejs-pod.yaml
$ kubectl describe pods/node-js-pod
$ kubectl exec node-js-pod—curl <private ip address>

By default, this runs a command in the first container it finds, but you can select a specific one using the -c argument.


Labels are just simple key-value pairs. You will see them on pods, replication controllers, services, and so on. The label acts as a selector and tells Kubernetes which resources to work with for a variety of operations. Think of it as a filtering option.


Services and replication controllers give us the ability to keep our applications running with little interruption and graceful recovery.

Services allow us to abstract access away from the consumers of our applications. Using a reliable endpoint, users and other programs can access pods running on your cluster seamlessly.

K8s achieves this by making sure that every node in the cluster runs a proxy named kube- proxy. As the name suggests, kube-proxy’s job is to proxy communication from a service endpoint back to the corresponding pod that is running the actual application.

Replication controllers (RCs)

As the name suggests, manage the number of nodes that a pod and included container images run on. They ensure that an instance of an image is being run with the specific number of copies.

RCs create a high-level mechanism to make sure that things are operating correctly across the entire application and cluster. RCs are simply charged with ensuring that you have the desired scale for your application. You define the number of pod replicas you want running and give it a template for how to create new pods. Just like services, we will use selectors and labels to define a pod’s membership in a replication controller.

Kubernetes doesn’t require the strict behavior of the replication controller. In fact, version 1.1 has a job controller in beta that can be used for short lived workloads which allow jobs to be run to a completion state


apiVersion: v1
kind: ReplicationController
    name: node-js
        name: node-js
deployment: demo
    replicas: 3
        name: node-js
        deployment: demo
                name: node-js
            -   name: node-js
                image: jonbaier/node-express-info:latest
                - containerPort: 80
  • Kind
    tells K8s what type of resource we are creating. In this case, the type is ReplicationController . The kubectl script uses a single create command for all types of resources. The benefit here is that you can easily create a number of resources of various types without needing to specify individual parameters for each type. However, it requires that the definition files can identify what it is they are specifying.
  • ApiVersion
    simply tells Kubernetes which version of the schema we are using. All examples in this book will be on v1 .
  • Metadata
    is where we will give the resource a name and also specify labels that willbe used to search and select resources for a given operation. The metadata element also allows you to create annotations, which are for nonidentifying information that might be useful for client tools and libraries.
  • spec
    which will vary based on the kind or type of resource we are creating. In this case, it’s ReplicationController , which ensures the desired number of pods are running. The replicas element defines the desired number of pods, the selector tells the controller which pods to watch, and finally, the template element defines a template to launch a new pod. The template section contains the same pieces we saw in our pod definition earlier. An important thing to note is that the selector values need to match the labels values specified in the pod template. Remember that this matching is used to select the pods being managed.
$ kubectl create -f nodejs-controller.yaml
$ kubectl create -f nodejs-rc-service.yaml

A Kubernetes cluster is formed out of 2 types of resources:

Master is coordinating the cluster Nodes are where we run applications


# docker run –net=host -d gcr.io/google_containers/etcd:2.0.9 /usr/local/bin/etcd –addr= –bind-addr= –data-dir=/var/etcd/data # docker run –net=host -d -v /var/run/docker.sock:/var/run/docker.sock gcr.io/google_containers/hyperkube:v0.21.2 /hyperkube kubelet –api_servers=http://localhost:8080 –v=2 –address= –enable_server –hostname_override= –config=/etc/kubernetes/manifests # docker run -d –net=host –privileged gcr.io/google_containers/hyperkube:v0.21.2 /hyperkube proxy –master= –v=2

Install manually

$ git clone --depth 1 https://github.com/kubernetes/kubernetes.git
$ export KUBERNETES_PROVIDER=vagrant
$ export KUBE_VERSION=1.2.0
$ export FLANNEL_VERSION=0.5.0
$ export ETCD_VERSION=2.2.0
$ export K8S_VERSION=$(curl -sS https://storage.googleapis.com/kubernetes-release/release/stable.txt)
$ export K8S_VERSION=$(curl -sS https://storage.googleapis.com/kubernetes-release/release/latest.txt)

Service Discovery

There are two ways Kubernetes can implement service discovery: through environment variables and through DNS.

Install kubectl binary via curl

$ curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/amd64/kubectl
# To download a specific version
$ curl -LO https://storage.googleapis.com/kubernetes-release/release/v1.4.6/bin/linux/amd64/kubectl
$ chmod +x ./kubectl
$ sudo mv ./kubectl /usr/local/bin/kubectl


Working with kubectl

$ kubectl config view
$ kubectl cluster-info
$ kubectl get nodes
$ kubectl get events
$ kubectl get services
$ kubectl get pods
$ kubectl get pods --namespace=kube-system

Interactive K8S starting guide



$ kubectl cluster-info
# Shows all nodes that can be used to host our applications on the nodes in the cluster
$ kubectl get nodes
# Show both the client and the server versions
$ kubectl version
# Deploy our app
$ kubectl run kubernetes-bootcamp --image=docker.io/jocatalin/kubernetes-bootcamp:v1 --port=80
    deployment "kubernetes-bootcamp" created
# List our deployments
$ kubectl get deployments
    kubernetes-bootcamp   1         1         1            1           4m
$ kubectl proxy
    Starting to serve on
$ export POD_NAME=$(kubectl get pods -o go-template --template '{{range .items}}{{.metadata.name}}{{"\n"}}{{end}}')
$ echo Name of the Pod: $POD_NAME
$ kubectl get pods
    NAME                                  READY     STATUS    RESTARTS   AGE
    kubernetes-bootcamp-390780338-rpcw8   1/1       Running   0          12m


Pods are the atomic unit on the Kubernetes platform. When we create a Deployment on Kubernetes, that Deployment creates Pods with containers inside them (as opposed to creating containers directly). Each Pod is tied to the Node where it is scheduled, and remains there until termination (according to restart policy) or deletion. In case of a Node failure, identical Pods are scheduled on other available Nodes in the cluster.


A Pod always runs on a Node. A Node is a worker machine in Kubernetes and may be either a virtual or a physical machine, depending on the cluster. Each Node is managed by the Master. A Node can have multiple pods, and the Kubernetes master automatically handles scheduling the pods across the Nodes in the cluster. The Master’s automatic scheduling takes into account the available resources on each Node

# To view what containers are inside that Pod and what images are used to build those containers
$ kubectl describe pods
# Anything that the application would normally send to STDOUT becomes logs for the container within the Pod.
$ kubectl logs $POD_NAME
# We can execute commands directly on the container once the Pod is up and running.
$ kubectl exec $POD_NAME
# Start a bash session in the Pod’s container
$ kubectl exec -ti $POD_NAME bash

A Service routes traffic across a set of Pods. Services are the abstraction that allow pods to die and replicate in Kubernetes without impacting your application. Discovery and routing among dependent Pods (such as the frontend and backend components in an application) is handled by Kubernetes Services.


A Service routes traffic across a set of Pods. Services are the abstraction that allow pods to die and replicate in Kubernetes without impacting your application. Discovery and routing among dependent Pods (such as the frontend and backend components in an application) is handled by Kubernetes Services.

Services match a set of Pods using labels and selectors, a grouping primitive that allows logical operation on objects in Kubernetes. Labels are key/value pairs attached to objects and can be used in any number of ways:

Designate objects for development, test, and production

Embed version tags

Classify an object using tags

# List the current Services from our cluster
$ kubectl get services
$ kubectl expose deployment/kubernetes-bootcamp --type="NodePort" --port 8080
$ kubectl get services
$ kubectl describe services/kubernetes-bootcamp