If you haven’t already, check out my previous post Kubernetes: Getting Started with Minikube.

helloworld — pod, service

Create a folder and cd into it

mkdir helloworld-example-01; cd helloworld-example-01

Create pod.yaml

Creates 1 pod

vi pod.yaml
---
apiVersion: v1
kind: Pod
metadata:
  name: itsmetommy
  labels:
    app: helloworld
spec:
  containers:
  - name: itsmetommy-httpd
    image: httpd
    ports:
    - name: web-port
      containerPort: 80

Create service.yaml

Creates 1 service

vi service.yaml
---
apiVersion: v1
kind: Service
metadata:
  name: itsmetommy-service
spec:
  ports:
  - port: 31000
    nodePort: 31000
    targetPort: web-port
    protocol: TCP
  selector:
    app: helloworld
  type: NodePort

Create your pod and service

The following command looks for all yaml files within the current directory and runs them.

kubectl create -f .

Example

kubectl create -f .
pod "itsmetommy" created
service "itsmetommy-service" created

View

curl "$(minikube ip):31000"

Example

curl "$(minikube ip):31000"
<html><body><h1>It works!</h1></body></html>

OR use minikube service.

Get services.

kubectl get services

Example

kubectl get services
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
itsmetommy-service NodePort 10.0.0.206 <none> 31000:31000/TCP 1m
kubernetes ClusterIP 10.0.0.1 <none> 443/TCP 39d

Launch browser.

minikube service itsmetommy-service

Example

minikube service itsmetommy-service
Opening kubernetes service default/itsmetommy-service in default browser...

Clean up

kubectl delete -f .

Example

kubectl delete -f .
pod "itsmetommy.com" deleted
service "itsmetommy-service" deleted

helloworld — deployment (replica set), pod, service

Note: Deployments manage Replica Sets and Replica Sets manage Pods.

Create a folder and cd into it

mkdir helloworld-example-02; cd helloworld-example-02

Create deployment.yaml

Creates 1 deployment
Creates 3 pods

vi deployment.yaml
---
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: itsmetommy-deployment
spec:
  replicas: 3
  template:
    metadata:
      labels:
        app: helloworld
    spec:
      containers:
      - name: itsmetommy-httpd
        image: httpd
        ports:
        - name: web-port
          containerPort: 80

Create service.yaml

Creates 1 service

vi service.yaml
---
apiVersion: v1
kind: Service
metadata:
  name: itsmetommy-service
spec:
  ports:
  - port: 32000
    nodePort: 32000
    targetPort: web-port
    protocol: TCP
  selector:
    app: helloworld
  type: NodePort

Create your deployment and service

The following command looks for all yaml files within the current directory and runs them.

kubectl create -f .

Example

kubectl create -f .
deployment "itsmetommy-deployment" created
service "itsmetommy-service" created

View

curl "$(minikube ip):32000"

Example

curl "$(minikube ip):32000"
<html><body><h1>It works!</h1></body></html>

OR use minikube service.

Get services.

kubectl get services

Example

kubectl get services
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
itsmetommy-service NodePort 10.0.0.206 <none> 31000:31000/TCP 1m
kubernetes ClusterIP 10.0.0.1 <none> 443/TCP 39d

Launch browser.

minikube service itsmetommy-service

Example

minikube service itsmetommy-service
Opening kubernetes service default/itsmetommy-service in default browser...

Clean up

kubectl delete -f .

Example

kubectl delete -f .
pod "itsmetommy.com" deleted
service "itsmetommy-service" deleted

web — HorizontalPodAutoscaler / Autoscaling

The Horizontal Pod Autoscaler automatically scales the number of pods in a replication controller, deployment or replica set based on observed CPU utilization (or, with custom metrics support, on some other application-provided metrics).

Enable metrics

minikube addons enable heapster

cAdvisor web UI

Analyzes resource usage and performance characteristics of running containers.

http://$(minikube ip):4194/containers/

Example

http://192.168.99.100:4194/containers/

open http://$(minikube ip):4194/containers/

Grafana Dashboard

Data visualization & Monitoring with support for Graphite, InfluxDB, Prometheus, Elasticsearch and many more databases.

minikube addons open heapster

Create pod

kubectl run web --image=k8s.gcr.io/hpa-example --requests=cpu=200m --port=80

Create deployment

kubectl expose deployment web --type=NodePort

Create Horizontal Pod Autoscaler

kubectl autoscale deployment web --min=2 --max=10 --cpu-percent=40

Watch

Install watch.

brew install watch

Watch hpa, pods and services.

watch kubectl get hpa,pods,services

Add load

Using your minikube IP along with the service port of web, we can write a while loop that sends http requests to the service, driving the CPU up.

Get your minikube IP.

minikube dashboard --url

Example

minikube dashboard --url
http://192.168.99.100:30000

Get your service port.

kubectl get service web

Example

kubectl get web
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
web NodePort 10.0.0.101 <none> 80:31663/TCP 17m

Run the while loop.

while true; do curl -s http://192.168.99.100:31663; done

You should now see pods spinning up automatically based on load.

Stop the while loop and watch the pods terminate back down to one.

Clean up

kubectl delete deployment,hpa,service web

helloworld — HorizontalPodAutoscaler / Autoscaling

Now that we’ve learned how to manually create a Horizontal Pod Autoscaler, let’s use YAML files this time.

If you followed along, you should already have metrics enabled. I’ll add the step here again just in case.

Enable metrics

minikube addons enable heapster

cAdvisor web UI

Analyzes resource usage and performance characteristics of running containers.

http://$(minikube ip):4194/containers/

Example

http://192.168.99.100:4194/containers/

open http://$(minikube ip):4194/containers/

Grafana Dashboard

Data visualization & Monitoring with support for Graphite, InfluxDB, Prometheus, Elasticsearch and many more databases.

minikube addons open heapster

Create a folder and cd into it

mkdir helloworld-example-03; cd helloworld-example-03

Create deployment.yaml

Creates 1 deployment

vi deployment.yaml
---
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: itsmetommy-deployment
spec:
  template:
    metadata:
      labels:
        app: helloworld
    spec:
      containers:
      - name: itsmetommy-httpd
        image: httpd
        ports:
        - name: web-port
          containerPort: 80
        resources:
          requests:
            cpu: 100m

Create autoscaler.yaml

Creates 1 hpa

I set the targetCPUUtilizationPercentage to 5 in order to get a faster autoscale.

vi autoscaler.yaml
---
apiVersion: autoscaling/v2beta1
kind: HorizontalPodAutoscaler
metadata:
  name: itsmetommy
  namespace: default
spec:
  scaleTargetRef:
    apiVersion: apps/v1beta1
    kind: Deployment
    name: itsmetommy-deployment
  minReplicas: 1
  maxReplicas: 10
  targetCPUUtilizationPercentage: 5

Create service.yaml

Creates 1 service

vi service.yaml
---
apiVersion: v1
kind: Service
metadata:
  name: itsmetommy-service
spec:
  ports:
  - port: 32500
    nodePort: 32500
    targetPort: web-port
    protocol: TCP
  selector:
    app: helloworld
  type: NodePort

Create your deployment, service and autoscaler

kubectl create -f .

View

View your application.

curl "$(minikube ip):32500"

Example

curl "$(minikube ip):32500"
<html><body><h1>It works!</h1></body></html>

OR use minikube service.

Get services.

kubectl get services

Example

kubectl get services
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
itsmetommy-service NodePort 10.0.0.247 <none> 32500:32500/TCP 38s
kubernetes ClusterIP 10.0.0.1 <none> 443/TCP 40d

Launch browser.

minikube service itsmetommy-service

Example

minikube service itsmetommy-service
Opening kubernetes service default/itsmetommy-service in default browser...

Watch

Watch hpa (Horizontal Pod Autoscaler), pods and services.

watch kubectl get hpa,pods,services

Add load

Using your minikube IP along with the service port of itsmetommy-service, we can write a while loop that sends http requests to the service, driving the CPU up.

Get your minikube IP.

minikube dashboard --url

Example

minikube dashboard --url
http://192.168.99.100:30000

Get your service port.

kubectl get service itsmetommy-service

Example

kubectl get services
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
itsmetommy-service NodePort 10.0.0.247 <none> 32500:32500/TCP 38s
kubernetes ClusterIP 10.0.0.1 <none> 443/TCP 40d

Run the while loop.

while true; do curl -s http://192.168.99.100:32500; done

Make sure to check out the watched pods within your terminal as the pods spin up.

Stop the while loop and watch the pods spin down.

Clean up

kubectl delete -f .

It's Me, Tommy

Kubernetes: Learning with Minikube

helloworld — pod, service

helloworld — deployment (replica set), pod, service

web — HorizontalPodAutoscaler / Autoscaling

helloworld — HorizontalPodAutoscaler / Autoscaling