Skip to content

Latest commit

 

History

History
132 lines (109 loc) · 3.33 KB

spring-k8s.md

File metadata and controls

132 lines (109 loc) · 3.33 KB

Setup Kubernetes in Windows & run Spring-Boot application on k8s cluster

  1. Start minikube with docker as a driver
minikube start --driver=docker

  1. To verify if minikube is started successfully or not
minikube status

  1. To verify that a single node cluster has been initiated
kubectl cluster-info

  1. To see that single node residing inside k8s cluster
kubectl get nodes

  1. To allow kubernetes to read my local docker repositories
minikube docker-env

  1. To view the current local docker images
docker images

  1. Move to the directory containing spring-boot application and build the docker image
docker build -t spring-boot-k8s:v0.0.1 .

docker images

  1. We want to run the built image inside the k8s-pod, for that we need a dployment object (dployment objects are k8s objects, which are used for managing the pods). So create a deployment object by running-
kubectl create deployment spring-boot-k8s-deploy --image spring-boot-k8s:v0.0.1 --port=8080

We're creating the D.O. that reads the docker image created previously by us and exposes port 8080


  1. To verify the deployment
kubectl get deployments

  1. To get description about the created deployment
kubectl describe deployment spring-boot-k8s-deploy

  1. To ensure that kubernetes successfully pulled the docker images and is able to run inside pods
kubectl get pods

  1. To verify that kubernetes is able to run a conatiner inside the above pod, we'll check the logs of the pod-
kubectl logs spring-boot-k8s-deploy-66d8b9d9fc-g8t2w

If the spring-boot application runs that means it is running inside the container residing in the pod.


  1. Now we need to create a service object and then we need to expose the current deployment with correct service tag, so that the application can be accessed from outside of this k8s-cluster (basically we need to expose our deployment to create a service object). Significance: All the traffics will come to the service first, then service will redirect it to the corresponding pods based on availability.
kubectl expose deployment spring-boot-k8s-deploy --type=NodePort

  1. Type the following to verify that the service object is successfully created-
kubectl get services

  1. Since we have one pod, we can directly get that service URL to point to that particular endpoint-
minikube service spring-boot-k8s-deploy --url

By doing this we will directly get the service url/proxy url to access our endpoint.


  1. We can also check health/status of each components (nodes, pods, deployments, proxy etc) of our k8s-cluster.
minikube dashboard

  1. After ending the above session, clean-up all the configs and components which we created (not mandatory but it's good to do the clean-up).
kubectl delete service spring-boot-k8s-deploy

kubectl delete deployment spring-boot-k8s-deploy

  1. Stop the minikube instance and then delete the minikube object
minikube stop

  1. Verify if it is stopped or not-
kubectl get nodes

  1. Delete the local kubernetes cluster.
minikube delete