Kubernetes Project

This repository contains the implementation of a RESTful API in Node.js using the Express framework. The API is containerized into a Docker image, with implementation details in the Dockerfile.

Navigating to the Docker Image

The Docker image is hosted on Docker Hub and can be found here.

Running Docker Container

Assumption: Docker must be installed on your system.

To run the project as a Docker container:

• Pull the Image:

  docker pull sareenv/test-repo:1

• Run the Container:

  docker run -p 8080:8080 -d test-repo

  The -p option maps the port, and -d runs it in detached mode.

Kubernetes Configuration Files

All Kubernetes specifications are in the k8s folder with .yaml extensions. Use the following command to apply these configurations:

kubectl apply -f file_name

The mongo-express folder also contains YAML configurations for learning purposes and is not directly linked to our project. This directory will be removed soon.

Google Kubernetes Engine (GKE)

We use GCP (Google Cloud Platform) tools for deploying node clusters, taking advantage of the $300 free credit available for 90 days. The Google Cloud command-line interface is used for managing the cluster.

• Install the CLI Tool:

  docker pull google/cloud-sdk
  docker run -it google/cloud-sdk:latest

  The above command runs the container in interactive mode.
• Authenticate to Your Project:

  gcloud auth login

  Follow the generated link to log in to your Google account.
• Create and Configure the Project:

  gcloud projects create gke-project-comp6231
  gcloud config set project gke-project-comp6231

• Select Region and Create the Cluster:

  gcloud container get-server-config --zone asia-east2-c

  gcloud container clusters create gke-project-comp6231 \
  --cluster-version 1.18.16-gke.500 \
  --disk-size 200 \
  --num-nodes 1 \
  --machine-type e2-small \
  --no-enable-cloud-logging \
  --no-enable-cloud-monitoring \
  --zone asia-east2-c

• Delete the Cluster After Testing:

  gcloud container clusters delete gke-project-comp6231 --zone asia-east2-c

Manual MongoDB ReplicaSet Scaling

kubectl scale statefulsets mongo --replicas=4

AutoScaling & Load Testing

We use HPA (Horizontal Pod Autoscaler) in Kubernetes:

kubectl autoscale deployment app-deployment --cpu-percent=60 --min=3 --max=10

For load testing, we use the Autocannon library from npm (Autocannon npm package). After installation, the following command sends traffic to the Node.js API's load-balancing service:

autocannon -c 2000 -d 10 -p 200 EXTERNAL_IP

Authors

This Kubernetes project is developed for the COMP6231 module and maintained by:

• Vinayak Sareen
• Nadip
• Tanzia
• Protim