- Voting App Deployment Documentation
This document provides a comprehensive guide on deploying the Example Voting App using Kubernetes, Helm charts, GitHub Actions, Argo CD, and Terraform. It covers the setup of CI/CD pipelines for both application deployment and infrastructure management.
- AWS account with access to EKS, S3, ECR, and IAM services.
- Experience with Docker and GitHub Actions for CI/CD pipelines.
- Understanding of microservices architecture and deployment on EKS.
- Proficiency in Infrastructure as Code (IaC) using Terraform and Helm charting.
- Familiarity with DevSecOps practices.
Utilized Terraform scripts to provision an AWS EKS cluster with an initial node count of 1, scaling up to a maximum of 5 nodes. Configured networking to allow traffic on ports 31000 (vote web page) and 31001 (result web page) via security group rules. Ensured security by implementing relevant firewall rules and IP restrictions.
Click here for Infrastructure Deployment Details
Developed Helm chart templates for microservices:
hpa.yaml: Horizontal Pod Autoscaler configuration.deployment.yaml: Deployment configurations for vote, result, and worker microservices.service.yaml: Service definitions for exposing microservices internally.cronjob.yaml: Scheduled jobs configuration.
Created values files to customize deployment configurations for different environments and scaling requirements.
Click here for more details about the helm charts Click here for details about the GitOps Deployment pipeline
Implemented GitHub Actions pipeline for:
- Building Docker images for vote, result, and worker microservices.
- Automated scaling script for vote microservice during weekdays.
Integrated Argo CD for continuous deployment of Helm charts to the EKS cluster upon updates to the Helm chart repository.
Established a CI/CD pipeline for managing infrastructure changes using Terraform scripts. Automated application of Terraform changes to the AWS infrastructure upon code updates. Click here for to see pipeline script
To ensure the application handles varying traffic patterns effectively, I have implemented a scaling strategy for the vote microservice. This includes an automated script that adjusts the service's replicas during weekdays to accommodate higher traffic volumes.
-
Automated Scaling Script: A GitHub Actions pipeline includes a script that automatically scales the vote microservice based on traffic patterns. The script increases the number of replicas during weekdays to manage higher traffic and scales down during weekends.
-
Horizontal Pod Autoscaler (HPA): Configured in the Helm charts (hpa.yaml) to adjust the number of pods based on CPU utilization and other relevant metrics.
Click here to view Scaling python script
Click the links to visit the application
This documentation outlines the deployment process of the Voting App, leveraging AWS infrastructure, Kubernetes, Azure Storage, Helm charts, and CI/CD pipelines. It covers the setup and configuration required for deploying and maintaining a scalable and resilient application architecture in a cloud environment.