π VProfile β GitOps Delivery Platform on AWS EKS
An end-to-end DevOps and GitOps implementation for building, testing, containerizing, and continuously deploying the VProfile application to Amazon EKS.
π― What This Project Is
This repository combines three focused components into one complete delivery platform:
| Component |
Responsibility |
vprofile-app |
Java application, tests, Docker build and GitHub Actions CI |
vprofile-helm |
Helm chart, Kubernetes desired state and Argo CD configuration |
vprofile-infra |
AWS infrastructure and EKS provisioning with Terraform |
The platform demonstrates a practical delivery chain:
Pull Request β Build & Test β Checkstyle β SonarQube β Quality Gate
and, after changes reach main:
Build VProfile Image β Amazon ECR β Update Helm Repository β Argo CD β Amazon EKS β AWS Load Balancer β Application
ποΈ Architecture
βββββββββββββββββββββββ
β Developer β
β Source Code
β
ββββββββββββ¬βββββββββββ
β
βΌ
βββββββββββββββββββββββ
β GitHub β
β vprofile-app β
ββββββββββββ¬βββββββββββ
β
βββββββββββββββββ΄βββββββββββββββββ
β β
Pull Request Push
to main
β β
βΌ βΌ
ββββββββββββββββββββ ββββββββββββββββββββ
β GitHub Actions β β GitHub Actions β
β Build
+ Test β β Docker Build
β
β Checkstyle β β VProfile App β
β SonarQube β ββββββββββ¬ββββββββββ
β Quality Gate β β
ββββββββββββββββββββ βΌ
ββββββββββββββββββ
β Amazon ECR β
β VProfile Image β
βββββββββ¬βββββββββ
β
βΌ
βββββββββββββββββββββββββββ
β vprofile-helm β
β β
β values.yaml image/tag β
ββββββββββββββ¬βββββββββββββ
β
β Git push
βΌ
ββββββββββββββ
β Argo CD β
β GitOps β
βββββββ¬βββββββ
β
βΌ
ββββββββββββββ
β AWS EKS β
β Kubernetes β
βββββββ¬βββββββ
β
βΌ
ββββββββββββββββββββ
β AWS Load Balancerβ
β / Kubernetes β
β Ingress β
ββββββββββ¬ββββββββββ
β
βΌ
π VProfile App
Runtime Application Components
The CI pipeline builds and publishes the VProfile application image. The supporting services are deployed from their configured container images through the Helm deployment.
AWS EKS
β
βββββββββββββββΌββββββββββββββ
β β β
βΌ βΌ βΌ
VProfile App Database RabbitMQ
(ECR image) (configured (configured
container
) container
)
β
ββββββββ¬βββββββ
βΌ
Memcached
(configured
container
)
+ application dependencies
defined by
the chart
The exact image sources, tags, replicas, services, and configuration are maintained in:
vprofile-helm/helm/
vprofile/values.yaml
π CI/CD Pipeline
The application workflow is:
vprofile-app
/.github/workflows/ci
.yml
The workflow uses two event-driven paths.
1. Pull Request β Build, Test & Quality Validation
When a pull request targets main:
Pull Request
β
βΌ
Checkout
β
βΌ
JDK
21
β
βΌ
Maven Build
+ Unit Tests
β
βΌ
Checkstyle
β
βΌ
SonarQube Analysis
β
βΌ
SonarQube Quality Gate
The workflow:
- Checks out the repository with full history.
- Sets up JDK 21.
- Caches Maven dependencies.
- Caches the Sonar scanner cache.
- Runs
mvn clean verify checkstyle:checkstyle -B.
- Runs the SonarQube scan.
- Waits for the SonarQube quality-gate result.
Evidence:
2. Push to main β Build & Publish VProfile Image
When changes are pushed to main, the delivery path builds the VProfile application image:
Push to main
β
βΌ
Checkout
β
βΌ
Configure AWS Credentials
β
βΌ
Create
ECR Repository if
Needed
β
βΌ
Build
VProfile Docker Image
β
βΌ
Tag with
7
-character
Git SHA
β
ββββββββββββββββΊ :<commit
-sha
>
β
ββββββββββββββββΊ :latest
β
βΌ
Push to
Amazon ECR
The application image is built from:
Docker-files/app/mu
ltistage/Dockerfile
The pipeline does not build separate Docker images for the database, RabbitMQ, or Memcached as part of this job. Those supporting services use their configured container images through the Kubernetes/Helm deployment.
Evidence:
π GitOps Deployment Flow
After the VProfile image is successfully pushed to ECR, the workflow updates the separate Helm repository.
VProfile image pushed to ECR
β
βΌ
Clone
vprofile-helm
β
βΌ
Install yq
β
βΌ
Update values.yaml
β
βββ app.image
βββ app.tag
β
βΌ
Commit
Helm change
β
βΌ
Push to Helm repo
β
βΌ
Argo CD
β
βΌ
Synchronize to EKS
The workflow updates:
app:
image:
<ECR image>
tag:
<7
-character Git SHA>
in:
vprofile-helm/helm/
vprofile/values.yaml
This creates a clear separation:
Application Repository
β
β build
βΌ
Amazon ECR
β
β image reference
βΌ
Helm / GitOps Repository
β
β desired state
βΌ
Argo CD
β
βΌ
AWS EKS
Evidence:
βΈοΈ Kubernetes & Helm
The Kubernetes delivery configuration lives in:
vprofile-helm/
The main Helm chart is:
vprofile-helm/helm/
vprofile/
It contains templates for the application stack, including:
- Application Deployment
- Database Deployment
- RabbitMQ Deployment
- Memcached Deployment
- Services
- PersistentVolumeClaim
- Secrets
- Ingress
- Docker registry authentication
The repository also retains Kubernetes definitions under:
vprofile-helm/kubedefs/
This makes the repository useful both as a record of the Kubernetes resources and as the Helm-based desired-state source used for GitOps delivery.
π Argo CD
Argo CD provides the GitOps deployment layer.
Git
β
β desired Kubernetes state
βΌ
Argo CD
β
β sync
βΌ
Amazon EKS
β
βΌ
VProfile workloads
Argo CD-related configuration is maintained under:
vprofile-helm/argocd/
vprofile-helm/apps/
vprofile-helm/projects/
Argo CD watches the configured Git source and applies the desired state to the Kubernetes cluster.
ποΈ AWS Infrastructure with Terraform
The AWS infrastructure is maintained in:
vprofile-infra/
Terraform configuration includes:
vprofile-infra/
βββ main.tf
βββ variables.tf
βββ outputs.tf
βββ backend.tf
βββ iam_policy.json
The repository also contains:
argocd-ingress
.yaml
for supporting Kubernetes/Argo CD ingress configuration.
The infrastructure layer provides the AWS/EKS foundation on which the GitOps deployment runs.
β‘οΈ vprofile-infra
π Ingress & Application Access
The application is exposed through Kubernetes Ingress and the AWS Load Balancer integration.
Internet
β
βΌ
AWS
Load Balancer
β
βΌ
Kubernetes
Ingress
β
βΌ
VProfile
Service
β
βΌ
VProfile
Pod
Evidence:
Final application verification:
π SonarQube Server Setup
The project includes:
resources/scripts/sonar-setup
.sh
This is a reusable setup script for installing and configuring the SonarQube server on an Ubuntu EC2 instance used by the CI quality-analysis workflow.
The script handles the server-side setup required for the SonarQube environment, including Java, PostgreSQL, SonarQube configuration, systemd service configuration, and Nginx reverse proxy configuration.
It is intentionally kept as an implementation artifact so the CI quality-analysis environment is reproducible.
π€ AI-Assisted Implementation Prompts
The project also contains a reusable prompt library:
resources/prompts/
These prompts document how AI assistance was used during implementation to generate or transform specific infrastructure and GitOps components.
| Prompt |
Purpose |
01-terraform-eks.md |
Generate Terraform configuration for AWS/EKS infrastructure |
02-ebs-csi-driver.md |
Configure the AWS EBS CSI Driver for Kubernetes storage |
03-helm-chart.md |
Transform Kubernetes definitions into a reusable Helm chart |
04-argocd-and-ingress.md |
Generate Argo CD and Ingress configuration |
05-argocd-vprofile-app.md |
Generate the Argo CD application configuration for VProfile |
AI-Assisted Workflow
Requirements / Existing
Configuration
β
βΌ
AI Implementation Prompt
β
βΌ
Generated Configuration
β
βΌ
Review & Modify
β
βΌ
Validate / Test
β
βΌ
Deploy
The prompts are provided as reproducible implementation references. Generated output was reviewed, modified where necessary, validated, and integrated into the project.
π Repository Structure
GitOps-Argo/
β
βββ README.md
βββ SECURITY.md
β
βββ resources/
β βββ commands/
β β βββ 01
-install-tools.md
β β βββ 02
-terraform-bootstrap.md
β β βββ 03
-aws-eks-and-kubeconfig.md
β β βββ 04
-alb-cert-manager.md
β β βββ 05
-argocd.md
β β βββ 06
-vprofile-verification.md
β β
β βββ prompts/
β β βββ 01
-terraform-eks.md
β β βββ 02
-ebs-csi-driver.md
β β βββ 03
-helm-chart.md
β β βββ 04
-argocd-and-ingress.md
β β βββ 05
-argocd-vprofile-app.md
β β
β βββ screenshots/
β βββ scripts/
β β βββ sonar-setup.sh
β βββ variables/
β βββ github-variables-and-secrets.example
.md
β
βββ vprofile-app/
β βββ .github/workflows/
β β βββ ci.yml
β βββ Docker-files/
β βββ src/
β βββ pom.xml
β βββ sonar-project.properties
β
βββ vprofile-helm/
β βββ apps/
β βββ argocd/
β βββ helm/
β β βββ vprofile/
β βββ kubedefs/
β βββ projects/
β
βββ vprofile-infra/
βββ main.tf
βββ variables.tf
βββ outputs.tf
βββ backend.tf
βββ iam_policy.json
βββ argocd-ingress.yaml
π§ Where to Start
If you are reviewing this project for the first time, follow this path:
01 β Understand the architecture
Start with:
resources/screenshots/
01
-architecture.png
Then read this README.
02 β Review the application CI workflow
Open:
vprofile-app
/.github/workflows/ci
.yml
Pay particular attention to:
- Pull request quality checks
- Maven build and tests
- Checkstyle
- SonarQube
- Docker image build
- Amazon ECR publishing
- Helm repository update
03 β Review the infrastructure
Open:
vprofile-infra/
Start with:
main.tf
variables.tf
outputs.tf
04 β Review Kubernetes and Helm
Open:
vprofile-helm/
Start with:
helm/vprofile/
values.yaml
helm/vprofile/
templates/
Then inspect:
kubedefs/
argocd/
apps/
projects/
05 β Follow the GitOps path
Trace the image reference from:
GitHub Actions
β
Amazon ECR
β
vprofile-helm/values.yaml
β
Argo CD
β
EKS
06 β Review deployment evidence
Open:
resources/screenshots/
The screenshots provide visual evidence of the major implementation stages.
π§° Resources
Commands
resources/commands/
Sequential command references covering:
- Tool installation
- Terraform bootstrap
- AWS EKS and kubeconfig
- ALB and certificate configuration
- Argo CD setup
- VProfile verification
Prompts
resources/prompts/
Reusable AI-assisted implementation prompts for the infrastructure and GitOps components.
Screenshots
resources/screenshots/
Implementation and deployment evidence.
Scripts
resources/scripts/
Reusable automation scripts, including the SonarQube EC2 setup script.
Variables
resources/variables/
Reference configuration for GitHub Actions variables and secrets.
π§ͺ Verification
The following checks can be used to verify the deployed environment.
terraform plan
terraform apply
Kubernetes
kubectl
get
nodes
kubectl
get
pods -A
kubectl
get
svc
-A
kubectl
get
ingress -A
Helm
helm lint ./vprofile-helm/
helm/vprofile
Argo CD
Verify that:
- The VProfile application exists.
- The Git repository is connected.
- The application is synchronized.
- The application health is healthy.
- The expected Kubernetes resources are running.
Application
Open the configured application endpoint through the Load Balancer/Ingress and verify the VProfile application.
πΈ Deployment Evidence
| # |
Evidence |
| 01 |
Architecture |
| 02 |
EKS cluster |
| 03 |
SonarQube quality gate |
| 04 |
Build and test |
| 05 |
Amazon ECR images |
| 06 |
ECR image + Helm update |
| 07 |
Argo CD dashboard |
| 08 |
Image tag update |
| 09 |
AWS Load Balancer |
| 10 |
VProfile application |
All screenshots are stored in:
resources/screenshots/
π― What This Project Demonstrates
This project demonstrates practical implementation of:
- Infrastructure as Code with Terraform
- Amazon EKS
- AWS networking and IAM
- Docker containerization
- Amazon ECR
- GitHub Actions
- Maven builds and automated tests
- Checkstyle
- SonarQube analysis
- SonarQube quality gates
- Kubernetes
- Helm
- Kubernetes Services
- Ingress
- Persistent storage
- Kubernetes Secrets
- Argo CD
- GitOps
- Multi-tier application deployment
- CI-driven container delivery
- Git-based Kubernetes desired state
- Operational verification and troubleshooting
π Security
Security guidance is available in:
β‘οΈ SECURITY.md
For production use:
- Replace demonstration credentials.
- Store sensitive values in an appropriate secret-management system.
- Apply least-privilege IAM policies.
- Restrict AWS Security Group access.
- Configure HTTPS/TLS appropriately.
- Never commit credentials, access keys, tokens, or private keys.
- Rotate credentials when required.
𧩠Separation of Responsibilities
The project intentionally separates application, deployment, and infrastructure responsibilities.
ββββββββββββββββββββββββββββββββββββ
β vprofile-app β
β β
β Java
+ Maven + Tests + CI β
β Docker Build
+ ECR Publishing β
ββββββββββββββββββ¬ββββββββββββββββββ
β
β Image
βΌ
ββββββββββββββββ
β Amazon ECR β
ββββββββ¬ββββββββ
β
β Image reference
βΌ
ββββββββββββββββββββββββββββββββββββ
β vprofile-helm β
β β
β Helm + Kubernetes + Argo CD β
β Git-based
Desired State β
ββββββββββββββββββ¬ββββββββββββββββββ
β
β Sync
βΌ
βββββββββββββ
β Argo CD β
βββββββ¬ββββββ
β
βΌ
βββββββββββββ
β AWS EKS β
βββββββββββββ
ββββββββββββββββββββββββββββββββββββ
β vprofile-infra β
β β
β Terraform + AWS Infrastructure β
ββββββββββββββββββββββββββββββββββββ
This separation makes the platform easier to understand, maintain, and evolve.
β End-to-End Summary
ββββββββββββββββ
β GitHub β
β vprofile-app β
ββββββββ¬ββββββββ
β
βββ Pull Request βββΊ Build
+ Test + Checkstyle
β β
β βΌ
β SonarQube
β β
β βΌ
β Quality Gate
β
βββ Push
to main βββΊ Docker Build
β
βΌ
Amazon ECR
β
βΌ
Update Helm values
β
βΌ
Git push
β
βΌ
Argo CD
β
βΌ
AWS EKS
β
βΌ
AWS Load Balancer
β
βΌ
VProfile App
Core Delivery Model
Application CI β Container Registry β Git-based Desired State β Argo CD β Kubernetes
π Quick Navigation
Infrastructure as Code β Continuous Integration β Container Registry β GitOps β Kubernetes β Application Delivery
