Project: KubeShip – GitOps Platform with ArgoCD and EKS

What Is KubeShip?

KubeShip is a real-world GitOps platform that automates the deployment and management of containerized applications using:

• Amazon EKS for Kubernetes orchestration
• ArgoCD for GitOps-based continuous delivery
• Terraform for infrastructure as code
• Helm for Kubernetes packaging
• GitHub Actions for CI pipelines
• React + FastAPI for microservices
• PostgreSQL + Redis for backend storage

This project shows how modern teams deploy applications at scale with automation and best DevOps practices.

---

Why KubeShip?

Because deploying microservices manually doesn’t scale. You need:

• Declarative infrastructure (Terraform)
• Automated provisioning (EKS)
• GitOps deployment model (ArgoCD)
• CI/CD pipelines (GitHub Actions)
• Monitoring and scalability

KubeShip combines all of these in one integrated stack.

---

Architecture Overview

Core Components

Layer	Tool
Infra Provision	Terraform
Cluster	Amazon EKS
GitOps CD	ArgoCD
Helm Packaging	Helm
CI/CD	GitHub Actions
Container Registry	Amazon ECR
Microservices	FastAPI, React, PostgreSQL
Monitoring	Prometheus + Grafana

---

Architecture Diagram

                        ┌──────────────────────────────┐
                        │     GitHub Repo (IaC + App)  │
                        └────────────┬─────────────────┘
                                     │ Push
                          ┌──────────▼──────────┐
                          │    GitHub Actions   │────────────┐
                          └──────────┬──────────┘            │
                                     │ Image Build/Push      │
                          ┌──────────▼──────────┐            │
                          │    Amazon ECR       │            │
                          └─────────────────────┘            │
                                                             ▼
                                               ┌────────────────────────┐
                                               │   Amazon EKS Cluster   │
                                               │   (Provisioned via TF) │
                                               └──────────┬─────────────┘
                                                          │
                                                ┌─────────▼─────────┐
                                                │     ArgoCD        │
                                                │  (GitOps CD Tool) │
                                                └──────┬────────────┘
                                                       │
                                           ┌───────────▼────────────┐
                                           │   Helm Deployments      │
                                           │  (Frontend, API, Redis) │
                                           └─────────────────────────┘

---

Project Structure

kubeship/
├── terraform/              # VPC + EKS via Terraform
├── microservices/          # React + FastAPI + DB
│   ├── api-gateway/
│   ├── auth-service/
│   └── frontend/
├── helm-charts/            # Helm charts per service
├── manifests/              # ArgoCD apps and projects
└── .github/workflows/      # CI build and push pipelines

---

Infrastructure Provisioning with Terraform

Provision:

• VPC with public and private subnets
• NAT Gateway for outbound traffic
• Amazon EKS cluster
• IAM roles and node groups

Modules used:

• terraform-aws-modules/vpc/aws
• terraform-aws-modules/eks/aws

Command:

terraform init
terraform apply

---

GitOps with ArgoCD

ArgoCD tracks:

• Git repo for Helm chart changes
• Helm charts from helm-charts/
• Syncs them to EKS cluster

To access ArgoCD:

kubectl port-forward svc/argocd-server -n argocd 8080:443

Open: https://localhost:8080

---

CI/CD with GitHub Actions

• Builds Docker images
• Pushes to ECR
• ArgoCD picks up and syncs latest deployment

CI defined in: .github/workflows/deploy.yaml

---

Testing and Observability

Post-deployment:

• Validate apps in ArgoCD UI
• kubectl get pods -A for health checks
• Test endpoints using ALB DNS

Monitoring with Prometheus + Grafana (optional):

helm install monitoring prometheus-community/kube-prometheus-stack -n monitoring --create-namespace

---

✅ Real-World Problem Solving

Concern	KubeShip Solution
Scalable Deployments	GitOps + ArgoCD
CI/CD	GitHub Actions + ECR
Infra as Code	Terraform
Secrets + Networking	EKS best practices
Observability	Prometheus + Grafana

---

What’s Next?

• Add custom domain with cert-manager + Route 53
• Use Sealed Secrets or External Secrets
• Setup staging and prod ArgoCD environments
• Enable autoscaling (HPA)

---

Summary

KubeShip lets you:

✅ Learn GitOps & EKS
✅ Use Terraform professionally
✅ Automate deployments
✅ Build full-stack apps in Kubernetes

---

Author

Built with ❤️ by Celestine — GitHub