Cloud Native DevOps Platform

A production-style Cloud Native DevOps platform demonstrating end-to-end CI/CD and GitOps using GitHub Actions, Docker, ArgoCD, and Kubernetes — with infrastructure provisioning via Terraform and observability via Prometheus & Grafana.

Code is automatically built, containerized, pushed to DockerHub, and deployed to Kubernetes through GitOps — no manual steps.

---

Architecture Overview

Developer pushes code
        │
        ▼
  GitHub Actions (CI)
        │
   ┌────┴────┐
   │  Docker  │  Build → Tag (Git SHA) → Push to DockerHub
   └────┬────┘
        │
        ▼
ArgoCD Image Updater
        │   Detects new image tag, updates K8s manifest
        ▼
   ArgoCD Sync
        │   Reconciles cluster state with Git
        ▼
Kubernetes Cluster
  ┌─────────────────────────────────┐
  │  Ingress → Service → Deployment │
  │           └── ReplicaSet        │
  │                └── Pods         │
  └─────────────────────────────────┘
        │
        ▼
  Prometheus + Grafana (Monitoring)

---

Tech Stack

Category	Tool
Version Control	Git, GitHub
CI Pipeline	GitHub Actions
Containerization	Docker
Container Registry	DockerHub
GitOps	ArgoCD + ArgoCD Image Updater
Orchestration	Kubernetes
Infrastructure as Code	Terraform (HCL)
Monitoring & Observability	Prometheus, Grafana
Application	Python (backend)

Language breakdown: Python 40% · HCL (Terraform) 31% · Dockerfile 29%

---

Project Structure

cloud-native-devops-platform/
├── .github/
│   └── workflows/
│       └── ci.yml              # GitHub Actions CI pipeline
├── app/
│   └── backend/
│       ├── Dockerfile          # Container image definition
│       └── ...                 # Python application code
├── kubernetes/
│   ├── deployment.yaml         # App deployment & ReplicaSet
│   ├── service.yaml            # ClusterIP / LoadBalancer
│   └── ingress.yaml            # Ingress routing rules
├── gitops/                     # ArgoCD application manifests
├── terraform/                  # Infrastructure provisioning
├── monitoring/
│   ├── prometheus/             # Scrape configs & rules
│   └── grafana/                # Dashboard definitions
├── docs/                       # Additional documentation
├── docker-compose.yml          # Local development setup
└── README.md

---

Getting Started

Prerequisites

• Docker & Docker Compose
• kubectl configured against a Kubernetes cluster
• Terraform ≥ 1.0
• ArgoCD installed on your cluster
• DockerHub account

1. Local Development

git clone https://github.com/dvanhu/cloud-native-devops-platform.git
cd cloud-native-devops-platform
docker-compose up --build

2. Provision Infrastructure (Terraform)

cd terraform
terraform init
terraform plan
terraform apply

3. Deploy to Kubernetes (Manual)

kubectl apply -f kubernetes/deployment.yaml
kubectl apply -f kubernetes/service.yaml
kubectl apply -f kubernetes/ingress.yaml

4. Set Up ArgoCD

# Install ArgoCD
kubectl create namespace argocd
kubectl apply -n argocd -f https://raw.githubusercontent.com/argoproj/argo-cd/stable/manifests/install.yaml

# Apply the GitOps app definition
kubectl apply -f gitops/

ArgoCD will then continuously sync the cluster state with this repository.

---

CI/CD Pipeline

Continuous Integration — GitHub Actions (.github/workflows/ci.yml)

Triggered on every push to main:

1. Checkout repository
2. Login to DockerHub
3. Build Docker image
4. Tag image with Git commit SHA
5. Push image to DockerHub → dvanhu/devops-backend:<commit-sha>

Continuous Deployment — GitOps with ArgoCD

New image pushed to DockerHub
          │
          ▼
ArgoCD Image Updater detects new tag
          │
          ▼
Auto-updates Kubernetes manifest in Git
          │
          ▼
ArgoCD syncs cluster → rolling update

Images are tagged by Git commit SHA for full traceability:

dvanhu/devops-backend:ed17084659da4ab08...

DockerHub repository: hub.docker.com/r/dvanhu/devops-backend

---

Kubernetes Resources

Resource	Purpose
Deployment	Manages pod lifecycle and rolling updates
ReplicaSet	Ensures desired number of pod replicas
Service	Exposes the application within the cluster
Ingress	Routes external HTTP traffic to the service

---

Monitoring

Prometheus scrapes application and cluster metrics; Grafana provides dashboards for:

• Pod health and restart counts
• Resource utilization (CPU, memory)
• Request rates and latency
• Deployment rollout status

---

Required Secrets (GitHub Actions)

Set these in your repository's Settings → Secrets and Variables → Actions:

Secret	Description
DOCKERHUB_USERNAME	Your DockerHub username
DOCKERHUB_TOKEN	DockerHub access token

---

Key DevOps Concepts Demonstrated

• CI/CD Automation — Zero-touch pipeline from commit to deployment
• GitOps — Git as the single source of truth for cluster state
• Containerization — Immutable, reproducible Docker images
• Kubernetes Orchestration — Scalable, self-healing deployments
• Infrastructure as Code — Reproducible infra via Terraform
• Monitoring & Observability — Metrics, dashboards, and alerting
• Image Traceability — SHA-tagged images for audit and rollback

---