FleetOps Platform

Plateforme cloud-native de gestion de flotte deployee sur AWS. Projet de fin d etudes — Cloud/DevOps Engineering.

Architecture

Internet → Load Balancer → EKS (FastAPI) → RDS PostgreSQL

• App : FastAPI + PostgreSQL + Alembic
• Infra : AWS VPC + EKS + RDS via Terraform
• CI/CD : GitHub Actions (lint, tests, security, build, deploy)
• Observabilite : Prometheus + Grafana + 5 alertes + SLOs
• Packaging : Helm chart avec values dev/prod

Stack technique

Composant	Technologie
API	FastAPI, Python 3.12
Base de donnees	PostgreSQL 16, SQLAlchemy, Alembic
Conteneurisation	Docker multi-stage
Infrastructure	Terraform, AWS (VPC, EKS, RDS, ECR, S3)
Orchestration	Kubernetes, Helm
CI/CD	GitHub Actions
Monitoring	Prometheus, Grafana
Securite	Bandit, pip-audit, Trivy, Gitleaks

Demarrage rapide

Prerequis

• Docker + Docker Compose
• Python 3.12
• Terraform >= 1.5
• AWS CLI configure
• kubectl + Helm

Dev local

git clone https://github.com/imane-ait/cloud-platform.git cd cloud-platform cp .env.example .env docker compose up -d docker compose exec app alembic upgrade head curl http://localhost:8000/health curl http://localhost:8000/docs

Acces aux interfaces

Interface	URL	Credentials
API Swagger	http://localhost:8000/docs	-
Prometheus	http://localhost:9090	-
Grafana	http://localhost:3000	admin/admin

Lancer les tests

python -m pytest tests/ --cov=app --cov-report=term-missing

Infrastructure AWS

cd infra terraform init terraform plan terraform apply

CI/CD

Le pipeline GitHub Actions se declenche automatiquement a chaque push :

1. Lint — ruff, black
2. Tests — pytest avec PostgreSQL
3. Security — Bandit, pip-audit
4. Build — image Docker poussee sur ECR
5. Deploy — Helm sur EKS (sur merge main)

Observabilite

• Metriques : /metrics expose les metriques Prometheus
• Dashboards : Grafana sur port 3000
• SLOs : disponibilite >= 99.5%, latence p95 < 300ms
• Runbooks : docs/runbooks/

Documentation

• docs/ARCHITECTURE.md — decisions d architecture (ADRs)
• docs/SLO.md — definition des SLOs
• docs/POSTMORTEM.md — postmortem incident DB
• docs/runbooks/ — runbooks par alerte
• docs/SECURITY.md — politique de securite