NexusCRM — Full Stack CRM on Aurora MySQL

React + Node.js/Express app deployed as a Kubernetes pod, connected to Aurora MySQL with separate Writer and Reader endpoints.

---

Architecture

Browser ──► K8s Service ──► Pod (React SPA + Express API)
                                    │
                    ┌───────────────┴────────────────┐
                    │ writerPool                     │ readerPool
                    ▼                                ▼
         Aurora Writer Endpoint          Aurora Reader Endpoint
         (INSERT / UPDATE / DELETE)      (SELECT / COUNT)

---

Quick Start

1. Run Schema

mysql -h <WRITER_HOST> -u admin -p < schema.sql

2. Configure K8s Secrets & ConfigMap

# Edit your endpoints and credentials
vim k8s/configmap.yaml   # DB_WRITER_HOST, DB_READER_HOST, DB_NAME
vim k8s/secret.yaml      # DB_USER, DB_PASSWORD

kubectl apply -f k8s/configmap.yaml
kubectl apply -f k8s/secret.yaml

3. Build & Push Docker Image

docker build -t your-registry/nexus-crm:latest .
docker push  your-registry/nexus-crm:latest

# Update image name in k8s/deployment.yaml

4. Deploy to Kubernetes

kubectl apply -f k8s/deployment.yaml
kubectl apply -f k8s/service.yaml
kubectl get pods -l app=nexus-crm

5. Check Readiness (both DB endpoints)

kubectl port-forward svc/nexus-crm-svc 8080:80
curl http://localhost:8080/readyz
# {"status":"ready","writer":"ok","reader":"ok"}

---

Local Development

# Backend
cd backend && npm install
DB_WRITER_HOST=localhost DB_READER_HOST=localhost \
DB_USER=root DB_PASSWORD=pass DB_NAME=crm_db \
npm run dev

# Frontend (separate terminal)
cd frontend && npm install && npm run dev
# → http://localhost:5173 (proxies /api to :3001)

---

API Reference

Method	Path	DB Pool	Description
GET	/api/contacts	Reader	List + search + filter
GET	/api/contacts/stats	Reader	Counts by status
GET	/api/contacts/:id	Reader	Single contact
POST	/api/contacts	Writer	Create contact
PUT	/api/contacts/:id	Writer	Update contact
DELETE	/api/contacts/:id	Writer	Delete contact
GET	/healthz	—	Liveness probe
GET	/readyz	Both	Readiness probe

---

K8s Notes

• Pod anti-affinity spreads replicas across AZs (matches Aurora multi-zone)
• Readiness probe hits /readyz which pings both writer & reader before accepting traffic
• HPA scales 2→8 pods at 65% CPU
• Rolling update with maxUnavailable: 0 = zero downtime deploys