DAGI Stack Deployment Guide¶

This guide covers deploying DAGI Stack in various environments.

🚀 Quick Start (Docker Compose)¶

Prerequisites¶

Docker 20.10+
Docker Compose 2.0+
4GB+ RAM
10GB+ disk space

Steps¶

Clone repository bash git clone https://github.com/daarion/dagi-stack.git cd dagi-stack
Configure environment bash cp .env.example .env # Edit .env with your tokens and settings
Start services bash docker-compose up -d
Verify health bash curl http://localhost:9102/health # Router curl http://localhost:8008/health # DevTools curl http://localhost:9010/health # CrewAI curl http://localhost:9200/health # RBAC curl http://localhost:9300/health # Gateway
View logs bash docker-compose logs -f router
Stop services bash docker-compose down

📋 Service Ports¶

Service	Port	Description
DAGI Router	9102	Main routing endpoint
DevTools	8008	File ops, tests, notebooks
CrewAI	9010	Multi-agent workflows
RBAC	9200	Role-based access control
Gateway	9300	Telegram/Discord webhooks
Ollama	11434	Local LLM (optional)

🔧 Production Deployment¶

Systemd Services (Linux)¶

Create service file bash sudo nano /etc/systemd/system/dagi-router.service
Service configuration ```ini [Unit] Description=DAGI Router Service After=network.target

[Service] Type=simple User=dagi WorkingDirectory=/opt/dagi-stack Environment="PATH=/opt/dagi-stack/.venv/bin" ExecStart=/opt/dagi-stack/.venv/bin/python main_v2.py --port 9102 Restart=always RestartSec=10

[Install] WantedBy=multi-user.target ```

Enable and start bash sudo systemctl daemon-reload sudo systemctl enable dagi-router sudo systemctl start dagi-router sudo systemctl status dagi-router

☸️ Kubernetes Deployment¶

Basic Deployment¶

apiVersion: apps/v1
kind: Deployment
metadata:
  name: dagi-router
  namespace: dagi-stack
spec:
  replicas: 3
  selector:
    matchLabels:
      app: dagi-router
  template:
    metadata:
      labels:
        app: dagi-router
    spec:
      containers:
      - name: router
        image: daarion/dagi-router:0.2.0
        ports:
        - containerPort: 9102
        env:
        - name: DAGI_ROUTER_CONFIG
          value: /config/router-config.yml
        volumeMounts:
        - name: config
          mountPath: /config
        livenessProbe:
          httpGet:
            path: /health
            port: 9102
          initialDelaySeconds: 10
          periodSeconds: 30
        readinessProbe:
          httpGet:
            path: /health
            port: 9102
          initialDelaySeconds: 5
          periodSeconds: 10
        resources:
          requests:
            memory: "512Mi"
            cpu: "500m"
          limits:
            memory: "1Gi"
            cpu: "1000m"
      volumes:
      - name: config
        configMap:
          name: dagi-router-config
---
apiVersion: v1
kind: Service
metadata:
  name: dagi-router
  namespace: dagi-stack
spec:
  selector:
    app: dagi-router
  ports:
  - protocol: TCP
    port: 9102
    targetPort: 9102
  type: ClusterIP

Deploy¶

kubectl create namespace dagi-stack
kubectl apply -f k8s/router-deployment.yaml
kubectl apply -f k8s/devtools-deployment.yaml
kubectl apply -f k8s/crewai-deployment.yaml
kubectl apply -f k8s/rbac-deployment.yaml
kubectl apply -f k8s/gateway-deployment.yaml
kubectl apply -f k8s/ingress.yaml

🔒 Security¶

Environment Variables¶

Never commit secrets to git. Use: - Docker secrets - Kubernetes secrets - Vault - AWS Secrets Manager

Example (Kubernetes):

apiVersion: v1
kind: Secret
metadata:
  name: dagi-secrets
  namespace: dagi-stack
type: Opaque
stringData:
  telegram-token: "your_token_here"
  deepseek-key: "your_key_here"

Network Security¶

Firewall rules
Allow: 9102 (Router), 9300 (Gateway)
Deny: 8008, 9010, 9200 (internal only)
TLS/SSL Use reverse proxy (Nginx, Traefik) for HTTPS
Rate limiting Configure in reverse proxy or API gateway

📊 Monitoring¶

Health Checks¶

All services expose /health endpoint:

#!/bin/bash
# health-check.sh
services=("9102" "8008" "9010" "9200" "9300")
for port in "${services[@]}"; do
  status=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:$port/health)
  if [ "$status" = "200" ]; then
    echo "✅ Port $port: healthy"
  else
    echo "❌ Port $port: unhealthy (HTTP $status)"
  fi
done

Prometheus Metrics (Future)¶

Add to router:

from prometheus_client import Counter, Histogram
requests_total = Counter('dagi_requests_total', 'Total requests')
request_duration = Histogram('dagi_request_duration_seconds', 'Request duration')

🔄 Updates & Rollback¶

Docker Compose¶

# Pull latest images
docker-compose pull

# Restart services
docker-compose up -d

# Rollback
docker-compose down
docker-compose up -d --force-recreate

Kubernetes¶

# Rolling update
kubectl set image deployment/dagi-router router=daarion/dagi-router:0.3.0

# Rollback
kubectl rollout undo deployment/dagi-router

# Check status
kubectl rollout status deployment/dagi-router

🐛 Troubleshooting¶

Service not starting¶

# Check logs
docker-compose logs router

# Or for systemd
sudo journalctl -u dagi-router -f

Connection refused¶

Check firewall rules
Verify service is running: systemctl status dagi-router
Check port binding: netstat -tulpn | grep 9102

LLM timeout¶

Increase timeout in router-config.yml
Check Ollama service: curl http://localhost:11434/api/tags
Consider using smaller model or GPU acceleration

RBAC errors¶

Verify RBAC service is running
Check database connection
Review RBAC logs: docker-compose logs rbac

📈 Scaling¶

Horizontal Scaling¶

# Docker Compose
docker-compose up -d --scale router=3

# Kubernetes
kubectl scale deployment/dagi-router --replicas=5

Load Balancing¶

Use: - Nginx - Traefik - AWS ALB - GCP Load Balancer

Example Nginx config:

upstream dagi_router {
    least_conn;
    server router-1:9102;
    server router-2:9102;
    server router-3:9102;
}

server {
    listen 80;
    location / {
        proxy_pass http://dagi_router;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
    }
}

🔧 Performance Tuning¶

Router¶

Adjust timeout_ms in config
Increase worker processes
Enable connection pooling

Database (RBAC)¶

Use PostgreSQL instead of SQLite
Add indexes on user_id, dao_id
Enable query caching

LLM¶

Use GPU for Ollama
Consider model quantization
Implement request queuing

📞 Support¶

Documentation: https://docs.daarion.city
Issues: https://github.com/daarion/dagi-stack/issues
Discord: https://discord.gg/daarion