simstudioai--sim/helm/sim/examples/values-production.yaml

# values-production.yaml
#
# When to use: generic production deployment on any cluster. Enables HA
# (replicaCount > 1, PDB), NetworkPolicy, HPA, ServiceMonitor, and
# PSS-restricted security defaults.
#
# Prerequisites:
#   - A premium StorageClass with reclaimPolicy: Retain
#   - cert-manager (for TLS) or pre-provisioned certs
#   - ingress-nginx (or equivalent) for ingress
#   - metrics-server in the cluster (HPA)
#   - (recommended) Prometheus Operator for ServiceMonitor
#   - (recommended) External Secrets Operator — see values-external-secrets.yaml
#
# Install:
#   helm install sim ./helm/sim \
#     --namespace sim --create-namespace \
#     --values ./helm/sim/examples/values-production.yaml \
#     --set app.env.BETTER_AUTH_SECRET="$BETTER_AUTH_SECRET" \
#     --set app.env.ENCRYPTION_KEY="$ENCRYPTION_KEY" \
#     --set app.env.INTERNAL_API_SECRET="$INTERNAL_API_SECRET" \
#     --set app.env.CRON_SECRET="$CRON_SECRET" \
#     --set postgresql.auth.password="$POSTGRES_PASSWORD"

# Global configuration
global:
  imageRegistry: "ghcr.io"
  # For production, use a StorageClass with reclaimPolicy: Retain
  storageClass: "managed-csi-premium"

# Main application
app:
  enabled: true
  replicaCount: 2

  resources:
    limits:
      memory: "8Gi"
      cpu: "2000m"
    requests:
      memory: "6Gi"
      cpu: "1000m"
  
  # Production URLs (REQUIRED - update with your actual domain names)
  env:
    NEXT_PUBLIC_APP_URL: "https://sim.acme.ai"
    BETTER_AUTH_URL: "https://sim.acme.ai"
    SOCKET_SERVER_URL: "https://sim-ws.acme.ai"
    NEXT_PUBLIC_SOCKET_URL: "https://sim-ws.acme.ai"
    
    # Security settings (REQUIRED - replace with your own secure secrets)
    # Generate using: openssl rand -hex 32
    BETTER_AUTH_SECRET: "your-production-auth-secret-here"
    ENCRYPTION_KEY: "your-production-encryption-key-here"
    INTERNAL_API_SECRET: "your-production-internal-api-secret-here"
    CRON_SECRET: "your-production-cron-secret-here"

    # Optional: API Key Encryption (RECOMMENDED for production)
    # Generate 64-character hex string using: openssl rand -hex 32
    API_ENCRYPTION_KEY: "your-64-char-hex-api-encryption-key-here"  # Optional but recommended
    
    # Email verification (set to true if you want to require email verification)
    EMAIL_VERIFICATION_ENABLED: "false"
    
    # Optional third-party service integrations (configure as needed)
    RESEND_API_KEY: "your-resend-api-key"
    GOOGLE_CLIENT_ID: "your-google-client-id"
    GOOGLE_CLIENT_SECRET: "your-google-client-secret"
    # DISABLE_GOOGLE_AUTH: "true"                      # Uncomment to hide Google OAuth login
    # DISABLE_GITHUB_AUTH: "true"                      # Uncomment to hide GitHub OAuth login

# Realtime service
realtime:
  enabled: true
  replicaCount: 2

  resources:
    limits:
      memory: "1Gi"
      cpu: "500m"
    requests:
      memory: "512Mi"
      cpu: "250m"
  
  env:
    NEXT_PUBLIC_APP_URL: "https://sim.acme.ai"
    BETTER_AUTH_URL: "https://sim.acme.ai"
    BETTER_AUTH_SECRET: "your-production-auth-secret-here"
    ALLOWED_ORIGINS: "https://sim.acme.ai"

# Database migrations
migrations:
  enabled: true
  resources:
    limits:
      memory: "2Gi"
      cpu: "1000m"
    requests:
      memory: "1Gi"
      cpu: "500m"

# PostgreSQL database
postgresql:
  enabled: true
  
  # Database authentication (REQUIRED - set secure credentials)
  auth:
    username: postgres
    password: "your-secure-postgres-password"
    database: simstudio
  
  # Resource allocation for production workloads
  resources:
    limits:
      memory: "4Gi"
      cpu: "2000m"
    requests:
      memory: "2Gi"
      cpu: "1000m"
  
  # Persistent storage configuration
  persistence:
    enabled: true
    storageClass: "managed-csi-premium"
    size: 50Gi
  
  # SSL/TLS configuration (recommended for production)
  tls:
    enabled: true
    certificatesSecret: postgres-tls-secret
  
  # PostgreSQL performance configuration for production
  config:
    maxConnections: 1000
    sharedBuffers: "2GB"
    maxWalSize: "8GB"
    minWalSize: "160MB"

# Ollama AI models (optional - enable if you need local AI model serving)
ollama:
  enabled: false

# Ingress configuration
ingress:
  enabled: true
  className: nginx
  
  annotations:
    nginx.ingress.kubernetes.io/force-ssl-redirect: "true"
    cert-manager.io/cluster-issuer: "letsencrypt-prod"
  
  # Main application
  app:
    host: sim.acme.ai
    paths:
      - path: /
        pathType: Prefix
  
  # Realtime service
  realtime:
    host: sim-ws.acme.ai
    paths:
      - path: /
        pathType: Prefix
  
  # TLS configuration
  tls:
    enabled: true
    secretName: sim-tls-secret

# Horizontal Pod Autoscaler (automatically scales pods based on CPU/memory usage)
autoscaling:
  enabled: true
  minReplicas: 2
  maxReplicas: 20
  targetCPUUtilizationPercentage: 70
  targetMemoryUtilizationPercentage: 80
  behavior:
    scaleDown:
      stabilizationWindowSeconds: 300
      policies:
      - type: Percent
        value: 50
        periodSeconds: 60
    scaleUp:
      stabilizationWindowSeconds: 60
      policies:
      - type: Percent
        value: 100
        periodSeconds: 15
      - type: Pods
        value: 2
        periodSeconds: 60

# Pod disruption budget (ensures minimum availability during cluster maintenance)
podDisruptionBudget:
  enabled: true
  minAvailable: null
  maxUnavailable: 1  
  unhealthyPodEvictionPolicy: AlwaysAllow 

# Monitoring integration with Prometheus
monitoring:
  serviceMonitor:
    enabled: true
    labels:
      monitoring: "prometheus"
    interval: 15s
    scrapeTimeout: 10s

# Network policies (restricts pod-to-pod communication for security)
networkPolicy:
  enabled: true

# Shared storage for data sharing between pods (enterprise feature)
sharedStorage:
  enabled: true
  storageClass: "managed-csi-premium"
  volumes:
    - name: output-share
      size: 100Gi
      accessModes:
        - ReadWriteMany
    - name: model-share
      size: 200Gi
      accessModes:
        - ReadWriteMany

# Telemetry and observability (comprehensive monitoring and tracing)
telemetry:
  enabled: true
  resources:
    limits:
      memory: "1Gi"
      cpu: "500m"
    requests:
      memory: "512Mi"
      cpu: "200m"
  # Configure endpoints based on your observability infrastructure
  prometheus:
    enabled: true
    endpoint: "http://prometheus-server/api/v1/write"
  jaeger:
    enabled: true
    endpoint: "http://jaeger-collector:14250"