# AAFP Docker Compose — 3-agent setup with Prometheus + Grafana
#
# Usage:
#   docker compose up -d
#   docker compose logs -f agent-1
#   docker compose down
#
# This starts:
#   - relay:      Public relay node (port 4433/udp)
#   - agent-1:    Agent behind NAT (uses relay for traversal)
#   - agent-2:    Another agent behind NAT (uses relay for traversal)
#   - prometheus: Metrics scraper (port 9090)
#   - grafana:    Dashboard UI (port 3000, admin/admin)
#
# After starting, open http://localhost:3000 for the AAFP dashboard.

services:
  # ── Relay Node ─────────────────────────────────────────────────────────────
  relay:
    build:
      context: ./implementations/rust
      dockerfile: Dockerfile
    container_name: aafp-relay
    ports:
      - "4433:4433/udp"
    environment:
      AAFP_BIND: "0.0.0.0:4433"
      AAFP_RELAY_MODE: "true"
      AAFP_CAPABILITIES: "relay"
      AAFP_METRICS: "0.0.0.0:9090"
      RUST_LOG: "info"
    volumes:
      - relay-keys:/data/keys
      - relay-dht:/data/dht
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "/aafp-agent", "healthcheck"]
      interval: 30s
      timeout: 5s
      retries: 3
      start_period: 10s
    networks:
      - aafp-net

  # ── Agent 1 (behind NAT) ───────────────────────────────────────────────────
  agent-1:
    build:
      context: ./implementations/rust
      dockerfile: Dockerfile
    container_name: aafp-agent-1
    environment:
      AAFP_BIND: "0.0.0.0:4433"
      AAFP_CAPABILITIES: "inference,translation"
      AAFP_RELAY: "quic://relay:4433"
      AAFP_METRICS: "0.0.0.0:9090"
      RUST_LOG: "info"
    volumes:
      - agent1-keys:/data/keys
      - agent1-dht:/data/dht
    depends_on:
      relay:
        condition: service_healthy
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "/aafp-agent", "healthcheck"]
      interval: 30s
      timeout: 5s
      retries: 3
      start_period: 15s
    networks:
      - aafp-net

  # ── Agent 2 (behind NAT) ───────────────────────────────────────────────────
  agent-2:
    build:
      context: ./implementations/rust
      dockerfile: Dockerfile
    container_name: aafp-agent-2
    environment:
      AAFP_BIND: "0.0.0.0:4433"
      AAFP_CAPABILITIES: "inference,summarization"
      AAFP_RELAY: "quic://relay:4433"
      AAFP_METRICS: "0.0.0.0:9090"
      RUST_LOG: "info"
    volumes:
      - agent2-keys:/data/keys
      - agent2-dht:/data/dht
    depends_on:
      relay:
        condition: service_healthy
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "/aafp-agent", "healthcheck"]
      interval: 30s
      timeout: 5s
      retries: 3
      start_period: 15s
    networks:
      - aafp-net

  # ── Prometheus (metrics scraper) ───────────────────────────────────────────
  prometheus:
    image: prom/prometheus:latest
    container_name: aafp-prometheus
    ports:
      - "9090:9090"
    volumes:
      - ./deploy/prometheus/prometheus.yml:/etc/prometheus/prometheus.yml
    networks:
      - aafp-net
    restart: unless-stopped

  # ── Grafana (dashboard UI) ─────────────────────────────────────────────────
  grafana:
    image: grafana/grafana:latest
    container_name: aafp-grafana
    ports:
      - "3000:3000"
    environment:
      - GF_SECURITY_ADMIN_PASSWORD=admin
      - GF_USERS_ALLOW_SIGN_UP=false
    volumes:
      - ./deploy/grafana/datasource.yml:/etc/grafana/provisioning/datasources/datasource.yml
      - ./deploy/grafana/dashboards.yml:/etc/grafana/provisioning/dashboards/dashboards.yml
      - ./deploy/grafana/aafp-dashboard.json:/etc/grafana/provisioning/dashboards/aafp-dashboard.json
      - grafana-data:/var/lib/grafana
    depends_on:
      - prometheus
    networks:
      - aafp-net
    restart: unless-stopped

# ── Volumes ──────────────────────────────────────────────────────────────────
volumes:
  relay-keys:
  relay-dht:
  agent1-keys:
  agent1-dht:
  agent2-keys:
  agent2-dht:
  grafana-data:

# ── Network ──────────────────────────────────────────────────────────────────
networks:
  aafp-net:
    driver: bridge