docker-compose.yml

version: "3.9"

# ── Shared logging driver ─────────────────────────────────────────────────────
# All app services use json-file so Promtail can tail them via the Docker socket.
x-logging: &default-logging
  driver: json-file
  options:
    max-size: "10m"
    max-file: "3"

# ── Resource limit anchors ────────────────────────────────────────────────────
# Override via env vars (e.g. BACKEND_MEM_LIMIT=1g) for non-testnet environments.
x-resources-backend: &resources-backend
  deploy:
    resources:
      limits:
        memory: ${BACKEND_MEM_LIMIT:-512m}
        cpus:   "${BACKEND_CPU_LIMIT:-0.75}"
      reservations:
        memory: ${BACKEND_MEM_RESERVATION:-128m}
        cpus:   "${BACKEND_CPU_RESERVATION:-0.25}"

x-resources-frontend: &resources-frontend
  deploy:
    resources:
      limits:
        memory: ${FRONTEND_MEM_LIMIT:-256m}
        cpus:   "${FRONTEND_CPU_LIMIT:-0.50}"
      reservations:
        memory: ${FRONTEND_MEM_RESERVATION:-64m}
        cpus:   "${FRONTEND_CPU_RESERVATION:-0.10}"

x-resources-postgres: &resources-postgres
  deploy:
    resources:
      limits:
        memory: ${POSTGRES_MEM_LIMIT:-1g}
        cpus:   "${POSTGRES_CPU_LIMIT:-1.00}"
      reservations:
        memory: ${POSTGRES_MEM_RESERVATION:-256m}
        cpus:   "${POSTGRES_CPU_RESERVATION:-0.25}"

x-resources-redis: &resources-redis
  deploy:
    resources:
      limits:
        memory: ${REDIS_MEM_LIMIT:-256m}
        cpus:   "${REDIS_CPU_LIMIT:-0.25}"
      reservations:
        memory: ${REDIS_MEM_RESERVATION:-32m}
        cpus:   "${REDIS_CPU_RESERVATION:-0.05}"

x-resources-oracle: &resources-oracle
  deploy:
    resources:
      limits:
        memory: ${ORACLE_MEM_LIMIT:-256m}
        cpus:   "${ORACLE_CPU_LIMIT:-0.50}"
      reservations:
        memory: ${ORACLE_MEM_RESERVATION:-64m}
        cpus:   "${ORACLE_CPU_RESERVATION:-0.10}"

x-resources-observability: &resources-observability
  deploy:
    resources:
      limits:
        memory: ${OBSERVABILITY_MEM_LIMIT:-256m}
        cpus:   "${OBSERVABILITY_CPU_LIMIT:-0.25}"
      reservations:
        memory: ${OBSERVABILITY_MEM_RESERVATION:-64m}
        cpus:   "${OBSERVABILITY_CPU_RESERVATION:-0.05}"

services:
  # ── Infrastructure ──────────────────────────────────────────────────────────

  postgres:
    image: postgres:16-alpine
    restart: unless-stopped
    <<: *resources-postgres
    environment:
      POSTGRES_DB:       carbonledger
      POSTGRES_USER:     carbonledger
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-changeme}
    volumes:
      - postgres_data:/var/lib/postgresql/data
    # Only expose directly for local development/debugging.
    # In production, remove this port binding and only connect via pgbouncer.
    ports:
      - "5433:5432"
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U carbonledger"]
      interval: 10s
      timeout: 5s
      retries: 5
    logging: *default-logging

  redis-primary:
    image: redis:7-alpine
    restart: unless-stopped
    <<: *resources-redis
    command: redis-server --requirepass ${REDIS_PASSWORD:-}
    ports:
      - "6379:6379"
    healthcheck:
      test: ["CMD", "redis-cli", "-a", "${REDIS_PASSWORD:-}", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5
    logging: *default-logging

  redis-replica:
    image: redis:7-alpine
    restart: unless-stopped
    <<: *resources-redis
    command: redis-server --replicaof redis-primary 6379 --masterauth ${REDIS_PASSWORD:-} --requirepass ${REDIS_PASSWORD:-}
    depends_on:
      redis-primary:
        condition: service_healthy
    logging: *default-logging

  redis-sentinel:
    image: bitnami/redis-sentinel:7.2
    restart: unless-stopped
    <<: *resources-redis
    environment:
      - REDIS_SENTINEL_QUORUM=1
      - REDIS_MASTER_HOST=redis-primary
      - REDIS_MASTER_PORT_NUMBER=6379
      - REDIS_MASTER_SET=mymaster
      - REDIS_SENTINEL_PASSWORD=${REDIS_PASSWORD:-}
      - REDIS_MASTER_PASSWORD=${REDIS_PASSWORD:-}
    healthcheck:
      test: ["CMD", "redis-cli", "-p", "26379", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5
    depends_on:
      redis-primary:
        condition: service_healthy
    logging: *default-logging

  # ── Application services ────────────────────────────────────────────────────

  # ── Connection Pool ────────────────────────────────────────────────────────────
  # PgBouncer sits between the application and PostgreSQL, multiplexing up to
  # 200 client connections down to a managed pool of 100 server connections.
  # Pool mode: transaction — each connection is returned to the pool after
  # every transaction, maximising concurrency for stateless NestJS requests.
  #
  # Monitor pool health at any time:
  #   psql -h localhost -p 6432 -U carbonledger pgbouncer -c "SHOW STATS;"
  #   psql -h localhost -p 6432 -U carbonledger pgbouncer -c "SHOW POOLS;"
  #   psql -h localhost -p 6432 -U carbonledger pgbouncer -c "SHOW CLIENTS;"
  pgbouncer:
    build:
      context: ./pgbouncer
      dockerfile: Dockerfile
    restart: unless-stopped
    depends_on:
      postgres:
        condition: service_healthy
    environment:
      # These override values in pgbouncer.ini at runtime via env substitution.
      # The DB_* vars allow the image to generate userlist.txt on first boot
      # (supported by edoburu/pgbouncer via environment variables).
      DB_USER:              carbonledger
      DB_PASSWORD:          ${POSTGRES_PASSWORD:-changeme}
      DB_HOST:              postgres
      DB_PORT:              5432
      DB_NAME:              carbonledger
      POOL_MODE:            ${PGBOUNCER_POOL_MODE:-transaction}
      MAX_CLIENT_CONN:      ${PGBOUNCER_MAX_CLIENT_CONN:-200}
      DEFAULT_POOL_SIZE:    ${PGBOUNCER_DEFAULT_POOL_SIZE:-100}
      MIN_POOL_SIZE:        10
      RESERVE_POOL_SIZE:    20
      RESERVE_POOL_TIMEOUT: 5
      SERVER_IDLE_TIMEOUT:  600
      LOG_CONNECTIONS:      1
      LOG_DISCONNECTIONS:   1
      STATS_PERIOD:         60
      AUTH_TYPE:            md5
      ADMIN_USERS:          carbonledger
      STATS_USERS:          carbonledger
    ports:
      # Expose PgBouncer admin port for monitoring tools (not for app traffic)
      - "6432:5432"
    healthcheck:
      # pg_isready targets PgBouncer which proxies to postgres
      test: ["CMD-SHELL", "pg_isready -h localhost -p 5432 -U carbonledger"]
      interval: 10s
      timeout: 5s
      retries: 5

  # ── Cache ──────────────────────────────────────────────────────────────────────
  # Redis is used for 5-minute TTL caching of expensive read queries:
  #   - GET /api/v1/projects
  #   - GET /api/v1/marketplace/listings
  #   - GET /api/v1/oracle/benchmark-price
  # Cache is invalidated on any write (create / update / delete) to those resources.
  redis:
    image: redis:7-alpine
    restart: unless-stopped
    command: >
      redis-server
      --maxmemory 256mb
      --maxmemory-policy allkeys-lru
      --save ""
      --appendonly no
    ports:
      - "6379:6379"
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
      timeout: 3s
      retries: 5

  # ── Backend API ────────────────────────────────────────────────────────────────
  backend:
    build:
      context: ./backend
      dockerfile: Dockerfile
    restart: unless-stopped
    <<: *resources-backend
    depends_on:
      pgbouncer:
        condition: service_healthy
      redis:
        condition: service_healthy
      redis-sentinel:
        condition: service_healthy
    environment:
      DATABASE_URL:      postgresql://carbonledger:${POSTGRES_PASSWORD:-changeme}@postgres:5432/carbonledger
      JWT_SECRET:        ${JWT_SECRET}
      JWT_EXPIRY:        7d
      REDIS_URL:         redis://redis:6379
      STELLAR_NETWORK:   ${STELLAR_NETWORK:-testnet}
      STELLAR_RPC_URL:   ${STELLAR_RPC_URL:-https://soroban-testnet.stellar.org}
      FRONTEND_URL:      http://localhost:3000
      ALLOWED_ORIGINS:   ${ALLOWED_ORIGINS:-http://localhost:3000}
      PORT:              3001
      REDIS_SENTINELS:   redis-sentinel:26379
      REDIS_SENTINEL_NAME: mymaster
      REDIS_PASSWORD:    ${REDIS_PASSWORD:-}
      LOG_LEVEL:         ${LOG_LEVEL:-info}
      OTEL_ENABLED:      ${OTEL_ENABLED:-true}
      OTEL_SERVICE_NAME: carbonledger-backend
      OTEL_EXPORTER_OTLP_ENDPOINT: http://jaeger:4318
      OTEL_TRACES_SAMPLER_ARG: ${OTEL_TRACES_SAMPLER_ARG:-0.1}
    ports:
      - "3001:3001"
    healthcheck:
      test: ["CMD-SHELL", "wget -q --spider http://localhost:3001/health/ready || exit 1"]
      interval: 15s
      timeout: 5s
      retries: 3
      start_period: 30s
    command: >
      sh -c "npx prisma migrate deploy && node dist/main"
    logging: *default-logging

  # ── Frontend ───────────────────────────────────────────────────────────────────
  frontend:
    build:
      context: ./frontend
      dockerfile: Dockerfile
    restart: unless-stopped
    <<: *resources-frontend
    depends_on:
      backend:
        condition: service_healthy
    environment:
      NEXT_PUBLIC_API_URL:              http://localhost:3001/api/v1
      NEXT_PUBLIC_STELLAR_NETWORK:      ${STELLAR_NETWORK:-testnet}
      NEXT_PUBLIC_HORIZON_URL:          ${STELLAR_HORIZON_URL:-https://horizon-testnet.stellar.org}
      NEXT_PUBLIC_SOROBAN_RPC_URL:      ${STELLAR_RPC_URL:-https://soroban-testnet.stellar.org}
      NEXT_PUBLIC_REGISTRY_CONTRACT:    ${CARBON_REGISTRY_CONTRACT_ID}
      NEXT_PUBLIC_CREDIT_CONTRACT:      ${CARBON_CREDIT_CONTRACT_ID}
      NEXT_PUBLIC_MARKETPLACE_CONTRACT: ${CARBON_MARKETPLACE_CONTRACT_ID}
      NEXT_PUBLIC_ORACLE_CONTRACT:      ${CARBON_ORACLE_CONTRACT_ID}
      NEXT_PUBLIC_USDC_CONTRACT:        ${USDC_CONTRACT_ID}
    ports:
      - "3000:3000"
    healthcheck:
      test: ["CMD-SHELL", "wget -q --spider http://localhost:3000/ || exit 1"]
      interval: 15s
      timeout: 5s
      retries: 3
      start_period: 30s
    logging: *default-logging

  # ── Oracle Services ───────────────────────────────────────────────────────────
  oracle_verification:
    build:
      context: ./oracle
      dockerfile: Dockerfile
    restart: unless-stopped
    <<: *resources-oracle
    environment:
      ORACLE_SECRET_KEY:           ${ORACLE_SECRET_KEY}
      CARBON_ORACLE_CONTRACT_ID:   ${CARBON_ORACLE_CONTRACT_ID}
      CARBON_REGISTRY_CONTRACT_ID: ${CARBON_REGISTRY_CONTRACT_ID}
      STELLAR_RPC_URL:             ${STELLAR_RPC_URL:-https://soroban-testnet.stellar.org}
      DATABASE_URL:                postgresql://carbonledger:${POSTGRES_PASSWORD:-changeme}@pgbouncer:5432/carbonledger
      ADMIN_ALERT_WEBHOOK:         ${ADMIN_ALERT_WEBHOOK}
      GOLD_STANDARD_API_URL:       ${GOLD_STANDARD_API_URL}
      GOLD_STANDARD_API_KEY:       ${GOLD_STANDARD_API_KEY}
      VERRA_VCS_API_URL:           ${VERRA_VCS_API_URL}
      VERRA_VCS_API_KEY:           ${VERRA_VCS_API_KEY}
      LOG_LEVEL:                   ${LOG_LEVEL:-INFO}
    command: python3 verification_listener.py
    depends_on:
      postgres:
        condition: service_healthy
    healthcheck:
      test: ["CMD-SHELL", "ps aux | grep verification_listener.py | grep -v grep || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    logging: *default-logging

  oracle_verification_standby:
    build:
      context: ./oracle
      dockerfile: Dockerfile
    restart: unless-stopped
    <<: *resources-oracle
    environment:
      ORACLE_SECRET_KEY:           ${ORACLE_SECRET_KEY}
      CARBON_ORACLE_CONTRACT_ID:   ${CARBON_ORACLE_CONTRACT_ID}
      CARBON_REGISTRY_CONTRACT_ID: ${CARBON_REGISTRY_CONTRACT_ID}
      STELLAR_RPC_URL:             ${STELLAR_RPC_URL:-https://soroban-testnet.stellar.org}
      DATABASE_URL:                postgresql://carbonledger:${POSTGRES_PASSWORD:-changeme}@postgres:5432/carbonledger
      ADMIN_ALERT_WEBHOOK:         ${ADMIN_ALERT_WEBHOOK}
      GOLD_STANDARD_API_URL:       ${GOLD_STANDARD_API_URL}
      GOLD_STANDARD_API_KEY:       ${GOLD_STANDARD_API_KEY}
      VERRA_VCS_API_URL:           ${VERRA_VCS_API_URL}
      VERRA_VCS_API_KEY:           ${VERRA_VCS_API_KEY}
      LOG_LEVEL:                   ${LOG_LEVEL:-INFO}
      ORACLE_STANDBY_MODE:         "true"
      FAILOVER_SERVICE_NAME:       verification_listener
    command: python3 verification_listener.py
    depends_on:
      postgres:
        condition: service_healthy
    healthcheck:
      test: ["CMD-SHELL", "ps aux | grep verification_listener.py | grep -v grep || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    logging: *default-logging

  oracle_price:
    build:
      context: ./oracle
      dockerfile: Dockerfile
    restart: unless-stopped
    <<: *resources-oracle
    environment:
      ORACLE_SECRET_KEY:         ${ORACLE_SECRET_KEY}
      CARBON_ORACLE_CONTRACT_ID: ${CARBON_ORACLE_CONTRACT_ID}
      STELLAR_RPC_URL:           ${STELLAR_RPC_URL:-https://soroban-testnet.stellar.org}
      XPANSIV_API_KEY:           ${XPANSIV_API_KEY}
      TOUCAN_API_KEY:            ${TOUCAN_API_KEY}
      ADMIN_ALERT_WEBHOOK:       ${ADMIN_ALERT_WEBHOOK}
      LOG_LEVEL:                 ${LOG_LEVEL:-INFO}
    command: python3 price_oracle.py
    healthcheck:
      test: ["CMD-SHELL", "ps aux | grep price_oracle.py | grep -v grep || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    logging: *default-logging

  oracle_price_standby:
    build:
      context: ./oracle
      dockerfile: Dockerfile
    restart: unless-stopped
    <<: *resources-oracle
    environment:
      ORACLE_SECRET_KEY:         ${ORACLE_SECRET_KEY}
      CARBON_ORACLE_CONTRACT_ID: ${CARBON_ORACLE_CONTRACT_ID}
      STELLAR_RPC_URL:           ${STELLAR_RPC_URL:-https://soroban-testnet.stellar.org}
      XPANSIV_API_KEY:           ${XPANSIV_API_KEY}
      TOUCAN_API_KEY:            ${TOUCAN_API_KEY}
      ADMIN_ALERT_WEBHOOK:       ${ADMIN_ALERT_WEBHOOK}
      LOG_LEVEL:                 ${LOG_LEVEL:-INFO}
      ORACLE_STANDBY_MODE:       "true"
      FAILOVER_SERVICE_NAME:     price_oracle
    command: python3 price_oracle.py
    healthcheck:
      test: ["CMD-SHELL", "ps aux | grep price_oracle.py | grep -v grep || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    logging: *default-logging

  oracle_satellite:
    build:
      context: ./oracle
      dockerfile: Dockerfile
    restart: unless-stopped
    <<: *resources-oracle
    environment:
      ORACLE_SECRET_KEY:         ${ORACLE_SECRET_KEY}
      CARBON_ORACLE_CONTRACT_ID: ${CARBON_ORACLE_CONTRACT_ID}
      STELLAR_RPC_URL:           ${STELLAR_RPC_URL:-https://soroban-testnet.stellar.org}
      BACKEND_API_URL:           http://backend:3001/api/v1
      GEE_WEBHOOK_SECRET:        ${GEE_WEBHOOK_SECRET}
      ADMIN_ALERT_WEBHOOK:       ${ADMIN_ALERT_WEBHOOK}
      SATELLITE_MONITOR_PORT:    5001
      LOG_LEVEL:                 ${LOG_LEVEL:-INFO}
    command: python3 satellite_monitor.py
    ports:
      - "5001:5001"
    healthcheck:
      test: ["CMD-SHELL", "wget -q --spider http://localhost:5001/health || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    logging: *default-logging

  oracle_satellite_standby:
    build:
      context: ./oracle
      dockerfile: Dockerfile
    restart: unless-stopped
    <<: *resources-oracle
    environment:
      ORACLE_SECRET_KEY:         ${ORACLE_SECRET_KEY}
      CARBON_ORACLE_CONTRACT_ID: ${CARBON_ORACLE_CONTRACT_ID}
      STELLAR_RPC_URL:           ${STELLAR_RPC_URL:-https://soroban-testnet.stellar.org}
      BACKEND_API_URL:           http://backend:3001/api/v1
      GEE_WEBHOOK_SECRET:        ${GEE_WEBHOOK_SECRET}
      ADMIN_ALERT_WEBHOOK:       ${ADMIN_ALERT_WEBHOOK}
      SATELLITE_MONITOR_PORT:    5002
      LOG_LEVEL:                 ${LOG_LEVEL:-INFO}
      ORACLE_STANDBY_MODE:       "true"
      FAILOVER_SERVICE_NAME:     satellite_monitor
    command: python3 satellite_monitor.py
    healthcheck:
      test: ["CMD-SHELL", "wget -q --spider http://localhost:5002/health || exit 1"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    logging: *default-logging

  # ── Log aggregation stack ───────────────────────────────────────────────────

  loki:
    image: grafana/loki:2.9.8
    restart: unless-stopped
    <<: *resources-observability
    ports:
      - "3100:3100"
    volumes:
      - ./logging/loki/loki.yml:/etc/loki/loki.yml:ro
      - loki_data:/loki
    command: -config.file=/etc/loki/loki.yml
    healthcheck:
      test: ["CMD-SHELL", "wget -q --spider http://localhost:3100/ready || exit 1"]
      interval: 10s
      timeout: 5s
      retries: 5
    logging: *default-logging

  promtail:
    image: grafana/promtail:2.9.8
    restart: unless-stopped
    <<: *resources-observability
    volumes:
      - ./logging/promtail/promtail.yml:/etc/promtail/promtail.yml:ro
      - /var/run/docker.sock:/var/run/docker.sock:ro
    command: -config.file=/etc/promtail/promtail.yml
    depends_on:
      loki:
        condition: service_healthy
    logging: *default-logging

  grafana:
    image: grafana/grafana:10.4.2
    restart: unless-stopped
    <<: *resources-observability
    ports:
      - "3200:3000"
    environment:
      GF_SECURITY_ADMIN_PASSWORD: ${GRAFANA_PASSWORD:-admin}
      GF_USERS_ALLOW_SIGN_UP:     "false"
      GF_ALERTING_ENABLED:        "true"
      GF_UNIFIED_ALERTING_ENABLED: "true"
    volumes:
      - ./logging/grafana/provisioning:/etc/grafana/provisioning:ro
      - grafana_data:/var/lib/grafana
    depends_on:
      loki:
        condition: service_healthy
    logging: *default-logging

  jaeger:
    image: jaegertracing/all-in-one:1.57
    restart: unless-stopped
    <<: *resources-observability
    environment:
      - COLLECTOR_OTLP_ENABLED=true
    ports:
      - "16686:16686"
      - "4317:4317"
      - "4318:4318"
      - "6831:6831/udp"
    healthcheck:
      test: ["CMD-SHELL", "wget -q --spider http://localhost:16686/ || exit 1"]
      interval: 10s
      timeout: 5s
      retries: 5
    logging: *default-logging

volumes:
  postgres_data:
  redis_data:
  loki_data:
  grafana_data: