docker-compose.yml

# ============================================================
#  Phase 2 — Kafka KRaft Single-Broker (No Zookeeper)
#  Phase 3 — Adds PostgreSQL 16 for alert persistence
# ============================================================
#
#  Services:
#    kafka      — single broker, KRaft mode, exposed on localhost:9092
#    postgres   — PostgreSQL 16, exposed on localhost:5432 (Phase 3+)
#
#  Usage:
#    docker compose up -d              # start all services
#    docker compose down               # stop and remove containers
#    docker compose logs -f kafka      # tail broker logs
#    docker compose logs -f postgres   # tail PostgreSQL logs
#
#  Topic auto-creation is ON (auto.create.topics.enable=true).
#  The 'market-trades' topic is created automatically on first produce.
#  Alert topics 'wash-alerts' and 'pump-dump-alerts' are also auto-created.
#
#  PostgreSQL:
#    Database: surveillance
#    User:     surveillance
#    Password: surveillance_local (dev only — override via PG_PASSWORD env var)
#
#  To connect via psql:
#    psql -h localhost -p 5432 -U surveillance -d surveillance
#
#  To manually create topics:
#    docker exec -it kafka kafka-topics.sh \
#      --bootstrap-server localhost:9092 \
#      --create --topic market-trades \
#      --partitions 3 --replication-factor 1
# ============================================================

services:
  kafka:
    image: apache/kafka:3.7.0
    container_name: kafka
    hostname: kafka
    ports:
      - "9092:9092"       # external (host) access
      - "9093:9093"       # KRaft controller port (internal)
    environment:
      # ---- KRaft identity ----
      KAFKA_NODE_ID: 1
      KAFKA_PROCESS_ROLES: broker,controller
      KAFKA_CONTROLLER_QUORUM_VOTERS: "1@kafka:9093"

      # ---- Listeners ----
      # PLAINTEXT  → broker-to-broker + Spark Structured Streaming (inside Docker network)
      # EXTERNAL   → host machine access (localhost:9092)
      # CONTROLLER → KRaft controller traffic
      KAFKA_LISTENERS: "PLAINTEXT://kafka:29092,EXTERNAL://0.0.0.0:9092,CONTROLLER://kafka:9093"
      KAFKA_ADVERTISED_LISTENERS: "PLAINTEXT://kafka:29092,EXTERNAL://localhost:9092"
      KAFKA_LISTENER_SECURITY_PROTOCOL_MAP: "PLAINTEXT:PLAINTEXT,EXTERNAL:PLAINTEXT,CONTROLLER:PLAINTEXT"
      KAFKA_INTER_BROKER_LISTENER_NAME: PLAINTEXT
      KAFKA_CONTROLLER_LISTENER_NAMES: CONTROLLER

      # ---- Performance & reliability ----
      KAFKA_AUTO_CREATE_TOPICS_ENABLE: "true"
      KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1       # single broker → RF=1
      KAFKA_TRANSACTION_STATE_LOG_REPLICATION_FACTOR: 1
      KAFKA_TRANSACTION_STATE_LOG_MIN_ISR: 1
      KAFKA_LOG_RETENTION_HOURS: 1                    # keep 1h of data (dev mode)
      KAFKA_LOG_SEGMENT_BYTES: 10485760               # 10 MB segments
      KAFKA_NUM_PARTITIONS: 3                         # 3 partitions for market-trades topic

      # ---- Cluster ID (KRaft requires a stable UUID) ----
      CLUSTER_ID: "MkU3OEVBNTcwNTJENDM2Qk"

    volumes:
      - kafka_data:/var/lib/kafka/data

    healthcheck:
      test: ["CMD", "/opt/kafka/bin/kafka-broker-api-versions.sh", "--bootstrap-server", "localhost:9092"]
      interval: 10s
      timeout: 10s
      retries: 10
      start_period: 30s

  # ---- PostgreSQL 16 (Phase 3+) ----
  postgres:
    image: postgres:16-alpine
    container_name: surveillance-postgres
    hostname: postgres
    ports:
      - "5432:5432"
    environment:
      POSTGRES_DB: surveillance
      POSTGRES_USER: surveillance
      POSTGRES_PASSWORD: surveillance_local
    volumes:
      - pgdata:/var/lib/postgresql/data
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U surveillance -d surveillance"]
      interval: 5s
      timeout: 5s
      retries: 10
      start_period: 10s

volumes:
  kafka_data:
    driver: local
  pgdata:
    driver: local