docker-compose.yml

version: "3.8"

services:
  app:
    build:
      context: .
    ports:
      - "3000:3000"
    volumes:
      - .:/app
      - /app/node_modules
    working_dir: /app
    command: nodemon app.js
    depends_on:
      - postgres
      - minio
      - spark
      - airflow-webserver
    environment:
      - DATABASE_URL=postgres://postgres:password@postgres:5432/opencap
      - MINIO_ENDPOINT=http://minio:9000
      - MINIO_ACCESS_KEY=minio
      - MINIO_SECRET_KEY=minio123
      - NODE_ENV=development

  postgres:
    image: postgres:15-alpine
    container_name: opencap_postgres
    restart: always
    ports:
      - "5432:5432"
    environment:
      POSTGRES_USER: postgres
      POSTGRES_PASSWORD: password
      POSTGRES_DB: opencap
    volumes:
      - postgres_data:/var/lib/postgresql/data

  minio:
    image: minio/minio:latest
    container_name: opencap_minio
    command: server /data
    ports:
      - "9000:9000"
      - "9001:9001"
    environment:
      MINIO_ROOT_USER: minio
      MINIO_ROOT_PASSWORD: minio123
    volumes:
      - minio_data:/data

  spark:
    image: bitnami/spark:latest
    container_name: opencap_spark
    environment:
      - SPARK_MODE=standalone
      - SPARK_MASTER_URL=spark://spark:7077
    ports:
      - "7077:7077"
      - "8080:8080"
    volumes:
      - spark_data:/opt/bitnami/spark

  airflow-webserver:
    image: apache/airflow:2.7.2
    container_name: opencap_airflow_webserver
    restart: always
    ports:
      - "8085:8080"
    depends_on:
      - airflow-scheduler
      - airflow-db
      - airflow-redis
    environment:
      - AIRFLOW__CORE__EXECUTOR=CeleryExecutor
      - AIRFLOW__CORE__DAGS_ARE_PAUSED_AT_CREATION=True
      - AIRFLOW__CORE__LOAD_EXAMPLES=False
      - AIRFLOW__DATABASE__SQL_ALCHEMY_CONN=postgresql+psycopg2://airflow:airflow@airflow-db:5432/airflow
      - AIRFLOW__CELERY__RESULT_BACKEND=db+postgresql://airflow:airflow@airflow-db:5432/airflow
      - AIRFLOW__CELERY__BROKER_URL=redis://airflow-redis:6379/0
      - AIRFLOW_UID=50000
    volumes:
      - airflow_data:/opt/airflow

  airflow-db:
    image: postgres:15-alpine
    container_name: opencap_airflow_db
    restart: always
    environment:
      POSTGRES_USER: airflow
      POSTGRES_PASSWORD: airflow
      POSTGRES_DB: airflow
    volumes:
      - airflow_db_data:/var/lib/postgresql/data

  airflow-redis:
    image: redis:6-alpine
    container_name: opencap_airflow_redis
    restart: always
    ports:
      - "6379:6379"

  airflow-scheduler:
    image: apache/airflow:2.7.2
    container_name: opencap_airflow_scheduler
    restart: always
    depends_on:
      - airflow-db
      - airflow-redis
    environment:
      - AIRFLOW__CORE__EXECUTOR=CeleryExecutor
      - AIRFLOW__CORE__DAGS_ARE_PAUSED_AT_CREATION=True
      - AIRFLOW__CORE__LOAD_EXAMPLES=False
      - AIRFLOW__DATABASE__SQL_ALCHEMY_CONN=postgresql+psycopg2://airflow:airflow@airflow-db:5432/airflow
      - AIRFLOW__CELERY__RESULT_BACKEND=db+postgresql://airflow:airflow@airflow-db:5432/airflow
      - AIRFLOW__CELERY__BROKER_URL=redis://airflow-redis:6379/0
      - AIRFLOW_UID=50000
    volumes:
      - airflow_data:/opt/airflow

volumes:
  postgres_data:
  minio_data:
  spark_data:
  airflow_data:
  airflow_db_data: