superset/docker-compose.yml

x-superset-image: &superset-image apachesuperset.docker.scarf.sh/apache/superset:${TAG:-latest-dev}
x-superset-build:
  &superset-build
  context: docker
x-superset-volumes:
  &superset-volumes # /app/pythonpath_docker will be appended to the PYTHONPATH in the final container
  - ./docker:/app/docker
  - superset_home:/app/superset_home

services:
  redis:
    image: redis:7
    container_name: superset_cache
    restart: unless-stopped
    volumes:
      - redis:/data

  superset:
    env_file:
      - path: docker/.env # default
        required: true
      - path: docker/.env-local # optional override
        required: false
    build: *superset-build
    command: ["run-server.sh"]
    container_name: superset_app
    user: "root"
    restart: unless-stopped
    ports:
      - 8088:8088
    depends_on:
      superset-init:
        condition: service_completed_successfully
    volumes: *superset-volumes
    environment:
      SUPERSET_LOG_LEVEL: "${SUPERSET_LOG_LEVEL:-info}"

  superset-init:
    build: *superset-build
    command: ["init_superset.sh"]
    container_name: superset_init
    env_file:
      - path: docker/.env # default
        required: true
      - path: docker/.env-local # optional override
        required: false
    depends_on:
      postgres:
        condition: service_started
      redis:
        condition: service_started
    user: "root"
    volumes: *superset-volumes
    healthcheck:
      disable: true
    environment:
      SUPERSET_LOAD_EXAMPLES: "${SUPERSET_LOAD_EXAMPLES:-no}"
      SUPERSET_LOG_LEVEL: "${SUPERSET_LOG_LEVEL:-info}"

  superset-worker:
    build: *superset-build
    container_name: superset_worker
    command: 
      - celery
      - --app=superset.tasks.celery_app:app 
      - worker 
      - -O 
      - fair 
      - -l
      - INFO 
      - --concurrency=2
    env_file:
      - path: docker/.env # default
        required: true
      - path: docker/.env-local # optional override
        required: false
    restart: unless-stopped
    depends_on:
      superset-init:
        condition: service_completed_successfully
    user: "root"
    volumes: *superset-volumes
    healthcheck:
      test:
        [
          "CMD-SHELL",
          "celery -A superset.tasks.celery_app:app inspect ping -d celery@$$HOSTNAME",
        ]
    environment:
      SUPERSET_LOG_LEVEL: "${SUPERSET_LOG_LEVEL:-info}"

  superset-worker-beat:
    build: *superset-build
    container_name: superset_worker_beat
    command: 
    - celery
    - --app=superset.tasks.celery_app:app
    - beat
    - --pidfile
    - /tmp/celerybeat.pid
    - -l
    - INFO
    - -s
    - /tmp/celerybeat-schedule
    env_file:
      - path: docker/.env # default
        required: true
      - path: docker/.env-local # optional override
        required: false
    restart: unless-stopped
    depends_on:
      superset-init:
        condition: service_completed_successfully
    user: "root"
    volumes: *superset-volumes
    healthcheck:
      disable: true
    environment:
      SUPERSET_LOG_LEVEL: "${SUPERSET_LOG_LEVEL:-info}"

  postgres:
    image: postgres:17-alpine
    container_name: postgres
    hostname: postgres
    env_file:
      - path: docker/.env # default
        required: true
      - path: docker/.env-local # optional override
        required: false
    environment:
      POSTGRES_USER: ${POSTGRES_USER:-postgres}
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:?database password required}
      POSTGRES_DB: ${POSTGRES_DB:-app}
      PGDATA: /var/lib/postgresql/data/pgdata
    volumes:
      - postgres_data:/var/lib/postgresql/data/pgdata  # Persistent data storage
      - ./init:/docker-entrypoint-initdb.d/  # Initialization scripts
      - ./backup:/backup  # Backup location
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER:-postgres}"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 30s
    shm_size: 256mb
    command:
      - "postgres"
      - "-c"
      - "max_connections=200"
      - "-c"
      - "shared_buffers=1GB"
      - "-c"
      - "effective_cache_size=3GB"
      - "-c"
      - "work_mem=16MB"
      - "-c"
      - "maintenance_work_mem=512MB"
      - "-c"
      - "random_page_cost=1.1"
      - "-c"
      - "temp_file_limit=10GB"
      - "-c"
      - "log_min_duration_statement=200ms"
      - "-c"
      - "idle_in_transaction_session_timeout=10s"
      - "-c"
      - "lock_timeout=1s"
      - "-c"
      - "statement_timeout=60s"
      - "-c"
      - "shared_preload_libraries=pg_stat_statements"
      - "-c"
      - "pg_stat_statements.max=10000"
      - "-c"
      - "pg_stat_statements.track=all"

  pgbackups:
    image: prodrigestivill/postgres-backup-local
    container_name: postgres_backup
    restart: unless-stopped
    volumes:
      - ./data/backup:/backups
    environment:
      - POSTGRES_HOST=postgres
      - POSTGRES_DB=${POSTGRES_DB:-app}
      - POSTGRES_USER=${POSTGRES_USER:-postgres}
      - POSTGRES_PASSWORD=${POSTGRES_PASSWORD:?database password required}
      - SCHEDULE=@daily
      - BACKUP_KEEP_DAYS=7
      - BACKUP_KEEP_WEEKS=4
      - BACKUP_KEEP_MONTHS=6
    depends_on:
      - postgres

volumes:
  postgres_data:
    name: postgres_data
  superset_home:
    external: false
  redis:
    external: false