docker-compose.yml

version: '2' 
services:
  
  namenode:
    build: ./components/namenode
    container_name: namenode
    volumes:
      - ./volume-data/namenode:/hadoop/dfs/name
    environment:
      - CLUSTER_NAME=bigdatagrapescluster
    env_file:
      - ./config/hadoop.env
    ports:
      - 50070:50070
      - 8020:8020
  
  datanode1:
    build: ./components/datanode
    container_name: datanode1
    depends_on: 
      - namenode
    volumes:
      - ./volume-data/datanode1:/hadoop/dfs/data
    env_file:
      - ./config/hadoop.env
    ports:
      - 50075:50075
  
  datanode2:
    build: ./components/datanode
    container_name: datanode2
    depends_on: 
      - namenode
    volumes:
      - ./volume-data/datanode2:/hadoop/dfs/data
    env_file:
      - ./config/hadoop.env
    ports:
      - 50076:50075

  spark-master:
    build: ./components/spark-master-notebook
    container_name: spark-master
    volumes:
      - ./volume-data/spark-master-notebook/notebooks:/notebooks
      - ./volume-data/spark-master-notebook/datasets:/datasets
    ports:
      - 8080:8080
      - 7077:7077
      - 9999:9999
    env_file:
      - ./config/hadoop.env

  spark-worker:
    build: ./components/spark-worker
    container_name: spark-worker
    depends_on:
      - spark-master
    environment:
      - SPARK_MASTER=spark://spark-master:7077
    ports:
      - 8081:8081
    env_file:
      - ./config/hadoop.env

  hue:
    build: ./components/hue
    container_name: hue
    ports:
      - 8088:8088
    environment:
      - NAMENODE_HOST=namenode

  git-sync:
    image: openweb/git-sync:0.0.1
    environment:
      GIT_SYNC_REPO: "https://github.com/BigDataGrapes-EU/deliverable-D4.3.git"
      GIT_SYNC_DEST: "/git"
      GIT_SYNC_BRANCH: "master"
      GIT_SYNC_REV: "FETCH_HEAD"
      GIT_SYNC_WAIT: "10"
    restart: always

  price-prediction:
    build: ./volume-data/spark-master-notebook/notebooks/Pilot5-FoodProtection/PricePrediction
    container_name: price-prediction
    depends_on:
      - git-sync