--- # ПРЕСЕТ: Big Data кластер (12-18 машин) # # Описание: Кластер для обработки больших данных с различными компонентами # - 3 Hadoop NameNode (активный/резервный/журнал) - метаданные # - 3 Hadoop DataNode - хранение данных # - 1 Hive Metastore - метаданные таблиц # - 1 Spark Master - управление Spark # - 2 Spark Worker - обработка данных # - 1 Kafka Broker - потоковая обработка # - 1 Zookeeper - координация сервисов # - 1 Elasticsearch - поиск и аналитика # - 1 Kibana - визуализация данных # # Использование: make lab-test SCENARIO=universal LAB_SPEC=molecule/presets/bigdata.yml # # Автор: Сергей Антропов # Сайт: https://devops.org.ru docker_network: labnet images: debian: "ghcr.io/ansible-community/molecule-ubuntu-systemd:jammy" rhel: "quay.io/centos/centos:stream9-systemd" systemd_defaults: privileged: true command: "/sbin/init" volumes: - "/sys/fs/cgroup:/sys/fs/cgroup:ro" tmpfs: - "/run" - "/run/lock" capabilities: - "SYS_ADMIN" hosts: - name: namenode1 group: namenodes family: debian publish: - "9870:9870" - name: namenode2 group: namenodes family: rhel publish: - "9871:9870" - name: namenode3 group: namenodes family: debian publish: - "9872:9870" - name: datanode1 group: datanodes family: rhel publish: - "9864:9864" - name: datanode2 group: datanodes family: debian publish: - "9865:9864" - name: datanode3 group: datanodes family: rhel publish: - "9866:9864" - name: hive-metastore group: metastores family: debian publish: - "9083:9083" - name: spark-master group: spark family: rhel publish: - "8080:8080" - name: spark-worker1 group: spark family: debian publish: - "8081:8080" - name: spark-worker2 group: spark family: rhel publish: - "8082:8080" - name: kafka group: streaming family: debian publish: - "9092:9092" - name: zookeeper group: coordination family: rhel publish: - "2181:2181" - name: elasticsearch group: search family: debian publish: - "9200:9200" - name: kibana group: visualization family: rhel publish: - "5601:5601"