vars.yml

---
hadoop_site_configuration:
  core-site:
    fs.default.name: hdfs://master:9000

  hdfs-site:
    dfs.namenode.name.dir: "/home/{{ ansible_user }}/data/nameNode"
    dfs.datanode.data.dir: "/home/{{ ansible_user }}/data/dataNode"
    dfs.replication: 1

  mapred-site:
    mapreduce.framework.name: yarn
    yarn.app.mapreduce.am.env: HADOOP_MAPRED_HOME=$HADOOP_HOME
    mapreduce.map.env: HADOOP_MAPRED_HOME=$HADOOP_HOME
    mapreduce.reduce.env: HADOOP_MAPRED_HOME=$HADOOP_HOME
    yarn.app.mapreduce.am.resource.mb: 512
    mapreduce.map.memory.mb: 256
    mapreduce.reduce.memory.mb: 256

  yarn-site:
    yarn.acl.enable: "false"
    yarn.resourcemanager.hostname: master
    yarn.nodemanager.aux-services: mapreduce_shuffle
    yarn.nodemanager.resource.memory-mb: 1536
    yarn.scheduler.maximum-allocation-mb: 1536
    yarn.scheduler.minimum-allocation-mb: 128
    yarn.nodemanager.vmem-check-enabled: "false"
    yarn.nodemanager.resource.detect-hardware-capabilities: "true"

# spark
spark_master: spark://master:7077
spark_driver_memory: 512m
spark_executor_memory: 512m
spark_history_update_interval: 10s