spark
org.apache.spark.deploy.history.SparkFSFetcher
hdfs://spark-history
100
webhdfs://10.50.0.247:50070/spark-history
/spark-history
#the values specified in namenode_addresses will be used for obtaining spark logs. The cluster configuration will be ignored.
ip-10-50-0-247.ec2.internal:50070