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