Configure data-fabric Client Node to Run Spark Applications

When Spark runs on YARN, data-fabric client nodes require the hadoop-yarn-server-web-proxy JAR file to run Spark applications. On Windows, the client node also requires an update to the SPARK_DIST_CLASSPATH. A data-fabric client node (a node with the mapr-client package, but without mapr-core packages) is also known as an edge node.

The mapr-client package does not include the JAR file required to run Spark applications. Therefore, you must copy the /opt/mapr/hadoop/hadoop-2.x.x/share/hadoop/yarn/hadoop-yarn-server-web-proxy-<version>.jar from a data-fabric cluster node to the same location on the data-fabric client node from which you want to run the Spark application.