I was able to resolve the AccessControLException by using "sudo -u hdfs" and pushing my data file to hdfs (using the full path) as below:   sudo -u hdfs spark-submit --class com.cloudera.sparkwordcount.JavaWordCount --master local target/sparkwordcount-0.0.1-SNAPSHOT.jar /user/cloudera/data/inputfile.txt 2   I thought it would then be simple to switch the user to 'cloudera' (e.g. "sudo -u cloudera"), since I was putting the data file under the cloudera user path it seemed reasonable, but gave me the same exception. Not sure why? ... View more