Copy the input file to HDFS. No need to copy WordCount-1.0-SNAPSHOT.jar to HDFS.
The command to run:
hadoop jar /tmp/WordCount-1.0-SNAPSHOT.jar com.yufeigu.WordCount /tmp/input.txt /tmp/output
Check the result:
hadoop fs -cat /tmp/output/part-r-00001
You can find the WordCount-1.0-SNAPSHOT.jar under local nodemanager directory, but it is renamed to job.jar instead of original name.