先在idea中maven install 打jar包
[houzhq@agent sparkjob]$ spark2-submit --class Sparkjob.count --name countownerjob --num-executors 5 --executor-memory 6G --driver-memory 6G --master yarn --deploy-mode cluster /home/houzhq/sparkjob/SparkJob-1.0-SNAPSHOT.jar /user/houzhq/trips.csv /user/houzhq/output