score:0

You run your application on a Spark cluster, so I guess the problem is that your code cannot find the users.txt file. Try using a path that can be reached independent of the current work directory and/or using local[3] as the master parameter.

score:1

You need not only to compile project, but package it as assembled jar file with all dependencies. It can be a little trick to remove all dependencies conflicts. You can take a look at my sample project for Spark that uses Spark-Streaming, it can be helpful to start with

https://github.com/pellucidanalytics/tweet-driven-comparable-companies

especially to Dependencies exclusion https://github.com/pellucidanalytics/tweet-driven-comparable-companies/blob/master/project/Dependency.scala

And read some articles:

http://prabstechblog.blogspot.com/2014/04/creating-single-jar-for-spark-project.html http://eugenezhulenev.com/blog/2014/11/20/twitter-analytics-with-spark/ http://eugenezhulenev.com/blog/2014/10/18/run-tests-in-standalone-spark-cluster/


Related Query

More Query from same tag