AnsweredAssumed Answered

No FileSystem for scheme: maprfs in Spark job

Question asked by tkrol1 on Mar 13, 2018
Latest reply on Mar 16, 2018 by maprcommunity



I hope you will be able to help me. Basically I was able to query Hive tables or access files on MaprFS from my Spark jobs with no issues. 


But after adding two new dependencies to my build.sbt file for MapRDB:


"com.mapr.db" % "maprdb" % "5.2.0-mapr"
"com.mapr.db" % "maprdb-spark" % "5.2.1-mapr",


I am getting following error:

18/03/13 16:56:26 INFO SparkContext: Created broadcast 0 from show at poc_scheduler.scala:19
Exception in thread "main" No FileSystem for scheme: maprfs
at org.apache.hadoop.fs.FileSystem.getFileSystemClass(
at org.apache.hadoop.fs.FileSystem.createFileSystem(
at org.apache.hadoop.fs.FileSystem.access$200(


When I remove them, then I can access Hive or Maprfs with no issues.


Any idea what would be the issue here?