this has been resolved by adding the following spark_conf (not thru --conf)

 "spark.hadoop.fs.file.impl": "org.apache.hadoop.fs.LocalFileSystem"

example:

------

"new_cluster": {

"spark_version": "9.1.x-scala2.12",

...

"spark_conf": {

"spark.hadoop.fs.file.impl": "org.apache.hadoop.fs.LocalFileSystem"

}

},

"spark_submit_task": {

"parameters": [

"--packages",

"org.apache.spark:spark-sql-kafka-0-10_2.12:3.0.1",

...

------------