Natalie created SPARK-20561:
-------------------------------
Summary: Running SparkR with no RHive installed in secured
environment
Key: SPARK-20561
URL: https://issues.apache.org/jira/browse/SPARK-20561
Project: Spark
Issue Type: Question
Components: Examples, Input/Output
Affects Versions: 2.1.0
Environment: Hadoop, Spark
Reporter: Natalie
I need to start running data mining analysis in secured environment (IP, Port,
and database name are given), where Spark runs on hive tables. So I have
installed R, SparkR, dplyr, and some other r libraries.
Now I understand that I need to point sparkR to that database(with
IP/Port/Name).
What should be my R code?
I start with evoking R,
then SparkR library.
Next I right sc<-sparkR.init()
it tells me immediately that spark-submit command:not found
Do I need to have RHive installed first?
Or should I actually point somehow to spark library and to that database?
I couldn't find any documentation on that.
Thank you
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
