Mario Briggs created TOREE-398:
----------------------------------
Summary: Toree - R : Error in getSparkSession(): SparkSession not
initialized when using 'spark'
Key: TOREE-398
URL: https://issues.apache.org/jira/browse/TOREE-398
Project: TOREE
Issue Type: Bug
Affects Versions: 0.2.0
Reporter: Mario Briggs
Using Toree R interpreter with Spark 2.0.
After kernel is initialized, 'spark' var is available
>>
spark
Out[19]:
Java ref type org.apache.spark.sql.SparkSession id 33
<<
However trying to use it , ends up with this
>>
localDF <- data.frame(name=c("a", "b", "c"), age=c(19, 23, 18))
df <- createDataFrame(spark, localDF)
Out[20]:
Message: Error in getSparkSession(): SparkSession not initialized
<<
Trying to do sparkR.session(), leads to eror that only one SparkContext can be
created
Seem zepellin hit same issue and fixed -
https://issues.apache.org/jira/browse/ZEPPELIN-1260
I see 2.0 fixes are around this line -
https://github.com/apache/incubator-toree/blob/master/sparkr-interpreter/src/main/resources/kernelR/sparkr_runner.R#L122
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
