Py4JError: SparkConf does not exist in the JVM

Question

I am running pyspark but it can be unstable at times. There are couple of times it crashes at this command

spark_conf = SparkConf()

with the following error message

 File "/home/user1/spark/spark-1.5.1-bin-hadoop2.6/python/pyspark/conf.py", line 106, in __init__ self._jconf = _jvm.SparkConf(loadDefaults) File "/home/user1/spark/spark-1.5.1-bin-hadoop2.6/python/lib/py4j-0.8.2.1-src.zip/py4j/java_gateway.py", line 772, in __getattr__ raise Py4JError('{0} does not exist in the JVM'.format(name)) Py4JError: SparkConf does not exist in the JVM

Any idea what is the problem? Thank you for your help!

Shawn Guo · Accepted Answer · 2015-12-11 05:20:55Z

1

SparkConf does not exist in the pyspark context, try:

from pyspark import SparkConf

in the pyspark console or code.

answered Dec 11, 2015 at 5:20

Shawn Guo

3,2283 gold badges24 silver badges28 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

Michael Over a year ago

Thanks, I found the problem. After closing a SparkContext, I will get the above error message when I try to call SparkConf() and initialize a new SparkContext again.

thebeancounter Over a year ago

What do you mean? I have the same error when using from pyspark import SparkContext and then sc = SparkContext()

Jatin Patel - JP Over a year ago

@Michael how to closed spark session?

Purushothaman Srikanth Over a year ago

@JatinPatel-JP Use spark.stop() to stop a SparkSession. Here spark is the name of the SparkSession object.

Collectives™ on Stack Overflow

Py4JError: SparkConf does not exist in the JVM

1 Answer 1

4 Comments

Hot Network Questions