diff --git a/spark-pyspark-kafka/Dockerfile b/spark-pyspark-kafka/Dockerfile index ca5f2df..a056eba 100644 --- a/spark-pyspark-kafka/Dockerfile +++ b/spark-pyspark-kafka/Dockerfile @@ -65,7 +65,7 @@ # More PySpark stuff ENV PYTHONPATH="$SPARK_HOME/python:$SPARK_HOME/python/lib/py4j-0.10.7-src.zip" \ - PYSPARK_SUBMIT_ARGS="--master $SPARK_MASTER pyspark-shell --py-files $SPARK_HOME/jars/graphframes-$GRAPHFRAMES_VERSION.jar" \ + PYSPARK_SUBMIT_ARGS="--master $SPARK_MASTER pyspark-shell" \ PYTHONSTARTUP="$SPARK_HOME/python/pyspark/shell.py" # Coalescing the RUNs saves quite a significant amount of space