diff options
Diffstat (limited to 'python/pyspark/java_gateway.py')
| -rw-r--r-- | python/pyspark/java_gateway.py | 12 |
1 files changed, 8 insertions, 4 deletions
diff --git a/python/pyspark/java_gateway.py b/python/pyspark/java_gateway.py index e503fb7621..26fbe0f080 100644 --- a/python/pyspark/java_gateway.py +++ b/python/pyspark/java_gateway.py @@ -17,6 +17,7 @@ import os import sys +import signal from subprocess import Popen, PIPE from threading import Thread from py4j.java_gateway import java_import, JavaGateway, GatewayClient @@ -28,9 +29,12 @@ SPARK_HOME = os.environ["SPARK_HOME"] def launch_gateway(): # Launch the Py4j gateway using Spark's run command so that we pick up the # proper classpath and SPARK_MEM settings from spark-env.sh - command = [os.path.join(SPARK_HOME, "run"), "py4j.GatewayServer", + command = [os.path.join(SPARK_HOME, "spark-class"), "py4j.GatewayServer", "--die-on-broken-pipe", "0"] - proc = Popen(command, stdout=PIPE, stdin=PIPE) + # Don't send ctrl-c / SIGINT to the Java gateway: + def preexec_function(): + signal.signal(signal.SIGINT, signal.SIG_IGN) + proc = Popen(command, stdout=PIPE, stdin=PIPE, preexec_fn=preexec_function) # Determine which ephemeral port the server started on: port = int(proc.stdout.readline()) # Create a thread to echo output from the GatewayServer, which is required @@ -49,7 +53,7 @@ def launch_gateway(): # Connect to the gateway gateway = JavaGateway(GatewayClient(port=port), auto_convert=False) # Import the classes used by PySpark - java_import(gateway.jvm, "spark.api.java.*") - java_import(gateway.jvm, "spark.api.python.*") + java_import(gateway.jvm, "org.apache.spark.api.java.*") + java_import(gateway.jvm, "org.apache.spark.api.python.*") java_import(gateway.jvm, "scala.Tuple2") return gateway |