aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorMarcelo Vanzin <vanzin@cloudera.com>2015-04-08 10:14:52 -0700
committerJosh Rosen <joshrosen@databricks.com>2015-04-08 10:14:52 -0700
commitf7e21dd1ec4541be54eb01d8b15cfcc6714feed0 (patch)
treef2a0a3473052c109b04e21498bcf0bb3659c4741
parent15e0d2bd1304db62fad286c1bb687e87c361e16c (diff)
downloadspark-f7e21dd1ec4541be54eb01d8b15cfcc6714feed0.tar.gz
spark-f7e21dd1ec4541be54eb01d8b15cfcc6714feed0.tar.bz2
spark-f7e21dd1ec4541be54eb01d8b15cfcc6714feed0.zip
[SPARK-6506] [pyspark] Do not try to retrieve SPARK_HOME when not needed...
.... In particular, this makes pyspark in yarn-cluster mode fail unless SPARK_HOME is set, when it's not really needed. Author: Marcelo Vanzin <vanzin@cloudera.com> Closes #5405 from vanzin/SPARK-6506 and squashes the following commits: e184507 [Marcelo Vanzin] [SPARK-6506] [pyspark] Do not try to retrieve SPARK_HOME when not needed.
-rw-r--r--python/pyspark/java_gateway.py3
1 files changed, 1 insertions, 2 deletions
diff --git a/python/pyspark/java_gateway.py b/python/pyspark/java_gateway.py
index 0a16cbd8bf..2a5e84a7df 100644
--- a/python/pyspark/java_gateway.py
+++ b/python/pyspark/java_gateway.py
@@ -29,11 +29,10 @@ from pyspark.serializers import read_int
def launch_gateway():
- SPARK_HOME = os.environ["SPARK_HOME"]
-
if "PYSPARK_GATEWAY_PORT" in os.environ:
gateway_port = int(os.environ["PYSPARK_GATEWAY_PORT"])
else:
+ SPARK_HOME = os.environ["SPARK_HOME"]
# Launch the Py4j gateway using Spark's run command so that we pick up the
# proper classpath and settings from spark-env.sh
on_windows = platform.system() == "Windows"