diff options
author | Patrick Wendell <pwendell@gmail.com> | 2014-01-12 17:01:13 -0800 |
---|---|---|
committer | Patrick Wendell <pwendell@gmail.com> | 2014-01-12 17:01:13 -0800 |
commit | 074f50232fd8d8cf05eb88db0ac6f03f61452810 (patch) | |
tree | 27dd81c6a94b9b214084d7ede44eabceac70d8da | |
parent | 82e2b92c6dfebc2f570dbef366696c150548be55 (diff) | |
parent | cfb1e6c13cd9021e29abe38edceded4b988fbe5b (diff) | |
download | spark-074f50232fd8d8cf05eb88db0ac6f03f61452810.tar.gz spark-074f50232fd8d8cf05eb88db0ac6f03f61452810.tar.bz2 spark-074f50232fd8d8cf05eb88db0ac6f03f61452810.zip |
Merge pull request #396 from pwendell/executor-env
Setting load defaults to true in executor
This preserves the behavior in earlier releases. If properties are set for the executors via `spark-env.sh` on the slaves, then they should take precedence over spark defaults. This is useful for if system administrators are setting properties for a standalone cluster, such as shuffle locations.
/cc @andrewor14 who initially reported this issue.
-rw-r--r-- | core/src/main/scala/org/apache/spark/executor/Executor.scala | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/core/src/main/scala/org/apache/spark/executor/Executor.scala b/core/src/main/scala/org/apache/spark/executor/Executor.scala index a7b2328a02..7f31d7e6f8 100644 --- a/core/src/main/scala/org/apache/spark/executor/Executor.scala +++ b/core/src/main/scala/org/apache/spark/executor/Executor.scala @@ -57,7 +57,7 @@ private[spark] class Executor( Utils.setCustomHostname(slaveHostname) // Set spark.* properties from executor arg - val conf = new SparkConf(false) + val conf = new SparkConf(true) conf.setAll(properties) // If we are in yarn mode, systems can have different disk layouts so we must set it |