diff options
author | Patrick Wendell <pwendell@gmail.com> | 2014-04-22 19:22:06 -0700 |
---|---|---|
committer | Patrick Wendell <pwendell@gmail.com> | 2014-04-22 19:22:06 -0700 |
commit | 995fdc96bcd2c540804401eaab009a777d7d7aa9 (patch) | |
tree | c57cf54c46a3a33cce065682fb26422a42c0ca56 | |
parent | ea8cea82a02099bb66f1e77b757e4d96cc31d6e2 (diff) | |
download | spark-995fdc96bcd2c540804401eaab009a777d7d7aa9.tar.gz spark-995fdc96bcd2c540804401eaab009a777d7d7aa9.tar.bz2 spark-995fdc96bcd2c540804401eaab009a777d7d7aa9.zip |
Assorted clean-up for Spark-on-YARN.
In particular when the HADOOP_CONF_DIR is not not specified.
Author: Patrick Wendell <pwendell@gmail.com>
Closes #488 from pwendell/hadoop-cleanup and squashes the following commits:
fe95f13 [Patrick Wendell] Changes based on Andrew's feeback
18d09c1 [Patrick Wendell] Review comments from Andrew
17929cc [Patrick Wendell] Assorted clean-up for Spark-on-YARN.
4 files changed, 18 insertions, 10 deletions
diff --git a/conf/spark-env.sh.template b/conf/spark-env.sh.template index 177a21cc03..f906be611a 100755 --- a/conf/spark-env.sh.template +++ b/conf/spark-env.sh.template @@ -5,6 +5,7 @@ # Options read when launching programs locally with # ./bin/run-example or ./bin/spark-submit +# - HADOOP_CONF_DIR, to point Spark towards Hadoop configuration files # - SPARK_LOCAL_IP, to set the IP address Spark binds to on this node # - SPARK_PUBLIC_DNS, to set the public dns name of the driver program # - SPARK_CLASSPATH, default classpath entries to append @@ -17,6 +18,7 @@ # - MESOS_NATIVE_LIBRARY, to point to your libmesos.so if you use Mesos # Options read in YARN client mode +# - HADOOP_CONF_DIR, to point Spark towards Hadoop configuration files # - SPARK_EXECUTOR_INSTANCES, Number of workers to start (Default: 2) # - SPARK_EXECUTOR_CORES, Number of cores for the workers (Default: 1). # - SPARK_EXECUTOR_MEMORY, Memory per Worker (e.g. 1000M, 2G) (Default: 1G) diff --git a/core/src/main/scala/org/apache/spark/deploy/SparkSubmitArguments.scala b/core/src/main/scala/org/apache/spark/deploy/SparkSubmitArguments.scala index 02502adfbd..cc976565cc 100644 --- a/core/src/main/scala/org/apache/spark/deploy/SparkSubmitArguments.scala +++ b/core/src/main/scala/org/apache/spark/deploy/SparkSubmitArguments.scala @@ -116,6 +116,15 @@ private[spark] class SparkSubmitArguments(args: Array[String]) { if (args.length == 0) printUsageAndExit(-1) if (primaryResource == null) SparkSubmit.printErrorAndExit("Must specify a primary resource") if (mainClass == null) SparkSubmit.printErrorAndExit("Must specify a main class with --class") + + if (master.startsWith("yarn")) { + val hasHadoopEnv = sys.env.contains("HADOOP_CONF_DIR") || sys.env.contains("YARN_CONF_DIR") + val testing = sys.env.contains("SPARK_TESTING") + if (!hasHadoopEnv && !testing) { + throw new Exception(s"When running with master '$master' " + + "either HADOOP_CONF_DIR or YARN_CONF_DIR must be set in the environment.") + } + } } override def toString = { diff --git a/docs/hadoop-third-party-distributions.md b/docs/hadoop-third-party-distributions.md index de6a2b0a43..454877a7fa 100644 --- a/docs/hadoop-third-party-distributions.md +++ b/docs/hadoop-third-party-distributions.md @@ -110,10 +110,5 @@ The location of these configuration files varies across CDH and HDP versions, bu a common location is inside of `/etc/hadoop/conf`. Some tools, such as Cloudera Manager, create configurations on-the-fly, but offer a mechanisms to download copies of them. -There are a few ways to make these files visible to Spark: - -* You can copy these files into `$SPARK_HOME/conf` and they will be included in Spark's -classpath automatically. -* If you are running Spark on the same nodes as Hadoop _and_ your distribution includes both -`hdfs-site.xml` and `core-site.xml` in the same directory, you can set `HADOOP_CONF_DIR` -in `$SPARK_HOME/spark-env.sh` to that directory. +To make these files visible to Spark, set `HADOOP_CONF_DIR` in `$SPARK_HOME/spark-env.sh` +to a location containing the configuration files. diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala b/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala index d264ecbeaf..b403292d9c 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala +++ b/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala @@ -375,9 +375,11 @@ object ClientBase { val classpathEntries = Option(conf.getStrings( YarnConfiguration.YARN_APPLICATION_CLASSPATH)).getOrElse( getDefaultYarnApplicationClasspath()) - for (c <- classpathEntries) { - YarnSparkHadoopUtil.addToEnvironment(env, Environment.CLASSPATH.name, c.trim, - File.pathSeparator) + if (classpathEntries != null) { + for (c <- classpathEntries) { + YarnSparkHadoopUtil.addToEnvironment(env, Environment.CLASSPATH.name, c.trim, + File.pathSeparator) + } } val mrClasspathEntries = Option(conf.getStrings( |