[SPARK-10916] [YARN] Set perm gen size when launching containers on YARN.

This makes YARN containers behave like all other processes launched by
Spark, which launch with a default perm gen size of 256m unless
overridden by the user (or not needed by the vm).

Author: Marcelo Vanzin <vanzin@cloudera.com>

Closes #8970 from vanzin/SPARK-10916.

6 files changed, 48 insertions, 27 deletions

diff --git a/core/src/main/scala/org/apache/spark/launcher/WorkerCommandBuilder.scala b/core/src/main/scala/org/apache/spark/launcher/WorkerCommandBuilder.scala
index 0c096656f9..a2add61617 100644
--- a/core/src/main/scala/org/apache/spark/launcher/WorkerCommandBuilder.scala
+++ b/core/src/main/scala/org/apache/spark/launcher/WorkerCommandBuilder.scala
@@ -40,7 +40,7 @@ private[spark] class WorkerCommandBuilder(sparkHome: String, memoryMb: Int, comm
     cmd.add(s"-Xms${memoryMb}M")
     cmd.add(s"-Xmx${memoryMb}M")
     command.javaOpts.foreach(cmd.add)
-    addPermGenSizeOpt(cmd)
+    CommandBuilderUtils.addPermGenSizeOpt(cmd)
     addOptionString(cmd, getenv("SPARK_JAVA_OPTS"))
     cmd
   }
diff --git a/launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java b/launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java
index 0a237ee73b..610e8bdaaa 100644
--- a/launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java
+++ b/launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java
@@ -116,29 +116,6 @@ abstract class AbstractCommandBuilder {
     return cmd;
   }
 
-  /**
-   * Adds the default perm gen size option for Spark if the VM requires it and the user hasn't
-   * set it.
-   */
-  void addPermGenSizeOpt(List<String> cmd) {
-    // Don't set MaxPermSize for IBM Java, or Oracle Java 8 and later.
-    if (getJavaVendor() == JavaVendor.IBM) {
-      return;
-    }
-    String[] version = System.getProperty("java.version").split("\\.");
-    if (Integer.parseInt(version[0]) > 1 || Integer.parseInt(version[1]) > 7) {
-      return;
-    }
-
-    for (String arg : cmd) {
-      if (arg.startsWith("-XX:MaxPermSize=")) {
-        return;
-      }
-    }
-
-    cmd.add("-XX:MaxPermSize=256m");
-  }
-
   void addOptionString(List<String> cmd, String options) {
     if (!isEmpty(options)) {
       for (String opt : parseOptionString(options)) {
diff --git a/launcher/src/main/java/org/apache/spark/launcher/CommandBuilderUtils.java b/launcher/src/main/java/org/apache/spark/launcher/CommandBuilderUtils.java
index a16c0d2b5c..d30c2ec5f8 100644
--- a/launcher/src/main/java/org/apache/spark/launcher/CommandBuilderUtils.java
+++ b/launcher/src/main/java/org/apache/spark/launcher/CommandBuilderUtils.java
@@ -313,4 +313,27 @@ class CommandBuilderUtils {
     return quoted.append('"').toString();
   }
 
+  /**
+   * Adds the default perm gen size option for Spark if the VM requires it and the user hasn't
+   * set it.
+   */
+  static void addPermGenSizeOpt(List<String> cmd) {
+    // Don't set MaxPermSize for IBM Java, or Oracle Java 8 and later.
+    if (getJavaVendor() == JavaVendor.IBM) {
+      return;
+    }
+    String[] version = System.getProperty("java.version").split("\\.");
+    if (Integer.parseInt(version[0]) > 1 || Integer.parseInt(version[1]) > 7) {
+      return;
+    }
+
+    for (String arg : cmd) {
+      if (arg.startsWith("-XX:MaxPermSize=")) {
+        return;
+      }
+    }
+
+    cmd.add("-XX:MaxPermSize=256m");
+  }
+
 }
diff --git a/yarn/src/main/scala/org/apache/spark/deploy/yarn/Client.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/Client.scala
index 8c53c24a79..f8748ef658 100644
--- a/yarn/src/main/scala/org/apache/spark/deploy/yarn/Client.scala
+++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/Client.scala
@@ -54,8 +54,9 @@ import org.apache.hadoop.yarn.conf.YarnConfiguration
 import org.apache.hadoop.yarn.exceptions.ApplicationNotFoundException
 import org.apache.hadoop.yarn.util.Records
 
-import org.apache.spark.deploy.SparkHadoopUtil
 import org.apache.spark.{Logging, SecurityManager, SparkConf, SparkContext, SparkException}
+import org.apache.spark.deploy.SparkHadoopUtil
+import org.apache.spark.launcher.YarnCommandBuilderUtils
 import org.apache.spark.util.Utils
 
 private[spark] class Client(
@@ -730,6 +731,7 @@ private[spark] class Client(
 
     // For log4j configuration to reference
     javaOpts += ("-Dspark.yarn.app.container.log.dir=" + ApplicationConstants.LOG_DIR_EXPANSION_VAR)
+    YarnCommandBuilderUtils.addPermGenSizeOpt(javaOpts)
 
     val userClass =
       if (isClusterMode) {
diff --git a/yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala
index 9abd09b3cc..2232ffba47 100644
--- a/yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala
+++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala
@@ -38,6 +38,7 @@ import org.apache.hadoop.yarn.ipc.YarnRPC
 import org.apache.hadoop.yarn.util.{ConverterUtils, Records}
 
 import org.apache.spark.{Logging, SecurityManager, SparkConf, SparkException}
+import org.apache.spark.launcher.YarnCommandBuilderUtils
 import org.apache.spark.network.util.JavaUtils
 import org.apache.spark.util.Utils
 
@@ -199,6 +200,7 @@ class ExecutorRunnable(
 
     // For log4j configuration to reference
     javaOpts += ("-Dspark.yarn.app.container.log.dir=" + ApplicationConstants.LOG_DIR_EXPANSION_VAR)
+    YarnCommandBuilderUtils.addPermGenSizeOpt(javaOpts)
 
     val userClassPath = Client.getUserClasspath(sparkConf).flatMap { uri =>
       val absPath =
diff --git a/yarn/src/main/scala/org/apache/spark/launcher/YarnCommandBuilderUtils.scala b/yarn/src/main/scala/org/apache/spark/launcher/YarnCommandBuilderUtils.scala
index 3ac36ef0a1..7d246bf407 100644
--- a/yarn/src/main/scala/org/apache/spark/launcher/YarnCommandBuilderUtils.scala
+++ b/yarn/src/main/scala/org/apache/spark/launcher/YarnCommandBuilderUtils.scala
@@ -17,11 +17,28 @@
 
 package org.apache.spark.launcher
 
+import scala.collection.JavaConverters._
+import scala.collection.mutable.ListBuffer
+
 /**
- * Exposes needed methods
+ * Exposes methods from the launcher library that are used by the YARN backend.
  */
 private[spark] object YarnCommandBuilderUtils {
-  def quoteForBatchScript(arg: String) : String = {
+
+  def quoteForBatchScript(arg: String): String = {
     CommandBuilderUtils.quoteForBatchScript(arg)
   }
+
+  /**
+   * Adds the perm gen configuration to the list of java options if needed and not yet added.
+   *
+   * Note that this method adds the option based on the local JVM version; if the node where
+   * the container is running has a different Java version, there's a risk that the option will
+   * not be added (e.g. if the AM is running Java 8 but the container's node is set up to use
+   * Java 7).
+   */
+  def addPermGenSizeOpt(args: ListBuffer[String]): Unit = {
+    CommandBuilderUtils.addPermGenSizeOpt(args.asJava)
+  }
+
 }