[SPARK-2778] [yarn] Add yarn integration tests.

pom.xml

@@ -710,6 +710,35 @@
           </exclusion>
         </exclusions>
       </dependency>
+      <dependency>
+        <groupId>org.apache.hadoop</groupId>
+        <artifactId>hadoop-yarn-server-tests</artifactId>
+        <version>${yarn.version}</version>
+        <classifier>tests</classifier>
+        <scope>test</scope>
+        <exclusions>
+          <exclusion>
+            <groupId>asm</groupId>
+            <artifactId>asm</artifactId>
+          </exclusion>
+          <exclusion>
+            <groupId>org.ow2.asm</groupId>
+            <artifactId>asm</artifactId>
+          </exclusion>
+          <exclusion>
+            <groupId>org.jboss.netty</groupId>
+            <artifactId>netty</artifactId>
+          </exclusion>
+          <exclusion>
+            <groupId>javax.servlet</groupId>
+            <artifactId>servlet-api</artifactId>
+          </exclusion>
+          <exclusion>
+            <groupId>commons-logging</groupId>
+            <artifactId>commons-logging</artifactId>
+          </exclusion>
+        </exclusions>
+      </dependency>
       <dependency>
         <groupId>org.apache.hadoop</groupId>
         <artifactId>hadoop-yarn-server-web-proxy</artifactId>
@@ -1182,7 +1211,7 @@
         <dependency>
           <groupId>org.apache.zookeeper</groupId>
           <artifactId>zookeeper</artifactId>
-          <version>3.4.5-mapr-1406</version> 
+          <version>3.4.5-mapr-1406</version>
         </dependency>
       </dependencies>
     </profile>

yarn/pom.xml

@@ -126,15 +126,14 @@
         <configuration>
           <environmentVariables>
             <SPARK_HOME>${basedir}/../..</SPARK_HOME>
-            <SPARK_CLASSPATH>${spark.classpath}</SPARK_CLASSPATH>
           </environmentVariables>
         </configuration>
       </plugin>
     </plugins>
 
     <outputDirectory>target/scala-${scala.binary.version}/classes</outputDirectory>
     <testOutputDirectory>target/scala-${scala.binary.version}/test-classes</testOutputDirectory>
-  
+
     <resources>
       <resource>
         <directory>../common/src/main/resources</directory>

yarn/stable/pom.xml

@@ -32,4 +32,13 @@
   <packaging>jar</packaging>
   <name>Spark Project YARN Stable API</name>
 
+  <dependencies>
+    <dependency>
+      <groupId>org.apache.hadoop</groupId>
+      <artifactId>hadoop-yarn-server-tests</artifactId>
+      <classifier>tests</classifier>
+      <scope>test</scope>
+    </dependency>
+  </dependencies>
+
 </project>

yarn/stable/src/test/resources/log4j.properties

@@ -0,0 +1,28 @@
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements.  See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License.  You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+# Set everything to be logged to the file core/target/unit-tests.log
+log4j.rootCategory=INFO, file
+log4j.appender.file=org.apache.log4j.FileAppender
+log4j.appender.file.append=false
+log4j.appender.file.file=target/unit-tests.log
+log4j.appender.file.layout=org.apache.log4j.PatternLayout
+log4j.appender.file.layout.ConversionPattern=%d{yy/MM/dd HH:mm:ss.SSS} %p %c{1}: %m%n
+
+# Ignore messages below warning level from Jetty, because it's a bit verbose
+log4j.logger.org.eclipse.jetty=WARN
+org.eclipse.jetty.LEVEL=WARN

yarn/stable/src/test/scala/org/apache/spark/deploy/yarn/YarnClusterSuite.scala

@@ -0,0 +1,133 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.deploy.yarn
+
+import java.io.File
+
+import scala.collection.JavaConversions._
+
+import com.google.common.base.Charsets
+import com.google.common.io.Files
+import org.scalatest.{BeforeAndAfterAll, FunSuite, Matchers}
+
+import org.apache.hadoop.yarn.conf.YarnConfiguration
+import org.apache.hadoop.yarn.server.MiniYARNCluster
+
+import org.apache.spark.{Logging, SparkConf, SparkContext}
+import org.apache.spark.deploy.SparkHadoopUtil
+import org.apache.spark.util.Utils
+
+class YarnClusterSuite extends FunSuite with BeforeAndAfterAll with Matchers {
+
+  private var oldConf: Map[String, String] = _
+  private var yarnCluster: MiniYARNCluster = _
+  private var tempDir: File = _
+  private var fakeSparkJar: File = _
+
+  override def beforeAll() {
+    tempDir = Utils.createTempDir()
+
+    yarnCluster = new MiniYARNCluster(getClass().getName(), 1, 1, 1)
+    yarnCluster.init(new YarnConfiguration())
+    yarnCluster.start()
+
+    oldConf = sys.props.toMap
+    yarnCluster.getConfig().foreach { e =>
+      sys.props += ("spark.hadoop." + e.getKey() -> e.getValue())
+    }
+
+    fakeSparkJar = File.createTempFile("sparkJar", null, tempDir)
+    sys.props += ("spark.yarn.jar" -> ("local:" + fakeSparkJar.getAbsolutePath()))
+    sys.props += ("spark.executor.instances" -> "1")
+    sys.props += ("spark.driver.extraClassPath" -> sys.props("java.class.path"))
+    sys.props += ("spark.executor.extraClassPath" -> sys.props("java.class.path"))
+
+    super.beforeAll()
+  }
+
+  override def afterAll() {
+    yarnCluster.stop()
+
+    val sysProps = sys.props.map { case (k, v) => (k, v) }
+    sysProps.foreach { case (k, v) =>
+      if (k.startsWith("spark.")) {
+        sys.props -= k
+      }
+    }
+
+    sys.props ++= oldConf
+
+    super.afterAll()
+  }
+
+  test("run Spark in yarn-client mode") {
+    var result = File.createTempFile("result", null, tempDir)
+    YarnClusterDriver.main(Array("yarn-client", result.getAbsolutePath()))
+    checkResult(result)
+  }
+
+  test("run Spark in yarn-cluster mode") {
+    val main = YarnClusterDriver.getClass.getName().stripSuffix("$")
+    var result = File.createTempFile("result", null, tempDir)
+
+    // The Client object will call System.exit() after the job is done, and we don't want
+    // that because it messes up the scalatest monitoring. So replicate some of what main()
+    // does here.
+    val args = Array("--class", main,
+      "--jar", "file:" + fakeSparkJar.getAbsolutePath(),
+      "--arg", "yarn-cluster",
+      "--arg", result.getAbsolutePath(),
+      "--num-executors", "1")
+    val sparkConf = new SparkConf()
+    val yarnConf = SparkHadoopUtil.get.newConfiguration(sparkConf)
+    val clientArgs = new ClientArguments(args, sparkConf)
+    new Client(clientArgs, yarnConf, sparkConf).run()
+    checkResult(result)
+  }
+
+  /**
+   * This is a workaround for an issue with yarn-cluster mode: the Client class will not provide
+   * any sort of error when the job process finishes successfully, but the job itself fails. So
+   * the tests enforce that something is written to a file after everything is ok to indicate
+   * that the job succeeded.
+   */
+  private def checkResult(result: File) = {
+    var resultString = Files.toString(result, Charsets.UTF_8)
+    resultString should be ("success")
+  }
+
+}
+
+private object YarnClusterDriver extends Logging with Matchers {
+
+  def main(args: Array[String]) = {
+    val sc = new SparkContext(new SparkConf().setMaster(args(0))
+      .setAppName("yarn \"test app\" 'with quotes' and \\back\\slashes and $dollarSigns"))
+    val status = new File(args(1))
+    var result = "failure"
+    try {
+      val data = sc.parallelize(1 to 4).map(i => i).collect().toSet
+      data should be (Set(1, 2, 3, 4))
+      result = "success"
+    } finally {
+      sc.stop()
+      Files.write(result, status, Charsets.UTF_8)
+    }
+  }
+
+}