Merge branch 'master' into spark-2706

.gitignore

@@ -15,11 +15,10 @@ out/
 third_party/libmesos.so
 third_party/libmesos.dylib
 conf/java-opts
-conf/spark-env.sh
-conf/streaming-env.sh
-conf/log4j.properties
-conf/spark-defaults.conf
-conf/hive-site.xml
+conf/*.sh
+conf/*.properties
+conf/*.conf
+conf/*.xml
 docs/_site
 docs/api
 target/
@@ -50,7 +49,6 @@ unit-tests.log
 /lib/
 rat-results.txt
 scalastyle.txt
-conf/*.conf
 scalastyle-output.xml
 
 # For Hive

core/src/test/scala/org/apache/spark/ui/UISuite.scala

@@ -23,7 +23,6 @@ import javax.servlet.http.HttpServletRequest
 import scala.io.Source
 import scala.util.{Failure, Success, Try}
 
-import org.eclipse.jetty.server.Server
 import org.eclipse.jetty.servlet.ServletContextHandler
 import org.scalatest.FunSuite
 import org.scalatest.concurrent.Eventually._
@@ -108,14 +107,8 @@ class UISuite extends FunSuite {
   }
 
   test("jetty selects different port under contention") {
-    val startPort = 4040
-    val server = new Server(startPort)
-
-    Try { server.start() } match {
-      case Success(s) =>
-      case Failure(e) =>
-      // Either case server port is busy hence setup for test complete
-    }
+    val server = new ServerSocket(0)
+    val startPort = server.getLocalPort
     val serverInfo1 = JettyUtils.startJettyServer(
       "0.0.0.0", startPort, Seq[ServletContextHandler](), new SparkConf)
     val serverInfo2 = JettyUtils.startJettyServer(
@@ -126,6 +119,9 @@ class UISuite extends FunSuite {
     assert(boundPort1 != startPort)
     assert(boundPort2 != startPort)
     assert(boundPort1 != boundPort2)
+    serverInfo1.server.stop()
+    serverInfo2.server.stop()
+    server.close()
   }
 
   test("jetty binds to port 0 correctly") {

docs/_layouts/global.html

@@ -111,6 +111,7 @@
                                 <li class="divider"></li>
                                 <li><a href="building-spark.html">Building Spark</a></li>
                                 <li><a href="https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark">Contributing to Spark</a></li>
+                                <li><a href="https://cwiki.apache.org/confluence/display/SPARK/Supplemental+Spark+Projects">Supplemental Projects</a></li>
                             </ul>
                         </li>
                     </ul>
@@ -151,7 +152,7 @@ <h1 class="title">{{ page.title }}</h1>
                     MathJax.Hub.Config({
                         tex2jax: {
                             inlineMath: [ ["$", "$"], ["\\\\(","\\\\)"] ],
-                            displayMath: [ ["$$","$$"], ["\\[", "\\]"] ], 
+                            displayMath: [ ["$$","$$"], ["\\[", "\\]"] ],
                             processEscapes: true,
                             skipTags: ['script', 'noscript', 'style', 'textarea', 'pre']
                         }

docs/configuration.md

@@ -520,10 +520,10 @@ Apart from these, the following properties are also available, and may be useful
 </tr>
 <tr>
   <td><code>spark.files.fetchTimeout</code></td>
-  <td>false</td>
+  <td>60</td>
   <td>
     Communication timeout to use when fetching files added through SparkContext.addFile() from
-    the driver.
+    the driver, in seconds.
   </td>
 </tr>
 <tr>

docs/ec2-scripts.md

@@ -156,6 +156,6 @@ If you have a patch or suggestion for one of these limitations, feel free to
 
 # Accessing Data in S3
 
-Spark's file interface allows it to process data in Amazon S3 using the same URI formats that are supported for Hadoop. You can specify a path in S3 as input through a URI of the form `s3n://<bucket>/path`. You will also need to set your Amazon security credentials, either by setting the environment variables `AWS_ACCESS_KEY_ID` and `AWS_SECRET_ACCESS_KEY` before your program or through `SparkContext.hadoopConfiguration`. Full instructions on S3 access using the Hadoop input libraries can be found on the [Hadoop S3 page](http://wiki.apache.org/hadoop/AmazonS3).
+Spark's file interface allows it to process data in Amazon S3 using the same URI formats that are supported for Hadoop. You can specify a path in S3 as input through a URI of the form `s3n://<bucket>/path`. To provide AWS credentials for S3 access, launch the Spark cluster with the option `--copy-aws-credentials`. Full instructions on S3 access using the Hadoop input libraries can be found on the [Hadoop S3 page](http://wiki.apache.org/hadoop/AmazonS3).
 
 In addition to using a single input file, you can also use a directory of files as input by simply giving the path to the directory.

docs/index.md

@@ -107,6 +107,7 @@ options for deployment:
   * [OpenStack Swift](storage-openstack-swift.html)
 * [Building Spark](building-spark.html): build Spark using the Maven system
 * [Contributing to Spark](https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark)
+* [Supplemental Projects](https://cwiki.apache.org/confluence/display/SPARK/Supplemental+Spark+Projects): related third party Spark projects
 
 **External Resources:**
 

ec2/deploy.generic/root/spark-ec2/ec2-variables.sh

@@ -30,3 +30,5 @@ export HADOOP_MAJOR_VERSION="{{hadoop_major_version}}"
 export SWAP_MB="{{swap}}"
 export SPARK_WORKER_INSTANCES="{{spark_worker_instances}}"
 export SPARK_MASTER_OPTS="{{spark_master_opts}}"
+export AWS_ACCESS_KEY_ID="{{aws_access_key_id}}"
+export AWS_SECRET_ACCESS_KEY="{{aws_secret_access_key}}"

ec2/spark_ec2.py

@@ -158,6 +158,9 @@ def parse_args():
     parser.add_option(
         "--additional-security-group", type="string", default="",
         help="Additional security group to place the machines in")
+    parser.add_option(
+        "--copy-aws-credentials", action="store_true", default=False,
+        help="Add AWS credentials to hadoop configuration to allow Spark to access S3")
 
     (opts, args) = parser.parse_args()
     if len(args) != 2:
@@ -714,6 +717,13 @@ def deploy_files(conn, root_dir, opts, master_nodes, slave_nodes, modules):
         "spark_master_opts": opts.master_opts
     }
 
+    if opts.copy_aws_credentials:
+        template_vars["aws_access_key_id"] = conn.aws_access_key_id
+        template_vars["aws_secret_access_key"] = conn.aws_secret_access_key
+    else:
+        template_vars["aws_access_key_id"] = ""
+        template_vars["aws_secret_access_key"] = ""
+
     # Create a temp directory in which we will place all the files to be
     # deployed after we substitue template parameters in them
     tmp_dir = tempfile.mkdtemp()

yarn/common/src/test/scala/org/apache/spark/deploy/yarn/ClientBaseSuite.scala

@@ -38,6 +38,7 @@ import org.scalatest.Matchers
 
 import scala.collection.JavaConversions._
 import scala.collection.mutable.{ HashMap => MutableHashMap }
+import scala.reflect.ClassTag
 import scala.util.Try
 
 import org.apache.spark.{SparkException, SparkConf}
@@ -200,9 +201,10 @@ class ClientBaseSuite extends FunSuite with Matchers {
 
 
     val knownDefMRAppCP: Seq[String] =
-      getFieldValue[String, Seq[String]](classOf[MRJobConfig],
-                                         "DEFAULT_MAPREDUCE_APPLICATION_CLASSPATH",
-                                         Seq[String]())(a => a.split(","))
+      getFieldValue2[String, Array[String], Seq[String]](
+        classOf[MRJobConfig],
+        "DEFAULT_MAPREDUCE_APPLICATION_CLASSPATH",
+        Seq[String]())(a => a.split(","))(a => a.toSeq)
 
     val knownYARNAppCP = Some(Seq("/known/yarn/path"))
 
@@ -232,6 +234,17 @@ class ClientBaseSuite extends FunSuite with Matchers {
   def getFieldValue[A, B](clazz: Class[_], field: String, defaults: => B)(mapTo: A => B): B =
     Try(clazz.getField(field)).map(_.get(null).asInstanceOf[A]).toOption.map(mapTo).getOrElse(defaults)
 
+  def getFieldValue2[A: ClassTag, A1: ClassTag, B](
+        clazz: Class[_],
+        field: String,
+        defaults: => B)(mapTo:  A => B)(mapTo1: A1 => B) : B = {
+    Try(clazz.getField(field)).map(_.get(null)).map {
+      case v: A => mapTo(v)
+      case v1: A1 => mapTo1(v1)
+      case _ => defaults
+    }.toOption.getOrElse(defaults)
+  }
+
   private class DummyClient(
       val args: ClientArguments,
       val conf: Configuration,