SPARK-1167: Remove metrics-ganglia from default build due to LGPL iss…

…ues.

This patch removes Ganglia integration from the default build. It
allows users willing to link against LGPL code to use Ganglia
by adding build flags or linking against a new Spark artifact called
spark-ganglia-lgpl.

This brings Spark in line with the Apache policy on LGPL code
enumerated here:

https://www.apache.org/legal/3party.html#options-optional

assembly/pom.xml

@@ -158,6 +158,16 @@
         </dependency>
       </dependencies>
     </profile>
+    <profile>
+      <id>spark-ganglia-lgpl</id>
+      <dependencies>
+        <dependency>
+          <groupId>org.apache.spark</groupId>
+          <artifactId>spark-ganglia-lgpl_${scala.binary.version}</artifactId>
+          <version>${project.version}</version>
+        </dependency>
+      </dependencies>
+    </profile>
     <profile>
       <id>bigtop-dist</id>
       <!-- This profile uses the assembly plugin to create a special "dist" package for BigTop

core/pom.xml

@@ -183,10 +183,6 @@
       <groupId>com.codahale.metrics</groupId>
       <artifactId>metrics-json</artifactId>
     </dependency>
-    <dependency>
-      <groupId>com.codahale.metrics</groupId>
-      <artifactId>metrics-ganglia</artifactId>
-    </dependency>
     <dependency>
       <groupId>com.codahale.metrics</groupId>
       <artifactId>metrics-graphite</artifactId>

dev/audit-release/README.md

@@ -0,0 +1,11 @@
+# Test Application Builds
+This directory includes test applications which are built when auditing releases. You can
+run them locally by setting appropriate environment variables.
+
+```
+$ cd sbt_app_core
+$ SCALA_VERSION=2.10.3 \
+  SPARK_VERSION=1.0.0-SNAPSHOT \
+  SPARK_RELEASE_REPOSITORY=file:///home/patrick/.ivy2/local \
+  sbt run
+```

dev/audit-release/sbt_app_core/src/main/scala/SparkApp.scala

@@ -17,6 +17,8 @@
 
 package main.scala
 
+import scala.util.Try
+
 import org.apache.spark.SparkContext
 import org.apache.spark.SparkContext._
 
@@ -31,6 +33,17 @@ object SimpleApp {
       println("Failed to parse log files with Spark")
       System.exit(-1)
     }
-    println("Test succeeded")
+
+    // Regression test for SPARK-1167: Remove metrics-ganglia from default build due to LGPL issue
+    val foundConsole = Try(Class.forName("org.apache.spark.metrics.sink.ConsoleSink")).isSuccess
+    val foundGanglia = Try(Class.forName("org.apache.spark.metrics.sink.GangliaSink")).isSuccess
+    if (!foundConsole) {
+      println("Console sink not loaded via spark-core")
+      System.exit(-1)
+    }
+    if (foundGanglia) {
+      println("Ganglia sink was loaded via spark-core")
+      System.exit(-1)
+    }
   }
 }

dev/audit-release/sbt_app_ganglia/build.sbt

@@ -0,0 +1,31 @@
+//
+// 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.
+//
+
+name := "Ganglia Test"
+
+version := "1.0"
+
+scalaVersion := System.getenv.get("SCALA_VERSION")
+
+libraryDependencies += "org.apache.spark" %% "spark-core" % System.getenv.get("SPARK_VERSION")
+
+libraryDependencies += "org.apache.spark" %% "spark-ganglia-lgpl" % System.getenv.get("SPARK_VERSION")
+
+resolvers ++= Seq(
+  "Spark Release Repository" at System.getenv.get("SPARK_RELEASE_REPOSITORY"),
+  "Akka Repository" at "http://repo.akka.io/releases/",
+  "Spray Repository" at "http://repo.spray.cc/")

dev/audit-release/sbt_app_ganglia/src/main/scala/SparkApp.scala

@@ -0,0 +1,39 @@
+/*
+ * 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 main.scala
+
+import scala.util.Try
+
+import org.apache.spark.SparkContext
+import org.apache.spark.SparkContext._
+
+object SimpleApp {
+  def main(args: Array[String]) {
+    // Regression test for SPARK-1167: Remove metrics-ganglia from default build due to LGPL issue
+    val foundConsole = Try(Class.forName("org.apache.spark.metrics.sink.ConsoleSink")).isSuccess
+    val foundGanglia = Try(Class.forName("org.apache.spark.metrics.sink.GangliaSink")).isSuccess
+    if (!foundConsole) {
+      println("Console sink not loaded via spark-core")
+      System.exit(-1)
+    }
+    if (!foundGanglia) {
+      println("Ganglia sink not loaded via spark-ganglia-lgpl")
+      System.exit(-1)
+    }
+  }
+}

dev/create-release/create-release.sh

@@ -49,14 +49,14 @@ mvn -DskipTests \
   -Darguments="-DskipTests=true -Dhadoop.version=2.2.0 -Dyarn.version=2.2.0 -Dgpg.passphrase=${GPG_PASSPHRASE}" \
   -Dusername=$GIT_USERNAME -Dpassword=$GIT_PASSWORD \
   -Dhadoop.version=2.2.0 -Dyarn.version=2.2.0 \
-  -Pyarn \
+  -Pyarn -Pspark-ganglia-lgpl \
   -Dtag=$GIT_TAG -DautoVersionSubmodules=true \
   --batch-mode release:prepare
 
 mvn -DskipTests \
   -Darguments="-DskipTests=true -Dhadoop.version=2.2.0 -Dyarn.version=2.2.0 -Dgpg.passphrase=${GPG_PASSPHRASE}" \
   -Dhadoop.version=2.2.0 -Dyarn.version=2.2.0 \
-  -Pyarn \
+  -Pyarn -Pspark-ganglia-lgpl\
   release:perform
 
 rm -rf spark

docs/monitoring.md

@@ -48,11 +48,22 @@ Each instance can report to zero or more _sinks_. Sinks are contained in the
 
 * `ConsoleSink`: Logs metrics information to the console.
 * `CSVSink`: Exports metrics data to CSV files at regular intervals.
-* `GangliaSink`: Sends metrics to a Ganglia node or multicast group.
 * `JmxSink`: Registers metrics for viewing in a JXM console.
 * `MetricsServlet`: Adds a servlet within the existing Spark UI to serve metrics data as JSON data.
 * `GraphiteSink`: Sends metrics to a Graphite node.
 
+Spark also supports a Ganglia sink which is not included in the default build due to
+licensing restrictions:
+
+* `GangliaSink`: Sends metrics to a Ganglia node or multicast group.
+
+To install the `GangliaSink` you'll need to perform a custom build of Spark. _**Note that
+by embedding this library you will include [LGPL](http://www.gnu.org/copyleft/lesser.html)-licensed 
+code in your Spark package**_. For sbt users, set the 
+`SPARK_GANGLIA_LGPL` environment varaible before building. For Maven users, enable 
+the `-Pspark-ganglia-lgpl` profile. For users linking applications against Spark, link 
+include the `spark-ganglia-lgpl` artifact as a dependency.
+
 The syntax of the metrics configuration file is defined in an example configuration file, 
 `$SPARK_HOME/conf/metrics.properties.template`.
 

extras/spark-ganglia-lgpl/pom.xml

@@ -0,0 +1,45 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!--
+~ 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.
+-->
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+  <modelVersion>4.0.0</modelVersion>
+  <parent>
+    <groupId>org.apache.spark</groupId>
+    <artifactId>spark-parent</artifactId>
+    <version>1.0.0-SNAPSHOT</version>
+    <relativePath>../../pom.xml</relativePath>
+  </parent>
+
+  <!-- Ganglia integration is not included by default due to LGPL-licensed code -->
+  <groupId>org.apache.spark</groupId>
+  <artifactId>spark-ganglia-lgpl_2.10</artifactId>
+  <packaging>jar</packaging>
+  <name>Spark Ganglia Integration</name>
+
+  <dependencies>
+    <dependency>
+      <groupId>org.apache.spark</groupId>
+      <artifactId>spark-core_${scala.binary.version}</artifactId>
+      <version>${project.version}</version>
+    </dependency>
+
+    <dependency>
+      <groupId>com.codahale.metrics</groupId>
+      <artifactId>metrics-ganglia</artifactId>
+    </dependency>
+  </dependencies>
+</project>

pom.xml

@@ -725,12 +725,19 @@
         <hadoop.version>0.23.7</hadoop.version>
         <!--<hadoop.version>2.0.5-alpha</hadoop.version> -->
       </properties>
-
       <modules>
         <module>yarn</module>
       </modules>
+    </profile>
 
+    <!-- Ganglia integration is not included by default due to LGPL-licensed code -->
+    <profile>
+      <id>spark-ganglia-lgpl</id>
+      <modules>
+        <module>extras/spark-ganglia-lgpl</module>
+      </modules>
     </profile>
+
     <profile>
       <id>java8-tests</id>
       <build>

project/SparkBuild.scala

@@ -65,7 +65,7 @@ object SparkBuild extends Build {
   lazy val mllib = Project("mllib", file("mllib"), settings = mllibSettings) dependsOn(core)
 
   lazy val assemblyProj = Project("assembly", file("assembly"), settings = assemblyProjSettings)
-    .dependsOn(core, graphx, bagel, mllib, repl, streaming) dependsOn(maybeYarn: _*)
+    .dependsOn(core, graphx, bagel, mllib, repl, streaming) dependsOn(maybeYarn: _*) dependsOn(maybeGanglia: _*)
 
   lazy val assembleDeps = TaskKey[Unit]("assemble-deps", "Build assembly of dependencies and packages Spark projects")
 
@@ -91,19 +91,26 @@ object SparkBuild extends Build {
   lazy val hadoopClient = if (hadoopVersion.startsWith("0.20.") || hadoopVersion == "1.0.0") "hadoop-core" else "hadoop-client"
   val maybeAvro = if (hadoopVersion.startsWith("0.23.") && isYarnEnabled) Seq("org.apache.avro" % "avro" % "1.7.4") else Seq()
 
-  // Conditionally include the java 8 sub-project
+  // Include Ganglia integration if the user has enabled Ganglia
+  // This is isolated from the normal build due to LGPL-licensed code in the library
+  lazy val isGangliaEnabled = Properties.envOrNone("SPARK_GANGLIA_LGPL").isDefined
+  lazy val gangliaProj = Project("spark-ganglia-lgpl", file("extras/spark-ganglia-lgpl"), settings = gangliaSettings).dependsOn(core)
+  val maybeGanglia: Seq[ClasspathDependency] = if (isGangliaEnabled) Seq(gangliaProj) else Seq()
+  val maybeGangliaRef: Seq[ProjectReference] = if (isGangliaEnabled) Seq(gangliaProj) else Seq()
+
+  // Include the Java 8 project if the JVM version is 8+
   lazy val javaVersion = System.getProperty("java.specification.version")
   lazy val isJava8Enabled = javaVersion.toDouble >= "1.8".toDouble
   val maybeJava8Tests = if (isJava8Enabled) Seq[ProjectReference](java8Tests) else Seq[ProjectReference]()
   lazy val java8Tests = Project("java8-tests", file("extras/java8-tests"), settings = java8TestsSettings).
     dependsOn(core) dependsOn(streaming % "compile->compile;test->test")
 
-  // Conditionally include the yarn sub-project
+  // Include the YARN project if the user has enabled YARN
   lazy val yarnAlpha = Project("yarn-alpha", file("yarn/alpha"), settings = yarnAlphaSettings) dependsOn(core)
   lazy val yarn = Project("yarn", file("yarn/stable"), settings = yarnSettings) dependsOn(core)
 
-  lazy val maybeYarn = if (isYarnEnabled) Seq[ClasspathDependency](if (isNewHadoop) yarn else yarnAlpha) else Seq[ClasspathDependency]()
-  lazy val maybeYarnRef = if (isYarnEnabled) Seq[ProjectReference](if (isNewHadoop) yarn else yarnAlpha) else Seq[ProjectReference]()
+  lazy val maybeYarn: Seq[ClasspathDependency] = if (isYarnEnabled) Seq(if (isNewHadoop) yarn else yarnAlpha) else Seq()
+  lazy val maybeYarnRef: Seq[ProjectReference] = if (isYarnEnabled) Seq(if (isNewHadoop) yarn else yarnAlpha) else Seq()
 
   lazy val externalTwitter = Project("external-twitter", file("external/twitter"), settings = twitterSettings)
     .dependsOn(streaming % "compile->compile;test->test")
@@ -127,7 +134,7 @@ object SparkBuild extends Build {
     .dependsOn(core, mllib, graphx, bagel, streaming, externalTwitter) dependsOn(allExternal: _*)
 
   // Everything except assembly, tools, java8Tests and examples belong to packageProjects
-  lazy val packageProjects = Seq[ProjectReference](core, repl, bagel, streaming, mllib, graphx) ++ maybeYarnRef
+  lazy val packageProjects = Seq[ProjectReference](core, repl, bagel, streaming, mllib, graphx) ++ maybeYarnRef ++ maybeGangliaRef
 
   lazy val allProjects = packageProjects ++ allExternalRefs ++
     Seq[ProjectReference](examples, tools, assemblyProj) ++ maybeJava8Tests
@@ -296,7 +303,6 @@ object SparkBuild extends Build {
         "com.codahale.metrics"       % "metrics-core"     % "3.0.0",
         "com.codahale.metrics"       % "metrics-jvm"      % "3.0.0",
         "com.codahale.metrics"       % "metrics-json"     % "3.0.0",
-        "com.codahale.metrics"       % "metrics-ganglia"  % "3.0.0",
         "com.codahale.metrics"       % "metrics-graphite" % "3.0.0",
         "com.twitter"               %% "chill"            % "0.3.1",
         "com.twitter"                % "chill-java"       % "0.3.1",
@@ -384,6 +390,11 @@ object SparkBuild extends Build {
     name := "spark-yarn"
   )
 
+  def gangliaSettings = sharedSettings ++ Seq(
+    name := "spark-ganglia-lgpl",
+    libraryDependencies += "com.codahale.metrics" % "metrics-ganglia" % "3.0.0"
+  )
+
   def java8TestsSettings = sharedSettings ++ Seq(
     name := "java8-tests",
     javacOptions := Seq("-target", "1.8", "-source", "1.8"),