[SPARK-4281][Build] Package Yarn shuffle service into its own jar

This is another addendum to #3082, which added the Yarn shuffle service to run inside the NM. This PR makes the feature much more usable by packaging enough dependencies into the jar to run the service inside an NM. After these changes, the user can run `./make-distribution.sh` and find a `spark-network-yarn*.jar` in their `lib` directory. The equivalent change is done in SBT by making the `network-yarn` module an assembly project. Author: Andrew Or <[email protected]> Closes #3147 from andrewor14/yarn-shuffle-build and squashes the following commits: bda58d0 [Andrew Or] Fix line too long 81e9705 [Andrew Or] Merge branch 'master' of github.com:apache/spark into yarn-shuffle-build fb7f398 [Andrew Or] Rename jar to spark-{VERSION}-yarn-shuffle.jar 65db822 [Andrew Or] Actually mark slf4j as provided abcefd1 [Andrew Or] Do the same for SBT c653028 [Andrew Or] Package network-yarn and its dependencies
apache · Nov 12, 2014 · aa43a8d · aa43a8d
1 parent 6e3c5a2
commit aa43a8d
Show file tree

Hide file tree

Showing 5 changed files with 53 additions and 10 deletions.
diff --git a/make-distribution.sh b/make-distribution.sh
@@ -181,6 +181,7 @@ echo "Spark $VERSION$GITREVSTRING built for Hadoop $SPARK_HADOOP_VERSION" > "$DI
 # Copy jars
 cp "$FWDIR"/assembly/target/scala*/*assembly*hadoop*.jar "$DISTDIR/lib/"
 cp "$FWDIR"/examples/target/scala*/spark-examples*.jar "$DISTDIR/lib/"
+cp "$FWDIR"/network/yarn/target/scala*/spark-*-yarn-shuffle.jar "$DISTDIR/lib/"
 
 # Copy example sources (needed for python and SQL)
 mkdir -p "$DISTDIR/examples/src/main"

diff --git a/network/common/pom.xml b/network/common/pom.xml
@@ -41,12 +41,13 @@
       <groupId>io.netty</groupId>
       <artifactId>netty-all</artifactId>
     </dependency>
+
+    <!-- Provided dependencies -->
     <dependency>
       <groupId>org.slf4j</groupId>
       <artifactId>slf4j-api</artifactId>
+      <scope>provided</scope>
     </dependency>
-
-    <!-- Provided dependencies -->
     <dependency>
       <groupId>com.google.guava</groupId>
       <artifactId>guava</artifactId>

diff --git a/network/shuffle/pom.xml b/network/shuffle/pom.xml
@@ -42,12 +42,13 @@
       <artifactId>spark-network-common_${scala.binary.version}</artifactId>
       <version>${project.version}</version>
     </dependency>
+
+    <!-- Provided dependencies -->
     <dependency>
       <groupId>org.slf4j</groupId>
       <artifactId>slf4j-api</artifactId>
+      <scope>provided</scope>
     </dependency>
-
-    <!-- Provided dependencies -->
     <dependency>
       <groupId>com.google.guava</groupId>
       <artifactId>guava</artifactId>

diff --git a/network/yarn/pom.xml b/network/yarn/pom.xml
@@ -54,5 +54,38 @@
   <build>
     <outputDirectory>target/scala-${scala.binary.version}/classes</outputDirectory>
     <testOutputDirectory>target/scala-${scala.binary.version}/test-classes</testOutputDirectory>
+    <plugins>
+      <plugin>
+        <groupId>org.apache.maven.plugins</groupId>
+        <artifactId>maven-shade-plugin</artifactId>
+        <configuration>
+          <shadedArtifactAttached>false</shadedArtifactAttached>
+          <outputFile>${project.build.directory}/scala-${scala.binary.version}/spark-${project.version}-yarn-shuffle.jar</outputFile>
+          <artifactSet>
+            <includes>
+              <include>*:*</include>
+            </includes>
+          </artifactSet>
+          <filters>
+            <filter>
+              <artifact>*:*</artifact>
+              <excludes>
+                <exclude>META-INF/*.SF</exclude>
+                <exclude>META-INF/*.DSA</exclude>
+                <exclude>META-INF/*.RSA</exclude>
+              </excludes>
+            </filter>
+          </filters>
+        </configuration>
+        <executions>
+          <execution>
+            <phase>package</phase>
+            <goals>
+              <goal>shade</goal>
+            </goals>
+          </execution>
+        </executions>
+      </plugin>
+    </plugins>
   </build>
 </project>
diff --git a/project/SparkBuild.scala b/project/SparkBuild.scala
@@ -38,12 +38,12 @@ object BuildCommons {
       "streaming-flume", "streaming-kafka", "streaming-mqtt", "streaming-twitter",
       "streaming-zeromq").map(ProjectRef(buildLocation, _))
 
-  val optionallyEnabledProjects@Seq(yarn, yarnStable, yarnAlpha, networkYarn, java8Tests,
-    sparkGangliaLgpl, sparkKinesisAsl) = Seq("yarn", "yarn-stable", "yarn-alpha", "network-yarn",
+  val optionallyEnabledProjects@Seq(yarn, yarnStable, yarnAlpha, java8Tests,
+    sparkGangliaLgpl, sparkKinesisAsl) = Seq("yarn", "yarn-stable", "yarn-alpha",
     "java8-tests", "ganglia-lgpl", "kinesis-asl").map(ProjectRef(buildLocation, _))
 
-  val assemblyProjects@Seq(assembly, examples) = Seq("assembly", "examples")
-    .map(ProjectRef(buildLocation, _))
+  val assemblyProjects@Seq(assembly, examples, networkYarn) =
+    Seq("assembly", "examples", "network-yarn").map(ProjectRef(buildLocation, _))
 
   val tools = ProjectRef(buildLocation, "tools")
   // Root project.
@@ -289,8 +289,15 @@ object Assembly {
 
   lazy val settings = assemblySettings ++ Seq(
     test in assembly := {},
-    jarName in assembly <<= (version, moduleName) map { (v, mName) => mName + "-"+v + "-hadoop" +
-      Option(System.getProperty("hadoop.version")).getOrElse("1.0.4") + ".jar" },
+    jarName in assembly <<= (version, moduleName) map { (v, mName) =>
+      if (mName.contains("network-yarn")) {
+        // This must match the same name used in maven (see network/yarn/pom.xml)
+        "spark-" + v + "-yarn-shuffle.jar"
+      } else {
+        mName + "-" + v + "-hadoop" +
+          Option(System.getProperty("hadoop.version")).getOrElse("1.0.4") + ".jar"
+      }
+    },
     mergeStrategy in assembly := {
       case PathList("org", "datanucleus", xs @ _*)             => MergeStrategy.discard
       case m if m.toLowerCase.endsWith("manifest.mf")          => MergeStrategy.discard