diff options
| author | Andrew Or <andrew@databricks.com> | 2014-11-12 13:39:45 -0800 |
|---|---|---|
| committer | Andrew Or <andrew@databricks.com> | 2014-11-12 13:39:54 -0800 |
| commit | f50c0881be943d8df98a88cc73d163b16169874e (patch) | |
| tree | fba7983386be2b3fe80d005cdbf369dd19039586 /project | |
| parent | 233f0377aaf1dafb8f7e0fb53fc6c09ea65743c3 (diff) | |
| download | spark-f50c0881be943d8df98a88cc73d163b16169874e.tar.gz spark-f50c0881be943d8df98a88cc73d163b16169874e.tar.bz2 spark-f50c0881be943d8df98a88cc73d163b16169874e.zip | |
[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 <andrew@databricks.com>
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
(cherry picked from commit aa43a8da012cf0dac7c7fcccde5f028a942599f0)
Signed-off-by: Andrew Or <andrew@databricks.com>
Diffstat (limited to 'project')
| -rw-r--r-- | project/SparkBuild.scala | 19 |
1 files changed, 13 insertions, 6 deletions
diff --git a/project/SparkBuild.scala b/project/SparkBuild.scala index 492607d558..bbba6423dc 100644 --- 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 |