Skip to content

Commit 6ccd6c5

Browse files
ScrapCodespwendell
ScrapCodes
authored and
pwendell
committed
SPARK-1121 Only add avro if the build is for Hadoop 0.23.X and SPARK_YARN is set
Author: Prashant Sharma <[email protected]> Closes #6 from ScrapCodes/SPARK-1121/avro-dep-fix and squashes the following commits: 9b29e34 [Prashant Sharma] Review feedback on PR 46ed2ad [Prashant Sharma] SPARK-1121-Only add avro if the build is for Hadoop 0.23.X and SPARK_YARN is set
1 parent 5a3ad10 commit 6ccd6c5

File tree

3 files changed

+39
-55
lines changed

3 files changed

+39
-55
lines changed

‎docs/building-with-maven.md‎

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -76,3 +76,7 @@ The maven build includes support for building a Debian package containing the as
7676
$ mvn -Pdeb -DskipTests clean package
7777

7878
The debian package can then be found under assembly/target. We added the short commit hash to the file name so that we can distinguish individual packages built for SNAPSHOT versions.
79+
80+
## A note about Hadoop version 0.23.x
81+
82+
For building spark with hadoop 0.23.x and also yarn, you will have to manually add a dependency on avro (org.apache.avro, avro, 1.7.4).

‎pom.xml‎

Lines changed: 0 additions & 21 deletions
Original file line numberDiff line numberDiff line change
@@ -505,27 +505,6 @@
505505
</exclusion>
506506
</exclusions>
507507
</dependency>
508-
<!-- Specify Avro version because Kafka also has it as a dependency -->
509-
<dependency>
510-
<groupId>org.apache.avro</groupId>
511-
<artifactId>avro</artifactId>
512-
<version>1.7.4</version>
513-
</dependency>
514-
<dependency>
515-
<groupId>org.apache.avro</groupId>
516-
<artifactId>avro-ipc</artifactId>
517-
<version>1.7.4</version>
518-
<exclusions>
519-
<exclusion>
520-
<groupId>org.jboss.netty</groupId>
521-
<artifactId>netty</artifactId>
522-
</exclusion>
523-
<exclusion>
524-
<groupId>io.netty</groupId>
525-
<artifactId>netty</artifactId>
526-
</exclusion>
527-
</exclusions>
528-
</dependency>
529508
</dependencies>
530509
</dependencyManagement>
531510

‎project/SparkBuild.scala‎

Lines changed: 35 additions & 34 deletions
Original file line numberDiff line numberDiff line change
@@ -87,7 +87,7 @@ object SparkBuild extends Build {
8787
case Some(v) => v.toBoolean
8888
}
8989
lazy val hadoopClient = if (hadoopVersion.startsWith("0.20.") || hadoopVersion == "1.0.0") "hadoop-core" else "hadoop-client"
90-
90+
val maybeAvro = if (hadoopVersion.startsWith("0.23.") && isYarnEnabled) Seq("org.apache.avro" % "avro" % "1.7.4") else Seq()
9191
// Conditionally include the yarn sub-project
9292
lazy val yarnAlpha = Project("yarn-alpha", file("yarn/alpha"), settings = yarnAlphaSettings) dependsOn(core)
9393
lazy val yarn = Project("yarn", file("yarn/stable"), settings = yarnSettings) dependsOn(core)
@@ -130,6 +130,8 @@ object SparkBuild extends Build {
130130
javacOptions := Seq("-target", JAVAC_JVM_VERSION, "-source", JAVAC_JVM_VERSION),
131131
unmanagedJars in Compile <<= baseDirectory map { base => (base / "lib" ** "*.jar").classpath },
132132
retrieveManaged := true,
133+
// This is to add convenience of enabling sbt -Dsbt.offline=true for making the build offline.
134+
offline := "true".equalsIgnoreCase(sys.props("sbt.offline")),
133135
retrievePattern := "[type]s/[artifact](-[revision])(-[classifier]).[ext]",
134136
transitiveClassifiers in Scope.GlobalScope := Seq("sources"),
135137
testListeners <<= target.map(t => Seq(new eu.henkelmann.sbt.JUnitXmlTestsListener(t.getAbsolutePath))),
@@ -254,39 +256,38 @@ object SparkBuild extends Build {
254256
),
255257

256258
libraryDependencies ++= Seq(
257-
"com.google.guava" % "guava" % "14.0.1",
258-
"com.google.code.findbugs" % "jsr305" % "1.3.9",
259-
"log4j" % "log4j" % "1.2.17",
260-
"org.slf4j" % "slf4j-api" % slf4jVersion,
261-
"org.slf4j" % "slf4j-log4j12" % slf4jVersion,
262-
"org.slf4j" % "jul-to-slf4j" % slf4jVersion,
263-
"org.slf4j" % "jcl-over-slf4j" % slf4jVersion,
264-
"commons-daemon" % "commons-daemon" % "1.0.10", // workaround for bug HADOOP-9407
265-
"com.ning" % "compress-lzf" % "1.0.0",
266-
"org.xerial.snappy" % "snappy-java" % "1.0.5",
267-
"org.ow2.asm" % "asm" % "4.0",
268-
"org.spark-project.akka" %% "akka-remote" % "2.2.3-shaded-protobuf" excludeAll(excludeNetty),
269-
"org.spark-project.akka" %% "akka-slf4j" % "2.2.3-shaded-protobuf" excludeAll(excludeNetty),
270-
"org.spark-project.akka" %% "akka-testkit" % "2.2.3-shaded-protobuf" % "test",
271-
"org.json4s" %% "json4s-jackson" % "3.2.6",
272-
"it.unimi.dsi" % "fastutil" % "6.4.4",
273-
"colt" % "colt" % "1.2.0",
274-
"org.apache.mesos" % "mesos" % "0.13.0",
275-
"net.java.dev.jets3t" % "jets3t" % "0.7.1" excludeAll(excludeCommonsLogging),
276-
"org.apache.derby" % "derby" % "10.4.2.0" % "test",
277-
"org.apache.hadoop" % hadoopClient % hadoopVersion excludeAll(excludeJackson, excludeNetty, excludeAsm, excludeCglib, excludeCommonsLogging, excludeSLF4J),
278-
"org.apache.avro" % "avro" % "1.7.4",
279-
"org.apache.avro" % "avro-ipc" % "1.7.4" excludeAll(excludeNetty),
280-
"org.apache.curator" % "curator-recipes" % "2.4.0" excludeAll(excludeNetty),
281-
"com.codahale.metrics" % "metrics-core" % "3.0.0",
282-
"com.codahale.metrics" % "metrics-jvm" % "3.0.0",
283-
"com.codahale.metrics" % "metrics-json" % "3.0.0",
284-
"com.codahale.metrics" % "metrics-ganglia" % "3.0.0",
285-
"com.codahale.metrics" % "metrics-graphite" % "3.0.0",
286-
"com.twitter" %% "chill" % "0.3.1",
287-
"com.twitter" % "chill-java" % "0.3.1",
288-
"com.clearspring.analytics" % "stream" % "2.5.1"
289-
)
259+
"com.google.guava" % "guava" % "14.0.1",
260+
"com.google.code.findbugs" % "jsr305" % "1.3.9",
261+
"log4j" % "log4j" % "1.2.17",
262+
"org.slf4j" % "slf4j-api" % slf4jVersion,
263+
"org.slf4j" % "slf4j-log4j12" % slf4jVersion,
264+
"org.slf4j" % "jul-to-slf4j" % slf4jVersion,
265+
"org.slf4j" % "jcl-over-slf4j" % slf4jVersion,
266+
"commons-daemon" % "commons-daemon" % "1.0.10", // workaround for bug HADOOP-9407
267+
"com.ning" % "compress-lzf" % "1.0.0",
268+
"org.xerial.snappy" % "snappy-java" % "1.0.5",
269+
"org.ow2.asm" % "asm" % "4.0",
270+
"org.spark-project.akka" %% "akka-remote" % "2.2.3-shaded-protobuf" excludeAll(excludeNetty),
271+
"org.spark-project.akka" %% "akka-slf4j" % "2.2.3-shaded-protobuf" excludeAll(excludeNetty),
272+
"org.spark-project.akka" %% "akka-testkit" % "2.2.3-shaded-protobuf" % "test",
273+
"org.json4s" %% "json4s-jackson" % "3.2.6",
274+
"it.unimi.dsi" % "fastutil" % "6.4.4",
275+
"colt" % "colt" % "1.2.0",
276+
"org.apache.mesos" % "mesos" % "0.13.0",
277+
"net.java.dev.jets3t" % "jets3t" % "0.7.1" excludeAll(excludeCommonsLogging),
278+
"org.apache.derby" % "derby" % "10.4.2.0" % "test",
279+
"org.apache.hadoop" % hadoopClient % hadoopVersion excludeAll(excludeJackson, excludeNetty, excludeAsm, excludeCglib, excludeCommonsLogging, excludeSLF4J),
280+
"org.apache.curator" % "curator-recipes" % "2.4.0" excludeAll(excludeNetty),
281+
"com.codahale.metrics" % "metrics-core" % "3.0.0",
282+
"com.codahale.metrics" % "metrics-jvm" % "3.0.0",
283+
"com.codahale.metrics" % "metrics-json" % "3.0.0",
284+
"com.codahale.metrics" % "metrics-ganglia" % "3.0.0",
285+
"com.codahale.metrics" % "metrics-graphite" % "3.0.0",
286+
"com.twitter" %% "chill" % "0.3.1",
287+
"com.twitter" % "chill-java" % "0.3.1",
288+
"com.clearspring.analytics" % "stream" % "2.5.1"
289+
),
290+
libraryDependencies ++= maybeAvro
290291
)
291292

292293
def rootSettings = sharedSettings ++ Seq(

0 commit comments

Comments
 (0)