From b16b5434ff44c42e4b3a337f9af147669ba44896 Mon Sep 17 00:00:00 2001
From: Marcelo Vanzin <vanzin@cloudera.com>
Date: Fri, 5 Jun 2015 14:11:38 +0200
Subject: [PATCH] [MINOR] [BUILD] Use custom temp directory during build.
Even with all the efforts to cleanup the temp directories created by
unit tests, Spark leaves a lot of garbage in /tmp after a test run.
This change overrides java.io.tmpdir to place those files under the
build directory instead.
After an sbt full unit test run, I was left with > 400 MB of temp
files. Since they're now under the build dir, it's much easier to
clean them up.
Also make a slight change to a unit test to make it not pollute the
source directory with test data.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes #6653 from vanzin/unit-test-tmp and squashes the following commits:
31e2dd5 [Marcelo Vanzin] Fix tests that depend on each other.
aa92944 [Marcelo Vanzin] [minor] [build] Use custom temp directory during build.
---
.../spark/deploy/SparkSubmitUtilsSuite.scala | 22 ++++++++++---------
pom.xml | 4 +++-
project/SparkBuild.scala | 1 +
3 files changed, 16 insertions(+), 11 deletions(-)
diff --git a/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala b/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala
index 8fda5c8b47..07d261cc42 100644
--- a/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala
+++ b/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala
@@ -28,9 +28,12 @@ import org.apache.ivy.plugins.resolver.IBiblioResolver
import org.apache.spark.SparkFunSuite
import org.apache.spark.deploy.SparkSubmitUtils.MavenCoordinate
+import org.apache.spark.util.Utils
class SparkSubmitUtilsSuite extends SparkFunSuite with BeforeAndAfterAll {
+ private var tempIvyPath: String = _
+
private val noOpOutputStream = new OutputStream {
def write(b: Int) = {}
}
@@ -47,6 +50,7 @@ class SparkSubmitUtilsSuite extends SparkFunSuite with BeforeAndAfterAll {
super.beforeAll()
// We don't want to write logs during testing
SparkSubmitUtils.printStream = new BufferPrintStream
+ tempIvyPath = Utils.createTempDir(namePrefix = "ivy").getAbsolutePath()
}
test("incorrect maven coordinate throws error") {
@@ -90,21 +94,20 @@ class SparkSubmitUtilsSuite extends SparkFunSuite with BeforeAndAfterAll {
}
test("ivy path works correctly") {
- val ivyPath = "dummy" + File.separator + "ivy"
val md = SparkSubmitUtils.getModuleDescriptor
val artifacts = for (i <- 0 until 3) yield new MDArtifact(md, s"jar-$i", "jar", "jar")
- var jPaths = SparkSubmitUtils.resolveDependencyPaths(artifacts.toArray, new File(ivyPath))
+ var jPaths = SparkSubmitUtils.resolveDependencyPaths(artifacts.toArray, new File(tempIvyPath))
for (i <- 0 until 3) {
- val index = jPaths.indexOf(ivyPath)
+ val index = jPaths.indexOf(tempIvyPath)
assert(index >= 0)
- jPaths = jPaths.substring(index + ivyPath.length)
+ jPaths = jPaths.substring(index + tempIvyPath.length)
}
val main = MavenCoordinate("my.awesome.lib", "mylib", "0.1")
IvyTestUtils.withRepository(main, None, None) { repo =>
// end to end
val jarPath = SparkSubmitUtils.resolveMavenCoordinates(main.toString, Option(repo),
- Option(ivyPath), true)
- assert(jarPath.indexOf(ivyPath) >= 0, "should use non-default ivy path")
+ Option(tempIvyPath), true)
+ assert(jarPath.indexOf(tempIvyPath) >= 0, "should use non-default ivy path")
}
}
@@ -123,13 +126,12 @@ class SparkSubmitUtilsSuite extends SparkFunSuite with BeforeAndAfterAll {
assert(jarPath.indexOf("mylib") >= 0, "should find artifact")
}
// Local ivy repository with modified home
- val dummyIvyPath = "dummy" + File.separator + "ivy"
- val dummyIvyLocal = new File(dummyIvyPath, "local" + File.separator)
+ val dummyIvyLocal = new File(tempIvyPath, "local" + File.separator)
IvyTestUtils.withRepository(main, None, Some(dummyIvyLocal), true) { repo =>
val jarPath = SparkSubmitUtils.resolveMavenCoordinates(main.toString, None,
- Some(dummyIvyPath), true)
+ Some(tempIvyPath), true)
assert(jarPath.indexOf("mylib") >= 0, "should find artifact")
- assert(jarPath.indexOf(dummyIvyPath) >= 0, "should be in new ivy path")
+ assert(jarPath.indexOf(tempIvyPath) >= 0, "should be in new ivy path")
}
}
diff --git a/pom.xml b/pom.xml
index e28d4b9fc2..a848deffe7 100644
--- a/pom.xml
+++ b/pom.xml
@@ -179,7 +179,7 @@
<parquet.deps.scope>compile</parquet.deps.scope>
<!--
- Overridable test home. So that you can call individual pom files directory without
+ Overridable test home. So that you can call individual pom files directly without
things breaking.
-->
<spark.test.home>${session.executionRootDirectory}</spark.test.home>
@@ -1256,6 +1256,7 @@
<systemProperties>
<derby.system.durability>test</derby.system.durability>
<java.awt.headless>true</java.awt.headless>
+ <java.io.tmpdir>${project.build.directory}/tmp</java.io.tmpdir>
<spark.test.home>${spark.test.home}</spark.test.home>
<spark.testing>1</spark.testing>
<spark.ui.enabled>false</spark.ui.enabled>
@@ -1289,6 +1290,7 @@
<systemProperties>
<derby.system.durability>test</derby.system.durability>
<java.awt.headless>true</java.awt.headless>
+ <java.io.tmpdir>${project.build.directory}/tmp</java.io.tmpdir>
<spark.test.home>${spark.test.home}</spark.test.home>
<spark.testing>1</spark.testing>
<spark.ui.enabled>false</spark.ui.enabled>
diff --git a/project/SparkBuild.scala b/project/SparkBuild.scala
index ef3a175bac..921f1599fe 100644
--- a/project/SparkBuild.scala
+++ b/project/SparkBuild.scala
@@ -496,6 +496,7 @@ object TestSettings {
"SPARK_DIST_CLASSPATH" ->
(fullClasspath in Test).value.files.map(_.getAbsolutePath).mkString(":").stripSuffix(":"),
"JAVA_HOME" -> sys.env.get("JAVA_HOME").getOrElse(sys.props("java.home"))),
+ javaOptions in Test += s"-Djava.io.tmpdir=$sparkHome/target/tmp",
javaOptions in Test += "-Dspark.test.home=" + sparkHome,
javaOptions in Test += "-Dspark.testing=1",
javaOptions in Test += "-Dspark.port.maxRetries=100",
--
GitLab