aboutsummaryrefslogtreecommitdiff
path: root/launcher
diff options
context:
space:
mode:
authorhyukjinkwon <gurwls223@gmail.com>2016-12-14 19:24:24 +0000
committerSean Owen <sowen@cloudera.com>2016-12-14 19:24:24 +0000
commitc6b8eb71a9638c9a8ce02d11d5fe26f4c5be531e (patch)
tree08466a3d1154edbc97d6be688e30c15130b05c8f /launcher
parentba4aab9b85688141d3d0c185165ec7a402c9fbba (diff)
downloadspark-c6b8eb71a9638c9a8ce02d11d5fe26f4c5be531e.tar.gz
spark-c6b8eb71a9638c9a8ce02d11d5fe26f4c5be531e.tar.bz2
spark-c6b8eb71a9638c9a8ce02d11d5fe26f4c5be531e.zip
[SPARK-18842][TESTS][LAUNCHER] De-duplicate paths in classpaths in commands for local-cluster mode to work around the path length limitation on Windows
## What changes were proposed in this pull request? Currently, some tests are being failed and hanging on Windows due to this problem. For the reason in SPARK-18718, some tests using `local-cluster` mode were disabled on Windows due to the length limitation by paths given to classpaths. The limitation seems roughly 32K (see the [blog in MS](https://blogs.msdn.microsoft.com/oldnewthing/20031210-00/?p=41553/) and [another reference](https://support.thoughtworks.com/hc/en-us/articles/213248526-Getting-around-maximum-command-line-length-is-32767-characters-on-Windows)) but in `local-cluster` mode, executors were being launched as processes with the command such as [here](https://gist.github.com/HyukjinKwon/5bc81061c250d4af5a180869b59d42ea) in (only) tests. This length is roughly 40K due to the classpaths given to `java` command. However, it seems duplicates are almost half of them. So, if we deduplicate the paths, it seems reduced to roughly 20K with the command, [here](https://gist.github.com/HyukjinKwon/dad0c8db897e5e094684a2dc6a417790). Maybe, we should consider as some more paths are added in the future but it seems better than disabling all the tests for now with minimised changes. Therefore, this PR proposes to deduplicate the paths in classpaths in case of launching executors as processes in `local-cluster` mode. ## How was this patch tested? Existing tests in `ShuffleSuite` and `BroadcastJoinSuite` manually via AppVeyor Author: hyukjinkwon <gurwls223@gmail.com> Closes #16266 from HyukjinKwon/disable-local-cluster-tests.
Diffstat (limited to 'launcher')
-rw-r--r--launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java8
1 files changed, 5 insertions, 3 deletions
diff --git a/launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java b/launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java
index ba43659d96..0622fef17c 100644
--- a/launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java
+++ b/launcher/src/main/java/org/apache/spark/launcher/AbstractCommandBuilder.java
@@ -26,9 +26,11 @@ import java.nio.charset.StandardCharsets;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.HashMap;
+import java.util.LinkedHashSet;
import java.util.List;
import java.util.Map;
import java.util.Properties;
+import java.util.Set;
import java.util.regex.Pattern;
import static org.apache.spark.launcher.CommandBuilderUtils.*;
@@ -135,7 +137,7 @@ abstract class AbstractCommandBuilder {
List<String> buildClassPath(String appClassPath) throws IOException {
String sparkHome = getSparkHome();
- List<String> cp = new ArrayList<>();
+ Set<String> cp = new LinkedHashSet<>();
addToClassPath(cp, getenv("SPARK_CLASSPATH"));
addToClassPath(cp, appClassPath);
@@ -201,7 +203,7 @@ abstract class AbstractCommandBuilder {
addToClassPath(cp, getenv("HADOOP_CONF_DIR"));
addToClassPath(cp, getenv("YARN_CONF_DIR"));
addToClassPath(cp, getenv("SPARK_DIST_CLASSPATH"));
- return cp;
+ return new ArrayList<>(cp);
}
/**
@@ -210,7 +212,7 @@ abstract class AbstractCommandBuilder {
* @param cp List to which the new entries are appended.
* @param entries New classpath entries (separated by File.pathSeparator).
*/
- private void addToClassPath(List<String> cp, String entries) {
+ private void addToClassPath(Set<String> cp, String entries) {
if (isEmpty(entries)) {
return;
}