You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@spark.apache.org by we...@apache.org on 2019/02/12 04:17:03 UTC
[spark] branch master updated: [SPARK-25158][SQL] Executor
accidentally exit because ScriptTransformationWriterThread throw Exception.
This is an automated email from the ASF dual-hosted git repository.
wenchen pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/master by this push:
new 5864e8e [SPARK-25158][SQL] Executor accidentally exit because ScriptTransformationWriterThread throw Exception.
5864e8e is described below
commit 5864e8e47496c3a841b97632e5137de87a91efea
Author: yangjie01 <ya...@baidu.com>
AuthorDate: Tue Feb 12 12:16:33 2019 +0800
[SPARK-25158][SQL] Executor accidentally exit because ScriptTransformationWriterThread throw Exception.
## What changes were proposed in this pull request?
Run Spark-Sql job use transform features(`ScriptTransformationExec`) with config `spark.speculation = true`, sometimes job fails and we found many Executor Dead through `Executor Tab`, through analysis log and code we found :
`ScriptTransformationExec` start a new thread(`ScriptTransformationWriterThread`), the new thread is very likely to throw `TaskKilledException`(from iter.map.foreach part) when speculation is on, this exception will captured by `SparkUncaughtExceptionHandler` which registered during Executor start, `SparkUncaughtExceptionHandler` will call `System.exit (SparkExitCode.UNCAUGHT_EXCEPTION)` to shutdown `Executor`, this is unexpected.
We should not kill the executor just because `ScriptTransformationWriterThread` fails. log the error(not only `TaskKilledException`) instead of throwing it is enough, Exception already pass to `ScriptTransformationExec` and handle by `TaskRunner`.
## How was this patch tested?
Register `TestUncaughtExceptionHandler` to test case in `ScriptTransformationSuite`, then assert there is no Uncaught Exception handled.
Before this patch "script transformation should not swallow errors from upstream operators (no serde)" and "script transformation should not swallow errors from upstream operators (with serde)" throwing `IllegalArgumentException` and handle by `TestUncaughtExceptionHandler` .
Closes #22149 from LuciferYang/fix-transformation-task-kill.
Authored-by: yangjie01 <ya...@baidu.com>
Signed-off-by: Wenchen Fan <we...@databricks.com>
---
.../hive/execution/ScriptTransformationExec.scala | 5 +++-
.../spark/sql/hive/execution/SQLQuerySuite.scala | 33 +++++++++++++++++++++-
.../hive/execution/ScriptTransformationSuite.scala | 32 ++++++++++++++++++++-
.../execution/TestUncaughtExceptionHandler.scala | 31 ++++++++++++++++++++
4 files changed, 98 insertions(+), 3 deletions(-)
diff --git a/sql/hive/src/main/scala/org/apache/spark/sql/hive/execution/ScriptTransformationExec.scala b/sql/hive/src/main/scala/org/apache/spark/sql/hive/execution/ScriptTransformationExec.scala
index 7b35a5f..905cb52 100644
--- a/sql/hive/src/main/scala/org/apache/spark/sql/hive/execution/ScriptTransformationExec.scala
+++ b/sql/hive/src/main/scala/org/apache/spark/sql/hive/execution/ScriptTransformationExec.scala
@@ -308,12 +308,15 @@ private class ScriptTransformationWriterThread(
}
threwException = false
} catch {
+ // SPARK-25158 Exception should not be thrown again, otherwise it will be captured by
+ // SparkUncaughtExceptionHandler, then Executor will exit because of this Uncaught Exception,
+ // so pass the exception to `ScriptTransformationExec` is enough.
case t: Throwable =>
// An error occurred while writing input, so kill the child process. According to the
// Javadoc this call will not throw an exception:
_exception = t
proc.destroy()
- throw t
+ logError("Thread-ScriptTransformation-Feed exit cause by: ", t)
} finally {
try {
Utils.tryLogNonFatalError(outputStream.close())
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/SQLQuerySuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/SQLQuerySuite.scala
index d506edc..ce7661a 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/SQLQuerySuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/SQLQuerySuite.scala
@@ -26,7 +26,7 @@ import java.util.{Locale, Set}
import com.google.common.io.Files
import org.apache.hadoop.fs.{FileSystem, Path}
-import org.apache.spark.TestUtils
+import org.apache.spark.{SparkException, TestUtils}
import org.apache.spark.sql._
import org.apache.spark.sql.catalyst.TableIdentifier
import org.apache.spark.sql.catalyst.analysis.{EliminateSubqueryAliases, FunctionRegistry}
@@ -2348,4 +2348,35 @@ class SQLQuerySuite extends QueryTest with SQLTestUtils with TestHiveSingleton {
}
}
+ test("SPARK-25158: " +
+ "Executor accidentally exit because ScriptTransformationWriterThread throw Exception") {
+ withTempView("test") {
+ val defaultUncaughtExceptionHandler = Thread.getDefaultUncaughtExceptionHandler
+ try {
+ val uncaughtExceptionHandler = new TestUncaughtExceptionHandler
+ Thread.setDefaultUncaughtExceptionHandler(uncaughtExceptionHandler)
+
+ // Use a bad udf to generate failed inputs.
+ val badUDF = org.apache.spark.sql.functions.udf((x: Int) => {
+ if (x < 1) x
+ else throw new RuntimeException("Failed to produce data.")
+ })
+ spark
+ .range(5)
+ .select(badUDF('id).as("a"))
+ .createOrReplaceTempView("test")
+ val scriptFilePath = getTestResourcePath("data")
+ val e = intercept[SparkException] {
+ sql(
+ s"""FROM test SELECT TRANSFORM(a)
+ |USING 'python $scriptFilePath/scripts/test_transform.py "\t"'
+ """.stripMargin).collect()
+ }
+ assert(e.getMessage.contains("Failed to produce data."))
+ assert(uncaughtExceptionHandler.exception.isEmpty)
+ } finally {
+ Thread.setDefaultUncaughtExceptionHandler(defaultUncaughtExceptionHandler)
+ }
+ }
+ }
}
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/ScriptTransformationSuite.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/ScriptTransformationSuite.scala
index 5f73b71..ed3b376 100644
--- a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/ScriptTransformationSuite.scala
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/ScriptTransformationSuite.scala
@@ -18,6 +18,7 @@
package org.apache.spark.sql.hive.execution
import org.apache.hadoop.hive.serde2.`lazy`.LazySimpleSerDe
+import org.scalatest.BeforeAndAfterEach
import org.scalatest.exceptions.TestFailedException
import org.apache.spark.{SparkException, TaskContext, TestUtils}
@@ -29,7 +30,8 @@ import org.apache.spark.sql.execution.{SparkPlan, SparkPlanTest, UnaryExecNode}
import org.apache.spark.sql.hive.test.TestHiveSingleton
import org.apache.spark.sql.types.StringType
-class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton {
+class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton with
+ BeforeAndAfterEach {
import spark.implicits._
private val noSerdeIOSchema = HiveScriptIOSchema(
@@ -49,6 +51,26 @@ class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton {
outputSerdeClass = Some(classOf[LazySimpleSerDe].getCanonicalName)
)
+ private var defaultUncaughtExceptionHandler: Thread.UncaughtExceptionHandler = _
+
+ private val uncaughtExceptionHandler = new TestUncaughtExceptionHandler
+
+ protected override def beforeAll(): Unit = {
+ super.beforeAll()
+ defaultUncaughtExceptionHandler = Thread.getDefaultUncaughtExceptionHandler
+ Thread.setDefaultUncaughtExceptionHandler(uncaughtExceptionHandler)
+ }
+
+ protected override def afterAll(): Unit = {
+ super.afterAll()
+ Thread.setDefaultUncaughtExceptionHandler(defaultUncaughtExceptionHandler)
+ }
+
+ override protected def afterEach(): Unit = {
+ super.afterEach()
+ uncaughtExceptionHandler.cleanStatus()
+ }
+
test("cat without SerDe") {
assume(TestUtils.testCommandAvailable("/bin/bash"))
@@ -63,6 +85,7 @@ class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton {
ioschema = noSerdeIOSchema
),
rowsDf.collect())
+ assert(uncaughtExceptionHandler.exception.isEmpty)
}
test("cat with LazySimpleSerDe") {
@@ -79,6 +102,7 @@ class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton {
ioschema = serdeIOSchema
),
rowsDf.collect())
+ assert(uncaughtExceptionHandler.exception.isEmpty)
}
test("script transformation should not swallow errors from upstream operators (no serde)") {
@@ -98,6 +122,8 @@ class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton {
rowsDf.collect())
}
assert(e.getMessage().contains("intentional exception"))
+ // Before SPARK-25158, uncaughtExceptionHandler will catch IllegalArgumentException
+ assert(uncaughtExceptionHandler.exception.isEmpty)
}
test("script transformation should not swallow errors from upstream operators (with serde)") {
@@ -117,6 +143,8 @@ class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton {
rowsDf.collect())
}
assert(e.getMessage().contains("intentional exception"))
+ // Before SPARK-25158, uncaughtExceptionHandler will catch IllegalArgumentException
+ assert(uncaughtExceptionHandler.exception.isEmpty)
}
test("SPARK-14400 script transformation should fail for bad script command") {
@@ -135,6 +163,7 @@ class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton {
SparkPlanTest.executePlan(plan, hiveContext)
}
assert(e.getMessage.contains("Subprocess exited with status"))
+ assert(uncaughtExceptionHandler.exception.isEmpty)
}
test("SPARK-24339 verify the result after pruning the unused columns") {
@@ -154,6 +183,7 @@ class ScriptTransformationSuite extends SparkPlanTest with TestHiveSingleton {
ioschema = serdeIOSchema
),
rowsDf.select("name").collect())
+ assert(uncaughtExceptionHandler.exception.isEmpty)
}
}
diff --git a/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/TestUncaughtExceptionHandler.scala b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/TestUncaughtExceptionHandler.scala
new file mode 100644
index 0000000..681eb4e
--- /dev/null
+++ b/sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/TestUncaughtExceptionHandler.scala
@@ -0,0 +1,31 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.sql.hive.execution
+
+class TestUncaughtExceptionHandler extends Thread.UncaughtExceptionHandler {
+
+ @volatile private var _exception: Throwable = _
+
+ def exception: Option[Throwable] = Option(_exception)
+
+ def cleanStatus(): Unit = _exception = null
+
+ override def uncaughtException(t: Thread, e: Throwable): Unit = {
+ _exception = e
+ }
+}
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@spark.apache.org
For additional commands, e-mail: commits-help@spark.apache.org