Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

On task failure catch some CUDA exceptions and kill executor [databricks] #5118

Merged
merged 10 commits into from
Apr 1, 2022
Merged

On task failure catch some CUDA exceptions and kill executor [databricks] #5118

Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
cecaaa4
On task failure catch some cuda exceptions and kill executor
tgravescs Mar 31, 2022
b49a64d
include other exceptions
tgravescs Mar 31, 2022
1192ba9
cleanup logs
tgravescs Mar 31, 2022
00492d3
fix message
tgravescs Mar 31, 2022
a9b169c
update other message to debug
tgravescs Mar 31, 2022
afb00a5
dont' call super
tgravescs Mar 31, 2022
cd35d1c
comment out checking all error string
tgravescs Mar 31, 2022
7492b1e
fxi extra space
tgravescs Mar 31, 2022
47c9416
Check the entire stack trace
tgravescs Apr 1, 2022
17f212b
remove extra comment
tgravescs Apr 1, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 20 additions & 1 deletion sql-plugin/src/main/scala/com/nvidia/spark/rapids/Plugin.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ import scala.util.matching.Regex
import com.nvidia.spark.rapids.python.PythonWorkerSemaphore
import com.nvidia.spark.rapids.shims.SparkShimImpl

import org.apache.spark.{SparkConf, SparkContext}
import org.apache.spark.{ExceptionFailure, SparkConf, SparkContext, TaskFailedReason}
import org.apache.spark.api.plugin.{DriverPlugin, ExecutorPlugin, PluginContext}
import org.apache.spark.internal.Logging
import org.apache.spark.serializer.{JavaSerializer, KryoSerializer}
Expand Down Expand Up @@ -275,6 +275,25 @@ class RapidsExecutorPlugin extends ExecutorPlugin with Logging {
GpuDeviceManager.shutdown()
Option(rapidsShuffleHeartbeatEndpoint).foreach(_.close())
}

override def onTaskFailed(failureReason: TaskFailedReason): Unit = {
failureReason match {
case ef: ExceptionFailure =>
val unrecoverableErrors = Seq("cudaErrorIllegalAddress", "cudaErrorLaunchTimeout",
"cudaErrorHardwareStackError", "cudaErrorIllegalInstruction",
"cudaErrorMisalignedAddress", "cudaErrorInvalidAddressSpace", "cudaErrorInvalidPc",
"cudaErrorLaunchFailure", "cudaErrorExternalDevice", "cudaErrorUnknown",
"cudaErrorECCUncorrectable")
if (unrecoverableErrors.exists(ef.description.contains(_)) ||
unrecoverableErrors.exists(ef.toErrorString.contains(_))) {
logError("Stopping the Executor based on exception being a fatal CUDA error: " +
s"${ef.toErrorString}")
System.exit(20)
}
case other =>
logDebug(s"Executor onTaskFailed not a CUDA fatal error: ${other.toString}")
}
}
}

object RapidsExecutorPlugin {
Expand Down