igreenfield commented on a change in pull request #26624:
URL: https://github.com/apache/spark/pull/26624#discussion_r419656913



##########
File path: core/src/main/scala/org/apache/spark/util/ThreadUtils.scala
##########
@@ -17,21 +17,106 @@
 
 package org.apache.spark.util
 
+import java.util
 import java.util.concurrent._
 import java.util.concurrent.locks.ReentrantLock
 
+import com.google.common.util.concurrent.{MoreExecutors, ThreadFactoryBuilder}
 import scala.concurrent.{Awaitable, ExecutionContext, 
ExecutionContextExecutor, Future}
 import scala.concurrent.duration.{Duration, FiniteDuration}
 import scala.language.higherKinds
 import scala.util.control.NonFatal
 
-import com.google.common.util.concurrent.ThreadFactoryBuilder
-
 import org.apache.spark.SparkException
 import org.apache.spark.rpc.RpcAbortException
 
 private[spark] object ThreadUtils {
 
+  object MDCAwareThreadPoolExecutor {
+    def newCachedThreadPool(threadFactory: ThreadFactory): ThreadPoolExecutor 
= {
+      // The values needs to be synced with `Executors.newCachedThreadPool`
+      new MDCAwareThreadPoolExecutor(
+        0,
+        Integer.MAX_VALUE,
+        60L,
+        TimeUnit.SECONDS,
+        new SynchronousQueue[Runnable],
+        threadFactory)
+    }
+
+    def newFixedThreadPool(nThreads: Int, threadFactory: ThreadFactory): 
ThreadPoolExecutor = {
+      // The values needs to be synced with `Executors.newFixedThreadPool`
+      new MDCAwareThreadPoolExecutor(
+        nThreads,
+        nThreads,
+        0L,
+        TimeUnit.MILLISECONDS,
+        new LinkedBlockingQueue[Runnable],
+        threadFactory)
+    }
+
+    def newSingleThreadExecutor(threadFactory: ThreadFactory): ExecutorService 
= {
+      // The values needs to be synced with `Executors.newSingleThreadExecutor`
+      Executors.unconfigurableExecutorService(
+        new MDCAwareThreadPoolExecutor(

Review comment:
       The only difference now is the finalize method. So I don't see any 
benefit from it.
   In order to propagate MDC to all places, we need to pass it to all threads.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

Reply via email to