Hello. I am running Spark 2.3.0 via Yarn. I have a Spark Streaming application where the driver threw an uncaught out of memory exception:
19/01/31 13:00:59 ERROR Utils: Uncaught exception in thread element-tracking-store-worker java.lang.OutOfMemoryError: GC overhead limit exceeded at org.apache.spark.util.kvstore.KVTypeInfo$MethodAccessor.get(KVTypeInfo.java:154) at org.apache.spark.util.kvstore.InMemoryStore$InMemoryView.compare(InMemoryStore.java:248) at org.apache.spark.util.kvstore.InMemoryStore$InMemoryView.lambda$iterator$0(InMemoryStore.java:203) at org.apache.spark.util.kvstore.InMemoryStore$InMemoryView$$Lambda$27/1691147907.compare(Unknown Source) at java.util.TimSort.binarySort(TimSort.java:296) at java.util.TimSort.sort(TimSort.java:239) at java.util.Arrays.sort(Arrays.java:1512) at java.util.ArrayList.sort(ArrayList.java:1462) at java.util.Collections.sort(Collections.java:175) at org.apache.spark.util.kvstore.InMemoryStore$InMemoryView.iterator(InMemoryStore.java:203) at scala.collection.convert.Wrappers$JIterableWrapper.iterator(Wrappers.scala:54) at scala.collection.IterableLike$class.foreach(IterableLike.scala:72) at scala.collection.AbstractIterable.foreach(Iterable.scala:54) at org.apache.spark.status.AppStatusListener$$anonfun$org$apache$spark$status$AppStatusListener$$cleanupStages$1.apply(AppStatusListener.scala:894) at org.apache.spark.status.AppStatusListener$$anonfun$org$apache$spark$status$AppStatusListener$$cleanupStages$1.apply(AppStatusListener.scala:874) at scala.collection.immutable.List.foreach(List.scala:381) at org.apache.spark.status.AppStatusListener.org $apache$spark$status$AppStatusListener$$cleanupStages(AppStatusListener.scala:874) at org.apache.spark.status.AppStatusListener$$anonfun$3.apply$mcVJ$sp(AppStatusListener.scala:84) at org.apache.spark.status.ElementTrackingStore$$anonfun$write$1$$anonfun$apply$1$$anonfun$apply$mcV$sp$1.apply(ElementTrackingStore.scala:109) at org.apache.spark.status.ElementTrackingStore$$anonfun$write$1$$anonfun$apply$1$$anonfun$apply$mcV$sp$1.apply(ElementTrackingStore.scala:107) at scala.collection.immutable.List.foreach(List.scala:381) at org.apache.spark.status.ElementTrackingStore$$anonfun$write$1$$anonfun$apply$1.apply$mcV$sp(ElementTrackingStore.scala:107) at org.apache.spark.status.ElementTrackingStore$$anonfun$write$1$$anonfun$apply$1.apply(ElementTrackingStore.scala:105) at org.apache.spark.status.ElementTrackingStore$$anonfun$write$1$$anonfun$apply$1.apply(ElementTrackingStore.scala:105) at org.apache.spark.util.Utils$.tryLog(Utils.scala:2001) at org.apache.spark.status.ElementTrackingStore$$anon$1.run(ElementTrackingStore.scala:91) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) at java.lang.Thread.run(Thread.java:748) Despite the uncaught exception the Streaming application never terminated. No new batches were started. As a result my job did not process data for some period of time (until our ancillary monitoring noticed the issue). *Ask: What can we do to ensure that the driver is shut down when this type of exception occurs?* Regards, Bryan Jeffrey