Community Conundrum 10: The Titanic is now live Learn more

Spark failure: out of memory in task

Level 1
Spark failure: out of memory in task

Does anyone experienced this error?  I run a join recipe on Spark engine and the run is taking ages, I checked logs, I always got this error. Also tried running on Hive engine, but got IOException.

 

Job aborted due to stage failure: Task 5 in stage 4.0 failed 1 times, most recent failure: Lost task 5.0 in stage 4.0 (TID 152, localhost): java.lang.OutOfMemoryError: GC overhead limit exceeded at org.apache.spark.unsafe.types.UTF8String.fromAddress(UTF8String.java:99) at org.apache.spark.sql.catalyst.expressions.UnsafeRow.getUTF8String(UnsafeRow.java:452) at org.apache.spark.sql.catalyst.expressions.GeneratedClass$SpecificOrdering.compare(Unknown Source) at org.apache.spark.sql.execution.UnsafeKVExternalSorter$KVComparator.compare(UnsafeKVExternalSorter.java:212) at org.apache.spark.util.collection.unsafe.sort.UnsafeInMemorySorter$SortComparator.compare(UnsafeInMemorySorter.java:60) at org.apache.spark.util.collection.unsafe.sort.UnsafeInMemorySorter$SortComparator.compare(UnsafeInMemorySorter.java:37) at org.apache.spark.util.collection.TimSort$SortState.mergeLo(TimSort.java:734) at org.apache.spark.util.collection.TimSort$SortState.mergeAt(TimSort.java:525) at org.apache.spark.util.collection.TimSort$SortState.mergeCollapse(TimSort.java:453) at org.apache.spark.util.collection.TimSort$SortState.access$200(TimSort.java:325) at org.apache.spark.util.collection.TimSort.sort(TimSort.java:153) at org.apache.spark.util.collection.Sorter.sort(Sorter.scala:37) at org.apache.spark.util.collection.unsafe.sort.UnsafeInMemorySorter.getSortedIterator(UnsafeInMemorySorter.java:235) at org.apache.spark.util.collection.unsafe.sort.UnsafeExternalSorter.spill(UnsafeExternalSorter.java:186) at org.apache.spark.util.collection.unsafe.sort.UnsafeExternalSorter.createWithExistingInMemorySorter(UnsafeExternalSorter.java:86) at org.apache.spark.sql.execution.UnsafeKVExternalSorter.<init>(UnsafeKVExternalSorter.java:116) at org.apache.spark.sql.execution.UnsafeFixedWidthAggregationMap.destructAndCreateExternalSorter(UnsafeFixedWidthAggregationMap.java:246) at org.apache.spark.sql.execution.aggregate.TungstenAggregationIterator.processInputs(TungstenAggregationIterator.scala:520) at org.apache.spark.sql.execution.aggregate.TungstenAggregationIterator.<init>(TungstenAggregationIterator.scala:686) at org.apache.spark.sql.execution.aggregate.TungstenAggregate$$anonfun$doExecute$1$$anonfun$2.apply(TungstenAggregate.scala:95) at org.apache.spark.sql.execution.aggregate.TungstenAggregate$$anonfun$doExecute$1$$anonfun$2.apply(TungstenAggregate.scala:86) at org.apache.spark.rdd.RDD$$anonfun$mapPartitions$1$$anonfun$apply$22.apply(RDD.scala:717) at org.apache.spark.rdd.RDD$$anonfun$mapPartitions$1$$anonfun$apply$22.apply(RDD.scala:717) at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:38) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:313) at org.apache.spark.rdd.RDD.iterator(RDD.scala:277) at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:38) at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:313) at org.apache.spark.rdd.RDD.iterator(RDD.scala:277) at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:73) at org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:41) at org.apache.spark.scheduler.Task.run(Task.scala:89) Driver stacktrace:, caused by: OutOfMemoryError: GC overhead limit exceeded

 

 

From Hive engine error log.

 

Oops: an unexpected error occurred

Process failure, caused by: IOException: Failed to execute Hive script, please check job logs (return code 2)

Please see our options for getting help

HTTP code: , type: java.io.IOException

1 Reply
Dataiker
Dataiker

Hi,

When you get an "OutOfMemoryError" that happens in a task (i.e. you get "Task failed ... OutOfMemoryError"), you typically need to increase the spark.executor.memory setting.

You can set this up in the recipe settings (Advanced > Spark config), add a key spark.executor.memory - If you have not overriden it, the default value is 2g, you may want to try with 4g for example, and keep increasing if it still fails.