[SPARK-30745] Spark streaming, kafka broker error, "Failed to get records for spark-executor- .... after polling for 512"

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Incomplete
• Affects Version/s: 2.0.2
• Component/s: Build, Deploy, DStreams, Kubernetes, Spark Core
• Labels: Spark2.0.2 cluster kafka-0.10 spark-streaming-kafka

Environment

Spark 2.0.2, Kafka 0.10

• Assignee: Unassigned
• Reporter: Harneet K
• Created: 06/Feb/20 03:01
• Updated: 12/Dec/22 18:10
• Resolved: 10/Feb/20 03:11

Description

We have a spark streaming application reading data from Kafka.
Data size: 15 Million

Below errors were seen:
java.lang.AssertionError: assertion failed: Failed to get records for spark-executor- .... after polling for 512 at scala.Predef$.assert(Predef.scala:170)

There were more errors seen pertaining to CachedKafkaConsumer
at org.apache.spark.streaming.kafka010.CachedKafkaConsumer.get(CachedKafkaConsumer.scala:74)
at org.apache.spark.streaming.kafka010.KafkaRDD$KafkaRDDIterator.next(KafkaRDD.scala:227)
at org.apache.spark.streaming.kafka010.KafkaRDD$KafkaRDDIterator.next(KafkaRDD.scala:193)
at scala.collection.Iterator$$anon$11.next(Iterator.scala:409)
at org.apache.spark.storage.memory.MemoryStore.putIteratorAsValues(MemoryStore.scala:214)
at org.apache.spark.storage.BlockManager$$anonfun$doPutIterator$1.apply(BlockManager.scala:935)
at org.apache.spark.storage.BlockManager$$anonfun$doPutIterator$1.apply(BlockManager.scala:926)
at org.apache.spark.storage.BlockManager.doPut(BlockManager.scala:866)
at org.apache.spark.storage.BlockManager.doPutIterator(BlockManager.scala:926) 
at org.apache.spark.storage.BlockManager.getOrElseUpdate(BlockManager.scala:670) 
at org.apache.spark.rdd.RDD.getOrCompute(RDD.scala:330)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:281)
at org.apache.spark.rdd.UnionRDD.compute(UnionRDD.scala:105)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:319)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:283) 
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:70)
at org.apache.spark.scheduler.Task.run(Task.scala:86)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:274)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)

 

The spark.streaming.kafka.consumer.poll.ms is set to default 512ms and other Kafka stream timeout settings are default.
"request.timeout.ms" 
"heartbeat.interval.ms" 
"session.timeout.ms" 
"max.poll.interval.ms" 

Also, the kafka is being recently updated to 0.10 from 0.8. In 0.8, this behavior was not seen.
No resource issues are seen.