[SPARK-22163] Design Issue of Spark Streaming that Causes Random Run-time Exception

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Not A Problem
• Affects Version/s: 2.2.0
• Component/s: DStreams, Structured Streaming

Environment

Spark Streaming
Kafka
Linux

• Flags: Important
• Assignee: Unassigned
• Reporter: Michael N (Inactive)
• Created: 28/Sep/17 23:55
• Updated: 11/Oct/17 06:43
• Resolved: 09/Oct/17 06:59

Description

The application objects can contain List and can be modified dynamically as well. However, Spark Streaming framework asynchronously serializes the application's objects as the application runs. Therefore, it causes random run-time exception on the List when Spark Streaming framework happens to serializes the application's objects while the application modifies a List in its own object.

In fact, there are multiple bugs reported about

Caused by: java.util.ConcurrentModificationException
at java.util.ArrayList.writeObject

that are permutation of the same root cause. So the design issue of Spark streaming framework is that it should do this serialization asynchronously. Instead, it should either

1. do this serialization synchronously. This is preferred to eliminate the issue completely. Or

2. Allow it to be configured per application whether to do this serialization synchronously or asynchronously, depending on the nature of each application.

Also, Spark documentation should describe the conditions that trigger Spark to do this type of serialization asynchronously, so the applications can work around them until the fix is provided.

===

Vadim Semenov and Steve Loughran, per your inquiries in ticket https://issues.apache.org/jira/browse/SPARK-21999, I am posting the reply here because this issue involves Spark's design and not necessarily its code implementation.

—

My application does not spin up its own thread. All the threads are controlled by Spark.

Batch interval = 5 seconds

Batch #3
1. Driver - Spark Thread #1 - starts batch #3 and blocks until all slave threads are done with this batch
2. Slave A - Spark Thread #2. Says it takes 10 seconds to complete
3. Slave B - Spark Thread #3. Says it takes 1 minutes to complete

4. Both thread #1 for the driver and thread 2# for Slave A do not jump ahead and process batch #4. Instead, they wait for thread #3 until it is done. => So there is already synchronization among the threads within the same batch. Also, batch to batch is synchronous.

5. After Spark Thread #3 is done, the driver does other processing to finish the current batch. In my case, it updates a list of objects.

The above steps repeat for the next batch #4 and subsequent batches.

Based on the exception stack trace, it looks like in step 5, Spark has another thread #4 that serializes application objects asynchronously. So it causes random occurrences of ConcurrentModificationException, because the list of objects is being changed by Spark own thread #1 for the driver.

So the issue is not that my application "is modifying a collection asynchronously w.r.t. Spark" as Sean kept claiming. Instead, it is Spark's asynchronous operations among its own different threads within the same batch that causes this issue.

Since Spark controls all the threads and their synchronization, it is a Spark design's issue for the lack of synchronization between threads #1 and #4, that triggers ConcurrentModificationException. That is the root cause of this issue.

Further, even if the application does not modify its list of objects, in step 5 the driver could be modifying multiple native objects say two integers. In thread #1 the driver could have updated integer X and before it could update integer Y, when Spark's thread #4 asynchronous serializes the application objects. So the persisted serialized data does not match with the actual data. This resulted in a permutation of this issue with a false positive condition where the serialized checkpoint data has partially correct data.

One solution for both issues is to modify Spark's design and allow the serialization of application objects by Spark's thread #4 to be configurable per application to be either asynchronous or synchronous with Spark's thread #1. That way, it is up to individual applications to decide based on the nature of their business requirements and needed throughput.

===

The code is listed below. Due to the asynchronous nature of Spark's thread operations and different hardware, the issue relating to this ticket occurs randomly. So you may need to tweak the batch duration.

package test;

import java.util.HashMap;
import java.util.HashSet;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
import java.util.Set;
import java.util.regex.Pattern;

import scala.Tuple2;

import kafka.serializer.StringDecoder;

import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.function.*;
import org.apache.spark.streaming.api.java.*;
import org.apache.spark.streaming.kafka.KafkaUtils;
import org.apache.spark.streaming.Durations;

/**
 * Consumes messages from one or more topics in Kafka and does wordcount.
 * Usage: JavaDirectKafkaWordCount <brokers> <topics>
 *   <brokers> is a list of one or more Kafka brokers
 *   <topics> is a list of one or more kafka topics to consume from
 *
 * Example:
 *    $ bin/run-example streaming.JavaDirectKafkaWordCount broker1-host:port,broker2-host:port \
 *      topic1,topic2
 */
//  ,    VoidFunction<Iterator<Tuple2<String, Integer>>>
public final class JavaDirectKafkaWordCount_Extended implements VoidFunction<JavaPairRDD<String, Integer>> {
   
  private static final Pattern SPACE = Pattern.compile(" ");
 
  private List<String> appStringList;

  public static void main(String[] args) throws Exception {
      try {
        if (args.length < 2) {
            System.err.println("Usage: JavaDirectKafkaWordCount <brokers> <topics>\n" +
                "  <brokers> is a list of one or more Kafka brokers\n" +
                "  <topics> is a list of one or more kafka topics to consume from\n\n");
            System.exit(1);           
          }
       
        JavaDirectKafkaWordCount javaDirectKafkaWordCount = new JavaDirectKafkaWordCount ();
        javaDirectKafkaWordCount.setupStreamApp(args);
      } catch (Throwable exc) {
          exc.printStackTrace();
      }
  }
 
  private void setupStreamApp (String[] args) throws InterruptedException {     
    // StreamingExamples.setStreamingLogLevels();

        String brokers = args[0];
        String topics = args[1];

        // create list of string with dummy values.
        appStringList = new ArrayList<>();
        for (int i = 0; i < 1000; ) {
            appStringList.add("a-"+ i++);           
        }
   
        SparkConf sparkConf = new SparkConf().setAppName("JavaDirectKafkaWordCount");
       
        // Create context with a 2 seconds batch interval
        JavaStreamingContext jssc = new JavaStreamingContext(sparkConf, Durations.seconds(10));
        jssc.checkpoint("./test-checkpoint/");
   
        Set<String> topicsSet = new HashSet<>(Arrays.asList(topics.split(",")));
        Map<String, String> kafkaParams = new HashMap<>();
        kafkaParams.put("metadata.broker.list", brokers);
   
        // Create direct kafka stream with brokers and topics
        JavaPairInputDStream<String, String> messages = KafkaUtils.createDirectStream(
            jssc,
            String.class,
            String.class,
            StringDecoder.class,
            StringDecoder.class,
            kafkaParams,
            topicsSet
        );
   
        // Get the lines, split them into words, count the words and print
        JavaDStream<String> lines = messages.map(new Function<Tuple2<String, String>, String>() {
          /**
             *
             */
            private static final long serialVersionUID = 3769940753726592424L;

            @Override
              public String call(Tuple2<String, String> tuple2) {
                return tuple2._2();
              }
        });
       
        JavaDStream<String> words = lines.flatMap(new FlatMapFunction<String, String>() {
              @Override
              public Iterator<String> call(String x) {
                return Arrays.asList(SPACE.split(x)).iterator();
              }
        });
       
        JavaPairDStream<String, Integer> wordCounts = words.mapToPair(
              new PairFunction<String, String, Integer>() {
                @Override
                public Tuple2<String, Integer> call(String s) {
                  return new Tuple2<>(s, 1);
                }
              }).reduceByKey(
                new Function2<Integer, Integer, Integer>() {
                @Override
                public Integer call(Integer i1, Integer i2) {
                  return i1 + i2;
                }
        });
       
        wordCounts.foreachRDD(this);
               
        // Start the computation
        jssc.start();
        jssc.awaitTermination();
    }

    @Override
    public void call(JavaPairRDD<String, Integer> dataStream) throws Exception {

        System.out.println("start foreachRDD");
       
        dataStream.foreachPartition(new VoidFunction<Iterator<Tuple2<String, Integer>>> () {

            // Assuming that there are two slave threads, this foreachPartition code corresponds to Steps 2 and 3 for threads #2 and #3
            // as described in the high-level sequence described of the textual description above the code
            @Override
            public void call(Iterator<Tuple2<String, Integer>> tuples) throws Exception {
                if (tuples == null || !tuples.hasNext()) {
                    return;
                }
               
                while (tuples.hasNext()) {
                    // The step below is not related to the issue. It is used just to simulate some operation in the slave threads for completeness
                    System.out.println(tuples.next()._1);
                }                   
            }               
        });           

         /* ===> 
          *  the steps below corresponds to Step 5 of the high-level sequence and Spark's thread #1 as described of the textual description above the code.
          * 
          *  These steps below are where ConcurrentModificationException occurs randomly as explained in Step 5 of the textual description of this ticket.
          *  For the purpose of this test, these steps update the list by simply rotating the entries.
         *
         * Based on the stack trace, Spark has another thread, i,e, thread #4, that asynchronously serializes the application objects during the
         * the next three operations. So it would randomly encounters ConcurrentModificationException  because Spark's thread #4 tries to serialize appStringList
         * while Spark's thread #1 is modifying the same list.
         */
        String tmp = appStringList.get(0);
        appStringList.remove(0);
        appStringList.add(tmp);
        System.out.println("end foreachRDD");
    }   
}

Issue Links

• duplicates: SPARK-21999 ConcurrentModificationException - Spark Streaming