Spark Streaming 1.6.0中Checkpointing/WAL的可靠性问题

Question

描述

我们在Scala中有一个Spark Streaming 1.5.2应用程序,它从Kinesis Stream中读取JSON事件,执行一些转换/聚合并将结果写入不同的S3前缀.当前批处理间隔为60秒.我们有3000-7000事件/秒.我们正在使用检查点来保护我们免于丢失聚合.

它已经运行了一段时间,从异常甚至群集重启中恢复.我们最近重新编译了Spark Streaming 1.6.0的代码,只更改了build.sbt文件中的库依赖项.在Spark 1.6.0群集中运行代码几个小时之后,我们注意到以下内容:

1. "投入率"和"处理时间"波动率在1.6.0中大幅增加(见下面的截图).
2. 每隔几个小时,在写记录时会抛出一个'Exception:BlockAdditionEvent ...到WriteAheadLog.java.util.concurrent.TimeoutException:[5000毫秒]之后的期货超时"异常(参见下面的完整堆栈跟踪)与特定批次(分钟)的下降到0事件/秒一致.

在做了一些挖掘之后,我认为第二个问题看起来与此Pull Request有关.PR的最初目标:"当使用S3作为WALs的目录时,写入时间太长.当多个接收器将AddBlock事件发送到ReceiverTracker时,驱动程序很容易受到瓶颈.此PR在ReceivedBlockTracker中添加事件批处理,以便接收器不会被驱动程序阻塞太长时间."

我们在Spark 1.5.2中的S3中检查点,并且没有性能/可靠性问题.我们在S3和本地NAS中测试了Spark 1.6.0中的检查点,在这两种情况下我们都收到了这个例外.看起来当检查点批次需要超过5秒时,会出现此异常,并且我们已检查该批次的事件是否永远丢失.

问题

• Spark Streaming 1.6.0中预计"输入速率"和"处理时间"波动的增加是否有任何已知的改进方法？

• 你知道除了这2个以外的任何解决方法吗？:

  1)保证检查点接收器写入所有文件所需的时间少于5秒.根据我的经验,即使是小批量,也无法保证使用S3.对于本地NAS,它取决于谁负责基础设施(云提供商很难).

  2)增加spark.streaming.driver.writeAheadLog.batchingTimeout属性值.

• 您是否期望在描述的场景中丢失任何事件？我认为如果批量检查点失败,则不会增加分片/接收器序列号,并且将在稍后重试.

Spark 1.5.2统计 - 截图

Spark 1.6.0统计 - 截图

全栈跟踪

16/01/19 03:25:03 WARN ReceivedBlockTracker: Exception thrown while writing record: BlockAdditionEvent(ReceivedBlockInfo(0,Some(3521),Some(SequenceNumberRanges(SequenceNumberRange(StreamEventsPRD,shardId-000000000003,49558087746891612304997255299934807015508295035511636018,49558087746891612304997255303224294170679701088606617650), SequenceNumberRange(StreamEventsPRD,shardId-000000000004,49558087949939897337618579003482122196174788079896232002,49558087949939897337618579006984380295598368799020023874), SequenceNumberRange(StreamEventsPRD,shardId-000000000001,49558087735072217349776025034858012188384702720257294354,49558087735072217349776025038332464993957147037082320914), SequenceNumberRange(StreamEventsPRD,shardId-000000000009,49558088270111696152922722880993488801473174525649617042,49558088270111696152922722884455852348849472550727581842), SequenceNumberRange(StreamEventsPRD,shardId-000000000000,49558087841379869711171505550483827793283335010434154498,49558087841379869711171505554030816148032657077741551618), SequenceNumberRange(StreamEventsPRD,shardId-000000000002,49558087853556076589569225785774419228345486684446523426,49558087853556076589569225789389107428993227916817989666))),BlockManagerBasedStoreResult(input-0-1453142312126,Some(3521)))) to the WriteAheadLog.
java.util.concurrent.TimeoutException: Futures timed out after [5000 milliseconds]
    at scala.concurrent.impl.Promise$DefaultPromise.ready(Promise.scala:219)
    at scala.concurrent.impl.Promise$DefaultPromise.result(Promise.scala:223)
    at scala.concurrent.Await$$anonfun$result$1.apply(package.scala:107)
    at scala.concurrent.BlockContext$DefaultBlockContext$.blockOn(BlockContext.scala:53)
    at scala.concurrent.Await$.result(package.scala:107)
    at org.apache.spark.streaming.util.BatchedWriteAheadLog.write(BatchedWriteAheadLog.scala:81)
    at org.apache.spark.streaming.scheduler.ReceivedBlockTracker.writeToLog(ReceivedBlockTracker.scala:232)
    at org.apache.spark.streaming.scheduler.ReceivedBlockTracker.addBlock(ReceivedBlockTracker.scala:87)
    at org.apache.spark.streaming.scheduler.ReceiverTracker.org$apache$spark$streaming$scheduler$ReceiverTracker$$addBlock(ReceiverTracker.scala:321)
    at org.apache.spark.streaming.scheduler.ReceiverTracker$ReceiverTrackerEndpoint$$anonfun$receiveAndReply$1$$anon$1$$anonfun$run$1.apply$mcV$sp(ReceiverTracker.scala:500)
    at org.apache.spark.util.Utils$.tryLogNonFatalError(Utils.scala:1230)
    at org.apache.spark.streaming.scheduler.ReceiverTracker$ReceiverTrackerEndpoint$$anonfun$receiveAndReply$1$$anon$1.run(ReceiverTracker.scala:498)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
    at java.lang.Thread.run(Thread.java:745)

源代码提取

...
     // Function to create a new StreamingContext and set it up
  def setupContext(): StreamingContext = {
    ...
    // Create a StreamingContext
    val ssc = new StreamingContext(sc, Seconds(batchIntervalSeconds))

    // Create a Kinesis DStream
    val data = KinesisUtils.createStream(ssc,
      kinesisAppName, kinesisStreamName,
      kinesisEndpointUrl, RegionUtils.getRegionByEndpoint(kinesisEndpointUrl).getName(),
      InitialPositionInStream.LATEST, Seconds(kinesisCheckpointIntervalSeconds),
      StorageLevel.MEMORY_AND_DISK_SER_2, awsAccessKeyId, awsSecretKey)
...
    ssc.checkpoint(checkpointDir)

    ssc
  }


  // Get or create a streaming context.
  val ssc = StreamingContext.getActiveOrCreate(checkpointDir, setupContext)

  ssc.start()
  ssc.awaitTermination()

Answer 1

遵循zero323关于发布我的评论作为答案的建议：

增加spark.streaming.driver.writeAheadLog.batchingTimeout解决了检查点超时问题。我们在确保有足够的空间后就这样做了。我们已经测试了一段时间。因此，我只建议在仔细考虑后再增加它。

细节

我们在$ SPARK_HOME / conf / spark-defaults.conf中使用了以下两个设置：

spark.streaming.driver.writeAheadLog.allowBatching true spark.streaming.driver.writeAheadLog.batchingTimeout 15000

最初，我们仅将spark.streaming.driver.writeAheadLog.allowBatching设置为true。

在进行更改之前，我们已经在测试环境中重现了问题中提到的问题（“ ... ReceivedBlockTracker：写入记录时引发了异常...”）。它每隔几个小时发生一次。更改后，问题消失了。我们将其运行了几天后才投入生产。

我们发现WriteAheadLogUtils类的getBatchingTimeout（）方法的默认值为5000ms，如下所示：

def getBatchingTimeout(conf: SparkConf): Long = {
    conf.getLong(DRIVER_WAL_BATCHING_TIMEOUT_CONF_KEY, defaultValue = 5000)
}