sparkStreaming 微批处理任务监控

sparkstreaming任务在处理数据时存在堆积情况，但是仍然会不断从kafka拉取数据
首先先说2个参数

spark.streaming.kafka.consumer.poll.ms

spark去kafka取数的时候，会有一个超时时间。如果两次尝试后都出现了超时，这个任务就会失败，然后spark会把这个任务分发到其它的executor上面去执行，这就会导致一定的调度耗时。
在spark中这个参数的默认值是512ms。如果超时时间很短，但是kafka响应的时间很长，这就会导致spark中有很多的任务失败。如果超时时间太长，spark在这段时间内什么都不做且最终还是会超时，就会产生很大的延迟（poll timeout+executor中的任务调度）。
如果spark作业中有很多的任务失败，你需要去增大这个值（或者你该去kafka那边看看为啥它那么久都没响应）。
如果不指定这个参数，spark可能就会使用spark.network.timeout（所有网络交互的默认超时，默认是120s）

spark.streaming.kafka.maxRatePerPartition

设置Spark Streaming从kafka分区每秒拉取的条数

spark.streaming.backpressure.enabled

参数设置为ture,开启背压机制后Spark Streaming会根据延迟动态去kafka消费数据,上限由spark.streaming.kafka.maxRatePerPartition参数控制，所以两个参数一般会一起使用

spark.streaming.stopGracefullyOnShutdown

参数设置成ture,Spark会在JVM关闭时正常关闭StreamingContext,而不是立马关闭

reconnect.backoff.ms

如果客户端与kafka的broker的连接断开了，客户端会等reconnect.backoff.ms之后重新连接。

reconnect.backoff.max.ms

重连的总的最大时间，每次连接失败，重连时间都会指数级增加，每次增加的时间会存在20%的随机抖动，以避免连接风暴。

正文主题

val ssc:  = new StreamingContext(spark.sparkContext, Seconds(interval))
ssc.addStreamingListener(new MyStreamingListener(appName, 10))

MyStreamingListener自定义的监听类

class MyStreamingListener(private val appName:String, private val duration: Int) extends StreamingListener {
 //在流处理启动时调用
  override def onStreamingStarted(streamingStarted: StreamingListenerStreamingStarted): Unit = {
    super.onStreamingStarted(streamingStarted)
  }
  //当一个接收器启动时调用
  override def onReceiverStarted(receiverStarted: StreamingListenerReceiverStarted): Unit = super.onReceiverStarted(receiverStarted)
  //当接收端报告错误时调用
  override def onReceiverError(receiverError: StreamingListenerReceiverError): Unit = {
    super.onReceiverError(receiverError)
  }
  //当一个接收器被停止时呼叫
  override def onReceiverStopped(receiverStopped: StreamingListenerReceiverStopped): Unit = {
  }
  //在提交一批作业以进行处理时调用。
  override def onBatchSubmitted(batchSubmitted: StreamingListenerBatchSubmitted): Unit = {
    super.onBatchSubmitted(batchSubmitted)
  }
  //在开始处理一批作业时调用。
  override def onBatchStarted(batchStarted: StreamingListenerBatchStarted): Unit = {
    super.onBatchStarted(batchStarted)
  }
  //当一批作业的处理完成时调用。
  override def onBatchCompleted(batchCompleted: StreamingListenerBatchCompleted): Unit = {
 
  }
  //在开始处理批处理作业时调用。
  override def onOutputOperationStarted(outputOperationStarted: StreamingListenerOutputOperationStarted): Unit = {}
  
  //当批处理作业的处理完成时调用。
  override def onOutputOperationCompleted(outputOperationCompleted: StreamingListenerOutputOperationCompleted): Unit = {
  }
}

监听类源码

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *    http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.spark.streaming.scheduler

import scala.collection.mutable.Queue

import org.apache.spark.annotation.DeveloperApi
import org.apache.spark.util.Distribution

/**
 * :: DeveloperApi ::
 * Base trait for events related to StreamingListener
 */
@DeveloperApi
sealed trait StreamingListenerEvent

@DeveloperApi
case class StreamingListenerStreamingStarted(time: Long) extends StreamingListenerEvent

@DeveloperApi
case class StreamingListenerBatchSubmitted(batchInfo: BatchInfo) extends StreamingListenerEvent

@DeveloperApi
case class StreamingListenerBatchCompleted(batchInfo: BatchInfo) extends StreamingListenerEvent

@DeveloperApi
case class StreamingListenerBatchStarted(batchInfo: BatchInfo) extends StreamingListenerEvent

@DeveloperApi
case class StreamingListenerOutputOperationStarted(outputOperationInfo: OutputOperationInfo)
  extends StreamingListenerEvent

@DeveloperApi
case class StreamingListenerOutputOperationCompleted(outputOperationInfo: OutputOperationInfo)
  extends StreamingListenerEvent

@DeveloperApi
case class StreamingListenerReceiverStarted(receiverInfo: ReceiverInfo)
  extends StreamingListenerEvent

@DeveloperApi
case class StreamingListenerReceiverError(receiverInfo: ReceiverInfo)
  extends StreamingListenerEvent

@DeveloperApi
case class StreamingListenerReceiverStopped(receiverInfo: ReceiverInfo)
  extends StreamingListenerEvent

/**
 * :: DeveloperApi ::
 * A listener interface for receiving information about an ongoing streaming
 * computation.
 */
@DeveloperApi
trait StreamingListener {

  /** Called when the streaming has been started */
  def onStreamingStarted(streamingStarted: StreamingListenerStreamingStarted) { }

  /** Called when a receiver has been started */
  def onReceiverStarted(receiverStarted: StreamingListenerReceiverStarted) { }

  /** Called when a receiver has reported an error */
  def onReceiverError(receiverError: StreamingListenerReceiverError) { }

  /** Called when a receiver has been stopped */
  def onReceiverStopped(receiverStopped: StreamingListenerReceiverStopped) { }

  /** Called when a batch of jobs has been submitted for processing. */
  def onBatchSubmitted(batchSubmitted: StreamingListenerBatchSubmitted) { }

  /** Called when processing of a batch of jobs has started.  */
  def onBatchStarted(batchStarted: StreamingListenerBatchStarted) { }

  /** Called when processing of a batch of jobs has completed. */
  def onBatchCompleted(batchCompleted: StreamingListenerBatchCompleted) { }

  /** Called when processing of a job of a batch has started. */
  def onOutputOperationStarted(
      outputOperationStarted: StreamingListenerOutputOperationStarted) { }

  /** Called when processing of a job of a batch has completed. */
  def onOutputOperationCompleted(
      outputOperationCompleted: StreamingListenerOutputOperationCompleted) { }
}


/**
 * :: DeveloperApi ::
 * A simple StreamingListener that logs summary statistics across Spark Streaming batches
 * @param numBatchInfos Number of last batches to consider for generating statistics (default: 10)
 */
@DeveloperApi
class StatsReportListener(numBatchInfos: Int = 10) extends StreamingListener {
  // Queue containing latest completed batches
  val batchInfos = new Queue[BatchInfo]()

  override def onBatchCompleted(batchStarted: StreamingListenerBatchCompleted) {
    batchInfos.enqueue(batchStarted.batchInfo)
    if (batchInfos.size > numBatchInfos) batchInfos.dequeue()
    printStats()
  }

  def printStats() {
    showMillisDistribution("Total delay: ", _.totalDelay)
    showMillisDistribution("Processing time: ", _.processingDelay)
  }

  def showMillisDistribution(heading: String, getMetric: BatchInfo => Option[Long]) {
    org.apache.spark.scheduler.StatsReportListener.showMillisDistribution(
      heading, extractDistribution(getMetric))
  }

  def extractDistribution(getMetric: BatchInfo => Option[Long]): Option[Distribution] = {
    Distribution(batchInfos.flatMap(getMetric(_)).map(_.toDouble))
  }
}

重点看上边代码块 70行 -100行