由于目前每天需要从kafka中消费20亿条左右的消息，集群压力有点大，会导致job不同程度的异常退出。原来使用spark1.1.0版本中的createStream函数，但是在数据处理速度跟不上数据消费速度且job异常退出的情况下，可能造成大量的数据丢失。幸好，Spark后续版本对这一情况有了很大的改进，1.2版本加入WAL特性，，但是性能应该会受到一些影响（本人未测试），1.3版本可以直接通过低阶API从kafka的topic消费消息，并且不再向zookeeper中更新consumer offsets，使得基于zookeeper的consumer offsets的监控工具都会失效。
官方只是非常简单的描述了可以用以下方法修改zookeeper中的consumer offsets（可以查看）：
// Hold a reference to the current offset ranges, so it can be used downstream var offsetRanges = Array[OffsetRange]() directKafkaStream.transform { rdd => offsetRanges = rdd.asInstanceOf[HasOffsetRanges].offsetRanges rdd }.map { ... }.foreachRDD { rdd => for (o <- offsetRanges) { println(s"${o.topic} ${o.partition} ${o.fromOffset} ${o.untilOffset}") } ... }
所以更新zookeeper中的consumer offsets还需要自己去实现，并且官方提供的两个createDirectStream重载并不能很好的满足我的需求，需要进一步封装。具体看以下KafkaManager类的代码：
package org.apache.spark.streaming.kafka import kafka.common.TopicAndPartition import kafka.message.MessageAndMetadata import kafka.serializer.Decoder import org.apache.spark.SparkException import org.apache.spark.rdd.RDD import org.apache.spark.streaming.StreamingContext import org.apache.spark.streaming.dstream.InputDStream import org.apache.spark.streaming.kafka.KafkaCluster.{LeaderOffset} import scala.reflect.ClassTag /** * Created by knowpigxia on 15-8-5. */ class KafkaManager(val kafkaParams: Map[String, String]) extends Serializable { private val kc = new KafkaCluster(kafkaParams) /** * 创建数据流 * @param ssc * @param kafkaParams * @param topics * @tparam K * @tparam V * @tparam KD * @tparam VD * @return */ def createDirectStream[K: ClassTag, V: ClassTag, KD <: Decoder[K]: ClassTag, VD <: Decoder[V]: ClassTag]( ssc: StreamingContext, kafkaParams: Map[String, String], topics: Set[String]): InputDStream[(K, V)] = { val groupId = kafkaParams.get("group.id").get // 在zookeeper上读取offsets前先根据实际情况更新offsets setOrUpdateOffsets(topics, groupId) //从zookeeper上读取offset开始消费message val messages = { val partitionsE = kc.getPartitions(topics) if (partitionsE.isLeft) throw new SparkException(s"get kafka partition failed: ${partitionsE.left.get}") val partitions = partitionsE.right.get val consumerOffsetsE = kc.getConsumerOffsets(groupId, partitions) if (consumerOffsetsE.isLeft) throw new SparkException(s"get kafka consumer offsets failed: ${consumerOffsetsE.left.get}") val consumerOffsets = consumerOffsetsE.right.get KafkaUtils.createDirectStream[K, V, KD, VD, (K, V)]( ssc, kafkaParams, consumerOffsets, (mmd: MessageAndMetadata[K, V]) => (mmd.key, mmd.message)) } messages } /** * 创建数据流前，根据实际消费情况更新消费offsets * @param topics * @param groupId */ private def setOrUpdateOffsets(topics: Set[String], groupId: String): Unit = { topics.foreach(topic => { var hasConsumed = true val partitionsE = kc.getPartitions(Set(topic)) if (partitionsE.isLeft) throw new SparkException(s"get kafka partition failed: ${partitionsE.left.get}") val partitions = partitionsE.right.get val consumerOffsetsE = kc.getConsumerOffsets(groupId, partitions) if (consumerOffsetsE.isLeft) hasConsumed = false if (hasConsumed) {// 消费过 /** * 如果streaming程序执行的时候出现kafka.common.OffsetOutOfRangeException， * 说明zk上保存的offsets已经过时了，即kafka的定时清理策略已经将包含该offsets的文件删除。 * 针对这种情况，只要判断一下zk上的consumerOffsets和earliestLeaderOffsets的大小， * 如果consumerOffsets比earliestLeaderOffsets还小的话，说明consumerOffsets已过时, * 这时把consumerOffsets更新为earliestLeaderOffsets */ val earliestLeaderOffsetsE = kc.getEarliestLeaderOffsets(partitions) if (earliestLeaderOffsetsE.isLeft) throw new SparkException(s"get earliest leader offsets failed: ${earliestLeaderOffsetsE.left.get}") val earliestLeaderOffsets = earliestLeaderOffsetsE.right.get val consumerOffsets = consumerOffsetsE.right.get // 可能只是存在部分分区consumerOffsets过时，所以只更新过时分区的consumerOffsets为earliestLeaderOffsets var offsets: Map[TopicAndPartition, Long] = Map() consumerOffsets.foreach({ case(tp, n) => val earliestLeaderOffset = earliestLeaderOffsets(tp).offset if (n < earliestLeaderOffset) { println("consumer group:" + groupId + ",topic:" + tp.topic + ",partition:" + tp.partition + " offsets已经过时，更新为" + earliestLeaderOffset) offsets += (tp -> earliestLeaderOffset) } }) if (!offsets.isEmpty) { kc.setConsumerOffsets(groupId, offsets) } } else {// 没有消费过 val reset = kafkaParams.get("auto.offset.reset").map(_.toLowerCase) var leaderOffsets: Map[TopicAndPartition, LeaderOffset] = null if (reset == Some("smallest")) { val leaderOffsetsE = kc.getEarliestLeaderOffsets(partitions) if (leaderOffsetsE.isLeft) throw new SparkException(s"get earliest leader offsets failed: ${leaderOffsetsE.left.get}") leaderOffsets = leaderOffsetsE.right.get } else { val leaderOffsetsE = kc.getLatestLeaderOffsets(partitions) if (leaderOffsetsE.isLeft) throw new SparkException(s"get latest leader offsets failed: ${leaderOffsetsE.left.get}") leaderOffsets = leaderOffsetsE.right.get } val offsets = leaderOffsets.map { case (tp, offset) => (tp, offset.offset) } kc.setConsumerOffsets(groupId, offsets) } }) } /** * 更新zookeeper上的消费offsets * @param rdd */ def updateZKOffsets(rdd: RDD[(String, String)]) : Unit = { val groupId = kafkaParams.get("group.id").get val offsetsList = rdd.asInstanceOf[HasOffsetRanges].offsetRanges for (offsets <- offsetsList) { val topicAndPartition = TopicAndPartition(offsets.topic, offsets.partition) val o = kc.setConsumerOffsets(groupId, Map((topicAndPartition, offsets.untilOffset))) if (o.isLeft) { println(s"Error updating the offset to Kafka cluster: ${o.left.get}") } } } }
接下来再给一个简单的例子：