kafka偏移量利用redis来管理

import Utils.RedisOffset
import day09.Jpools
import org.apache.kafka.common.serialization.StringDeserializer
import org.apache.spark.SparkConf
import org.apache.spark.streaming.kafka010.{ConsumerStrategies, HasOffsetRanges, KafkaUtils, LocationStrategies}
import org.apache.spark.streaming.{Seconds, StreamingContext}

object SSCDirectKafka010_Redis_Offset {

  def main(args: Array[String]): Unit = {

    val conf = new SparkConf().setAppName("SSCDirectKafka010_Redis_Offset").setMaster("local[*]")

    //配置在kafka中每次拉取的数据量，这里配置的2并不是每次在kafka拉取2条数据，而是：2*分区数量*采样时间(12)
    conf.set("spark.streaming.kafka.maxRatePerPartition", "2")
    //是否优雅的停止你的SparkStreaming，如果不加这个参数的话，服务停止的时候可能会造成数据的丢失
    conf.set("spark.streaming.stopGracefullyOnShutdown", "true")

    val ssc = new StreamingContext(conf,Seconds(2))

    val groupId = "day11_09"

    val kafkaParams = Map[String, Object](
      "bootstrap.servers" -> "bigdata01:9092,bigdata02:9092,bigdata03:9092",
      "key.deserializer" -> classOf[StringDeserializer],
      "value.deserializer" -> classOf[StringDeserializer],
      "group.id" -> groupId,
      "auto.offset.reset" -> "earliest",
      "enable.auto.commit" -> (false: java.lang.Boolean) //是否自动递交偏移量
    )

    val topic = "helloTopic"
    val topics = Array(topic)

    //获取偏移量
    val redisManage = RedisOffset(topic)
    val result = if (redisManage.size > 0){
      KafkaUtils.createDirectStream[String,String](
        ssc,
        LocationStrategies.PreferConsistent,
        ConsumerStrategies.Subscribe[String,String](topics,kafkaParams,redisManage)
      )
    }else {
      KafkaUtils.createDirectStream[String, String](
        ssc,
        LocationStrategies.PreferConsistent,
        ConsumerStrategies.Subscribe[String, String](topics, kafkaParams)
      )
    }

    result.foreachRDD(foreachFunc = rdd => {
      val jedis = Jpools.getJedis
      val offsetRange = rdd.asInstanceOf[HasOffsetRanges].offsetRanges


      //拉取到driver端
      val reduced = rdd.map(t=>(t.value(),1)).reduceByKey(_+_).collect()

      //设置回滚
      val transaction = jedis.multi()//返回一个事务控制对象
      try{
        for (i <- reduced){
          transaction.hincrBy("helloTopic",i._1,i._2)
        }
        for(i <- offsetRange){
          println(i)
          transaction.hset(groupId,i.topic+"-"+i.partition,i.untilOffset.toString)
        }
        transaction.exec()
      }catch {
        case _ => println("你报错了，需要回滚")
        transaction.discard()
      }

      jedis.close()
    })

    ssc.start()

    ssc.awaitTermination()
  }
}

工具包

import java.util

import day09.Jpools
import org.apache.kafka.common.TopicPartition
import scala.collection.mutable._

object RedisOffset {

  def apply(groupId:String) = {
    val redisOffset = Map[TopicPartition,Long]()

    //获取jedis连接
    val jedis = Jpools.getJedis
    val tpOffset: util.Map[String, String] = jedis.hgetAll(groupId)

    import scala.collection.JavaConversions._
    val tpOffsetList = tpOffset.toList
    for (i <- tpOffsetList){
      val s = i._1.split("-")
      redisOffset += (new TopicPartition(s(0),s(1).toInt) -> i._2.toLong)
    }
    redisOffset
  }
}

import org.apache.commons.pool2.impl.GenericObjectPoolConfig
import redis.clients.jedis.{Jedis, JedisPool}

object Jpools {

  private val poolConfig = new GenericObjectPoolConfig
  poolConfig.setMaxIdle(5)
  poolConfig.setMaxTotal(2000)

  private val jedisPool = new JedisPool(poolConfig,"hadoop01")

  def getJedis:Jedis = {
    val jedis = jedisPool.getResource
    jedis.select(1)
    jedis
  }
}