zoukankan      html  css  js  c++  java
  • kafka使用getOffsetsBefore()获取获取offset异常分析

    根据时间戳获取kafka的topic的偏移量,结果获取的偏移量量数据组的长度为0,就会出现如下的数组下标越界的异常,实现的原理是使用了kafka的getOffsetsBefore()方法:

    Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException : 0

         at co.gridport.kafka.hadoop.KafkaInputFetcher.getOffset(KafkaInputFetcher.java:126)

         at co.gridport.kafka.hadoop.TestKafkaInputFetcher.testGetOffset(TestKafkaInputFetcher.java:68)

         at co.gridport.kafka.hadoop.TestKafkaInputFetcher.main(TestKafkaInputFetcher.java:80)

    OffsetResponse(0,Map([barrage_detail,0] -> error: kafka.common.UnknownException offsets: ))

    源码如下:

    /*

          * 得到partition的offset Finding Starting Offset for Reads

          */

    public Long getOffset(Long time) throws IOException {

               TopicAndPartition topicAndPartition = new TopicAndPartition(this.topic , this.partition );

               Map<TopicAndPartition, PartitionOffsetRequestInfo> requestInfo = new HashMap<TopicAndPartition, PartitionOffsetRequestInfo>();

    requestInfo.put( topicAndPartition, new PartitionOffsetRequestInfo(time, 1));

               kafka.javaapi.OffsetRequest request = new kafka.javaapi.OffsetRequest(

    requestInfo, kafka.api.OffsetRequest.CurrentVersion(), this. client_id);

               OffsetResponse response = this. kafka_consumer.getOffsetsBefore( request);

    if ( response.hasError()) {

    log.error( "Error fetching data Offset Data the Broker. Reason: " + response.errorCode(this.topic, this. partition));

    throw new IOException ( "Error fetching kafka Offset by time:" + response.errorCode(this.topic, this. partition));

               }

    //         if (response.offsets(this.topic, this.partition).length == 0){

    //              return getOffset(kafka.api.OffsetRequest

    //                         .EarliestTime());

    //         }

    return response.offsets( this. topic, this. partition)[0];

         }

    返回的response对象会有error: kafka.common.UnknownException offsets如下异常:

    OffsetResponse(0,Map([barrage_detail,0] -> error: kafka.common.UnknownException offsets: ))

    同时呢,response.hasError()检查不到error。

    是什么原因造成了response.offsets(this.topic,this.partition)的返回数组的长度为0呢?

    分析了getOffsetsBefore()方法的源码,并做源码了大量的测试,终于重现了这种情况?

    1.getOffsetsBefore()的功能以及实现原理:

    getOffsetsBefore的功能是返回某个时间点前的maxOffsetNum个offset(时间点指的是kafka日志文件的最后修改时间,offset指的是log文件名中的offset,这个offset是该日志文件的第一条记录的offset,即base offset;maxNumOffsets参数即返回结果的最大个数,如果该参数为2,就返回指定时间点前的2个offset,如果是负数,就报逻辑错误,怎么能声明一个长度为负数的数组呢,呵呵);

    根据这个实现原理,所以返回的结果长度为0是合理的,反映的是这个时间点前没有kafka日志这种情况,该情况自然就没有offset了。

    说明我们指定的时间参数太早了,正常的时间范围为:最早的时间之后的时间参数都可以有返回值。

    其实合理的处理方式应该为如果这个时间点前没有值,就返回最早的offset了,对api的使用者就友好多了我们可以自己来实现这个功能。

    代码如下:

    /*

          * 得到partition的offset Finding Starting Offset for Reads

          */

    public Long getOffset(Long time ) throws IOException {

               TopicAndPartition topicAndPartition = new TopicAndPartition(this .topic , this.partition);

               Map<TopicAndPartition, PartitionOffsetRequestInfo> requestInfo = new HashMap<TopicAndPartition, PartitionOffsetRequestInfo>();

    requestInfo.put( topicAndPartition, new PartitionOffsetRequestInfo(time, 1));

               kafka.javaapi.OffsetRequest request = new kafka.javaapi.OffsetRequest(

    requestInfo, kafka.api.OffsetRequest.CurrentVersion(), this. client_id);

               OffsetResponse response = this. kafka_consumer.getOffsetsBefore( request);

    if ( response.hasError()) {

    log.error( "Error fetching data Offset Data the Broker. Reason: " + response.errorCode(this.topic, this. partition));

    throw new IOException ( "Error fetching kafka Offset by time:" + response.errorCode(this.topic, this. partition));

               }

                //如果返回的数据长度为0,就获取最早的offset。

    if ( response.offsets( this. topic, this. partition). length == 0){

    return getOffset(kafka.api.OffsetRequest

                               . EarliestTime());

               }

    return response.offsets( this. topic, this. partition)[0];

         }

  • 相关阅读:
    Ibatis入门基本语法(转) good
    zip文件压缩(转)
    联合创始人股权分配,五五分是最糟糕的做法(转)
    家长如何检查孩子的家庭作业
    oracle存储过程实例
    MachineKey
    写写我那天參加过的《文明之光》书友会
    各种加解密算法比較
    算法分析---查找最大回文子串
    随机数是骗人的,.Net、Java、C为我作证
  • 原文地址:https://www.cnblogs.com/honeybee/p/5249714.html
Copyright © 2011-2022 走看看