kafka max.poll.interval.ms配置太短

2021-10-27 01:34:52 阅读：249 来源： 互联网

标签：max interval kafka props put poll consumer

max.poll.interval.ms这个应该是消费者每次去kafka拉取数据最大间隔，如果超过这个间隔，服务端会认为消费者已离线。触发rebalance

demo

 1     public ConsumerDemo(String topicName) {
 2         Properties props = new Properties();
 3         props.put("bootstrap.servers", "localhost:9092");
 4         props.put("group.id", GROUPID);
 5         props.put("enable.auto.commit", "false");
 6         props.put("max.poll.interval.ms", "1000");
 7         props.put("auto.offset.reset", "earliest");
 8         props.put("key.deserializer", StringDeserializer.class.getName());
 9         props.put("value.deserializer", StringDeserializer.class.getName());
10         this.consumer = new KafkaConsumer<String, String>(props);
11         this.topic = topicName;
12         this.consumer.subscribe(Arrays.asList(topic));
13     }

5行配置自动提交为false，手动提交。6行配置 max.poll.interval.ms为1秒

 1     public void receiveMsg() {
 2         int messageNo = 1;
 3         System.out.println("---------开始消费---------");
 4         try {
 5             for (;;) {
 6                 msgList = consumer.poll(1000);
 7                 System.out.println("start sleep" + System.currentTimeMillis() / 1000);
 8                 Thread.sleep(10000);
 9                 System.out.println("end sleep" + System.currentTimeMillis() / 1000);
10                 if(null!=msgList&&msgList.count()>0){
11                     for (ConsumerRecord<String, String> record : msgList) {
12                         System.out.println(messageNo+"=======receive: key = " + record.key() + ", value = " + record.value()+" offset==="+record.offset());
13                     }
14                 }else{
15                     Thread.sleep(1000);
16                 }
17                 consumer.commitSync();
18             }
19         } catch (InterruptedException e) {
20             e.printStackTrace();
21         } finally {
22             consumer.close();
23         }
24     }

8行slepp 10秒，模拟处理消息耗时。提交消息的时候报错

Exception in thread "main" org.apache.kafka.clients.consumer.CommitFailedException: Commit cannot be completed since the group has already rebalanced and assigned the partitions to another member. This means that the time between subsequent calls to poll() was longer than the configured max.poll.interval.ms, which typically implies that the poll loop is spending too much time message processing. You can address this either by increasing the session timeout or by reducing the maximum size of batches returned in poll() with max.poll.records.
    at org.apache.kafka.clients.consumer.internals.ConsumerCoordinator.sendOffsetCommitRequest(ConsumerCoordinator.java:722)
    at org.apache.kafka.clients.consumer.internals.ConsumerCoordinator.commitOffsetsSync(ConsumerCoordinator.java:600)
    at org.apache.kafka.clients.consumer.KafkaConsumer.commitSync(KafkaConsumer.java:1211)
    at com.gxf.kafka.ConsumerDemo.receiveMsg(ConsumerDemo.java:49)
    at com.gxf.kafka.ConsumerDemo.main(ConsumerDemo.java:59)

max.poll.interval.ms 可以配置稍微大点，或者减少处理时间，每次少拉取数据，异步处理等

标签：max,interval,kafka,props,put,poll,consumer
来源： https://www.cnblogs.com/luckygxf/p/15468693.html

本站声明： 1. iCode9 技术分享网（下文简称本站）提供的所有内容，仅供技术学习、探讨和分享；
2. 关于本站的所有留言、评论、转载及引用，纯属内容发起人的个人观点，与本站观点和立场无关；
3. 关于本站的所有言论和文字，纯属内容发起人的个人观点，与本站观点和立场无关；
4. 本站文章均是网友提供，不完全保证技术分享内容的完整性、准确性、时效性、风险性和版权归属；如您发现该文章侵犯了您的权益，可联系我们第一时间进行删除；
5. 本站为非盈利性的个人网站，所有内容不会用来进行牟利，也不会利用任何形式的广告来间接获益，纯粹是为了广大技术爱好者提供技术内容和技术思想的分享性交流网站。

ICode9

kafka max.poll.interval.ms配置太短