1、FETCH_SESSION_TOPIC_ID_ERROR
用java客户端消费topic的数据的时候,会报错The fetch session encountered inconsistent topic ID usage Caused by: org.apache.kafka.common.KafkaException: Unexpected error in join group response: The fetch session encountered inconsistent topic ID usage
如何处理
FETCH_SESSION_TOPIC_ID_ERROR 关闭了自动创建group，可以打开或先手动创建完group再消费
2、问题现象：客户原问题x使用ip：port无法连接，报kafka: client has run out of available brokers to talk to: EOF
网络也是通的
加了这个config.Version = sarama.V1_1_1_0
就可以发送成功了
3、kafka 使用 github.com/Shopify/sarama v1.32.0 --连接 0.10.2 报错
panic: kafka: client has run out of available brokers to talk to (Is your cluster reachable?)
client/metadata fetching metadata for all topics from broker [xxx.xxx.xxx.xxx:6169]n
Connected to broker at [xxx.xxx.xxx.xxx:6169] (unregistered)n
client/metadata got error from broker [-1 824634286320] while fetching metadata: %!v(MISSING)n
Closed connection to broker [xxx.xxx.xxx.xxx:6169]n
[client/metadata no available broker to send metadata request to]
client/brokers resurrecting [1] dead seed brokers
[Closing Client]
服务端的报错：
[2023-07-26 14:03:40,853] ERROR Closing socket for xxx.xxx.xxx.xxx:57918-92605684 because of error (kafka.network.Processor)
org.apache.kafka.common.errors.InvalidRequestException: Error getting request for apiKey: 3 and apiVersion: 5
Caused by: java.lang.IllegalArgumentException: Invalid version for API key 3: 5
需要客户端指定版本：
默认是：
4、消费者报错报retriable exception
报错信息：[Consumer clientId=consumer-6, groupId=-test] Asynchrnous auto-commit of offsets {gactravel.charge.estimate.result.bi.topic-0=OffsetAndMetadata{offset=5499679, metadata=''}} failed: Offset commit failed with a retriable exception. You should retry committing the latest consumed offsets.
辛苦大佬协助看下报错原因 使用的异步提交导致的
可以看一下这个：https://baijiahao.baidu.com/s?id=1761383239525043271&wfr=spider&for=pc
可以改成同步提交再观察一下
5、消息队列CKafKa报错kafka.Error {"event": "Application maximum poll interval (300000ms) exceeded by xxx ms"}的解决方案
报错明显提示maximum poll interval超时。指的是超过消费者超过max.poll.interval.ms=300000这个时间退出了
表示##### 使用 Kafka 消费分组机制时，再次调用 poll 允许的最大间隔。如果在该时间内没有再次调用 poll，则认为该消费者已经失败，Broker 会重新发起 Rebalance 把分配给它的 partition 分配给其他消费者
处理建议：增大参数max.poll.interval.ms
后续有相关的报错会继续更新，欢迎持续关注
