我已经配置了3个Kafka经纪人:
/opt/kafka/bin/kafka-topics.sh --describe --topic logs --zookeeper=localhost:2181
Topic:logs PartitionCount:8 ReplicationFactor:3 Configs:
Topic: logs Partition: 0 Leader: 1 Replicas: 1,2,0 Isr: 0,1,2
Topic: logs Partition: 1 Leader: 2 Replicas: 2,0,1 Isr: 0,1,2
Topic: logs Partition: 2 Leader: 0 Replicas: 0,1,2 Isr: 0,1,2
Topic: logs Partition: 3 Leader: 1 Replicas: 1,0,2 Isr: 0,1,2
Topic: logs Partition: 4 Leader: 2 Replicas: 2,1,0 Isr: 0,1,2
Topic: logs Partition: 5 Leader: 0 Replicas: 0,2,1 Isr: 0,1,2
Topic: logs Partition: 6 Leader: 1 Replicas: 1,2,0 Isr: 0,1,2
Topic: logs Partition: 7 Leader: 2 Replicas: 2,0,1 Isr: 0,1,2
我注意到所有时间最小偏移滞后值都是1.
/opt/kafka/bin/kafka-consumer-offset-checker.sh --group logreaders --topic logs --zookeeper=localhost:2181
Group Topic Pid Offset logSize Lag Owner
logreaders logs 0 255851 255852 1 none
logreaders logs 1 254514 254515 1 none
logreaders logs 2 252885 252888 3 none
logreaders logs 3 255904 255905 1 none
logreaders logs 4 253897 253898 1 none
logreaders logs 5 253517 253519 2 none
logreaders logs 6 255098 255099 1 none
logreaders logs 7 255412 255414 2 none
我预计最小偏移延迟应该是-0,所以我想弄清楚这是否正常。它看起来不像我正在丢失消息,但这让我感到困惑。
Kafka版本为0.8.2.1
答案 0 :(得分:-2)
此问题与您的 KAFKA PERFORMANCE TUNING 有关。很明显,你的消费者无法赶上制片人
这可以通过调整消费者配置来实现。 Apache Kafka已就此提供了非常好的documentation,请参考并调整您的消费者。
如果您对消费者调整有任何疑问,请告诉我。