一 需求描述
一个分区可以有多个副本，这些副本保存在不同的broker上，可实现kafka高可用。每个分区的副本中都会有一个作为Leader。当一个broker失败时，Leader在这台broker上的分区都会变得不可用，kafka会自动移除Leader，再其他副本中选一个作为新的Leader。
假如我们目前kafka集群环境中的topic没有副本，可通过如下方法增加topic的副本
二 操作步骤
2.1 查看某个topic当前的分区及副本信息
[root@Centos7-Mode-V8 kafka]# bin/kafka-topics.sh --describe --zookeeper 192.168.144.247:3292,192.168.144.251:3292,192.168.144.253:3292 --topic mubai
Topic:mubai PartitionCount:3 ReplicationFactor:1 Configs:
Topic: mubai Partition: 0 Leader: 0 Replicas: 0 Isr: 0
Topic: mubai Partition: 1 Leader: 1 Replicas: 1 Isr: 1
Topic: mubai Partition: 2 Leader: 0 Replicas: 0 Isr: 0
2.2 制定分区及副本分配策略
注意修改下面文件中的topic为自己的topic，注意修改里面的partition，replicas，值需要与上面2.1步骤中查询出来的自己topic的现有值保持一致
[root@Centos7-Mode-V8 opt]# vi /opt/test.json
{
"version": 1,
"partitions": [
{
"topic": "mubai",
"partition": 0,
"replicas": [
0,
1,
2
]
},
{
"topic": "mubai",
"partition": 1,
"replicas": [
1,
2,
0
]
},
{
"topic": "mubai",
"partition": 2,
"replicas": [
2,
0,
1
]
}
]
}2.3 扩副本
[root@Centos7-Mode-V8 kafka]# bin/kafka-reassign-partitions.sh --zookeeper 192.168.144.247:3292,192.168.144.251:3292,192.168.144.253:3292 --reassignment-json-file /opt/test.json --execute
Current partition replica assignment
{"version":1,"partitions":[{"topic":"mubai","partition":0,"replicas":[0],"log_dirs":["any"]},{"topic":"mubai","partition":2,"replicas":[0],"log_dirs":["any"]},{"topic":"mubai","partition":1,"replicas":[1],"log_dirs":["any"]}]}
Save this to use as the --reassignment-json-file option during rollback
Successfully started reassignment of partitions.
2.4 验证副本是否添加成功
bin/kafka-topics.sh --describe --zookeeper 192.168.144.247:3292,192.168.144.251:3292,192.168.144.253:3292 --topic mubai
可以看到输出结果和我们上面规划的是一样的：
————————————————
版权声明：本文为CSDN博主「shykevin」的原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/shykevin/article/details/90103360