# kafka 部署 - 单机

https://abc.htmltoo.com/thread-44967.htm

# kafka 编译安装

https://abc.htmltoo.com/thread-45973.htm

# zookeeper 集群部署

https://abc.htmltoo.com/thread-45885.htm

server-1：

/usr/local/kafka/config# vim server.properties

···

broker.id=1 #默认是0，这里自定义

host.name=192.168.0.166 #ip地址

zookeeper.connect=192.168.0.169:2181,192.168.0.170:2181,192.168.0.166:2181 #zookeeper集群地址

···

同样的操作在server-2和server-3上修改一下broker.id和host.name：

scp -r /usr/local/kafka  root@192.168.0.170:/usr/local/

scp -r /usr/local/kafka  root@192.168.0.170:/usr/local/

# 启动

nohup  bin/kafka-server-start.sh  -daemon  config/server.properties  &   #   > test.log 2>&1 & 

-注意：不要关了这个窗口，启用Kafka前请确保ZooKeeper实例已经准备好并开始运行

# 关闭

bin/kafka-server-stop.sh config/server.properties

-三台机器 都开启

-注意的问题是 host.name不要错

# 连接zookeeper测试

[zk: localhost:2181(CONNECTED) 0] ls /

[cluster, controller_epoch, controller, brokers, zookeeper, admin, isr_change_notification, consumers, log_dir_event_notification, latest_producer_id_block, config]

[zk: localhost:2181(CONNECTED) 1] ls /brokers

[ids, topics, seqid]

[zk: localhost:2181(CONNECTED) 2] ls /brokers/ids

[1, 2, 3]

[zk: localhost:2181(CONNECTED) 3]

-zookeeper集群建好之后，通过“ls /”出来的只有zookeeper，连接kafka使用后，/ 下面多了不少东西，其中通过查看/brokers/ids可以发现已经检查到了已经安装的三台kafka的broker.id[1,2,3]