一 搭建单节点单broker的kafka集群
注意：请打开不同的终端分别执行以下步骤
1.复制安装包到/usr/local目录下，解压缩，重命名（或者软链接），配置环境变量
[root@hadoop ~]# cd /usr/local/ [root@hadoop local]# tar xzvf kafka_2.11-2.0.0.tgz [root@hadoop local]# mv kafka_2.11-2.0.0 kafka [root@hadoop local]# ln -s kafka_2.11-2.0.0 kafka #软链接或者重命名执行一条即可 [root@hadoop local]# vi /etc/profile 添加变量 export KAFKA_HOME=/usr/local/kafka 在PATH后添加 :$KAFKA_HOME/bin [root@hadoop local]# source /etc/profile [root@hadoop kafka]# echo $KAFKA_HOME #查看环境变量 /usr/local/kafka
2.启动服务器
启动zookeeper
[root@hadoop kafka]# zookeeper-server-start.sh config/zookeeper.properties [root@hadoop kafka]# jps #打开另一个终端查看是否启动成功 3892 Jps 3566 QuorumPeerMain
启动kafka
[root@hadoop kafka]# kafka-server-start.sh config/server.properties
3.创建topic
#创建一个分区，一个副本的主题 #副本数无法修改，只能在创建主题时指定 [root@hadoop kafka]# kafka-topics.sh --create --zookeeper localhost:2181 --replication-factor 1 --partitions 1 --topic test Created topic "test". [root@hadoop kafka]# kafka-topics.sh --list --zookeeper localhost:2181 #列出主题 test
可以通过zk的客户端观察zk的数据结构
[root@hadoop kafka]# zkCli.sh -server localhost:2181 #进入zk客户端 Connecting to localhost:2181 2018-07-31 14:27:24,876 [myid:] - INFO [main:Environment@100] - Client environment:zookeeper.version=3.4.9-1757313, built on 08/23/2016 06:50 GMT 2018-07-31 14:27:24,879 [myid:] - INFO [main:Environment@100] - Client environment:host.name=hadoop 2018-07-31 14:27:24,880 [myid:] - INFO [main:Environment@100] - Client environment:java.version=1.8.0_11 2018-07-31 14:27:24,882 [myid:] - INFO [main:Environment@100] - Client environment:java.vendor=Oracle Corporation 2018-07-31 14:27:24,882 [myid:] - INFO [main:Environment@100] - Client environment:java.home=/usr/java/jre 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:java.class.path=/usr/local/zookeeper/bin/../build/classes:/usr/local/zookeeper/bin/../build/lib/*.jar:/usr/local/zookeeper/bin/../lib/slf4j-log4j12-1.6.1.jar:/usr/local/zookeeper/bin/../lib/slf4j-api-1.6.1.jar:/usr/local/zookeeper/bin/../lib/netty-3.10.5.Final.jar:/usr/local/zookeeper/bin/../lib/log4j-1.2.16.jar:/usr/local/zookeeper/bin/../lib/jline-0.9.94.jar:/usr/local/zookeeper/bin/../zookeeper-3.4.9.jar:/usr/local/zookeeper/bin/../src/java/lib/*.jar:/usr/local/zookeeper/bin/../conf:.:/usr/java/lib/dt.jar:/usr/java/lib/tools.jar:/usr/java/jre/lib 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:java.library.path=/usr/local/hadoop/lib/native:/usr/java/packages/lib/amd64:/usr/lib64:/lib64:/lib:/usr/lib 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:java.io.tmpdir=/tmp 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:java.compiler=<NA> 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:os.name=Linux 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:os.arch=amd64 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:os.version=3.10.0-514.el7.x86_64 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:user.name=root 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:user.home=/root 2018-07-31 14:27:24,883 [myid:] - INFO [main:Environment@100] - Client environment:user.dir=/usr/local/kafka 2018-07-31 14:27:24,888 [myid:] - INFO [main:ZooKeeper@438] - Initiating client connection, connectString=localhost:2181 sessionTimeout=30000 watcher=org.apache.zookeeper.ZooKeeperMain$MyWatcher@67424e82 Welcome to ZooKeeper! 2018-07-31 14:27:25,037 [myid:] - INFO [main-SendThread(localhost:2181):ClientCnxn$SendThread@1032] - Opening socket connection to server localhost/127.0.0.1:2181. Will not attempt to authenticate using SASL (unknown error) JLine support is enabled 2018-07-31 14:27:25,131 [myid:] - INFO [main-SendThread(localhost:2181):ClientCnxn$SendThread@876] - Socket connection established to localhost/127.0.0.1:2181, initiating session 2018-07-31 14:27:25,153 [myid:] - INFO [main-SendThread(localhost:2181):ClientCnxn$SendThread@1299] - Session establishment complete on server localhost/127.0.0.1:2181, sessionid = 0x10000ded7830002, negotiated timeout = 30000 WATCHER:: WatchedEvent state:SyncConnected type:None path:null [zk: localhost:2181(CONNECTED) 0] ls / [cluster, controller, controller_epoch, brokers, zookeeper, admin, isr_change_notification, consumers, log_dir_event_notification, latest_producer_id_block, config] [zk: localhost:2181(CONNECTED) 1] ls /brokers [ids, topics, seqid] [zk: localhost:2181(CONNECTED) 2] ls /brokers/topics [test] [zk: localhost:2181(CONNECTED) 3] get /brokers/topics/test {"version":1,"partitions":{"0":[0]}} cZxid = 0x22 ctime = Tue Jul 31 14:22:42 CST 2018 mZxid = 0x22 mtime = Tue Jul 31 14:22:42 CST 2018 pZxid = 0x24 cversion = 1 dataVersion = 0 aclVersion = 0 ephemeralOwner = 0x0 dataLength = 36 numChildren = 1 [zk: localhost:2181(CONNECTED) 4] ls /brokers/topics/test [partitions] [zk: localhost:2181(CONNECTED) 5] ls /brokers/topics/test/partitions [0] [zk: localhost:2181(CONNECTED) 6] ls /brokers/topics/test/partitions/0 [state] [zk: localhost:2181(CONNECTED) 7] ls /brokers/topics/test/partitions/0/state [] [zk: localhost:2181(CONNECTED) 8] get /brokers/topics/test/partitions/0/state {"controller_epoch":1,"leader":0,"version":1,"leader_epoch":0,"isr":[0]} cZxid = 0x26 ctime = Tue Jul 31 14:22:42 CST 2018 mZxid = 0x26 mtime = Tue Jul 31 14:22:42 CST 2018 pZxid = 0x26 cversion = 0 dataVersion = 0 aclVersion = 0 ephemeralOwner = 0x0 dataLength = 72 numChildren = 0 [zk: localhost:2181(CONNECTED) 9] quit #退出 Quitting... 2018-07-31 15:01:53,761 [myid:] - INFO [main:ZooKeeper@684] - Session: 0x10000ded7830002 closed 2018-07-31 15:01:53,789 [myid:] - INFO [main-EventThread:ClientCnxn$EventThread@519] - EventThread shut down for session: 0x10000ded7830002 [root@hadoop kafka]#
4.发送消息
[root@hadoop kafka]# kafka-console-producer.sh --broker-list localhost:9092 --topic test >hello world >how are you
5.启动消费者
[root@hadoop kafka]# kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic test --from-beginning hello world how are you
最后查看一下进程
[root@hadoop kafka]# jps 3905 Kafka 7637 ConsoleConsumer 5702 ConsoleProducer 3566 QuorumPeerMain 9135 Jps
二 搭建单节点多broker的kafka集群
首先把上述终端全部关闭，重新打开不同的终端分别执行以下步骤
1.创建并配置多个server配置文件
注意：以下3个broker中，必须有一个broker的端口号是9092。
这个是我踩过的坑，我刚开始建立了3个broker端口号分别为9093,9094,9095，然后消费消息时无论如何都不显示结果，在这卡了半天。
我猜是因为consumer.properties、producer.properties、connect-distributed.properties等文件中默认bootstrap.servers=localhost:9092，如果3个broker中没有一个端口号是9092，需要将这些配置文件中的bootstrap.servers参数值的端口号均改为其中一个（比如9093），否则消费者无法消费消息。但是我试过修改这3个配置文件，竟然没用！！！难道是我猜错了？郁闷。。。
后来我把其中一个端口号改为9092就好了。
[root@hadoop ~]# cd /usr/local/kafka/config/ [root@hadoop config]# cp server.properties server0.properties [root@hadoop config]# cp server.properties server1.properties [root@hadoop config]# cp server.properties server2.properties [root@hadoop config]# vi server0.properties broker.id=0 listeners=PLAINTEXT://:9092 log.dirs=/tmp/kafka-logs0 [root@hadoop config]# vi server1.properties broker.id=1 listeners=PLAINTEXT://:9093 log.dirs=/tmp/kafka-logs1 [root@hadoop config]# vi server2.properties broker.id=2 listeners=PLAINTEXT://:9094 log.dirs=/tmp/kafka-logs2
2.启动服务器
启动zookeeper
[root@hadoop kafka]# zookeeper-server-start.sh config/zookeeper.properties
...
分别启动3个kafka服务器（可以在每条命令之后加 & 使命令在后台运行）
[root@hadoop kafka]# kafka-server-start.sh config/server0.properties &
[root@hadoop kafka]# kafka-server-start.sh config/server1.properties &
[root@hadoop kafka]# kafka-server-start.sh config/server2.properties &
查看
[root@hadoop config]# jps #显示启动了zookeeper和3个kafka服务器 12161 QuorumPeerMain 13154 Kafka 15940 Jps 15609 Kafka 12828 Kafka
3.创建主题（3个副本）
注意：副本数必须小于等于broker数
[root@hadoop kafka]# kafka-topics.sh --create --zookeeper localhost:2181 --replication-factor 3 --partitions 1 --topic test02 Created topic "test02".
查看主题内容
[root@hadoop kafka]# kafka-topics.sh --describe --zookeeper localhost:2181 --topic test02 Topic:test02 PartitionCount:1 ReplicationFactor:3 Configs: Topic: test02 Partition: 0 Leader: 2 Replicas: 2,1,0 Isr: 2,1,0
还可以查看上一个主题内容
[root@hadoop kafka]# kafka-topics.sh --describe --zookeeper localhost:2181 --topic test Topic:test PartitionCount:1 ReplicationFactor:1 Configs: Topic: test Partition: 0 Leader: 0 Replicas: 0 Isr: 0
4.发布新消息给主题
[root@hadoop kafka]# netstat -ano | grep 909 #查看3个kafka server的端口号（9092,9093,9094）是否存在 tcp6 0 0 :::9092 :::* LISTEN off (0.00/0/0) tcp6 0 0 :::9093 :::* LISTEN off (0.00/0/0) tcp6 0 0 :::9094 :::* LISTEN off (0.00/0/0) tcp6 0 0 192.168.42.133:49292 192.168.42.133:9093 ESTABLISHED keepalive (7058.77/0/0) tcp6 0 0 192.168.42.133:50262 192.168.42.133:9094 ESTABLISHED keepalive (7075.16/0/0) tcp6 0 0 192.168.42.133:9092 192.168.42.133:54622 ESTABLISHED keepalive (7058.77/0/0) tcp6 0 0 192.168.42.133:9093 192.168.42.133:49292 ESTABLISHED keepalive (7058.77/0/0) tcp6 0 0 192.168.42.133:9094 192.168.42.133:50262 ESTABLISHED keepalive (7075.16/0/0) tcp6 0 0 192.168.42.133:50268 192.168.42.133:9094 ESTABLISHED keepalive (7157.08/0/0) tcp6 0 0 192.168.42.133:54622 192.168.42.133:9092 ESTABLISHED keepalive (7058.77/0/0) tcp6 0 0 192.168.42.133:9094 192.168.42.133:50264 ESTABLISHED keepalive (7075.16/0/0) tcp6 0 0 192.168.42.133:9094 192.168.42.133:50268 ESTABLISHED keepalive (7157.08/0/0) tcp6 0 0 192.168.42.133:50264 192.168.42.133:9094 ESTABLISHED keepalive (7075.16/0/0) [root@hadoop ~]# kafka-console-producer.sh --broker-list localhost:9092,localhost:9093,localhost:9094 --topic test02 >hello kafka
5.消费消息
[root@hadoop kafka]# kafka-console-consumer.sh --bootstrap-server localhost:9092,localhost:9093,localhost:9094 --topic test02 --from-beginning hello kafka
注意：旧版本的kafka消费消息命令如下：
[root@hadoop kafka]# kafka-console-consumer.sh --zookeeper localhost:2181 --topic test02 --from-beginning zookeeper is not a recognized option #新版本kafka移除了这个参数
百度后发现，从 0.9.0.0版本后，kafka的consumer配置作了很大的改变，不仅增加了很多配置，而且把原来的zookeeper相关配置统统取消了。感兴趣的朋友可以查看官方文档中新旧consumer配置的对比。
6.容错测试
在容错性测试之前，可以先将生产者和消费者关掉。
a) 找到并杀死server2进程（broker 2是原来3个broker中的leader，现在我们杀死leader）
[root@hadoop kafka]# ps -ef|grep server2.properties #找到server2的进程号为13154 root 13154 12486 1 11:02 pts/4 00:00:33 ...信息太多，忽略.../server2.properties root 16228 14977 0 11:56 pts/6 00:00:00 grep --color=auto server2.properties
[root@hadoop kafka]# kill 13154 #杀死进程
[root@hadoop kafka]# ps -ef|grep server2.properties #再次查看，进程号没有了 root 16239 14977 0 11:57 pts/6 00:00:00 grep --color=auto server2.properties
[root@hadoop kafka]# jps #只剩下2个kafka了 16240 Jps 12161 QuorumPeerMain 15609 Kafka 12828 Kafka
b) 查看主题描述
#主题还在，只是leader由原来的broker 2 变为了broker 1 [root@hadoop kafka]# kafka-topics.sh --describe --zookeeper localhost:2181 --topic test02 Topic:test02 PartitionCount:1 ReplicationFactor:3 Configs: Topic: test02 Partition: 0 Leader: 1 Replicas: 2,1,0 Isr: 1,0
c) 启动消费者消费主题消息
[root@hadoop kafka]# kafka-console-consumer.sh --bootstrap-server localhost:9092,localhost:9093,localhost:9094 --topic test02 --from-beginning [2018-08-03 12:05:47,143] WARN [Consumer clientId=consumer-1, groupId=console-consumer-16224] Connection to node -3 could not be established. Broker may not be available. (org.apache.kafka.clients.NetworkClient) hello kafka
可以发现，消费者依然可以消费原来的消息。
当然，如果生产者重新发布新消息，消费者同样也是可以消费的。
三 搭建多节点多broker的kafka集群
假设现在我们有3台服务器，分别为s101,s102,s103。
首先在3台服务器上先建立zookeeper完全分布式集群，步骤参考我以前写的博客。
接着安装kafka。
1.先在s101服务器上安装kafka（步骤参考 一 搭建单节点单broker的kafka集群 第1步）
2.将 kafka + 环境变量 拷贝到s102,s103服务器上。
3.分别修改3个服务器下的server.properties文件
[root@hadoop config]# vi server.properties #s101修改配置如下 broker.id=101 #注意3个broker id不冲突即可 log.dirs=/usr/local/kafka/kafka-logs #自己设置一个目录，但注意不要放在/tmp/临时目录下 zookeeper.connect=s101:2181,s102:2181,s103:2181
[root@hadoop config]# vi server.properties #s102修改配置如下 broker.id=102 log.dirs=/usr/local/kafka/kafka-logs zookeeper.connect=s101:2181,s102:2181,s103:2181
[root@hadoop config]# vi server.properties #s103修改配置如下 broker.id=103 log.dirs=/usr/local/kafka/kafka-logs zookeeper.connect=s101:2181,s102:2181,s103:2181
4.创建log目录 mkdir /usr/local/kafka/kafka-logs
5.分别启动3个zookeeper以及3个kafka server即可。