1.启动kakfa自带的zk
参考：https://blog.csdn.net/justlpf/article/details/127261664?utm_medium=distribute.pc_relevant.none-task-blog-2defaultbaidujs_baidulandingword~default-0-127261664-blog-127495317.pc_relevant_3mothn_strategy_recovery&spm=1001.2101.3001.4242.1&utm_relevant_index=3
选择三台机器，该三台机器上已经搭建了ssh免密登录
vi /etc/hosts
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
10.1.218.22 node1
10.1.218.26 node2
10.1.218.24 node3
环境
1.三台服务器安装java环境，jdk1.8
2.Kafka安装包版本：kafka_2.12-3.3.1.tgz
3、假设3台服务器分别为：node1、node2、node3
将 kafka_2.12-3.0.0.tgz上传到node1上
/opt/server/kafka下，并解压
编写同步脚本
vi xsync
#!/bin/bash
#1. 判断参数个数
if [ $# -lt 1 ]
then
echo Not Enough Arguement!
exit;
fi
#2. 遍历集群所有机器
for host in node2 node3
do
echo ==================== $host ====================
#3. 遍历所有目录，挨个发送
for file in $@
do
#4. 判断文件是否存在
if [ -e $file ]
then
#5. 获取父目录
pdir=$(cd -P $(dirname $file); pwd)
#6. 获取当前文件的名称
fname=$(basename $file)
ssh $host "mkdir -p $pdir"
rsync -av $pdir/$fname $host:$pdir
else
echo $file does not exists!
fi
done
done
授予权限
chmod +777 xsync
该脚本用于向node2 和node3上同步文件
在node1、node2、node3上，都新建
cd /opt/server/kafka/kafka_2.12-3.0.0
mkdir zk_kfk_data
cd zk_kfk_data
vi myid
node1，node2，node3的myid中内容分别是 1 2 3
修改node1，node2，node3上的zk配置文件
vi zookeeper.properties
dataDir=/opt/server/kafka/kafka_2.12-3.0.0/zk_kfk_data
maxClientCnxns=0
tickTime=2000
initLimit=10
syncLimit=5
server.1=node1:2888:3888
server.2=node2:2888:3888
server.3=node3:2888:3888
三台机器上的zookeeper.properties文件配置相同，data.Dir 为zk的数据目录，server.1、server.2、server.3 为集群信息。
2888端口号是zookeeper服务之间通信的端口
3888端口是zookeeper与其他应用程序通信的端口。
tickTime：CS通信心跳数
Zookeeper 服务器之间或客户端与服务器之间维持心跳的时间间隔，也就是每个 tickTime 时间就会发送一个心跳。
tickTime以毫秒为单位。
tickTime：该参数用来定义心跳的间隔时间，zookeeper的客户端和服务端之间也有和web开发里类似的session的概念，而zookeeper里最小的session过期时间就是tickTime的两倍。
initLimit：LF初始通信时限
集群中的follower服务器(F)与leader服务器(L)之间 初始连接 时能容忍的最多心跳数（tickTime的数量）
syncLimit：LF同步通信时限
集群中的follower服务器(F)与leader服务器(L)之间 请求和应答 之间能容忍的最多心跳数（tickTime的数量）
在每台机器上启动zk
nohup bin/zookeeper-server-start.sh config/zookeeper.properties &>>zookeeper.log &
[root@node1 kafka_2.12-3.0.0]# netstat -ant |grep 2181
tcp6 0 0 :::2181 :::* LISTEN
说明启动成功了
2.启动kafka集群
在node1上
cd /opt/server/kafka/kafka_2.12-3.0.0
mkdir kafka-logs-1
修改server.properties
broker.id=0
advertised.host.name=node1
advertised.port=9092
log.dirs=/opt/server/kafka/kafka_2.12-3.0.0/kafka-logs-1
num.partitions=40
# The minimum age of a log file to be eligible for deletion due to age
log.retention.hours=24
zookeeper.connect=node1:2181,node2:2181,node3:2181/kafka
在node2和node3上同样的操作，node2的broker.id=1,node3的broker.id=2
启动命令
kafka_2.12-3.0.0]# nohup bin/kafka-server-start.sh config/server.properties &>>kafka.log &
[root@node1 kafka_2.12-3.0.0]# netstat -ant |grep 9092
tcp6 0 0 :::9092 :::* LISTEN
tcp6 0 0 10.1.218.22:51018 10.1.218.22:9092 ESTABLISHED
tcp6 0 0 10.1.218.22:9092 10.1.218.22:51018 ESTABLISHED
3.查看zookeeper的kafka节点信息
[root@node1 kafka_2.12-3.0.0]# bin/zookeeper-shell.sh node1:2181,node2:2181,node3:2181
Connecting to node1:2181,node2:2181,node3:2181
Welcome to ZooKeeper!
JLine support is disabled
WATCHER::
WatchedEvent state:SyncConnected type:None path:null
ls
ls [-s] [-w] [-R] path
ls /
[kafka, zookeeper]
ls /kafka
[admin, brokers, cluster, config, consumers, controller, controller_epoch, feature, isr_change_notification, latest_producer_id_block, log_dir_event_notification]
ls /brokers/ids
Node does not exist: /brokers/ids
ls /kafka/brokers/ids
[0, 1, 2]
ls /kafka/brokers/topics
[]
get /kafka/brokers/ids/0
{"listener_security_protocol_map":{"PLAINTEXT":"PLAINTEXT"},"endpoints":["PLAINTEXT://node1:9092"],"jmx_port":-1,"features":{},"host":"node1","timestamp":"1673862098693","port":9092,"version":5}
启停kafka集群脚本
#!/bin/bash
basePath=/opt/server/kafka/kafka_2.12-3.0.0
case $1 in
"start"){
for host in node1 node2 node3
do
echo "启动$host的kafka服务"
ssh $host "$basePath/bin/kafka-server-start.sh -daemon $basePath/config/server.properties"
done
};;
"stop"){
for host in node1 node2 node3
do
echo "停止$host的kafka服务"
ssh $host "$basePath/bin/kafka-server-stop.sh"
done
};;
esac
创建主题
[root@node1 kafka_2.12-3.0.0]# bin/kafka-topics.sh --bootstrap-server node1:9092,node2:9092,node3:9092 --topic first --create --partitions 1 --replication-factor 3
Created topic first.
[root@node1 kafka_2.12-3.0.0]# bin/kafka-topics.sh --bootstrap-server node1:9092,node2:9092,node3:9092 --topic first --describe
Topic: first TopicId: IcU6kbglSSWOkruuSmgryg PartitionCount: 1 ReplicationFactor: 3 Configs: segment.bytes=1073741824
Topic: first Partition: 0 Leader: 1 Replicas: 1,0,2 Isr: 1,0,2
[root@node1 kafka_2.12-3.0.0]#
增加分区
[root@node1 kafka_2.12-3.0.0]# bin/kafka-topics.sh --bootstrap-server node1:9092,node2:9092,node3:9092 --topic first --alter --partitions 3
[root@node1 kafka_2.12-3.0.0]# bin/kafka-topics.sh --bootstrap-server node1:9092,node2:9092,node3:9092 --topic first --describe
Topic: first TopicId: IcU6kbglSSWOkruuSmgryg PartitionCount: 3 ReplicationFactor: 3 Configs: segment.bytes=1073741824
Topic: first Partition: 0 Leader: 1 Replicas: 1,0,2 Isr: 1,0,2
Topic: first Partition: 1 Leader: 2 Replicas: 2,1,0 Isr: 2,1,0
Topic: first Partition: 2 Leader: 0 Replicas: 0,2,1 Isr: 0,2,1
Leader节点负责给定Partition的所有读写请求。
Replicas表示某个Partition在哪几个Broker上存在备份。不管这个几点是不是”Replicas“，甚至这个节点挂了，也会列出。
Isr是Replicas的一个子集，它只列出当前还存活着的，并且已同步备份了该Partition的节点。
模拟生产消费
node1上
[root@node1 kafka_2.12-3.0.0]# bin/kafka-console-producer.sh --topic quickstart-events --bootstrap-server node1:9092
abc
hello
在node2上
[root@node2 kafka_2.12-3.0.0]# bin/kafka-console-consumer.sh --topic quickstart-events --from-beginning --bootstrap-server node2:9092
abc
hello
模拟消费组消费主题，查看offset
启动生产者,test-topic创建了40个分区
kafka_2.12-3.0.0]# bin/kafka-console-producer.sh --topic test-topic --bootstrap-server node1:9092,node2:9092,node3:9092
启动消费者
bin/kafka-console-consumer.sh --topic test-topic --group my-group --bootstrap-server node1:9092,node2:9092,node3:9092
查看offset
GROUP TOPIC PARTITION CURRENT-OFFSET LOG-END-OFFSET LAG CONSUMER-ID HOST CLIENT-ID
my-group test-topic 38 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 15 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 8 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 17 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 31 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 22 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 25 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 4 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 5 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 18 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
......忽略，共计40个分区
当生产者发送消息后，消费者可以收到，且offset显示为
[root@node3 kafka_2.12-3.0.0]# bin/kafka-consumer-groups.sh --describe --bootstrap-server node1:9092,node2:9092,node3:9092 --group my-group
GROUP TOPIC PARTITION CURRENT-OFFSET LOG-END-OFFSET LAG CONSUMER-ID HOST CLIENT-ID
my-group test-topic 38 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 15 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 8 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 17 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 31 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 22 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 25 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 4 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 5 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 18 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 34 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 32 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 16 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 29 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 39 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 2 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 23 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 13 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 6 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 28 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 3 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 12 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 24 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 10 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 1 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 11 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 36 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 33 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 14 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 27 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 20 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 21 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 7 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 9 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 30 1 1 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 0 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 35 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 26 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 19 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
my-group test-topic 37 0 0 0 consumer-my-group-1-aacccf11-d714-47ed-852f-32123d88eebe /10.1.218.26 consumer-my-group-1
[root@node3 kafka_2.12-3.0.0]#
关闭消费者，生产者陆续发送6条消息
GROUP TOPIC PARTITION CURRENT-OFFSET LOG-END-OFFSET LAG CONSUMER-ID HOST CLIENT-ID
my-group test-topic 38 0 0 0 - - -
my-group test-topic 15 0 0 0 - - -
my-group test-topic 8 0 0 0 - - -
my-group test-topic 17 0 1 1 - - -
my-group test-topic 31 0 0 0 - - -
my-group test-topic 22 0 0 0 - - -
my-group test-topic 25 0 0 0 - - -
my-group test-topic 4 0 2 2 - - -
my-group test-topic 5 0 0 0 - - -
my-group test-topic 18 0 0 0 - - -
my-group test-topic 34 0 0 0 - - -
my-group test-topic 32 0 0 0 - - -
my-group test-topic 16 0 0 0 - - -
my-group test-topic 29 0 0 0 - - -
my-group test-topic 39 0 1 1 - - -
my-group test-topic 2 0 0 0 - - -
my-group test-topic 23 0 0 0 - - -
my-group test-topic 13 0 0 0 - - -
my-group test-topic 6 0 0 0 - - -
my-group test-topic 28 0 0 0 - - -
my-group test-topic 3 0 0 0 - - -
my-group test-topic 12 0 0 0 - - -
my-group test-topic 24 0 0 0 - - -
my-group test-topic 10 0 0 0 - - -
my-group test-topic 1 0 0 0 - - -
my-group test-topic 11 0 0 0 - - -
my-group test-topic 36 0 0 0 - - -
my-group test-topic 33 0 0 0 - - -
my-group test-topic 14 0 0 0 - - -
my-group test-topic 27 0 0 0 - - -
my-group test-topic 20 0 0 0 - - -
my-group test-topic 21 0 1 1 - - -
my-group test-topic 7 0 0 0 - - -
my-group test-topic 9 0 1 1 - - -
my-group test-topic 30 1 1 0 - - -
my-group test-topic 0 0 0 0 - - -
my-group test-topic 35 0 0 0 - - -
my-group test-topic 26 0 0 0 - - -
my-group test-topic 19 0 0 0 - - -
my-group test-topic 37 0 0 0 - - -
lag 延后共计6个
offset是相对于consumer来说的，offset用来记录某个分区，某个topic的消费情况，每次提交offset，都是消费者向kafka提交一次消费进度
提交的记录会被保存到服务端的_consumer_offsets中
提交又分为同步提交和异步提交，首先来说同步提交，这种方式会阻塞消费端的消费性能，所以一般采用的都是异步提交，而异步提交的一个问题就是可能会提交失败，从而导致消费消息重复；
虽然在生产消息环节，我们可以通过幂等性、消息事务、ack机制等方式来保证消息的可靠性和幂等性。但是在消费环境，同样由于offset提交问题，会导致消息无法幂等消费。
所以消费环节要做到业务去重，或者事务消费
577
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
