zookeeper集群搭建-优快云博客

本文详细介绍Zookeeper 3.4.11集群的搭建步骤，包括环境配置、节点设置及启动过程。并演示如何通过命令行和Java进行客户端操作与连接测试。

Zookeeper版本：zookeeper-3.4.11

JDK版本：1.8.0_161

服务器：192.168.1.150

服务器节点	zoo.conf配置
192.168.1.150:2811	/usr/local/zookeeper-3.4.11/conf/zoo_1.cfg配置 dataDir=/usr/local/zookeeper-3.4.11/data_1 dataLogDir=/usr/local/zookeeper-3.4.11/logs_1 clientPort=2181 server.1=192.168.1.150:2287:3387 server.2=192.168.1.150:2288:3388 server.2=192.168.1.150:2289:3389 命令：echo "1">/usr/local/zookeeper-3.4.11/data_1/myid 启动：/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_1.cfg
192.168.1.150:2812	/usr/local/zookeeper-3.4.11/conf/zoo_2.cfg配置 dataDir=/usr/local/zookeeper-3.4.11/data_2 dataLogDir=/usr/local/zookeeper-3.4.11/logs_2 clientPort=2182 server.1=192.168.1.150:2287:3387 server.2=192.168.1.150:2288:3388 server.2=192.168.1.150:2289:3389 命令：echo "2">/usr/local/zookeeper-3.4.11/data_2/myid 启动：/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_2.cfg
192.168.1.150:2813	/usr/local/zookeeper-3.4.11/conf/zoo_3.cfg配置 dataDir=/usr/local/zookeeper-3.4.11/data_3 dataLogDir=/usr/local/zookeeper-3.4.11/logs_3 clientPort=2183 server.1=192.168.1.150:2287:3387 server.2=192.168.1.150:2288:3388 server.2=192.168.1.150:2289:3389 命令：echo "3">/usr/local/zookeeper-3.4.11/data_3/myid 启动：/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_3.cfg
	注意：线上部署的时候使用内网地址，同时防火墙开放2181/2182/2183端口，安全组的入站规则也要开放该三个端口

1、解压文件

解压到/usr/local/目录下

命令：

tar -zxvf zookeeper-3.4.11.tar.gz -C /usr/local/

2、配置环境变量

vim /etc/profile

#zookeeper path
export ZOOKEEPER_HOME=/usr/local/zookeeper-3.4.11
PATH=$PATH:$JAVA_HOME/bin:$JRE_HOME/bin:$MYSQL_HOME/bin:ZOOKEEPER_HOME/bin

保存退出后执行

source /etc/profile

使配置的环境变量生效

3、创建zoo.cfg配置文件

（1）创建节点1配置文件：

cp /usr/local/zookeeper-3.4.11/conf/zoo_sample.cfg /usr/local/zookeeper-3.4.11/conf/zoo_1.cfg

vim zoo_1.cfg

修改配置文件内容：

# the directory where the snapshot is stored.
dataDir=/usr/local/zookeeper-3.4.11/data_1

# the port at which the clients will connect
clientPort=2181

# the log file path
dataLogDir=/usr/local/zookeeper-3.4.11/logs_1      

# the server node
server.1=127.0.0.1:2287:3387
server.2=127.0.0.1:2288:3388
server.3=127.0.0.1:2289:3389

（2）创建节点2配置文件：

cp /usr/local/zookeeper-3.4.11/conf/zoo_1.cfg /usr/local/zookeeper-3.4.11/conf/zoo_2.cfg

vim zoo_2.cfg

修改配置文件内容：

# the directory where the snapshot is stored.
dataDir=/usr/local/zookeeper-3.4.11/data_2

# the port at which the clients will connect
clientPort=2182

# the log file path
dataLogDir=/usr/local/zookeeper-3.4.11/logs_2      

# the server node
server.1=127.0.0.1:2287:3387
server.2=127.0.0.1:2288:3388
server.3=127.0.0.1:2289:3389

（3）创建节点3配置文件：

cp /usr/local/zookeeper-3.4.11/conf/zoo_1.cfg /usr/local/zookeeper-3.4.11/conf/zoo_3.cfg

vim zoo_3.cfg

修改配置文件内容：

# the directory where the snapshot is stored.
dataDir=/usr/local/zookeeper-3.4.11/data_3

# the port at which the clients will connect
clientPort=2183

# the log file path
dataLogDir=/usr/local/zookeeper-3.4.11/logs_3   

# the server node
server.1=127.0.0.1:2287:3387
server.2=127.0.0.1:2288:3388
server.3=127.0.0.1:2289:3389

4、创建存放zookeeper数据和日志文件的目录

mkdir -p /usr/local/zookeeper-3.4.11/data_1
mkdir -p /usr/local/zookeeper-3.4.11/data_2
mkdir -p /usr/local/zookeeper-3.4.11/data_3
mkdir -p /usr/local/zookeeper-3.4.11/logs_1
mkdir -p /usr/local/zookeeper-3.4.11/logs_2
mkdir -p /usr/local/zookeeper-3.4.11/logs_3

#设置权限（root用户可以省略）

sudo chmod -R 777 /usr/local/zookeeper-3.4.11/data_{1/2/3}
sudo chmod -R 777 /usr/local/zookeeper-3.4.11/logs_{1/2/3}

5、分别在data_{1/2/3}文件夹下创建myid文件，并写入1/2/3（数字对应server.1的数字）

echo "1" > /usr/local/zookeeper-3.4.11/data_1/myid
echo "2" > /usr/local/zookeeper-3.4.11/data_2/myid
echo "3" > /usr/local/zookeeper-3.4.11/data_3/myid

6、启动

/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_1.cfg
/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_2.cfg
/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_3.cfg

启动结果验证：

命令：

jps

查看状态：

/usr/local/zookeeper-3.4.11/bin/zkServer.sh status /usr/local/zookeeper-3.4.11/conf/zoo_1.cfg

启动服务时，可以创建shell脚本启动：

vim start-zk-cluster.sh

输入内容：

/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_1.cfg
/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_2.cfg
/usr/local/zookeeper-3.4.11/bin/zkServer.sh start /usr/local/zookeeper-3.4.11/conf/zoo_3.cfg

创建停止服务：

vim stop-zk-cluster.sh

输入内容：

/usr/local/zookeeper-3.4.11/bin/zkServer.sh stop /usr/local/zookeeper-3.4.11/conf/zoo_1.cfg
/usr/local/zookeeper-3.4.11/bin/zkServer.sh stop /usr/local/zookeeper-3.4.11/conf/zoo_2.cfg
/usr/local/zookeeper-3.4.11/bin/zkServer.sh stop /usr/local/zookeeper-3.4.11/conf/zoo_3.cfg

授权：

chmod -R 777 start-zk-cluster.sh
chmod -R 777 stop-zk-cluster.sh

启动时执行这两个文件即可

7、连接客户端操作测试

/usr/local/zookeeper-3.4.11/bin/zkCli.sh -server 192.168.1.150:2181

ls / :列出根节点下所有的子节点信息

stat path :获取指定节点的状态信息

czxid 创建该节点的事物ID

ctime 创建该节点的时间

mZxid 更新该节点的事物ID

mtime 更新该节点的时间

pZxid 操作当前节点的子节点列表的事物ID(这种操作包含增加子节点，删除子节点)

cversion 当前节点的子节点版本号

dataVersion 当前节点的数据版本号

aclVersion 当前节点的acl权限版本号

ephemeralowner 当前节点的如果是临时节点，该属性是临时节点的事物ID

dataLength 当前节点的d的数据长度

numchildren 当前节点的子节点个数

get path 获取当前节点的数据内容

ls2 path :是ls 和 stat两个命令的结合

create [-s] [-e] path data acl

-s 表示是顺序节点

-e 标识是临时节点（注：临时节点在客户端结束与服务器的会话后，自动消失）

path 节点路径

data 节点数据

acl 节点权限

set path data [version] :修改当前节点的数据内容如果指定版本，需要和当前节点的数据版本一致

delete path [version] 删除指定路径的节点如果有子节点要先删除子节点

命令：

delete /node_1/node1_1

rmr path 删除当前路径节点及其所有子节点

8、Java连接测试

public class ZookeeperClusterTest {
    public static void main(String[] args) throws Exception {
        String zkAddress = "192.168.1.150:2181,192.168.1.150:2182,192.168.1.150:2183";
        
        CuratorFramework client = null;
        try{
            RetryPolicy retryPolicy = new ExponentialBackoffRetry(1000,5);
            client = CuratorFrameworkFactory.builder().
                    connectString(zkAddress).
                    retryPolicy(retryPolicy).sessionTimeoutMs(20000).
                    connectionTimeoutMs(6000).
                    build();

            client.start();
            List<String> list = client.getChildren().forPath("/");

            System.out.println("all node>>>" + Arrays.asList(list));
            Stat stat = client.checkExists().forPath("/node_1");
            System.out.println("stat>>" +stat);
            if(stat == null){
                client.create().forPath("/node_1","node001".getBytes());
                System.out.println("node_1 data>>>" + client.getData().forPath("/node_1"));
            }else {
                System.out.println("node_1 data>>>" + new String(client.getData().forPath("/node_1"),"UTF-8"));
            }
        }finally {
            CloseableUtils.closeQuietly(client);//一定得关闭，不然一直占用
        }
    }
}