zookeeper02

1 zookeeper的事件监听机制

1.1 watcher概念

• zookeeper提供了数据的发布/订阅功能，多个订阅者可以同时监听某一个特定主题对象，当该主题对象的自身状态发生改变的时候，例如：节点内容改变、节点下的子节点列表发生改变等等，会实时、主动通知所有订阅者。
• zookeeper采用了watcher机制实现数据的发布/订阅功能。该机制在被订阅对象发生改变的时候会异步通知客户端，因此客户端不必在watcher注册后轮询阻塞，从而减轻了客户端的压力。
• watcher机制实际上和观察者模式类似，也可以看作是一种观察者模式在分布式场景下的实现方式。

1.2 watcher架构

• watcher实现由三部分组成：
  • 1️⃣zookeeper客户端。
  • 2️⃣zookeeper服务端。
  • 3️⃣客户端的ZKWatchManger对象。
• 客户单首先将watcher注册到服务器端，同时将watcher对象保存到客户端的watch管理器中。当zookeeper服务器端监听的数据状态发生改变的时候，服务器端会主动通知客户端，接着客户端的watch管理器会触发相关的watcher来回调相应的处理逻辑，从而完成整体的数据发布/订阅流程。

1.3 watcher特性

特性	说明
一次性	watcher是一次性的，一旦被触发就会移除，再次使用时需要重新注册
客户端顺序回调	watcher回调是顺序串行化执行的，只有回调后客户端才能看到最新的数据状态。一个watcher回调逻辑不应该太多，以免影响到别的watcher的执行。
轻量级	WatchEvent是最小的通信单元，结构上只包含通知状态、事件类型和节点路径，并不会告诉 数据节点变化前后的具体内容。
时效性	watcher只有在当前session彻底失效时才会生效，如果在session有效期内快速重连成功，则watcher依然存在，依然可以接受到通知。

1.4 watcher接口设计

• Watcher是一个接口，任何实现了Watcher接口的类就是一个新的Watcher。Watcher内部包含了两个枚举：KeeperState和EventType。

1.4.1 Watcher通知状态（KeeperState）

• KeeperState是客户端和服务端连接状态发生变化时对应的通知类型。路径为：org.apache.zookeeper.Watcher.Event.KeeperState，是一个枚举类，其枚举属性如下：

枚举属性	说明
SyncConnected	客户端和服务器端正常连接时
Disconnected	客户端和服务器端断开连接时
Expired	会话session失效时
AuthFailed	身份认证失败时

1.4.2 Watcher事件类型（EventType）

• EventType是数据节点（Znode）发生变化时对应的通知类型。EventType变化时KeeperState永远处于SyncConnected通知状态下；当KeeperState发生变化时，EventType永远为None。其路径为org.apache.zookeeper.Watcher.Event.EventType，是一个枚举类，枚举类属性如下：

枚举属性	说明
None	无
NodeCreated	Watcher监听的数据节点被创建时
NodeDeleted	Watcher监听的数据节点被删除时
NodeDataChanged	Watcher监听的数据节点内容发生变更时（无论内容数据是否变化）
NodeChildrenChanged	Watcher监听的数据节点的子节点列表发生变更时

注意：客户端接收到的相关事件通知中只包含状态、类型等信息，不包含节点变化前后的具体内容，变化前的数据需要业务自身存储，变化后的数据需要调用get等方法重新获取。

1.5 捕获相应的事件

• 在zookeeper中采用zk.getChildren(path,watch)、zk.exists(path,watch)、zk.getData(path,watcher,stat)这样的方式为某个znode注册监听。
• 下面以node-x节点为例，说明调用的注册方式和可监听事件间的关系。

注册方式	Created	ChildrenChanged	DataChanged	Deleted
zk.exists(node-x,watch)	可以监控		可以监控	可以监控
zk.getData(node-x,watcher,stat)			可以监控	可以监控
zk.getChildren(path,watch)		可以监控		可以监控

1.6 注册watcher的方法

1.6.1 客户端和服务器端的连接状态

事件类型为：None。

KeeperState：通知状态。

• SyncConnected：客户端和服务器端正常连接时。
• Disconnected：客户端和服务器端断开连接时。
• Expired： 会话session失效时。
• AuthFailed：身份认证失败时。

• 示例：

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @Test
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                //事件类型
                if (event.getType().equals(Event.EventType.None)) {
                    if (event.getState() == Event.KeeperState.SyncConnected) {
                        System.out.println("连接创建成功");
                        countDownLatch.countDown();
                    }else if(event.getState() == Event.KeeperState.Disconnected){
                        System.out.println("断开连接");
                    }else if(event.getState() == Event.KeeperState.Expired){
                        System.out.println("会话超时");
                        //一般而言，zookeeper会在我们设置的会话超时时间内重新连接服务器端了，一般超过了我们自己设置的会话超时时间，那么zookeeper就不会再重新连接服务器端了。
                        //通常而言，我们一般在会话超时的时候，让其重新连接zookeeper服务器
                        //代码略
                    }else if(event.getState() == Event.KeeperState.AuthFailed){
                        System.out.println("身份认证失败");
                    }
                }
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);

        Thread.sleep(50000);

        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());

        zooKeeper.close();

    }
}

1.6.2 检查节点是否存在

//watch为true时，使用连接对象的监视器 NodeCreated、NodeDeleted、NodeDataChanged 
public Stat exists(String path, boolean watch){}

//自定义监视器  NodeCreated、NodeDeleted、NodeDataChanged  
public Stat exists(final String path, Watcher watcher){}

• 示例：使用连接对象的监视器

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests2 {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @BeforeEach
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("连接对象的参数");
                //事件类型
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    countDownLatch.countDown();
                }
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);
        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }


    @Test
    public void test() throws KeeperException, InterruptedException {
        //节点的路径
        String path = "/watcher";
        //使用连接对象的watcher
        boolean watch = true;
        zooKeeper.exists(path, watch);
        Thread.sleep(50000);
        System.out.println("结束");
    }
}

• 示例：自定义监视器

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests2 {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @BeforeEach
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("连接对象的参数");
                //事件类型
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    countDownLatch.countDown();
                }
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);
        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }


    @Test
    public void test() throws KeeperException, InterruptedException {
        //节点的路径
        String path = "/watcher";
        //自定义watcher对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("自定义watcher对象");
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);

            }
        };
        zooKeeper.exists(path, watcher);
        Thread.sleep(50000);
        System.out.println("结束");
    }
}

• 示例：修改watcher机制的一次性，实现watcher机制的一直监听

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests2 {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @BeforeEach
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("连接对象的参数");
                //事件类型
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    countDownLatch.countDown();
                }
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);
        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }


    @Test
    public void test() throws KeeperException, InterruptedException {
        //节点的路径
        String path = "/watcher";
        //自定义watcher对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("自定义watcher对象");
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
                try {
                    zooKeeper.exists(path,this);
                } catch (KeeperException e) {
                    e.printStackTrace();
                } catch (InterruptedException e) {
                    e.printStackTrace();
                }

            }
        };
        zooKeeper.exists(path, watcher);
        Thread.sleep(50000);
        System.out.println("结束");
    }
}

• 示例：注册多个监听器对象

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests2 {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @BeforeEach
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("连接对象的参数");
                //事件类型
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    countDownLatch.countDown();
                }
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);
        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }


    /**
     * 注册多个监听器对象
     *
     * @throws KeeperException
     * @throws InterruptedException
     */
    @Test
    public void test() throws KeeperException, InterruptedException {
        //节点的路径
        String path = "/watcher";
        //自定义watcher对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("自定义watcher对象");
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
                try {
                    zooKeeper.exists(path, this);
                } catch (KeeperException e) {
                    e.printStackTrace();
                } catch (InterruptedException e) {
                    e.printStackTrace();
                }

            }
        };
        //注册第一个监听器对象
        zooKeeper.exists(path, watcher);
        //注册第二个监听器对象
        zooKeeper.exists(path, watcher);
        Thread.sleep(50000);
        System.out.println("结束");
    }
}

1.6.3 查看节点

//使用连接对象的监视器  NodeDeleted NodeDataChanged
public byte[] getData(String path, boolean watch, Stat stat)
            throws KeeperException, InterruptedException {}

//自定义监视器
public byte[] getData(final String path, Watcher watcher, Stat stat){}

• 示例：使用连接对象的监视器

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests3 {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @BeforeEach
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("连接对象的参数");
                //事件类型
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    countDownLatch.countDown();
                }
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);
        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }


    @Test
    public void test() throws KeeperException, InterruptedException {
        //节点的路径
        String path = "/watcher";
        //使用连接对象中的watcher
        boolean watch = true;
        byte[] data = zooKeeper.getData(path, watch, null);
        System.out.println("data = " + new String(data));
        Thread.sleep(50000);
        System.out.println("结束");

    }
}

• 示例：自定义监视器

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests3 {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @BeforeEach
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("连接对象的参数");
                //事件类型
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    countDownLatch.countDown();
                }
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);
        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }


    @Test
    public void test() throws KeeperException, InterruptedException {
        //节点的路径
        String path = "/watcher";
        //自定义监视器
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper.getData(path, watcher, null);
        Thread.sleep(50000);
        System.out.println("结束");

    }
}

1.6.4 查看子节点

//使用连接对象的监视器 NodeDeleted NodeChildrenChanged
public List<String> getChildren(String path, boolean watch)
    throws KeeperException, InterruptedException {}

public List<String> getChildren(final String path, Watcher watcher)
    throws KeeperException, InterruptedException{}

• 示例：使用连接对象的监视器

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.List;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests4 {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @BeforeEach
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("连接对象的参数");
                //事件类型
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    countDownLatch.countDown();
                }
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);
        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }


    @Test
    public void test() throws KeeperException, InterruptedException {
        String path = "/hadoop";
        boolean watch = true;
        List<String> children = zooKeeper.getChildren(path, watch);
        System.out.println("children = " + children);
        Thread.sleep(50000);
        System.out.println("结束");
    }
}

• 示例：自定义监视器

package com.sunxiaping.zookeeper02;

import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import org.junit.jupiter.api.AfterEach;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;

import java.io.IOException;
import java.util.List;
import java.util.concurrent.CountDownLatch;

public class ZookeeperApplicationTests4 {

    ZooKeeper zooKeeper;

    @AfterEach
    public void after() throws InterruptedException {
        if (zooKeeper != null) {
            zooKeeper.close();
        }
    }


    /**
     * 连接zookeeper
     *
     * @throws IOException
     * @throws InterruptedException
     */
    @BeforeEach
    public void testZookeeperConnection() throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口
        String connectString = "192.168.179.100:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                System.out.println("连接对象的参数");
                //事件类型
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    countDownLatch.countDown();
                }
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);
        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }


    @Test
    public void test() throws KeeperException, InterruptedException {
        String path = "/hadoop";
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                String path = event.getPath();
                System.out.println("path = " + path);
                Event.EventType type = event.getType();
                System.out.println("type = " + type);
            }
        };
        List<String> children = zooKeeper.getChildren(path, watcher);
        System.out.println("children = " + children);
        Thread.sleep(50000);
        System.out.println("结束");
    }
}

1.7 配置中心案例

• 工作中有这样的一个场景：数据库用户名和密码等信息放在一个配置文件中，应用读取配置文件，配置文件信息放入缓存。

• 如果数据库的用户名和密码等改变的时候，还需要重新加载缓存，比较麻烦，通过zookeeper可以轻松完成，当数据发生变化的时候自动完成缓存同步。

• 设计思路：

  • 1️⃣连接zookeeper服务器。
  • 2️⃣读取zookeeper中的配置信息，注册watcher监视器，存入本地变量。
  • 3️⃣当zookeeper中的配置信息发生变化的时候，通过watcher的回调方法捕获数据变化事件。
  • 4️⃣重新获取配置信息。

• 示例：

• 1️⃣在zookeeper中创建数据库所需要的配置信息：

create /config "config"
create /config/url "192.168.179.100:3306"
create /config/username "root"
create /config/password "123456"

• 2️⃣使用watcher机制实现当zookeeper服务器中的配置信息发生改变时，客户端重新获取配置信息：

package com.sunxiaping.zookeeper03;

import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;

import java.util.concurrent.CountDownLatch;

/**
 * 配置中心
 */
public class MyConfigCenter implements Watcher {
    //zookeeper服务器的ip和端口
    private String connectString = "192.168.179.100:2181";
    //计数器对象
    private CountDownLatch countDownLatch = new CountDownLatch(1);
    //连接对象
    private ZooKeeper zooKeeper;
    //会话超时时间
    private int timeout = 60000;

    //用于本地化存储配置信息
    private String url;
    private String username;
    private String password;

    //构造方法
    public MyConfigCenter() {
        initValue();
    }


    @Override
    public void process(WatchedEvent event) {
        try {
            //捕获事件
            if (event.getType() == Event.EventType.None) {
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    System.out.println("连接成功");
                    countDownLatch.countDown();
                } else if (event.getState() == Event.KeeperState.Disconnected) {
                    System.out.println("连接断开");
                } else if (event.getState() == Event.KeeperState.Expired) {
                    System.out.println("会话超时");
                    zooKeeper = new ZooKeeper(connectString, timeout, this);
                } else if (event.getState() == Event.KeeperState.AuthFailed) {
                    System.out.println("验证失败");
                }
            } else if (event.getType() == Event.EventType.NodeDataChanged) { //当节点的数据发生变化的时候
                initValue();
            }
        } catch (Exception e) {
            e.printStackTrace();
        }

    }

    //连接zookeeper服务器，读取配置信息
    public void initValue() {
        try {
            //创建连接对象
            zooKeeper = new ZooKeeper(connectString, timeout, this);
            //阻塞线程，等待连接的创建成功
            countDownLatch.await();
            //读取配置信息
            this.url = new String(zooKeeper.getData("/config/url", true, null));
            this.username = new String(zooKeeper.getData("/config/username", true, null));
            this.password = new String(zooKeeper.getData("/config/password", true, null));
        } catch (Exception e) {
            e.printStackTrace();
        }
    }


    public String getUrl() {
        return url;
    }

    public void setUrl(String url) {
        this.url = url;
    }

    public String getUsername() {
        return username;
    }

    public void setUsername(String username) {
        this.username = username;
    }

    public String getPassword() {
        return password;
    }

    public void setPassword(String password) {
        this.password = password;
    }

    //模拟客户端
    public static void main(String[] args) {
        try {
            MyConfigCenter configCenter = new MyConfigCenter();

            //每5秒去服务器查询一次
            for (int i = 0; i < 30; i++) {
                Thread.sleep(5000);
                System.out.println("url:" + configCenter.getUrl());
                System.out.println("username:" + configCenter.getUsername());
                System.out.println("password:" + configCenter.getPassword());
                System.out.println("+++++++++++++++++++++++++++++++++++++++");
            }

        } catch (Exception e) {
            e.printStackTrace();
        }

    }
}

1.8 生成分布式唯一ID

• 在过去的单库单表型系统中，通常可以使用数据库字段自带的auto_increment属性来自动为每条记录生成一个唯一的ID。但是当分库分表后，就无法再依靠数据库的auto_increment属性来唯一标识一条记录了。此时我们就可以用zookeeper在分布式环境下生成全局唯一ID。

• 设计思路：

  • 1️⃣连接zookeeper服务器。
  • 2️⃣指定路径生成临时有序节点。
  • 3️⃣取序列号，即为分布式环境下的唯一ID。

• 示例：

package com.sunxiaping.zookeeper03;

import org.apache.zookeeper.*;

import java.util.concurrent.CountDownLatch;

/**
 * 全局唯一ID
 */
public class GloballyUniqueID implements Watcher {

    //zookeeper服务器的ip和端口
    private String connectString = "192.168.179.100:2181";
    //计数器对象
    private CountDownLatch countDownLatch = new CountDownLatch(1);
    //连接对象
    private ZooKeeper zooKeeper;
    //会话超时时间
    private int timeout = 60000;
    //用户生成序号的节点
    String defaultPath = "/uniqueId";

    @Override
    public void process(WatchedEvent event) {
        try {
            //捕获事件
            if (event.getType() == Event.EventType.None) {
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    System.out.println("连接成功");
                    countDownLatch.countDown();
                } else if (event.getState() == Event.KeeperState.Disconnected) {
                    System.out.println("连接断开");
                } else if (event.getState() == Event.KeeperState.Expired) {
                    System.out.println("会话超时");

                    zooKeeper = new ZooKeeper(connectString, timeout, this);

                } else if (event.getState() == Event.KeeperState.AuthFailed) {
                    System.out.println("验证失败");
                }
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }

    public GloballyUniqueID() {
        try {
            zooKeeper = new ZooKeeper(connectString, timeout, this);
            countDownLatch.await();
        } catch (Exception e) {
            e.printStackTrace();
        }
    }


    /**
     * 生成唯一ID
     *
     * @return
     */
    public String generateUniqueId() {

        String path = "";

        try {
            //创建临时有序节点
            path = zooKeeper.create(defaultPath, new byte[0], ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL_SEQUENTIAL);

            return path.substring(defaultPath.length() + 1);
        } catch (Exception e) {
            e.printStackTrace();
        }

        return null;
    }


    public static void main(String[] args) {
        GloballyUniqueID globallyUniqueID = new GloballyUniqueID();
        for (int i = 0; i < 100; i++) {
            String uniqueId = globallyUniqueID.generateUniqueId();
            System.out.println("uniqueId = " + uniqueId);
        }
    }

}

1.9 分布式锁

• 分布式锁有多种实现方式，比如通过数据库、redis都可以实现。作为分布式协同工具Zookeeper，当然也有着标准的实现方式。下面介绍在zookeeper中如下实现排他锁。

• 设计思路：

  • 1️⃣每个客户端往/Locks下创建临时有序节点/Locks/Lock_，创建成功后/Locks下面会有每个客户端对应的节点，如/Locks/Lock_000000001。
  • 2️⃣客户端获取/Locks下的子节点，并进行排序，判断排在最前面的是否为自己。如果自己的锁节点在第一位，则说明取锁成功。
  • 3️⃣如果自己的锁节点不是第一位，则监听自己前一位的锁节点。例如，自己的锁节点/Locks/Lock_000000002，那么则监听/Locks/Lock_000000001。
  • 4️⃣当前一位锁节点（/Locks/Lock_000000001）对应的客户端执行完成，释放了锁，将会触发监听客户端（/Locks/Lock_000000002）的逻辑。
  • 5️⃣监听客户端重新执行第2️⃣步的逻辑，判断自己是否获取了锁。

• 示例：

package com.sunxiaping.zookeeper03;

import org.apache.zookeeper.*;
import org.apache.zookeeper.data.Stat;

import java.util.Collections;
import java.util.List;
import java.util.concurrent.CountDownLatch;

public class LockDemo {
    //zookeeper服务器的ip和端口
    private String connectString = "192.168.179.100:2181";
    //计数器对象
    private CountDownLatch countDownLatch = new CountDownLatch(1);
    //连接对象
    private ZooKeeper zooKeeper;
    //会话超时时间
    private int timeout = 60000;

    private static final String LOCK_ROOT_PATH = "/Locks";
    private static final String LOCK_NODE_PATH = "/Lock_";

    private String lockPath;

    public LockDemo() {
        try {
            zooKeeper = new ZooKeeper(connectString, timeout, new Watcher() {
                @Override
                public void process(WatchedEvent event) {
                    try {
                        //捕获事件
                        if (event.getType() == Event.EventType.None) {
                            if (event.getState() == Event.KeeperState.SyncConnected) {
                                System.out.println("连接成功");
                                countDownLatch.countDown();
                            } else if (event.getState() == Event.KeeperState.Disconnected) {
                                System.out.println("连接断开");
                            } else if (event.getState() == Event.KeeperState.Expired) {
                                System.out.println("会话超时");

                            } else if (event.getState() == Event.KeeperState.AuthFailed) {
                                System.out.println("验证失败");
                            }
                        }
                    } catch (Exception e) {
                        e.printStackTrace();
                    }
                }
            });
            countDownLatch.await();
        } catch (Exception e) {
            e.printStackTrace();
        }
    }

    //获取锁
    public void acquireLock() throws KeeperException, InterruptedException {
        //创建锁
        createLock();
        //尝试获取锁
        attemptLock();
    }


    //创建锁
    public void createLock() throws KeeperException, InterruptedException {
        //判断/Locks是否存在
        Stat stat = zooKeeper.exists(LOCK_ROOT_PATH, false);
        if (stat == null) {
            zooKeeper.create(LOCK_ROOT_PATH, new byte[0], ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT);
        }
        //创建临时有序节点
        lockPath = zooKeeper.create(LOCK_ROOT_PATH + LOCK_NODE_PATH, new byte[0], ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL_SEQUENTIAL);
        System.out.println("lockPath = " + lockPath);
    }

    Watcher watcher = new Watcher() {
        @Override
        public void process(WatchedEvent event) {
            //监视上一个节点是否被删除
            if (event.getType() == Event.EventType.NodeDeleted) {
                synchronized (this) {
                    notifyAll();
                }
            }
        }
    };

    //尝试获取锁
    public void attemptLock() throws KeeperException, InterruptedException {
        //获取/Locks下的子节点  /Lock_xxxx
        List<String> lockList = zooKeeper.getChildren(LOCK_ROOT_PATH, false);
        //对子节点进行排序
        Collections.sort(lockList);
        //    /Locks/Lock_xxx
        int index = lockList.indexOf(lockPath.substring(LOCK_ROOT_PATH.length() + 1));
        if (0 == index) { //说明该临时有序节点在锁列表的第一位
            System.out.println("获取锁成功");
            return;
        } else {
            //获取上一个节点的路径
            String previousPath = lockList.get(index - 1);
            Stat stat = zooKeeper.exists(LOCK_ROOT_PATH + "/" + previousPath, watcher);
            if (null == stat) {
                attemptLock();
            } else {
                synchronized (watcher) {
                    watcher.wait();
                }
                attemptLock();
            }
        }
    }

    //释放锁
    public void releaseLock() throws KeeperException, InterruptedException {
        //删除临时有序节点
        zooKeeper.delete(lockPath, -1);
        if (zooKeeper != null) {
            zooKeeper.close();
        }
        System.out.println("锁已经释放");
    }
}

package com.sunxiaping.zookeeper03;

import org.apache.zookeeper.KeeperException;

public class SellTicketDemo {
    private void sell() {
        System.out.println("售票开始");
        int sleepMillis = 5000;
        try {
            Thread.sleep(sleepMillis);
        } catch (Exception e) {
            e.printStackTrace();
        }
        System.out.println("售票结束");
    }

    public void sellTicketWithLock() throws KeeperException, InterruptedException {
        LockDemo lockDemo = new LockDemo();
        //获取锁
        lockDemo.acquireLock();
        sell();
        //释放锁
        lockDemo.releaseLock();
    }

    public static void main(String[] args) throws KeeperException, InterruptedException {
        SellTicketDemo ticketDemo = new SellTicketDemo();
        for (int i = 0; i < 10; i++) {
            ticketDemo.sellTicketWithLock();
        }
    }
}

2 zookeeper集群搭建

2.1 准备工作

• 准备三台虚拟机（都装有CentOS7系统），IP分别为192.168.179.101、192.168.179.102和192.168.179.103。
• 分别在这三台虚拟机装单独的JDK（jdk-8u131-linux-x64.tar.gz）和Zookeeper（zookeeper-3.4.10.tar.gz）。

2.2 zookeeper集群的搭建

• 1️⃣关闭三台虚拟机的防火墙，防止不能通信：

# 192.168.107.101
systemctl stop firewalld
systemctl disable firewalld

# 192.168.107.102
systemctl stop firewalld
systemctl disable firewalld

# 192.168.107.103
systemctl stop firewalld
systemctl disable firewalld

• 2️⃣在三台虚拟机中的/usr/local目录中安装JDK和Zookeeper。

• 3️⃣分别在192.168.179.101、192.168.179.102和192.168.179.103三台机器上的zookeeper目录中新建data文件夹。

mkdir -pv data

• 4️⃣修改192.168.179.101、192.168.179.102、192.168.179.103机器上的zookeeper的conf目录中的zoo.cfg文件：

# 服务器对应的端口号
clientPort=2181
# 数据快照文件所在路径
dataDir=/usr/local/zookeeper-3.4.10/data
# 集群配置信息
# server.A=B:C:D
# A：是一个数字，表示当前服务器的编号
# B：当前服务器的IP地址
# C：Leader选举的端口
# D：Zookeeper服务器之间通信的端口
server.1=192.168.179.101:2888:3888
server.2=192.168.179.102:2888:3888
server.3=192.168.179.103:2888:3888

• 5️⃣在上一步dataDir指定的目录下，创建myid文件，然后在该文件中添加上一步server配置的编号数字：

# 192.168.107.101对应的数字是1
# /usr/local/zookeeper-3.4.10/data/
echo "1" > myid

# 192.168.107.102对应的数字是1
# /usr/local/zookeeper-3.4.10/data/
echo "2" > myid

# 192.168.107.103对应的数字是1
# /usr/local/zookeeper-3.4.10/data/
echo "3" > myid

• 6️⃣分别启动三台虚拟机中的zookeeper，检查集群状态：

#192.168.179.101
./zkServer.sh start

#192.168.179.102
./zkServer.sh start

#192.168.179.103
./zkServer.sh start

3 一致性协议：ZAB协议

• ZAB协议的全称是Zookeeper Atomic BroadCast（zookeeper原子广播）。zookeeper是通过ZAB协议来保证分布式事务的最终一致性。
• 基于ZAB协议，zookeeper集群中的角色主要有以下的三类。如下表所示：

• ZAB广播模式工作原理，通过类似于两阶段提交协议的方式解决数据一致性。

• 1️⃣Leader从客户端收到一个写请求。
• 2️⃣Leader生成一个新的事务并为这个事务生成一个唯一的Zxid。
• 3️⃣Leader将这个事务提议(propose)发送给所有的follower节点。
• 4️⃣follower节点将收到的事务请求加入到历史队列(history queue)中，并发送ack到Leader。
• 5️⃣当Leader收到大多数的follower（半数以上节点）的ack消息，Leader会发送commit请求。
• 6️⃣当follower收到commit请求的时候，从历史队列中将事务请求commit。

4 zookeeper的leader选举

4.1 服务器状态

• looking：寻找Leader状态。当服务器处于该状态的时候，它会认为当前集群中没有Leader，因此需要进入Leader选举状态。
• leading：领导者状态。表明当前服务器的角色是Leader。
• following：跟随者状态。表明当前服务器的角色是follower。
• observing：观察者状态。表明当前服务器的角色是observer。

4.2 服务器启动时期的Leader选举

• 在集群初始化阶段，当有一台服务器server1启动时，其单独无法进行和完成Leader的选举，当第二台服务器server2启动时，此时两台机器可以相互通信，每台机器都视图找到Leader，于是进入Leader选举状态。选举过程如下：
• 1️⃣每个server发出一个投票。由于是初始情况，server1和server2都会将自己作为Leader服务器来进行投票，每次投票会包含所推举的服务器的myid和zxid，使用(myid,zxid)来表示，此时server1的投票为(1,0)，server2的投票为(2,0)，然后各自将这个投票发给集群中的其他机器。
• 2️⃣集群中的每台服务器接收到来自集群中的各个服务器的投票。
• 3️⃣处理投票。针对每一个投票，服务器都需要将别人的投票和自己的投票进行PK，PK规则如下：
  • 优先检查zxid。zxid比较的服务器优先作为Leader。
  • 如果zxid相同，那么就比较myid。myid较大的服务器最为Leader服务器。

对于server1而言，它的投票是(1,0)，接受server2的投票为(2,0)，首先会比较两者的zxid，均为0，再比较myid，此时server2的myid最大，于是更新自己的投票为(2,0)，然后重新投票，对于server2而言，其无需更新自己的投票，只是再次向集群中的所有机器发出上一次投票信息即可。

• 4️⃣统计投票。每次投票后，服务器都会统计投票信息，判断是否已经有过半机器接收到相同的投票新，对于server1和server2而言，都统计出集群中已经有两台机器接收了(2,0)的投票信息，此时便认为已经选出了Leader。
• 5️⃣改变服务器状态。一旦确定了Leader，每个服务器都会更新自己的状态，如果是follower，那么就变更为following，如果是Leader，就变更为leading。

4.3 服务器运行时期的Leader选举

• 在zookeeper运行期间，Leader和follower服务器都各司其职，即便有非Leader服务器宕机或新加入，此时也不会影响Leader，但是一旦Leader服务器挂掉，那么整个集群将暂停对外提供服务，进入新一轮Leader选举，其过程和启动时期的Leader选举过程基本一致。
• 假设正在运行的有server1、server2和server3三台服务器，假设当前Leader是server2，如果某一时刻Leader挂了，此时便开始Leader选举。选举过程如下：
• 1️⃣变更状态。Leader挂后，余下的服务器都会将自己的状态变为looking，然后开始进入Leader选举过程。
• 2️⃣每个server会发出一个投票。在运行期间，每个服务器上的zxid可能不同，此时假定server1的zxid为122，server3的zxid为122，在第一轮投票中，server1和server3都会投自己一票，产生(1,122)和(3,122)，然后各自将投票发给集群中的所有机器。
• 3️⃣接收来自各个服务器的投票。和启动时过程相同。
• 4️⃣处理投票。和启动时过程相同，此时Leader3将会成为Leader。
• 5️⃣统计投票。和启动时过程相同。
• 6️⃣改变服务器的状态。和启动时过程相同。

5 observer角色及其配置

• observer角色特点：
  • 1️⃣不参与集群的Leader的选举。
  • 2️⃣不参与集群中写数据的ack反馈。
• 为了使用observer角色，在任何想要变成observer角色的配置文件中加入如下配置：

peerType=observer

• 并在所有sever的配置文件中，配置成observer模式的server的哪行配置追加:observer，例如：

server.3=192.168.179.103:2888:3888:observer

6 zookeeper API连接集群

/**
* @param connectString
*            comma separated host:port pairs, each corresponding to a zk
*            server. e.g. "127.0.0.1:3000,127.0.0.1:3001,127.0.0.1:3002" If
*            the optional chroot suffix is used the example would look
*            like: "127.0.0.1:3000,127.0.0.1:3001,127.0.0.1:3002/app/a"
*            where the client would be rooted at "/app/a" and all paths
*            would be relative to this root - ie getting/setting/etc...
*            "/foo/bar" would result in operations being run on
*            "/app/a/foo/bar" (from the server perspective).
* @param sessionTimeout
*            session timeout in milliseconds
* @param watcher
*            a watcher object which will be notified of state changes, may
*            also be notified for node events
*
* @throws IOException
*             in cases of network failure
* @throws IllegalArgumentException
*             if an invalid chroot path is specified
*/
public ZooKeeper(String connectString, int sessionTimeout, Watcher watcher)
    throws IOException
{
    this(connectString, sessionTimeout, watcher, false);
}

• 示例：

package com.sunxiaping.zookeeper04;

import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;

import java.io.IOException;
import java.util.concurrent.CountDownLatch;

public class ZookeeperClusterDemo {

    public static void main(String[] args) throws IOException, InterruptedException {
        //计数器对象
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //服务器的ip地址和端口  集群
        String connectString = "192.168.179.101:2181,192.168.179.102:2181,192.168.179.103:2181";
        //客户端和服务器之间的会话超时时间（以毫秒为单位）
        int sessionTimeout = 5000;
        //监视器对象
        Watcher watcher = new Watcher() {
            @Override
            public void process(WatchedEvent event) {
                //客户端和服务器连接创建成功
                if (event.getState() == Event.KeeperState.SyncConnected) {
                    System.out.println("event = " + event);
                    System.out.println("连接创建成功");
                    countDownLatch.countDown();
                }
            }
        };
        ZooKeeper zooKeeper = new ZooKeeper(connectString, sessionTimeout, watcher);

        //主线程阻塞等待连接对象的创建成功
        countDownLatch.await();

        System.out.println("zooKeeper = " + zooKeeper.getSessionId());
    }
}