zookeeper 实战操作

一：监听服务端zookeeper节点数据改变

import java.io.IOException;

import java.util.concurrent.CountDownLatch;

import java.util.concurrent.TimeUnit;

import org.apache.zookeeper.CreateMode;

import org.apache.zookeeper.KeeperException;

import org.apache.zookeeper.WatchedEvent;

import org.apache.zookeeper.Watcher;

import org.apache.zookeeper.Watcher.Event.EventType;

import org.apache.zookeeper.Watcher.Event.KeeperState;

import org.apache.zookeeper.ZooDefs.Ids;

import org.apache.zookeeper.ZooKeeper;

public class ConfigApp1 {

    private static CountDownLatch connectedSemaphore = new CountDownLatch(1);

    public static void main(String[] args) throws IOException, KeeperException, InterruptedException {

        //连接zookeeper服务器

        ZooKeeper zk = new ZooKeeper("127.0.0.1:2181", 5000,

                new Watcher() {

                    public void process(WatchedEvent event) {

                        if (KeeperState.SyncConnected == event.getState()) {    //zk连接成功通知事件

                            if ( EventType.None == event.getType() && null == event.getPath() ) {

                                connectedSemaphore.countDown();

                                System.out.println("===========");

                            }

                        }

                    }

                });

        connectedSemaphore.await();

        //创建节点app1，不进行ACL权限控制，EPHEMERAL：临时节点

        zk.create("/app1", "app1Date".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL);

        //注册对该节点的监听

        zk.exists("/app1", new WatcherClass(zk));

        //延迟2秒后 开始改变数据

        TimeUnit.SECONDS.sleep(2);

        for(int i = 0; i < 10; i++) {

            TimeUnit.SECONDS.sleep(1);

            String s = ("app" + i * 10);

            zk.setData("/app1", s.getBytes(), -1);

            System.out.println("数据改变了:"+s);

        }

        System.in.read();

    }

    static class WatcherClass implements Watcher {

        private ZooKeeper zk;

        public WatcherClass(ZooKeeper zk) {

            this.zk = zk;

        }

        @Override

        public void process(WatchedEvent arg0) {

            try {

                byte[] b = zk.getData("/app1", false, null);

                System.out.println("改变数据通知：" + new String(b));

                //获取数据后，再次对节点进行监听

                zk.exists("/app1", new WatcherClass(zk));

            } catch (KeeperException e) {

                e.printStackTrace();

            } catch (InterruptedException e) {

                e.printStackTrace();

            }

        }

    }

}

console结果截图：

二：集群管理

　　应用集群中，我们常常需要让每一个机器知道集群中（或依赖的其他某一个集群）哪些机器是活着的，并且在集群机器因为宕机，网络断链等原因能够不在人工介入的情况下迅速通知到每一个机器

思路：用三个类模拟成三个服务器，去连接zookeeper，这三个服务器监控zookeeper节点root，每个服务器上线都会在zookeeper的节点root下创建一个临时节点，这样，这三个服务器watcher这个zookeeper的root节点就可以动态感知服务器的上下线情况。

import java.io.IOException;

import java.util.List;

import java.util.concurrent.CountDownLatch;

import org.apache.zookeeper.CreateMode;

import org.apache.zookeeper.KeeperException;

import org.apache.zookeeper.WatchedEvent;

import org.apache.zookeeper.Watcher;

import org.apache.zookeeper.Watcher.Event.EventType;

import org.apache.zookeeper.Watcher.Event.KeeperState;

import org.apache.zookeeper.ZooDefs.Ids;

import org.apache.zookeeper.data.Stat;

import org.apache.zookeeper.ZooKeeper;

public class Cluster1 {

    private static final int zkSessionTimeOut = 5000;

    private static CountDownLatch connectedSemaphore = new CountDownLatch(1);

    public static void main(String[] args) throws IOException, InterruptedException, KeeperException {

        //连接zookeeper服务器

        ZooKeeper zk = new ZooKeeper("127.0.0.1:2181", zkSessionTimeOut,

                new Watcher() {

                    public void process(WatchedEvent event) {

                        if (KeeperState.SyncConnected == event.getState()) {    //zk连接成功通知事件

                            if ( EventType.None == event.getType() && null == event.getPath() ) {

                                connectedSemaphore.countDown();

                                System.out.println("===========");

                            }

                        }

                    }

                });

        connectedSemaphore.await();

        Stat stat = zk.exists("/root", true);

         if(stat == null) {

             System.out.println("/root" + "路径不存在，请先创建该节点");

             //创建节点root，不进行ACL权限控制，PERSISTENTAL：永久节点  只有永久节点才可以创建子节点的临时节点

             zk.create("/root", "rootDate".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT);

         }

        String clusterPath = zk.create("/root/cluster1", "cluster1Date".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL);

        System.out.println(clusterPath);

        ZkWatcher zkWatcher = new ZkWatcher(zk);

        List<String> clusterList = zk.getChildren("/root", zkWatcher);

        System.out.println("****************");

        for(String str : clusterList) {

            System.out.println("cluster:" + str);

        }

        System.out.println("****************");

        while(true) {

        }

    }

}

import java.io.IOException;

import java.util.List;

import java.util.concurrent.CountDownLatch;

import org.apache.zookeeper.CreateMode;

import org.apache.zookeeper.KeeperException;

import org.apache.zookeeper.WatchedEvent;

import org.apache.zookeeper.Watcher;

import org.apache.zookeeper.ZooKeeper;

import org.apache.zookeeper.Watcher.Event.EventType;

import org.apache.zookeeper.Watcher.Event.KeeperState;

import org.apache.zookeeper.ZooDefs.Ids;

import org.apache.zookeeper.data.Stat;

public class Cluster2 {

    private static final int zkSessionTimeOut = 5000;

    private static CountDownLatch connectedSemaphore = new CountDownLatch(1);

    public static void main(String[] args) throws IOException, InterruptedException, KeeperException {

        //连接zookeeper服务器

        ZooKeeper zk = new ZooKeeper("127.0.0.1:2181", zkSessionTimeOut,

                new Watcher() {

                    public void process(WatchedEvent event) {

                        if (KeeperState.SyncConnected == event.getState()) {    //zk连接成功通知事件

                            if ( EventType.None == event.getType() && null == event.getPath() ) {

                                connectedSemaphore.countDown();

                                System.out.println("===========");

                            }

                        }

                    }

                });

        connectedSemaphore.await();

        Stat stat = zk.exists("/root", true);

         if(stat == null) {

             System.out.println("/root" + "路径不存在，请先创建该节点");

             zk.create("/root", "rootDate".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT);

         }

        String clusterPath = zk.create("/root/cluster2", "cluster2Date".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL);

        System.out.println(clusterPath);

        ZkWatcher zkWatcher = new ZkWatcher(zk);

        List<String> clusterList = zk.getChildren("/root", zkWatcher);

        System.out.println("****************");

        for(String str : clusterList) {

            System.out.println("cluster:" + str);

        }

        System.out.println("****************");

        while(true) {

        }

    }

}

import java.io.IOException;

import java.util.List;

import java.util.concurrent.CountDownLatch;

import org.apache.zookeeper.CreateMode;

import org.apache.zookeeper.KeeperException;

import org.apache.zookeeper.WatchedEvent;

import org.apache.zookeeper.Watcher;

import org.apache.zookeeper.ZooKeeper;

import org.apache.zookeeper.data.Stat;

import org.apache.zookeeper.Watcher.Event.EventType;

import org.apache.zookeeper.Watcher.Event.KeeperState;

import org.apache.zookeeper.ZooDefs.Ids;

public class Cluster3 {

    private static final int zkSessionTimeOut = 5000;

    private static CountDownLatch connectedSemaphore = new CountDownLatch(1);

    public static void main(String[] args) throws IOException, InterruptedException, KeeperException {

        //连接zookeeper服务器

        ZooKeeper zk = new ZooKeeper("127.0.0.1:2181", zkSessionTimeOut,

                new Watcher() {

                    public void process(WatchedEvent event) {

                        if (KeeperState.SyncConnected == event.getState()) {    //zk连接成功通知事件

                            if ( EventType.None == event.getType() && null == event.getPath() ) {

                                connectedSemaphore.countDown();

                                System.out.println("===========");

                            }

                        }

                    }

                });

        connectedSemaphore.await();

         Stat stat = zk.exists("/root", true);

         if(stat == null) {

             System.out.println("/root" + "路径不存在，请先创建该节点");

             zk.create("/root", "rootDate".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT);

         }

        String clusterPath = zk.create("/root/cluster3", "cluster3Date".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL);

        System.out.println(clusterPath);

        ZkWatcher zkWatcher = new ZkWatcher(zk);

        List<String> clusterList = zk.getChildren("/root", zkWatcher);

        System.out.println("****************");

        for(String str : clusterList) {

            System.out.println("cluster:" + str);

        }

        System.out.println("****************");

        while(true) {

        }

    }

}

import java.util.List;

import org.apache.zookeeper.KeeperException;

import org.apache.zookeeper.WatchedEvent;

import org.apache.zookeeper.Watcher;

import org.apache.zookeeper.Watcher.Event.EventType;

import org.apache.zookeeper.ZooKeeper;

public class ZkWatcher implements Watcher{

    private ZooKeeper zk;

    public ZkWatcher(ZooKeeper zk) {

        this.zk = zk;

    }

    @Override

    public void process(WatchedEvent event) {

        if(EventType.NodeChildrenChanged.equals(event.getType())) {

            List<String> clusterList = null;

            try {

                clusterList = zk.getChildren("/root", this);

            } catch (KeeperException e) {

                e.printStackTrace();

            } catch (InterruptedException e) {

                e.printStackTrace();

            }

            System.out.println("****************");

            System.out.println("changed");

            for(String str : clusterList) {

                System.out.println("cluster:" + str);

            }

            System.out.println("****************");

        }

    }

}

zookeeper 实战操作的更多相关文章

架构设计：远程调用服务架构设计及zookeeper技术详解（下篇）
一.下篇开头的废话终于开写下篇了,这也是我写远程调用框架的第三篇文章,前两篇都被博客园作为[编辑推荐]的文章,很兴奋哦,嘿嘿~~~~,本人是个很臭美的人,一定得要截图为证: 今天是2014年的第一天 ...
[译]ZOOKEEPER RECIPES-Leader Election
选主使用ZooKeeper选主的一个简单方法是,在创建znode时使用Sequence和Ephemeral标志.主要思想是,使用一个znode,比如"/election",每个客 ...
zookeeper源码分析之六session机制
zookeeper中session意味着一个物理连接,客户端连接服务器成功之后,会发送一个连接型请求,此时就会有session 产生. session由sessionTracker产生的,sessio ...
zookeeper源码分析之五服务端(集群leader)处理请求流程
leader的实现类为LeaderZooKeeperServer,它间接继承自标准ZookeeperServer.它规定了请求到达leader时需要经历的路径: PrepRequestProcesso ...
zookeeper源码分析之四服务端(单机)处理请求流程
上文: zookeeper源码分析之一服务端启动过程中,我们介绍了zookeeper服务器的启动过程,其中单机是ZookeeperServer启动,集群使用QuorumPeer启动,那么这次我们分析 ...
zookeeper源码分析之三客户端发送请求流程
znode 可以被监控,包括这个目录节点中存储的数据的修改,子节点目录的变化等,一旦变化可以通知设置监控的客户端,这个功能是zookeeper对于应用最重要的特性,通过这个特性可以实现的功能包括配置的 ...
zookeeper源码分析之二客户端启动
ZooKeeper Client Library提供了丰富直观的API供用户程序使用,下面是一些常用的API: create(path, data, flags): 创建一个ZNode, path是其 ...
zookeeper源码分析之一服务端启动过程
zookeeper简介 zookeeper是为分布式应用提供分布式协作服务的开源软件.它提供了一组简单的原子操作,分布式应用可以基于这些原子操作来实现更高层次的同步服务,配置维护,组管理和命名.zoo ...
zookeeper集群的搭建以及hadoop ha的相关配置
1.环境 centos7 hadoop2.6.5 zookeeper3.4.9 jdk1.8 master作为active主机,data1作为standby备用机,三台机器均作为数据节点,yarn资源 ...

随机推荐

洛谷P3294 [SCOI2016]背单词——题解
题目传送阅读理解题题意解释可以看这位大佬的博客. 发现求后缀与倒序求前缀是等价的,而找前缀自然就想到了trie树.将所有字符串翻转后再建入trie树中,再对每一个字符串翻转后从trie树中找前缀,就 ...
「TJOI2019」甲苯先生的滚榜
题目链接问题分析参照数据范围,我们需要一个能够在\(O(n\log n)\)复杂度内维护有序数列的数据结构.那么平衡树是很好的选择.参考程序中使用带旋Treap. 参考程序 #pragma GCC ...
hexo的流程
1.基于node 的npm包管理工具2.npm install -g hexo(hexo-cli)3.下载的资源是国外的可能比较慢,可以使用淘宝镜像(代理的仓库)npm install -g hexo ...
Mongo rs
概念 rs中存储了每一次对mongo数据库的CUD操作,对rs的大小进行resize. 主要有两种方法: 在未开启过rs的情况下,修改配置文件已使用rs一段时间后,发现rs太大,重新进行大小配置启 ...
重温IOC，DI的理解
IOC和DI其实它们是同一个概念的不同角度描述 IOC强调的是程序控制对象(创建销毁),变换成了容器来控制对象(创建销毁) DI:即IoC容器帮对象找相应的依赖对象通过反射注入从Spring ...
JSP之Bean
<jsp:useBean id=" " class" "/>创建JavaBean对象,并把创建的对象保存到域对象比如:<jsp:useBea ...
idea 更新后和新的直接安装前，都需要配置 idea64.exe.vmoptions 后再使用
配置合适的参数提升性能,默认的性能不高
套接字之recvmsg系统调用
recvmsg系统调用允许用户指定msghdr结构来接收数据,可以将数据接收到多个缓冲区中,并且可以接收控制信息:接收信息过程与其他接收系统调用核心一致,都是调用传输层的接收函数进行数据接收: SYS ...
怎样用 Bash 编程：语法和工具
让我们通过本系列文章来学习基本的 Bash 编程语法和工具,以及如何使用变量和控制运算符,这是三篇中的第一篇. Shell 是操作系统的命令解释器,其中 Bash 是我最喜欢的.每当用户或者系统管理员 ...
SQL Server2016 AlwaysOn无域高可用
https://blog.csdn.net/qq_41981651/article/details/90314817 https://blog.csdn.net/roven257/article/de ...

zookeeper 实战操作

zookeeper 实战操作的更多相关文章

随机推荐

热门专题