基于ZK實現分布式锁-白红宇

引言

ZooKeeper是一个分布式的，开放源码的分布式应用程序协调服务，是Google的Chubby一个开源的实现，是Hadoop和Hbase的重要组件。它是一个为分布式应用提供一致性服务的软件，提供的功能包括：配置维护、域名服务、分布式同步、组服务等。

ZooKeeper的架构通过冗余服务实现高可用性。因此，如果第一次无应答，客户端就可以询问另一台ZooKeeper主机。ZooKeeper节点将它们的数据存储于一个分层的命名空间，非常类似于一个文件系统或一个前缀树结构。客户端可以在节点读写，从而以这种方式拥有一个共享的配置服务。更新是全序的。

基于ZooKeeper分布式锁的原理

在指定节点（locks）下创建临时顺序节点node_n

获取locks下所有子节点children

对子节点按节点自增序号从小到大排序

判断本节点是不是第一个子节点，若是，则获取锁；若不是，则监听比该节点小的那个节点的删除事件

若监听事件生效，则回到第二步重新进行判断，直到获取到锁

具体实现

下面就具体使用java和zookeeper实现分布式锁，操作zookeeper使用的是apache提供的zookeeper的包。

通过实现Watch接口，实现process(WatchedEvent event)方法来实施监控，使CountDownLatch来完成监控，在等待锁的时候使用CountDownLatch来计数，等到后进行countDown，停止等待，继续运行。

以下整体流程基本与上述描述流程一致，只是在监听的时候使用的是CountDownLatch来监听前一个节点。

package org.java.base.zookeeper;import java.io.IOException;import java.util.ArrayList;import java.util.Collections;import java.util.List;import java.util.concurrent.CountDownLatch;import java.util.concurrent.TimeUnit;import java.util.concurrent.locks.Condition;import java.util.concurrent.locks.Lock;import org.apache.zookeeper.CreateMode;import org.apache.zookeeper.KeeperException;import org.apache.zookeeper.WatchedEvent;import org.apache.zookeeper.Watcher;import org.apache.zookeeper.ZooDefs;import org.apache.zookeeper.ZooKeeper;import org.apache.zookeeper.data.Stat;import org.codehaus.jettison.json.JSONString;import com.fasterxml.jackson.databind.util.JSONPObject;import com.google.gson.JsonObject;public class DistributedLock implements Lock, Watcher { private ZooKeeper zk = null; // 根节点 private String ROOT_LOCK = "/locks"; // 竞争的资源 private String lockName; // 等待的前一个锁 private String WAIT_LOCK; // 当前锁 private String CURRENT_LOCK; // 计数器 private CountDownLatch countDownLatch; private int sessionTimeout = 30000; private List
    
      exceptionList = new ArrayList
     
      (); /** * 配置分布式锁 * @param config 连接的url * @param lockName 竞争资源 */ public DistributedLock(String config, String lockName) { this.lockName = lockName; try { // 连接zookeeper zk = new ZooKeeper(config, sessionTimeout, this); Stat stat = zk.exists(ROOT_LOCK, false); if (stat == null) { // 如果根节点不存在，则创建根节点 zk.create(ROOT_LOCK, new byte[0], ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT); } } catch (IOException e) { e.printStackTrace(); } catch (InterruptedException e) { e.printStackTrace(); } catch (KeeperException e) { e.printStackTrace(); } } // 节点监视器 public void process(WatchedEvent event) { if (this.countDownLatch != null) { this.countDownLatch.countDown(); } } public void lock() { if (exceptionList.size() > 0) { throw new LockException(exceptionList.get(0)); } try { if (this.tryLock()) { System.out.println(Thread.currentThread().getName() + " " + lockName + "获得了锁"); return; } else { // 等待锁 waitForLock(WAIT_LOCK, sessionTimeout); } } catch (InterruptedException e) { e.printStackTrace(); } catch (KeeperException e) { e.printStackTrace(); } } public boolean tryLock() { try { String splitStr = "_lock_"; if (lockName.contains(splitStr)) { throw new LockException("锁名有误"); } // 创建临时有序节点 CURRENT_LOCK = zk.create(ROOT_LOCK + "/" + lockName + splitStr, new byte[0], ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL_SEQUENTIAL); System.out.println(CURRENT_LOCK + " 已经创建"); // 取所有子节点 List
      
        subNodes = zk.getChildren(ROOT_LOCK, false); // 取出所有lockName的锁 List
       
         lockObjects = new ArrayList
        
         (); for (String node : subNodes) { String _node = node.split(splitStr)[0]; if (_node.equals(lockName)) { lockObjects.add(node); } } Collections.sort(lockObjects); System.out.println(Thread.currentThread().getName() + " 的锁是 " + CURRENT_LOCK); // 若当前节点为最小节点，则获取锁成功 if (CURRENT_LOCK.equals(ROOT_LOCK + "/" + lockObjects.get(0))) { return true; } // 若不是最小节点，则找到自己的前一个节点 String prevNode = CURRENT_LOCK.substring(CURRENT_LOCK.lastIndexOf("/") + 1); WAIT_LOCK = lockObjects.get(Collections.binarySearch(lockObjects, prevNode) - 1); } catch (InterruptedException e) { e.printStackTrace(); } catch (KeeperException e) { e.printStackTrace(); } return false; } public boolean tryLock(long timeout, TimeUnit unit) { try { if (this.tryLock()) { return true; } return waitForLock(WAIT_LOCK, timeout); } catch (Exception e) { e.printStackTrace(); } return false; } // 等待锁 private boolean waitForLock(String prev, long waitTime) throws KeeperException, InterruptedException { Stat stat = zk.exists(ROOT_LOCK + "/" + prev, true); if (stat != null) { System.out.println(Thread.currentThread().getName() + "等待锁 " + ROOT_LOCK + "/" + prev); this.countDownLatch = new CountDownLatch(1); // 计数等待，若等到前一个节点消失，则precess中进行countDown，停止等待，获取锁 this.countDownLatch.await(waitTime, TimeUnit.MILLISECONDS); this.countDownLatch = null; System.out.println(Thread.currentThread().getName() + " 等到了锁"); } return true; } public void unlock() { try { System.out.println("释放锁 " + CURRENT_LOCK); zk.delete(CURRENT_LOCK, -1); CURRENT_LOCK = null; zk.close(); } catch (InterruptedException e) { e.printStackTrace(); } catch (KeeperException e) { e.printStackTrace(); } } public Condition newCondition() { return null; } public void lockInterruptibly() throws InterruptedException { this.lock(); } public class LockException extends RuntimeException { private static final long serialVersionUID = 1L; public LockException(String e){ super(e); } public LockException(Exception e){ super(e); } }}复制代码

测试类

package org.java.base.zookeeper;/** * 这段代码有一个问题，看大家能否发现？ * @author Liuhaihua * */public class Test { static int n = 500;  public static void secskill() { System.out.println(--n); } public static void main(String[] args) {  Runnable runnable = new Runnable() { public void run() { DistributedLock lock = null; try { lock = new DistributedLock("10.152.17.73:2181", "test1"); lock.lock(); secskill(); System.out.println(Thread.currentThread().getName() + "正在运行"); } finally { if (lock != null) { lock.unlock(); } } } }; for (int i = 0; i < 10; i++) { Thread t = new Thread(runnable); t.start(); } }}复制代码