一、客户端 API 操作
1.1 准备一个项目
前提:保证 hadoop101、hadoop102、hadoop103 服务器上 Zookeeper 集群服务端启动。
1)创建一个工程:zookeeper
2)添加pom文件
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<version>RELEASE</version>
</dependency>
<dependency>
<groupId>org.apache.logging.log4j</groupId>
<artifactId>log4j-core</artifactId>
<version>2.8.2</version>
</dependency>
<dependency>
<groupId>org.apache.zookeeper</groupId>
<artifactId>zookeeper</artifactId>
<version>3.6.3</version>
</dependency>
3)拷贝log4j.properties文件到项目根目录
需要在项目的 src/main/resources 目录下,新建一个文件,命名为“log4j.properties”,在文件中填入。
log4j.rootLogger=INFO, stdout
log4j.appender.stdout=org.apache.log4j.ConsoleAppender
log4j.appender.stdout.layout=org.apache.log4j.PatternLayout
log4j.appender.stdout.layout.ConversionPattern=%d %p [%c] - %m%n
log4j.appender.logfile=org.apache.log4j.FileAppender
log4j.appender.logfile.File=target/spring.log
log4j.appender.logfile.layout=org.apache.log4j.PatternLayout
log4j.appender.logfile.layout.ConversionPattern=%d %p [%c] - %m%n
4)创建包名com.jsxl.zk
5)创建类名称zkClient
1.2 创建 ZooKeeper 客户端
package com.jsxl.zk;
import org.apache.zookeeper.CreateMode;
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.junit.Before;
import org.junit.Test;
import java.util.List;
public class zkClient {
// 注意:逗号前后不能有空格
private static String connectString = "hadoop101:2181,hadoop102:2181,hadoop103:2181";
private static int sessionTimeout = 2000;
private ZooKeeper zkClient = null;
@Before
public void init() throws Exception {
zkClient = new ZooKeeper(connectString, sessionTimeout, new
Watcher() {
@Override
public void process(WatchedEvent watchedEvent) {
// 收到事件通知后的回调函数(用户的业务逻辑)
System.out.println(watchedEvent.getType() + "--" + watchedEvent.getPath());
// 再次启动监听
try {
List<String> children = zkClient.getChildren("/", true);
for (String child : children) {
System.out.println(child);
}
} catch (Exception e) {
e.printStackTrace();
}
}
});
}
}
1.3 创建子节点
// 创建子节点
@Test
public void create() throws Exception {
// 参数 1:要创建的节点的路径; 参数 2:节点数据 ; 参数 3:节点权限 ;参数 4:节点的类型
String nodeCreated = zkClient.create("/jsxl","one".getBytes(), ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT);
}
节点权限源码
@InterfaceAudience.Public
public interface Perms {
int READ = 1 << 0;
int WRITE = 1 << 1;
int CREATE = 1 << 2;
int DELETE = 1 << 3;
int ADMIN = 1 << 4;
int ALL = READ | WRITE | CREATE | DELETE | ADMIN;
}
@InterfaceAudience.Public
public interface Ids {
/**
* This Id represents anyone.
* 这个Id代表任何人。
*/
Id ANYONE_ID_UNSAFE = new Id("world", "anyone");
/**
* This Id is only usable to set ACLs. It will get substituted with the
* Id's the client authenticated with.
* 他的Id只能用于设置acl。它会被取代
* Id是通过身份验证的客户端。
*/
Id AUTH_IDS = new Id("auth", "");
/**
* This is a completely open ACL .
* 这是一个完全开放的ACL
*/
@SuppressFBWarnings(value = "MS_MUTABLE_COLLECTION", justification = "Cannot break API")
ArrayList<ACL> OPEN_ACL_UNSAFE = new ArrayList<ACL>(Collections.singletonList(new ACL(Perms.ALL, ANYONE_ID_UNSAFE)));
/**
* This ACL gives the creators authentication id's all permissions.
* 此ACL向创建者提供身份验证id的所有权限。
*/
@SuppressFBWarnings(value = "MS_MUTABLE_COLLECTION", justification = "Cannot break API")
ArrayList<ACL> CREATOR_ALL_ACL = new ArrayList<ACL>(Collections.singletonList(new ACL(Perms.ALL, AUTH_IDS)));
/**
* This ACL gives the world the ability to read.
* 这个ACL提供了阅读的能力。
*/
@SuppressFBWarnings(value = "MS_MUTABLE_COLLECTION", justification = "Cannot break API")
ArrayList<ACL> READ_ACL_UNSAFE = new ArrayList<ACL>(Collections.singletonList(new ACL(Perms.READ, ANYONE_ID_UNSAFE)));
}
查看控制台输出
在 hadoop101 的 zk 客户端上查看创建节点情况
[zk: hadoop101:2181(CONNECTED) 0] ls /
[dunzan, jsxl, zookeeper]
[zk: hadoop101:2181(CONNECTED) 1] get /jsxl
one
1.4 获取子节点并监听节点变化
// 获取子节点
@Test
public void getChildren() throws Exception {
List<String> children = zkClient.getChildren("/", true);
for (String child : children) {
System.out.println("获取子节点"+child);
}
// 延时阻塞
Thread.sleep(Long.MAX_VALUE);
}
(2)在 hadoop101的客户端上创建再创建一个节点/presistenet,观察 IDEA 控制台
(3)在 hadoop101 的客户端上删除节点/presistenet,观察 IDEA 控制台
[zk: hadoop101:2181(CONNECTED) 2] create /presistent "创建持久节点"
Created /presistent
[zk: hadoop101:2181(CONNECTED) 3] delete /presistent
查看ide控制台
1.5 判断 Znode 是否存在
// 判断 znode 是否存在
@Test
public void exist() throws Exception {
Stat stat = zkClient.exists("/jsxl", false);
System.out.println(stat == null ? "not exist" : "exist");
}
二、客户端向服务端写数据流程
写流程之写入请求直接发送给Leader节点
写流程之写入请求发送给follower节点
三、服务器动态上下线监听案例
主节点可以有多台,可以动态上下线,任意一台客户端都能实时感知到主节点服务器的上下线。
3.1 实现
先在集群上创建/servers 节点
[zk: hadoop101:2181(CONNECTED) 4] create /servers "servers"
Created /servers
3.2 服务器端向 Zookeeper 注册代码
在 Idea 中创建包名:com.jsxl.case1
package com.jsxl.case1;
import org.apache.zookeeper.*;
import java.io.IOException;
public class DistributeServer {
private String connectString = "hadoop101:2181,hadoop102:2181,hadoop103:2181";
private int sessionTimeout = 2000;
private ZooKeeper zk;
public static void main(String[] args) throws IOException, KeeperException, InterruptedException {
DistributeServer server = new DistributeServer();
// 1 获取zk连接
server.getConnect();
// 2 注册服务器到zk集群
server.regist(args[0]);
// 3 启动业务逻辑(睡觉)
server.business();
}
private void business() throws InterruptedException {
Thread.sleep(Long.MAX_VALUE);
}
private void regist(String hostname) throws KeeperException, InterruptedException {
String create = zk.create("/servers/"+hostname, hostname.getBytes(), ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL_SEQUENTIAL);
System.out.println(hostname +" is online") ;
}
private void getConnect() throws IOException {
zk = new ZooKeeper(connectString, sessionTimeout, new Watcher() {
@Override
public void process(WatchedEvent watchedEvent) {
}
});
}
}
3.3 客户端代码
package com.jsxl.case1;
import org.apache.zookeeper.KeeperException;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
public class DistributeClient {
private String connectString = "hadoop101:2181,hadoop102:2181,hadoop103:2181";
private int sessionTimeout = 2000;
private ZooKeeper zk;
public static void main(String[] args) throws IOException, KeeperException, InterruptedException {
DistributeClient client = new DistributeClient();
// 1 获取zk连接
client.getConnect();
// 2 监听/servers下面子节点的增加和删除
client.getServerList();
// 3 业务逻辑(睡觉)
client.business();
}
private void business() throws InterruptedException {
Thread.sleep(Long.MAX_VALUE);
}
private void getServerList() throws KeeperException, InterruptedException {
List<String> children = zk.getChildren("/servers", true);
ArrayList<String> servers = new ArrayList<>();
for (String child : children) {
byte[] data = zk.getData("/servers/" + child, false, null);
servers.add(new String(data));
}
// 打印
System.out.println(servers);
}
private void getConnect() throws IOException {
zk = new ZooKeeper(connectString, sessionTimeout, new Watcher() {
@Override
public void process(WatchedEvent watchedEvent) {
try {
getServerList();
} catch (KeeperException e) {
e.printStackTrace();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
});
}
}
3.4 测试
在 Linux 命令行上操作增加减少服务器
(1)启动 DistributeClient 客户端
(2)在 hadoop101 上 zk 的客户端/servers 目录上创建临时带序号节点
[zk: hadoop101:2181(CONNECTED) 5] create -e -s /servers/hadoop101 "hadoop101"
Created /servers/hadoop1010000000000
[zk: hadoop101:2181(CONNECTED) 6] create -e -s /servers/hadoop102 "hadoop102"
Created /servers/hadoop1020000000001
(3)观察 Idea 控制台变化
(4)执行删除操作
[zk: hadoop101:2181(CONNECTED) 7] delete /servers/hadoop1010000000000
[zk: hadoop101:2181(CONNECTED) 8] delete /servers/hadoop1020000000001
(5)观察 Idea 控制台变化
3.5 在 Idea 上操作增加减少服务器
(1)启动 DistributeClient 客户端(如果已经启动过,不需要重启)
(2)启动 DistributeServer 服务
①点击 Edit Configurations…
②在弹出的窗口中(Program arguments)输入想启动的主机,例如,hadoop101
③回到 DistributeServer 的 main 方 法 , 右 键 , 在 弹 出 的 窗 口 中 点 击 Run
“DistributeServer.main()”
④观察 DistributeServer 控制台,提示 hadoop101 is working
⑤观察 DistributeClient 控制台,提示 hadoop101 已经上线
四、ZooKeeper 分布式锁案例
什么叫做分布式锁呢?
比如说"进程 1"在使用该资源的时候,会先去获得锁,"进程 1"获得锁以后会对该资源保持独占,这样其他进程就无法访问该资源,"进程 1"用完该资源以后就将锁释放掉,让其他进程来获得锁,那么通过这个锁机制,我们就能保证了分布式系统中多个进程能够有序的访问该临界资源。那么我们把这个分布式环境下的这个锁叫作分布式锁。
4.1 原生 Zookeeper 实现分布式锁案例
分布式锁实现
package com.jsxl.case2;
import org.apache.zookeeper.*;
import org.apache.zookeeper.data.Stat;
import java.io.IOException;
import java.util.Collections;
import java.util.List;
import java.util.concurrent.CountDownLatch;
public class DistributedLock {
private final String connectString = "hadoop101:2181,hadoop102:2181,hadoop103:2181";
private final int sessionTimeout = 2000;
private final ZooKeeper zk;
private CountDownLatch connectLatch = new CountDownLatch(1);
private CountDownLatch waitLatch = new CountDownLatch(1);
private String waitPath;
private String currentMode;
public DistributedLock() throws IOException, InterruptedException, KeeperException {
// 获取连接
zk = new ZooKeeper(connectString, sessionTimeout, new Watcher() {
@Override
public void process(WatchedEvent watchedEvent) {
// connectLatch 如果连接上zk 可以释放
if (watchedEvent.getState() == Event.KeeperState.SyncConnected){
connectLatch.countDown();
}
// waitLatch 需要释放
if (watchedEvent.getType()== Event.EventType.NodeDeleted && watchedEvent.getPath().equals(waitPath)){
waitLatch.countDown();
}
}
});
// 等待zk正常连接后,往下走程序
connectLatch.await();
// 判断根节点/locks是否存在
Stat stat = zk.exists("/locks", false);
if (stat == null) {
// 创建一下根节点
zk.create("/locks", "locks".getBytes(), ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT);
}
}
// 对zk加锁
public void zklock() {
// 创建对应的临时带序号节点
try {
currentMode = zk.create("/locks/" + "seq-", null, ZooDefs.Ids.OPEN_ACL_UNSAFE, CreateMode.EPHEMERAL_SEQUENTIAL);
// wait一小会, 让结果更清晰一些
Thread.sleep(10);
// 判断创建的节点是否是最小的序号节点,如果是获取到锁;如果不是,监听他序号前一个节点
List<String> children = zk.getChildren("/locks", false);
// 如果children 只有一个值,那就直接获取锁; 如果有多个节点,需要判断,谁最小
if (children.size() == 1) {
return;
} else {
Collections.sort(children);
// 获取节点名称 seq-00000000
String thisNode = currentMode.substring("/locks/".length());
// 通过seq-00000000获取该节点在children集合的位置
int index = children.indexOf(thisNode);
// 判断
if (index == -1) {
System.out.println("数据异常");
} else if (index == 0) {
// 就一个节点,可以获取锁了
return;
} else {
// 需要监听 他前一个节点变化
waitPath = "/locks/" + children.get(index - 1);
zk.getData(waitPath,true,new Stat());
// 等待监听
waitLatch.await();
return;
}
}
} catch (KeeperException e) {
e.printStackTrace();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
// 解锁
public void unZkLock() {
// 删除节点
try {
zk.delete(this.currentMode,-1);
} catch (InterruptedException e) {
e.printStackTrace();
} catch (KeeperException e) {
e.printStackTrace();
}
}
}
分布式锁测试
(1)创建两个线程
package com.jsxl.case2;
import org.apache.zookeeper.KeeperException;
import java.io.IOException;
public class DistributedLockTest {
public static void main(String[] args) throws InterruptedException, IOException, KeeperException {
final DistributedLock lock1 = new DistributedLock();
final DistributedLock lock2 = new DistributedLock();
new Thread(new Runnable() {
@Override
public void run() {
try {
lock1.zklock();
System.out.println("线程1 启动,获取到锁");
Thread.sleep(5 * 1000);
lock1.unZkLock();
System.out.println("线程1 释放锁");
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}).start();
new Thread(new Runnable() {
@Override
public void run() {
try {
lock2.zklock();
System.out.println("线程2 启动,获取到锁");
Thread.sleep(5 * 1000);
lock2.unZkLock();
System.out.println("线程2 释放锁");
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}).start();
}
}
观察控制台变化:
4.2 Curator 框架实现分布式锁案例
1)原生的 Java API 开发存在的问题
(1)会话连接是异步的,需要自己去处理。比如使用 CountDownLatch
(2)Watch 需要重复注册,不然就不能生效
(3)开发的复杂性还是比较高的
(4)不支持多节点删除和创建。需要自己去递归
2)Curator 是一个专门解决分布式锁的框架,解决了原生 JavaAPI 开发分布式遇到的问题。
详情请查看官方文档:https://curator.apache.org/index.html.
3)Curator 案例实操
(1)添加依赖
<dependency>
<groupId>org.apache.curator</groupId>
<artifactId>curator-framework</artifactId>
<version>4.3.0</version>
</dependency>
<dependency>
<groupId>org.apache.curator</groupId>
<artifactId>curator-recipes</artifactId>
<version>4.3.0</version>
</dependency>
<dependency>
<groupId>org.apache.curator</groupId>
<artifactId>curator-client</artifactId>
<version>4.3.0</version>
</dependency>
(2)代码实现
package com.jsxl.case3;
import org.apache.curator.framework.CuratorFramework;
import org.apache.curator.framework.CuratorFrameworkFactory;
import org.apache.curator.framework.recipes.locks.InterProcessMutex;
import org.apache.curator.retry.ExponentialBackoffRetry;
public class CuratorLockTest {
public static void main(String[] args) {
// 创建分布式锁1
InterProcessMutex lock1 = new InterProcessMutex(getCuratorFramework(), "/locks");
// 创建分布式锁2
InterProcessMutex lock2 = new InterProcessMutex(getCuratorFramework(), "/locks");
new Thread(new Runnable() {
@Override
public void run() {
try {
lock1.acquire();
System.out.println("线程1 获取到锁");
lock1.acquire();
System.out.println("线程1 再次获取到锁");
Thread.sleep(5 * 1000);
lock1.release();
System.out.println("线程1 释放锁");
lock1.release();
System.out.println("线程1 再次释放锁");
} catch (Exception e) {
e.printStackTrace();
}
}
}).start();
new Thread(new Runnable() {
@Override
public void run() {
try {
lock2.acquire();
System.out.println("线程2 获取到锁");
lock2.acquire();
System.out.println("线程2 再次获取到锁");
Thread.sleep(5 * 1000);
lock2.release();
System.out.println("线程2 释放锁");
lock2.release();
System.out.println("线程2 再次释放锁");
} catch (Exception e) {
e.printStackTrace();
}
}
}).start();
}
private static CuratorFramework getCuratorFramework() {
ExponentialBackoffRetry policy = new ExponentialBackoffRetry(3000, 3);
CuratorFramework client = CuratorFrameworkFactory.builder().connectString("hadoop102:2181,hadoop103:2181,hadoop104:2181")
.connectionTimeoutMs(2000)
.sessionTimeoutMs(2000)
.retryPolicy(policy).build();
// 启动客户端
client.start();
System.out.println("zookeeper 启动成功");
return client;
}
}
(2)观察控制台变化:
五、企业面试真题(面试重点)
5.1 选举机制
半数机制,超过半数的投票通过,即通过。
(1)第一次启动选举规则:
投票过半数时,服务器 id 大的胜出
(2)第二次启动选举规则:
- EPOCH 大的直接胜出
- EPOCH 相同,事务 id 大的胜出
- 事务 id 相同,服务器 id 大的胜出
5.2 生产集群安装多少 zk 合适?
安装奇数台。
生产经验:
- 10 台服务器:3 台 zk;
- 20 台服务器:5 台 zk;
- 100 台服务器:11 台 zk;
- 200 台服务器:11 台 zk
服务器台数多:好处,提高可靠性;坏处:提高通信延时
5.3 常用命令
ls、get、create、delete