1 分布式安装部署
0 集群规划
在 hadoop102、 hadoop103 和 hadoop104 三个节点上部署 Zookeeper。
1 解压安装
1)解压 zookeeper 安装包到/opt/module/目录下
[joker@hadoop102 software]$ tar -zxvf zookeeper-3.4.10.tar.gz -C /opt/module/
2) 在/opt/module/zookeeper-3.4.10/这个目录下创建 zkData
mkdir -p zkData
3)重命名/opt/module/zookeeper-3.4.10/conf 这个目录下的 zoo_sample.cfg 为zoo.cfg
mv zoo_sample.cfg zoo.cfg
2 配置 zoo.cfg 文件
1)具体配置
dataDir=/opt/module/zookeeper-3.4.10/zkData
增加如下配置
#######################cluster##########################
server.2=hadoop102:2888:3888
server.3=hadoop103:2888:3888
server.4=hadoop104:2888:3888
2)配置参数解读
Server.A=B:C:D。
A 是一个数字,表示这个是第几号服务器;
B 是这个服务器的 ip 地址;
C 是这个服务器与集群中的 Leader 服务器交换信息的端口;
D 是万一集群中的 Leader 服务器挂了,需要一个端口来重新进行选举,选出一个新的Leader,而这个端口就是用来执行选举时服务器相互通信的端口。
集群模式下配置一个文件 myid,这个文件在 dataDir目录下,这个文件里面有一个数据就是 A 的值, Zookeeper 启动时读取此文件,拿到里面的数据与 zoo.cfg 里面的配置信息比较从而判断到底是哪个 server。
3 集群操作
1)在/opt/module/zookeeper-3.4.10/zkData 目录下创建一个 myid 的文件
touch myid
添加 myid 文件,注意一定要在 linux 里面创建,在 notepad++里面很可能乱码
2)编辑 myid 文件
vi myid
在文件中添加与 server 对应的编号:如 2
3)拷贝配置好的 zookeeper 到其他机器上
xsync zookeeper-3.4.10/
也可以
scp -r zookeeper-3.4.10/ root@hadoop103:/opt/module/
scp -r zookeeper-3.4.10/ root@hadoop104:/opt/module/
并分别修改 myid 文件中内容为 3、 4
4)分别启动 zookeeper
[joker@hadoop102 zookeeper-3.4.10]# bin/zkServer.sh start
[joker@hadoop103 zookeeper-3.4.10]# bin/zkServer.sh start
[joker@hadoop104 zookeeper-3.4.10]# bin/zkServer.sh start
5)查看状态
[joker@hadoop102 zookeeper-3.4.10]# bin/zkServer.sh status
JMX enabled by default
Using config: /opt/module/zookeeper-3.4.10/bin/../conf/zoo.cfg
Mode: follower
[joker@hadoop103 zookeeper-3.4.10]# bin/zkServer.sh status
JMX enabled by default
Using config: /opt/module/zookeeper-3.4.10/bin/../conf/zoo.cfg
Mode: leader
[joker@hadoop104 zookeeper-3.4.5]# bin/zkServer.sh status
JMX enabled by default
Using config: /opt/module/zookeeper-3.4.10/bin/../conf/zoo.cfg
Mode: follower
2 客户端命令行操作
1 启动客户端
[joker@hadoop103 zookeeper-3.4.10]$ bin/zkCli.sh
2 显示所有操作命令
[zk: localhost:2181(CONNECTED) 1] help
3 查看当前 znode 中所包含的内容
[zk: localhost:2181(CONNECTED) 0] ls /
[zookeeper]
4 查看当前节点数据并能看到更新次数等数据
[zk: localhost:2181(CONNECTED) 1] ls2 /
[zookeeper]
cZxid = 0x0
ctime = Thu Jan 01 08:00:00 CST 1970
mZxid = 0x0
mtime = Thu Jan 01 08:00:00 CST 1970
pZxid = 0x0
cversion = -1
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 0
numChildren = 1
5 创建普通节点
[zk: localhost:2181(CONNECTED) 2] create /app1 "hello app1"
Created /app1
[zk: localhost:2181(CONNECTED) 4] create /app1/server101 "192.168.25.101"
Created /app1/server101
6 获得节点的值
[zk: localhost:2181(CONNECTED) 6] get /app1
hello app1
cZxid = 0x20000000a
ctime = Mon Jul 17 16:08:35 CST 2017
mZxid = 0x20000000a
mtime = Mon Jul 17 16:08:35 CST 2017
pZxid = 0x20000000b
cversion = 1
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 10
numChildren = 1
[zk: localhost:2181(CONNECTED) 8] get /app1/server101
192.168.25.101
cZxid = 0x20000000b
ctime = Mon Jul 17 16:11:04 CST 2017
mZxid = 0x20000000b
mtime = Mon Jul 17 16:11:04 CST 2017
pZxid = 0x20000000b
cversion = 0
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 13
numChildren = 0
7 创建短暂节点
[zk: localhost:2181(CONNECTED) 9] create -e /app-emphemeral 8888
1)在当前客户端是能查看到的
[zk: localhost:2181(CONNECTED) 10] ls /
[app1, app-emphemeral, zookeeper]
2)退出当前客户端然后再重启客户端
[zk: localhost:2181(CONNECTED) 12] quit
[atguigu@hadoop104 zookeeper-3.4.10]$ bin/zkCli.sh
3)再次查看根目录下短暂节点已经删除
[zk: localhost:2181(CONNECTED) 0] ls /
[app1, zookeeper]
8 创建带序号的节点
1)先创建一个普通的根节点 app2
[zk: localhost:2181(CONNECTED) 11] create /app2 "app2"
2)创建带序号的节点
[zk: localhost:2181(CONNECTED) 13] create -s /app2/aa 888
Created /app2/aa0000000000
[zk: localhost:2181(CONNECTED) 14] create -s /app2/bb 888
Created /app2/bb0000000001
[zk: localhost:2181(CONNECTED) 15] create -s /app2/cc 888
Created /app2/cc0000000002
如果原节点下有 1 个节点,则再排序时从 1 开始,以此类推。
[zk: localhost:2181(CONNECTED) 16] create -s /app1/aa 888
Created /app1/aa0000000001
9 修改节点数据值
[zk: localhost:2181(CONNECTED) 2] set /app1 999
10 节点的值变化监听
1)在 104 主机上注册监听/app1 节点数据变化
[zk: localhost:2181(CONNECTED) 26] get /app1 watch
2)在 103 主机上修改/app1 节点的数据
[zk: localhost:2181(CONNECTED) 5] set /app1 777
3)观察 104 主机收到数据变化的监听
WATCHER::
WatchedEvent state:SyncConnected type:NodeDataChanged path:/app1
11 节点的子节点变化监听(路径变化)
1)在 104 主机上注册监听/app1 节点的子节点变化
[zk: localhost:2181(CONNECTED) 1] ls /app1 watch
[aa0000000001, server101]
2)在 103 主机/app1 节点上创建子节点
[zk: localhost:2181(CONNECTED) 6] create /app1/bb 666
Created /app1/bb
3)观察 104 主机收到子节点变化的监听
WATCHER::
WatchedEvent state:SyncConnected type:NodeChildrenChanged path:/app1
12 删除节点
[zk: localhost:2181(CONNECTED) 4] delete /app1/bb
13 递归删除节点
[zk: localhost:2181(CONNECTED) 7] rmr /app2
14 查看节点状态
[zk: localhost:2181(CONNECTED) 12] stat /app1
cZxid = 0x20000000a
ctime = Mon Jul 17 16:08:35 CST 2017
mZxid = 0x200000018
mtime = Mon Jul 17 16:54:38 CST 2017
pZxid = 0x20000001c
cversion = 4
dataVersion = 2
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 3
numChildren = 2
3 API 应用
1 Eclipse 环境搭建
1)创建一个maven工程
2)加入pom文件依赖
<dependency>
<groupId>org.apache.zookeepergroupId>
<artifactId>zookeeperartifactId>
<version>3.4.10version>
dependency>
3)加入log4j.properties文件
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
2 创建 ZooKeeper 客户端
private static String connectString = "hadoop102:2181,hadoop103:2181,hadoop104:2181";
private static int sessionTimeout = 2000;
private ZooKeeper zkClient = null;
// 初始化客户端
@Before
public void initClient() throws Exception {
zkClient = new ZooKeeper(connectString, sessionTimeout, new Watcher() {
@Override
public void process(WatchedEvent event) {
// 收到事件通知后的回调函数(用户的业务逻辑)
System.out.println(event.getType() + "--" + event.getPath());
// 再次启动监听
try {
zkClient.getChildren("/", true);
} catch (Exception e) {
e.printStackTrace();
}
}
});
}
3 创建子节点
// 创建子节点
@Test
public void create() throws Exception {
// 数据的增删改查
// 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 path = zkClient.create("/eclipse", "hello joker".getBytes(), Ids.OPEN_ACL_UNSAFE, CreateMode.PERSISTENT);
System.out.println(path);
}
4 获取子节点并监听
// 获取子节点
@Test
public void getChildren() throws Exception {
List children = zkClient.getChildren("/", true);
for (String child : children) {
System.out.println(child);
}
// 延时阻塞
Thread.sleep(Long.MAX_VALUE);
}
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("/eclipse", false);
System.out.println(stat == null ? "not exist" : "exist");
}
4 案例实战
1 监听服务器节点动态上下线案例
1)需求:某分布式系统中,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)具体实现:
(0)现在集群上创建/servers 节点
[zk: localhost:2181(CONNECTED) 10] create /servers "servers"
Created /servers
(1)服务器端代码
package com.da;
import java.io.IOException;
import org.apache.zookeeper.CreateMode;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooDefs.Ids;
import org.apache.zookeeper.ZooKeeper;
public class ZkServer {
private static String connectString = "hadoop102:2181,hadoop103:2181,hadoop104:2181";
private static int sessionTimeout = 2000;
private ZooKeeper zk = null;
private String parentNode = "/servers";
// 创建到 zk 的客户端连接
public void getConnect() throws IOException {
zk = new ZooKeeper(connectString, sessionTimeout, new Watcher() {
@Override
public void process(WatchedEvent event) {
}
});
}
// 注册服务器
public void registServer(String hostname) throws Exception {
String create = zk.create(parentNode + "/server", hostname.getBytes(), Ids.OPEN_ACL_UNSAFE,
CreateMode.EPHEMERAL_SEQUENTIAL);
System.out.println(hostname + " is noline " + create);
}
// 业务功能
public void business(String hostname) throws Exception {
System.out.println(hostname + " is working ...");
Thread.sleep(Long.MAX_VALUE);
}
public static void main(String[] args) throws Exception {
// 获取 zk 连接
ZkServer server = new ZkServer();
server.getConnect();
// 利用 zk 连接注册服务器信息
server.registServer(args[0]);
// 启动业务功能
server.business(args[0]);
}
}
(2)客户端代码
package com.da;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import org.apache.zookeeper.WatchedEvent;
import org.apache.zookeeper.Watcher;
import org.apache.zookeeper.ZooKeeper;
public class ZkClient {
private static String connectString = "hadoop102:2181,hadoop103:2181,hadoop104:2181";
private static int sessionTimeout = 2000;
private ZooKeeper zk = null;
private String parentNode = "/servers";
private volatile ArrayList serversList = new ArrayList<>();
// 创建到 zk 的客户端连接
public void getConnect() throws IOException {
zk = new ZooKeeper(connectString, sessionTimeout, new Watcher() {
@Override
public void process(WatchedEvent event) {
// 再次启动监听
try {
getServerList();
} catch (Exception e) {
e.printStackTrace();
}
}
});
}
//
public void getServerList() throws Exception {
// 获取服务器子节点信息,并且对父节点进行监听
List children = zk.getChildren(parentNode, true);
// 存储服务器列表
ArrayList servers = new ArrayList<>();
// 获取每个节点的数据
for (String child : children) {
byte[] data = zk.getData(parentNode + "/" + child, false, null);
servers.add(new String(data));
}
// 把 servers 赋值给成员 serverList,已提供给各业务线程使用
serversList = servers;
System.out.println(serversList);
}
// 业务功能
public void business() throws Exception {
System.out.println("client is working ...");
Thread.sleep(Long.MAX_VALUE);
}
public static void main(String[] args) throws Exception {
// 获取 zk 连接
ZkClient client = new ZkClient();
client.getConnect();
// 获取 servers 的子节点信息,从中获取服务器信息列表
client.getServerList();
// 业务进程启动
client.business();
}
}