Python 官方文档:入门教程 => 点击学习
目录1.理解全过程1.1.概述1.2.整个流程2.代码实现2.1.节点类2.2.代理配置类2.3.负载均衡算法接口2.4.平滑加权轮询算法2.5.代理服务线程类2.6.代理服务类2.
需要具备的知识
目的:实现Socket 集群服务的平滑加权轮询负载。
业务实现:客户端通过用户名来查询集群服务中的用户信息。
几个细节点
用于保存服务节点相关信息
package com.yty.proxy.lba;
public class node implements Comparable<Node>{
private String ip;
private Integer port;
private final Integer weight;
private Integer effectiveWeight;
private Integer currentWeight;
// 默认权重为:1
public Node(String ip,Integer port){
this(ip,port,1);
}
public Node(String ip,Integer port, Integer weight){
this.ip = ip;
this.port = port;
this.weight = weight;
this.effectiveWeight = weight;
this.currentWeight = weight;
}
public String getIp() {
return ip;
}
public void setIp(String ip) {
this.ip = ip;
}
public Integer getPort() {
return port;
}
public void setPort(Integer port) {
this.port = port;
}
public Integer getWeight() {
return weight;
}
public Integer getEffectiveWeight() {
return effectiveWeight;
}
public void setEffectiveWeight(Integer effectiveWeight) {
this.effectiveWeight = effectiveWeight;
}
public Integer getCurrentWeight() {
return currentWeight;
}
public void setCurrentWeight(Integer currentWeight) {
this.currentWeight = currentWeight;
}
// 每成功一次,恢复有效权重1,不超过配置的起始权重
public void onInvokeSuccess(){
if(effectiveWeight < weight) effectiveWeight++;
}
// 每失败一次,有效权重减少1,无底线的减少
public void onInvokeFault(){
effectiveWeight--;
}
@Override
public int compareTo(Node node) {
return currentWeight > node.currentWeight ? 1 : (currentWeight.equals(node.currentWeight) ? 0 : -1);
}
@Override
public String toString() {
return "Node{" +
"ip='" + ip + '\'' +
", port=" + port +
", weight=" + weight +
", effectiveWeight=" + effectiveWeight +
", currentWeight=" + currentWeight +
'}';
}
}
一般都是放在配置文件中配置,然后读取指定key的配置文件信息来完成配置。模拟为了简单就直接写代码里了。
package com.yty.proxy;
import com.yty.proxy.lba.Node;
import java.util.ArrayList;
import java.util.List;
public class ProxyConfig {
private static List<Node> nodes = new ArrayList<>();
// 在配置文件中读取:节点集合信息。如果在同一台服务器测试,那就将ip配成一样
static {
nodes.add(new Node("192.168.233.100",8001,2));
nodes.add(new Node("127.0.0.1",8002,5));
nodes.add(new Node("127.0.0.1",8003,3));
}
public static List<Node> getProxyConfig(){
return nodes;
}
}
package com.yty.proxy.lba;
public interface Robin {
Node selectNode();
}
详细介绍可以阅读前两篇负载均衡算法的文章
package com.yty.proxy.lba;
import com.yty.proxy.ProxyConfig;
import java.util.List;
public class WeightedRoundRobin implements Robin {
private static List<Node> nodes;
// 读取配置信息
static {
nodes = ProxyConfig.getProxyConfig();
}
public Node selectNode(){
if (nodes ==null || nodes.size()<=0) return null;
if (nodes.size() == 1) return nodes.get(0);
// 权重之和
Integer totalWeight = 0;
for(Node node : nodes){
totalWeight += node.getEffectiveWeight();
}
synchronized (nodes){
// 选出当前权重最大的节点
Node nodeOfMaxWeight = null;
for (Node node : nodes) {
if (nodeOfMaxWeight == null)
nodeOfMaxWeight = node;
else
nodeOfMaxWeight = nodeOfMaxWeight.compareTo(node) > 0 ? nodeOfMaxWeight : node;
}
// 平滑负载均衡
nodeOfMaxWeight.setCurrentWeight(nodeOfMaxWeight.getCurrentWeight() - totalWeight);
nodes.forEach(node -> node.setCurrentWeight(node.getCurrentWeight()+node.getEffectiveWeight()));
return nodeOfMaxWeight;
}
}
}
用于处理代理服务请求的线程类,不同请求创建不同线程来处理
package com.yty.proxy;
import java.io.*;
import java.net.Socket;
import java.util.ArrayList;
import java.util.List;
public class ProxyServerThread implements Runnable {
private Socket proxySocket;
private OutputStream proxyOut;
private InputStream proxyIn;
private Socket socket;
private OutputStream serverOut;
private InputStream serverIn;
public ProxyServerThread(Socket proxySocket) throws IOException {
this.proxySocket = proxySocket;
this.proxySocket.setSoTimeout(6000);
this.proxyOut = proxySocket.getOutputStream();
this.proxyIn = proxySocket.getInputStream();
}
@Override
public void run() {
try {
this.proxyService();
} catch (IOException e) {
e.printStackTrace();
}finally {
this.close();
}
}
private void proxyService() throws IOException {
// 代理接收客户端请求
byte[] proxyDataBytes =null;
proxyDataBytes = getData(proxyIn);
System.out.println("代理收到请求数据:"+new String(proxyDataBytes));
if (proxyDataBytes == null){
proxyOut.write("请求内容异常".getBytes());
}
byte[] serverData = this.dispatcherService(proxyDataBytes);
// 代理响应客户端
assert serverData != null;
proxyOut.write(serverData);
proxySocket.shutdownOutput();
System.out.println("代理响应客户端数据:"+new String(proxyDataBytes));
}
private byte[] dispatcherService(byte[] proxyDataBytes){
// 选择节点:发送请求和接收响应信息
Robin wrr = new WeightedRoundRobin();
Node node = wrr.selectNode();
byte[] serverData = null;
try {
this.socket = new Socket(node.getIp(), node.getPort());
socket.setSoTimeout(6000);
serverIn = socket.getInputStream();
serverOut= socket.getOutputStream();
serverOut.write(proxyDataBytes);
socket.shutdownOutput();
serverData = getData(serverIn);
System.out.println("真实服务端响应数据:"+ new String(serverData));
node.onInvokeSuccess();//提权
} catch (IOException e) {
node.onInvokeFault();//降权
serverData = "代理的下游服务器异常".getBytes();
}
System.out.println("负载均衡到:" + node);
return serverData;
}
private byte[] getData(InputStream in) throws IOException {
List<Byte> byteList = new ArrayList<>();
int temp = -1;
while (true) {
temp = in.read();
if (temp != -1)
byteList.add((byte) temp);
else
break;
}
byte[] bytes = new byte[byteList.size()];
for (int i=0;i<byteList.size();i++){
bytes[i]=byteList.get(i);
}
return bytes;
}
private void close() {
try {
if (proxySocket!=null){
proxySocket.shutdownInput();
proxySocket.close();
}
if (socket!=null){
socket.shutdownInput();
socket.close();
}
}catch (IOException e){
e.printStackTrace();
System.out.println("代理服务关闭socket资源异常");
}
}
}
通过线程池来管理代理服务线程,不同的请求分发到不同线程处理。这里用的是newCachedThreadPool 线程池。
代理服务器在本地启动。这里也可以新建一个类来启动服务,这样可以启动多个代理服务,这里为了简单就直接在本类的main方法启动。可以对比后面的业务服务类,业务服务类就是这么起的,因为要放到不同的服务器启动业务服务。
package com.yty.proxy;
import java.io.IOException;
import java.net.ServerSocket;
import java.net.Socket;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
public class ProxyServer {
private final Integer port;
private ServerSocket serverSocket;
public ProxyServer(Integer port) {
this.port = port;
}
public void start(){
ExecutorService threadPool = Executors.newCachedThreadPool();
try {
serverSocket = new ServerSocket(port);
while (true){
Socket socket = serverSocket.accept();
threadPool.execute(new ProxyServerThread(socket));
}
} catch (IOException e) {
e.printStackTrace();
}
}
public static void main(String[] args) {
Integer proxyPort=8000;
ProxyServer proxyServer = new ProxyServer(proxyPort);
System.out.println("开启代理服务……");
proxyServer.start();
}
}
package com.yty.proxy.server;
public class MyUser {
private String name;
private int age;
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
public int getAge() {
return age;
}
public void setAge(int age) {
this.age = age;
}
public MyUser(String name, int age) {
this.name = name;
this.age = age;
}
public MyUser() {
}
@Override
public String toString() {
return "{" +"name='" + name + '\'' + ", age=" + age + '}';
}
}
处理具体业务的类,通过用户名称简单获取信息
package com.yty.proxy.server;
import java.util.ArrayList;
import java.util.List;
public class MyUserService {
private static List<MyUser> list = new ArrayList<>();
static{
list.add(new MyUser("张三",18));
list.add(new MyUser("张三丰",38));
list.add(new MyUser("小白",18));
}
public MyUser findByUsername(String username){
for (MyUser user:list){
if (user.getName().equals(username)){
return user;
}
}
return null;
}
}
用于处理业务服务请求的线程类,不同请求创建不同线程来处理
package com.yty.proxy.server;
import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.net.Socket;
import java.util.ArrayList;
import java.util.List;
public class SocketServerThread implements Runnable {
private Socket socket;
private OutputStream serverOut;
private InputStream serverIn;
public SocketServerThread(Socket socket) throws IOException {
this.socket = socket;
socket.setSoTimeout(6000);
this.serverOut = socket.getOutputStream();
this.serverIn = socket.getInputStream();
}
@Override
public void run() {
try {
this.service();
} catch (IOException e) {
e.printStackTrace();
}finally {
this.close();
}
}
private void service() throws IOException {
// 接收客户端请求
byte[] dataBytes =null;
dataBytes = getData(serverIn);
if (dataBytes == null){
serverOut.write("请求内容异常".getBytes());
}
String username = new String(dataBytes);
System.out.println("收到请求数据:"+username);
// 具体业务代码
MyUserService myUserService = new MyUserService();
MyUser user = myUserService.findByUsername(new String(dataBytes));
String serverData = "没有查询到用户" + username + "的数据";
if(user!=null){
serverData = user.toString();
serverOut.write(user.toString().getBytes());
}
System.out.println("响应客户端数据:" + serverData);
}
private byte[] getData(InputStream in) throws IOException {
List<Byte> byteList = new ArrayList<>();
int temp = -1;
while (true) {
temp = in.read();
if (temp != -1)
byteList.add((byte) temp);
else
break;
}
byte[] bytes = new byte[byteList.size()];
for (int i=0;i<byteList.size();i++){
bytes[i]=byteList.get(i);
}
return bytes;
}
private void close() {
try {
if (socket!=null){
socket.shutdownInput();
socket.shutdownOutput();
socket.close();
}
}catch (IOException e){
e.printStackTrace();
System.out.println("服务关闭socket资源异常");
}
}
}
通过线程池来管理业务服务线程,不同的请求分发到不同线程处理。这里用的也是newCachedThreadPool 线程池。
package com.yty.proxy.server;
import java.io.IOException;
import java.net.ServerSocket;
import java.net.Socket;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
public class SocketServer {
private final Integer port;
private ServerSocket serverSocket;
private Integer threads = 3;
public SocketServer(Integer port) {
this.port = port;
}
public void start(){
ExecutorService threadPool = Executors.newFixedThreadPool(threads);
try {
serverSocket = new ServerSocket(port);
while (true){
Socket socket = serverSocket.accept();
threadPool.execute(new SocketServerThread(socket));
}
} catch (IOException e) {
e.printStackTrace();
}
}
}
本次测试是分在两台服务器测试,1台【192.168.233.100】、另外都是本地【127.0.0.1】。如果觉得麻烦,那就都配成本地【127.0.0.1】,然后起服务都在本地起。
服务1:在IP为192.168.233.100 的服务器启动
package com.yty.proxy.test;
import com.yty.proxy.server.SocketServer;
public class StartServer1 {
public static void main(String[] args) {
System.out.println("开启后端服务8001……");
new SocketServer(8001).start();
}
}
服务2:在本地服务器启动
package com.yty.proxy.test;
import com.yty.proxy.server.SocketServer;
public class StartServer2 {
public static void main(String[] args) {
System.out.println("开启后端服务8002……");
new SocketServer(8002).start();
}
}
服务3:在本地服务器启动
package com.yty.proxy.test;
import com.yty.proxy.server.SocketServer;
public class StartServer3 {
public static void main(String[] args) {
System.out.println("开启后端服务8003……");
new SocketServer(8003).start();
}
}
package com.yty.proxy.test;
import java.io.BufferedInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.net.Socket;
import java.util.ArrayList;
import java.util.List;
public class Client {
public static void main(String[] args) throws IOException {
String ip = "127.0.0.1";
int port = 8000;
Socket socket = new Socket(ip, port);
socket.setSoTimeout(6000);
OutputStream out = socket.getOutputStream();
InputStream in = socket.getInputStream();
// 发送数据
out.write("小白".getBytes());
out.flush();
socket.shutdownOutput();
// 读取数据
byte[] data = new Client().getData(in);
System.out.println("响应数据:"+new String(data));
out.close();
}
private byte[] getData(InputStream in) throws IOException {
BufferedInputStream bin = new BufferedInputStream(in);
List<Byte> byteList = new ArrayList<>();
while (true) {
int temp = bin.read();
if (temp != -1)
byteList.add((byte) temp);
else
break;
}
byte[] bytes = new byte[byteList.size()];
for (int i=0;i<byteList.size();i++){
bytes[i]=byteList.get(i);
}
return bytes;
}
}
在代理配置类(ProxyConfig)中指定的服务器启动三个业务服务;
在你喜欢的服务器中启动代理服务(ProxyServer),这里在本地启动【127.0.0.1】;
客户端在本地测试咯(IP必须是代理服务器的IP,这里测试的代理服务器IP是【127.0.0.1】。
所有服务启动后的截图:
正常命中权重最高的节点2服务:节点信息在代理服务器中打印出来了【127.0.0.1、8002】。这些日志信息正常情况是写入到日志文件,这里只在控制台打印出来。
改了用户名再请求试试,发现忘记打印请求数据了……
第二次命中节点3服务,跟平滑加权算法预定的结果一样。
这次命中了节点2:192.168.233.100,8001的服务。到此可以看到平滑加权轮询算法正常运作中。
通过平滑加权轮询算法运算,我们知道这次肯定命中节点2服务。所以,在发起请求前,先关闭节点2服务,再由客户端发起请求。
细心的应该发现,有效权重没变小啊,是不是降权有问题?
其实不是,是打印信息的位置没放对……。要在下次访问才可以看到上一次的降权结果,额,有点呆(上面的代码我已经改了)。
先把宕机的服务启动起来,然后多测试几次,看看测试结果。可以看到,权重降低后又提起来了,说明测试提权成功。
还有两个点没测:第一个是一直降权后,会不会出现宕机的服务不再分配到?这就起到”剔除“宕机服务的效果?第二个是服务恢复后,会不会出现宕机再起的服务需要慢慢恢复权重,直到一定值后才可以分配到?
使用自己编写的平滑加权轮询算法,结合线程池和Socket 网络编程等,实现了反向代理集群服务的平滑分配,并通过降权/提权实现宕机服务的”剔除“和缓冲恢复。
到此这篇关于通过Java实现反向代理集群服务的平滑分配的文章就介绍到这了,更多相关Java集群服务平滑分配内容请搜索编程网以前的文章或继续浏览下面的相关文章希望大家以后多多支持编程网!
--结束END--
本文标题: 通过Java实现反向代理集群服务的平滑分配
本文链接: https://lsjlt.com/news/146071.html(转载时请注明来源链接)
有问题或投稿请发送至: 邮箱/279061341@qq.com QQ/279061341
2024-03-01
2024-03-01
2024-03-01
2024-02-29
2024-02-29
2024-02-29
2024-02-29
2024-02-29
2024-02-29
2024-02-29
回答
回答
回答
回答
回答
回答
回答
回答
回答
回答
0