RPC 笔记（08）— socket 通信（多进程多线程服务器）

在上一节中如果并行的客户端连接数超过了默认开启进程的数量，那么后来的客户端请求将会阻塞，为了不阻塞新的客户端，我们可以将进程的单线程改成多线程即可。
​

服务端代码：

import json
import struct
import socket
import threading
import multiprocessing


def handle_conn(conn, ip, handlers):
    print("{} connect ...".format(ip))
    # 循环读写
    while True:
        length_prefix = conn.recv(4)  # 请求长度前缀
        if not length_prefix:         # 连接关闭了
            conn.close()
            print("{} close ...".format(ip))
            break  # 退出循环，处理下一个连接

        length, = struct.unpack("I", length_prefix)
        body = conn.recv(length)  # 请求消息体
        request = json.loads(body)

        client_method = request['client']
        client_parameter = request['params']
        print("client request method is {}, params is {}".format(client_method, client_parameter))

        handler = handlers[client_method]  # 查找请求处理器
        handler(conn, client_parameter)  # 处理请求


def process_request(sock, handlers):
    while True:
        conn, host_ip = sock.accept()  # 接收连接
        handle_conn(conn, host_ip, handlers)  # 处理连接


def multi_thread(sock, handlers):
    threads = []
    thread_nums = 2
    for i in range(thread_nums):
        t = threading.Thread(target=process_request, args=(sock, handlers))
        threads.append(t)
    for i in range(thread_nums):
        threads[i].start()
    for i in range(thread_nums):
        threads[i].join()


def func1(conn, params):
    res = json.dumps({"response": "OK", "result": params})  # 响应消息体
    length_prefix = struct.pack("I", len(res))  # 响应长度前缀
    conn.sendall(length_prefix)
    # conn.sendall(response)  # python2
    conn.sendall(str.encode(res))  # python3


def main():
    s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)  # 创建一个 TCP 套接字
    s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)  # 打开 reuse addr 选项
    s.bind(("localhost", 8080))  # 绑定端口
    s.listen(1)  # 监听客户端连接
    handlers = {
        "func1": func1    # 注册请求处理器
    }
    # process_request(s, handlers)  # 进入服务循环
    process = []
    process_nums = 2
    for i in range(process_nums):
        p = multiprocessing.Process(target=multi_thread, args=(s, handlers))
        process.append(p)
    for i in range(process_nums):
        process[i].start()
    for i in range(process_nums):
        process[i].join()


if __name__ == '__main__':
    main()

示例代码中开启 2 个进程，每个进程中又开了 2 个线程，那么此时最多可以同时开启 4 个客户端，超过 4 个客户端后多余的请求会被阻塞处理。
​

​