chatgpt增加心跳和检查子进程状态

增加心跳和检查子进程状态.md

好的，以下是一个增加心跳和检查子进程状态的示例代码：

#include <iostream>
#include <vector>
#include <queue>
#include <unistd.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <sys/un.h>
#include <sys/wait.h>
#include <signal.h>

class ProcessPool {
public:
    ProcessPool(size_t numProcesses) {
        for (size_t i = 0; i < numProcesses; ++i) {
            int sockets[2];
            if (socketpair(AF_UNIX, SOCK_STREAM, 0, sockets) == -1) {
                std::cerr << "Failed to create socket pair" << std::endl;
                exit(1);
            }

            pid_t pid = fork();
            if (pid == -1) {
                std::cerr << "Failed to fork process" << std::endl;
                exit(1);
            } else if (pid == 0) {
                // Child process
                close(sockets[0]);
                handleClientConnection(sockets[1]);
                exit(0);
            } else {
                // Parent process
                close(sockets[1]);
                processIds.push_back(pid);
                workerSockets.push_back(sockets[0]);
            }
        }
    }

    ~ProcessPool() {
        for (pid_t pid : processIds) {
            kill(pid, SIGTERM);
        }
        for (pid_t pid : processIds) {
            waitpid(pid, nullptr, 0);
        }
    }

    void sendRequest(const std::string& request) {
        if (currentWorker < workerSockets.size()) {
            int socket = workerSockets[currentWorker];
            send(socket, request.c_str(), request.length(), 0);
            ++currentWorker;
        } else {
            std::cerr << "No available worker" << std::endl;
        }
    }

    void checkWorkers() {
        for (size_t i = 0; i < workerSockets.size(); ++i) {
            int socket = workerSockets[i];
            char buffer[2];
            ssize_t bytesRead = recv(socket, buffer, sizeof(buffer), MSG_DONTWAIT);
            if (bytesRead == -1 && (errno == EAGAIN || errno == EWOULDBLOCK)) {
                // No data available, worker is still alive
                continue;
            } else if (bytesRead == 0) {
                // Worker has closed the connection, restart it
                restartWorker(i);
            } else {
                // Received data from worker, handle it if needed
                // ...
            }
        }
    }

private:
    void handleClientConnection(int socket) {
        while (true) {
            char buffer[1024];
            ssize_t bytesRead = recv(socket, buffer, sizeof(buffer) - 1, 0);
            if (bytesRead <= 0) {
                break;
            }
            buffer[bytesRead] = '\0';
            std::cout << "Received request: " << buffer << std::endl;
            // Handle client request
            // ...
        }
        close(socket);
    }

    void restartWorker(size_t index) {
        pid_t pid = fork();
        if (pid == -1) {
            std::cerr << "Failed to fork process" << std::endl;
            exit(1);
        } else if (pid == 0) {
            // Child process
            int socket = workerSockets[index];
            close(socket);
            handleClientConnection(socket);
            exit(0);
        } else {
            // Parent process
            processIds[index] = pid;
            workerSockets[index] = createWorkerSocket();
        }
    }

    int createWorkerSocket() {
        int sockets[2];
        if (socketpair(AF_UNIX, SOCK_STREAM, 0, sockets) == -1) {
            std::cerr << "Failed to create socket pair" << std::endl;
            exit(1);
        }
        return sockets[0];
    }

    std::vector<pid_t> processIds;
    std::vector<int> workerSockets;
    size_t currentWorker = 0;
};

int main() {
    ProcessPool pool(4);

    pool.sendRequest("Request 1");
    pool.sendRequest("Request 2");
    pool.sendRequest("Request 3");

    while (true) {
        pool.checkWorkers();
        // Do other work
        // ...
        sleep(1);
    }

    return 0;
}

在这个示例中，我们添加了一个checkWorkers函数，用于检查子进程的状态。在这个函数中，我们遍历workerSockets向量，对每个子进程的socket进行非阻塞接收。如果接收到的字节数为0，表示子进程已经关闭连接，我们调用restartWorker函数重新启动该子进程。如果接收到的字节数大于0，表示从子进程接收到了数据，我们可以根据需要进行处理。

在main函数中，我们创建了一个包含4个子进程的进程池pool。然后，我们使用sendRequest函数向进程池中的子进程发送请求。在这个示例中，我们发送了3个请求，每个请求都会被发送给一个可用的子进程。然后，我们在一个无限循环中调用checkWorkers函数来检查子进程的状态，并在需要时重新启动子进程。你可以根据需要在主进程中进行其他工作。

希望对你有帮助！