c - C套接字客户端程序中2个进程之间使用共享内存会导致段错误

Question

我正在创建一个有2个进程的客户端程序:一个父进程定期向服务器传输数据以使其知道客户端仍然处于连接状态，以及一个子进程接收用户输入并将其发送到服务器，如果该输入恰好是“#EXIT”，则子进程将让父进程知道程序需要通过共享内存中的变量终止，然后退出。问题是在程序打印任何内容之前我就遇到了段错误。

奇怪的是，我第一次运行它时，它给出了预期的输出，并在我输入“#EXIT”时终止，但我随后运行它时，它立即出现了 seg 错误。一开始我以为是因为最后没有调用shm_unlink，但是我加上了，结果是一样的。然后我认为这是因为我没有在子进程上调用 exit，但我添加了它，结果再次是相同的。

客户端代码:

#include <stdio.h>
#include <sys/socket.h>
#include <arpa/inet.h>
#include <unistd.h>
#include <string.h>
#include <stdbool.h>
#include <sys/mman.h>
#include <fcntl.h>
#include <sys/stat.h>
#include <stdlib.h>

int main(int argc, char const *argv[])
{
    if (argc-1 != 2){
        perror("Usage: ./351ChatClient [address] [port]");
        return -1;
    }

    int sock = 0, valread;
    struct sockaddr_in serv_addr;
    char buffer[1024] = {0};

    if ((sock = socket(AF_INET, SOCK_STREAM, 0)) < 0)
    {
        printf("\n Socket creation error \n");
        return -1;
    }

    serv_addr.sin_family = AF_INET;
    serv_addr.sin_port = htons(atoi(argv[2]));

    // Convert IPv4 and IPv6 addresses from text to binary form 
    if(inet_pton(AF_INET, argv[1], &serv_addr.sin_addr)<=0)
    {
        printf("\nInvalid address/ Address not supported \n");
        return -1;
    }

    if (connect(sock, (struct sockaddr *)&serv_addr, sizeof(serv_addr)) < 0)
    {
        printf("\nConnection Failed \n");
        return -1;
    }

    int* shared_memory;

    int shm_fd = shm_open("Transmitting", O_CREAT | O_EXCL | O_RDWR, S_IRWXU | S_IRWXG);

    ftruncate(shm_fd, sizeof(int));

    shared_memory = (int *) mmap(NULL, sizeof(int), PROT_READ | PROT_WRITE, MAP_SHARED, shm_fd, 0);

    shared_memory[0] = 0;

    int pid = fork();

    if (pid == 0){
        char data[256];
        while (strcmp(data, "#EXIT") != 0){
            fgets(data, sizeof(data), stdin);
            data[strlen(data)-1] = '\0';
            send(sock , data, strlen(data) , 0 );
            valread = read( sock , buffer, 1024);
            printf("%s\n",buffer );
        }
        shared_memory[0] = 1;
        exit(EXIT_SUCCESS);
     }
    else if (pid > 0){
        while(shared_memory[0] == 0){
            // transmit data every couple seconds to let server know client is still connected
        }
        shm_unlink("Transmitting");
        printf("Stopped transmitting\n");
    }
    return 0;
}

服务器代码:

#include <stdio.h>
#include <sys/socket.h> //For Sockets
#include <stdlib.h>
#include <netinet/in.h> //For the AF_INET (Address Family)
#include <string.h>
#include <stdbool.h>
#include <unistd.h>

#define MAXSIZE 256

void add_user(char name[], char password[]){
    if (strlen(name) > MAXSIZE || strlen(password) > MAXSIZE){
        // send error message to client
        return;
    }

    FILE* fp = fopen("users.txt", "a");
    char s[strlen(name)+strlen(password)+2];
    strcpy(s, name);
    strcat(s, "\n");
    strcat(s, password);
    strcat(s, "\n");
    fputs(s, fp);
    fclose(fp);
}

void connect_user(char name[], char password[]){

    FILE* fp = fopen("users.txt", "r");

    char line[MAXSIZE];
    char cur_name[MAXSIZE];

    bool match = false;

    int line_no = 1;
    while (fgets(line, sizeof(line), fp)){

        line[strlen(line)-1] = '\0'; // remove trailing newline

        if (line_no % 2 != 0){

            strcpy(cur_name, line);

            if (strcmp(cur_name, name) == 0){
                match = true;
            }

        }
        else if(match){

            if (strcmp(line, password) == 0){
                char s[strlen(cur_name)+strlen(" is entering the queue\n")];
                strcpy(s, cur_name);
                strcat(s, " is entering the queue\n");
                printf("%s", s);
                // add user to the queue
                return;
            }
            else{
                // send invalid password message to client
                return;
            }
        }
        line_no = line_no + 1;
    }
    add_user(name, password); // if user not found, add it to the DB 
}

int main(int argc, char *argv[]){
    FILE* fp = fopen("users.txt", "a"); // creates users file if it does not exist
    fclose(fp);

    int sockfd, newsockfd, portno;
    socklen_t clilen;
    char buffer[256];
    struct sockaddr_in serv_addr, cli_addr;
    int n;
    if (argc < 2) {
         fprintf(stderr, "ERROR, no port provided\n");
         exit(1);
    }
    sockfd = socket(AF_INET, SOCK_STREAM, 0);
    if (sockfd < 0)
         perror("ERROR opening socket");
    bzero((char *) &serv_addr, sizeof(serv_addr));
    portno = atoi(argv[1]);
    serv_addr.sin_family = AF_INET;
    serv_addr.sin_addr.s_addr = INADDR_ANY;
    serv_addr.sin_port = htons(portno);
    if (bind(sockfd, (struct sockaddr *) &serv_addr, sizeof(serv_addr)) < 0)
        perror("ERROR on binding");
    listen(sockfd, 5);

    clilen = sizeof(cli_addr);
    //Below code is modified to handle multiple clients using fork
    //------------------------------------------------------------------
    int pid;
    while (1) {
         newsockfd = accept(sockfd, (struct sockaddr *) &cli_addr, &clilen);
         if (newsockfd < 0)
              perror("ERROR on accept");
         //fork new process
         pid = fork();
         if (pid < 0) {
              perror("ERROR in new process creation");
         }
         if (pid == 0) {
            //child process
            close(sockfd);
            //do whatever you want
            bzero(buffer, 256);
            while(strcmp(buffer, "#EXIT") != 0){

                memset(buffer, 0, sizeof(buffer)); // clear the read buffer

                n = read(newsockfd, buffer, 255);
                if (n < 0)
                    perror("ERROR reading from socket");
                printf("Here is the message: %s\n", buffer);

                n = write(newsockfd, "I got your message", 18);
                if (n < 0)
                    perror("ERROR writing to socket");
            }
            n = write(newsockfd, "See you soon!", 18);
            if (n < 0)
                perror("ERROR writing to socket");
            printf("Closing connection with client");
            close(newsockfd);
            exit(EXIT_SUCCESS);
          } else {
             //parent process
             close(newsockfd);
          }
    }
    //-------------------------------------------------------------------
   return 0;
}

当我输入“#EXIT”时，我期望收到的输出是

#EXIT
Stopped transmitting

但是我收到的输出很简单

segmentation fault (core dumped)

在我有机会输入任何内容之前。

我对 c 套接字和一般的 c 都很陌生，所以请毫不犹豫地指出我是否做了其他完全错误的事情。谢谢。

编辑:我注意到的另一件事是，当客户端出现段错误时，服务器会收到大量消息。但不知道为什么。

编辑2:我注释掉了客户端中的所有套接字代码，重新编译并执行它，仍然发生段错误。

编辑3:我在定义数据之前修复了 while 循环检查数据，但问题仍然存在。

这是简化的客户端代码:

#include <stdio.h>
#include <sys/socket.h>
#include <arpa/inet.h>
#include <unistd.h>
#include <string.h>
#include <stdbool.h>
#include <sys/mman.h>
#include <fcntl.h>
#include <sys/stat.h>
#include <stdlib.h>

int main(int argc, char const *argv[])
{
    if (argc-1 != 2){
        perror("Usage: ./351ChatClient [address] [port]");
        return -1;
    }

    int* shared_memory;

    int shm_fd = shm_open("Transmitting", O_CREAT | O_EXCL | O_RDWR, S_IRWXU | S_IRWXG);

    ftruncate(shm_fd, sizeof(int));

    shared_memory = (int *) mmap(NULL, sizeof(int), PROT_READ | PROT_WRITE, MAP_SHARED, shm_fd, 0);

    shared_memory[0] = 0;

    int pid = fork();

    if (pid == 0){
        char data[256];
        while (true){
            fgets(data, sizeof(data), stdin);
            data[strlen(data)-1] = '\0';
            if(strcmp(data, "#EXIT") != 0)
                break;  
        }
        shared_memory[0] = 1;
        exit(EXIT_SUCCESS);
     }
    else if (pid > 0){
        while(shared_memory[0] == 0){
            // transmit data every couple seconds to let server know client is still connected
        }
        shm_unlink("Transmitting");
        printf("Stopped transmitting\n");
    }
    return 0;
}

编辑4:我使用命令valgrind --leak-check=full --track-origins=yes ./351ChatClient 127.0.0.1 8080通过valgrind运行客户端，这就是我得到的:

==11098== Memcheck, a memory error detector
==11098== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==11098== Using Valgrind-3.13.0 and LibVEX; rerun with -h for copyright info
==11098== Command: ./351ChatClient 127.0.0.1 8080
==11098== 
==11098== Invalid write of size 4
==11098==    at 0x108AA4: main (in /home/jp/Courses/csci351/projects/project1/351ChatClient)
==11098==  Address 0xffffffffffffffff is not stack'd, malloc'd or (recently) free'd
==11098== 
==11098== 
==11098== Process terminating with default action of signal 11 (SIGSEGV)
==11098==  Access not within mapped region at address 0xFFFFFFFFFFFFFFFF
==11098==    at 0x108AA4: main (in /home/jp/Courses/csci351/projects/project1/351ChatClient)
==11098==  If you believe this happened as a result of a stack
==11098==  overflow in your program's main thread (unlikely but
==11098==  possible), you can try to increase the size of the
==11098==  main thread stack using the --main-stacksize= flag.
==11098==  The main thread stack size used in this run was 8388608.
==11098== 
==11098== HEAP SUMMARY:
==11098==     in use at exit: 0 bytes in 0 blocks
==11098==   total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==11098== 
==11098== All heap blocks were freed -- no leaks are possible
==11098== 
==11098== For counts of detected and suppressed errors, rerun with: -v
==11098== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 0 from 0)
Segmentation fault (core dumped)

Answer 1

您的段错误错误可能源于add_user(...)和connect_user(中的free(s) ...) 函数，因为 char s[...] 是堆栈上可变大小的数组，没有分配给堆来空闲。你需要摆脱它们。

在服务器端和客户端，您不需要使用 wait(..) 或 waitpid(..) 来等待子进程

此外，最好使用 0 来初始化数组，例如char arr[SIZE] = {0}，也适用于可变大小的数组。

AFAI 还知道，bzero 已被弃用并降低了可移植性。所以你应该使用 memset(..) 来代替。

n = write(newsockfd, "See you Soon!", 18); 在第 128 行，剩下的 5 个字符(字节)怎么样？垃圾文章？

更好地使用 -Weverything 标志。

<小时/>

让我们关注以下一点，

shared_memory[0] = 0; ?

如果shm_open(..)返回-1怎么办？在这种情况下，您确定 shared_memory 指向您可以写入的合法位置吗？您应该始终特别注意返回值及其结果。

int shm_fd = shm_open("Transmitting", O_CREAT | O_EXCL | O_RDWR, S_IRWXU | S_IRWXG);
    if (shm_fd == -1) {
        perror("shm_fd error");
        exit(-1);
    }

在我的计算机中，一次性运行后会产生 shm_fd 错误:文件存在。

请记住，非匿名共享映射对象会保留在计算机中，除非手动删除或通过 unlink(..) 或 remove(..)

或者，只需删除 O_EXCL 标志即可。