发送未感知套接字挂起错误

Question

我的 TCP 客户端使用 tcp_keepalive_interval = 10s、tcp_keepalive_time = 1s 和 tcp_keepalive_probes = 10 实现了一个 keepalive 功能。

但是send 函数在连接断开超过 20 秒后没有发现任何错误。理想情况下，启用 keepalive，20 秒后 (keepalive_interval + keepalive_probes*keepalive_time) E_TIMEDOUT 应添加到套接字挂起错误中。

正如this 回答所说，

"read(2) 和 write(2) 都首先检索任何挂起的错误 甚至在尝试处理任何数据之前的套接字。”

当连接被keepalive关闭时，发送应该选择E_TIMEDOUT或任何套接字错误，但在下面的代码中没有发生。

    int sockfd = -1;
    if ((sockfd = socket(AF_INET, SOCK_STREAM, IPPROTO_TCP)) == -1) {
        cerr<<"ERROR: Failed to obtain Socket Descriptor!"<<endl;
        return -1;
    }

    //setting keepalive
    int optval;
    socklen_t optlen = sizeof(optval);

    //setting keepalive
    optval = 1;
    optlen = sizeof(optval);
    if(setsockopt(sockfd, SOL_SOCKET, SO_KEEPALIVE, &optval, optlen) < 0) {
        cerr<<"set keepalive failed"<<endl;
        return -1;
    }
    //setting tcp_keepalive_intvl
    optval = 10;
    optlen = sizeof(optval);
    if(setsockopt(sockfd, SOL_TCP, TCP_KEEPINTVL, &optval, optlen) < 0) {
        cerr<<"set tcp_keepalive_interval failed"<<endl;
        return -1;
    }
    //setting tcp_keepalive_time
    optval = 1;
    optlen = sizeof(optval);
    if(setsockopt(sockfd, SOL_TCP, TCP_KEEPIDLE, &optval, optlen) < 0) {
        cerr<<"set tcp_keepalive_time failed"<<endl;
        return -1;
    }
    //setting tcp_keepalive_probes
    optval = 10;
    optlen = sizeof(optval);
    if(setsockopt(sockfd, SOL_TCP, TCP_KEEPCNT, &optval, optlen) < 0) {
        cerr<<"set tcp_keepalive_probe failed"<<endl;
        return -1;
    }

    struct sockaddr_in remote_addr;
    remote_addr.sin_family = AF_INET;
    remote_addr.sin_port = htons(#port_no);
    remote_addr.sin_addr.s_addr = inet_addr(#ip_addr);
    memset(&remote_addr.sin_zero, 0, 8);
    if (connect(sockfd, (struct sockaddr *)&remote_addr, sizeof(struct sockaddr)) == -1) {
        cerr<<"Connect failed with ERRORNO "<<errno<<endl;
        return -1;
    } else {
        cout<<"TcpClient.cpp connected to server"<<endl;
    }
    while(1) {
        char data[20] = "hi hello";
        int ret = -1;
        if((ret = send(sockfd, data, 20, 0)) < 0) {
            cerr<<"TcpClient.cpp:- failed to send_data, ERROR CODE: "<<errno<<endl;
            return -1;
        } else if (ret == 0) {
            cout<<"send returns 0"<<endl;
        } else {
            cout<<"data sent"<<endl;
        }        
        sleep(1);
    }
    getchar();
    return 0;

我在一台带有 gcc 编译器的 linux 机器上测试了这段代码。 注意：我用recv 尝试了相同的代码，它巧妙地选择了E_TIMEDOUT 错误。

Answer 1

你还没等多久。来自http://tldp.org/HOWTO/TCP-Keepalive-HOWTO/usingkeepalive.html：

tcp_keepalive_time

    the interval between the last data packet sent (simple ACKs are not considered data) and the first keepalive probe; after the connection is marked to need keepalive, this counter is not used any further 

tcp_keepalive_intvl

    the interval between subsequential keepalive probes, regardless of what the connection has exchanged in the meantime 

tcp_keepalive_probes

    the number of unacknowledged probes to send before considering the connection dead and notifying the application layer 

因此，如果您的值为 (10, 10, 10)，keepalive 将在从对等方收到最后一个数据后 10 秒后开始。然后需要 10 个探测，每个探测间隔 10 秒，然后才能宣布连接失效。