.NET-请求在特定时间段后超时

Question

I'm writing a simple application to query a certain URL over a period of time to detect changes, but I notice that after 5,000 requests or so, the requests will start to time out.

It does cycle through hosts (has about 200 different hosts to cycle through)

The target and source connections are both extremely stable.

After restarting the program, there are no issues with timing out.

Here is the relevant code:

public static string download_data()
{
    String ret = String.Empty;
    try
    {
        Uri newUri = new Uri(FinalURL.Replace(Host,Hosts[bestHostIndex]));
        var request = (HttpWebRequest)WebRequest.Create(newUri);
        request.Method = "GET";
        request.AllowAutoRedirect = true;
        request.KeepAlive = true;
        request.Referer = EscapedURL;
        request.UserAgent = UA;
        request.Headers.Add("Pragma", "no-cache");
        request.Timeout = 3000;
        request.Host = HostMain;
        request.Proxy = null;
        request.CookieContainer = Session.Cookies;
        using (var resp = request.GetResponse())
        {
            var reader = new StreamReader(resp.GetResponseStream());
            ret = reader.ReadToEnd();
        }

            return ret;
    }
    catch(WebException e)
    {
        Console.WriteLine("Error: "+e.Message);
        if (e.Status == WebExceptionStatus.ProtocolError)
        {
            WebResponse resp = e.Response;
            using (StreamReader sr = new StreamReader(resp.GetResponseStream()))
            {
                ret = sr.ReadToEnd();
            }
        }
        return ret;
    }
}

Very strange because I think for there to be an issue with connectivity, then restarting the application wouldn't fix the issue. It seems like the network is being "jammed" the longer the application runs. Perhaps I'm not properly flushing the resource or something like that.

Also it may be relevant to mention that the first request ALWAYS times out.

Help would be great

Answer 1

You need couple more using statements:

using (var resp = request.GetResponse())
using (var responseStream = resp.GetResponseStream())
using (var reader = new StreamReader(responseStream))
{
    ret = reader.ReadToEnd();
}

Without calling Close or disposing the stream returned by GetResponseStream the connection won't be released, and that's why you're running out of connections after a while. MSDN is clear about that:

You must call either the Stream.Close or the HttpWebResponse.Close method to close the stream and release the connection for reuse. It is not necessary to call both Stream.Close and HttpWebResponse.Close , but doing so does not cause an error. Failure to close the stream will cause your application to run out of connections.