Search code examples
c#memory-managementhttpwebrequestwebbrowser-controlhttpwebresponse

C# - Memory management in a app that is periodicaly calling HttpWebRequest and WebBrowser

About:

I have this Windows Form application which every 60 seconds it captures information from two common web pages, do some simple string treatment with the result and do something (or not) based in the result. One of those sites doesn't have any protection, so I can easily get it's HTML code using HttpWebRequest and it's HttpWebResponse.GetResponseStream(). The other one has some code protection and I can't use the same approach. The solution was use the WebBrowser class to select all text of the site and copy to the clipboard, as Jake Drew posted here (method 1).

Extra information:

When the timer reachs 1 min, each method is asynchronously execuded using Task. At the end of each Task the main thread will search some information in those texts and take or not some decisions based in the result. After this process, not even the captured text will relevant anymore. Basically everything can be wipe out from memory, since I'll get everything new and process it in about 1 minute.

Problem:

Everything is working fine but the problem is that it's gradually increasing the memory usage (about 20mb for each ticking), which are unecessary as I said before I don't need to maintain data in running in memory more than I had in the begin of app's execution:

enter image description here

and after comparing two snapshots I've found these 3 objects. Apparently they're responsible for that excess of memory usage:

enter image description here

So, even after I put the main execution in Tasks and do everything I could to help the Garbage Collector, I still have this issue.

What else could I do to avoid this issue or dump the trash from memory??

Edit:

Here's the code that is capturing the HTML of the page using HttpWebRequest:

    HttpWebRequest request = (HttpWebRequest)WebRequest.Create(URL);
    using (HttpWebResponse response = (HttpWebResponse)request.GetResponse()) {

        if (response.StatusCode == HttpStatusCode.OK) {
            Stream receiveStream = response.GetResponseStream();
            StreamReader readStream = null;
            if (response.CharacterSet == null) {
                readStream = new StreamReader(receiveStream);
            } else {
                readStream = new StreamReader(receiveStream, Encoding.GetEncoding(response.CharacterSet));
            }
            PB_value = readStream.ReadToEnd();

            readStream.Close(); //Ensure
        }
        response.Close(); //Ensure
    }
Solution

  • Solved:

    After some research I've found a solution. I actually feel kind of ashamed because it is a quite simple solution that I haven't tried before, still, it's important to share.

    The first thing I did was create an Event to identify when my two Tasks were finished then I assigned two functions to this event. The first function forced the Garbage Collector (GC.Collect()). The second function disposed the two Tasks, since all the main processes were done inside them (T.Dispose()). Then I've got the result I wanted:

    enter image description here