C# – HttpModule Init method is called several times – why

chttpmoduleinitnet

I was creating a http module and while debugging I noticed something which at first (at least) seemed like weird behaviour.

When I set a breakpoint in the init method of the httpmodule I can see that the http module init method is being called several times even though I have only started up the website for debugging and made one single request (sometimes it is hit only 1 time, other times as many as 10 times).

I know that I should expect several instances of the HttpApplication to be running and for each the http modules will be created, but when I request a single page it should be handled by a single http application object and therefore only fire the events associated once, but still it fires the events several times for each request which makes no sense – other than it must have been added several times within that httpApplication – which means it is the same httpmodule init method which is being called every time and not a new http application being created each time it hits my break point (see my code example at the bottom etc.).

What could be going wrong here? Is it because I am debugging and set a breakpoint in the http module?

It have noticed that it seems that if I startup the website for debugging and quickly step over the breakpoint in the httpmodule it will only hit the init method once and the same goes for the eventhandler. If I instead let it hang at the breakpoint for a few seconds the init method is being called several times (seems like it depends on how long time I wait before stepping over the breakpoint). Maybe this could be some build in feature to make sure that the httpmodule is initialized and the http application can serve requests , but it also seems like something that could have catastrophic consequences.

This could seem logical, as it might be trying to finish the request and since I have set the break point it thinks something have gone wrong and try to call the init method again? Soo it can handle the request?

But is this what is happening and is everything fine (I am just guessing), or is it a real problem?

What I am specially concerned about is that if something makes it hang on the "production/live" server for a few seconds a lot of event handlers are added through the init and therefore each request to the page suddenly fires the eventhandler several times.

This behaviour could quickly bring any site down.

I have looked at the "original" .net code used for the httpmodules for formsauthentication and the rolemanagermodule, etc… But my code isn't any different that those modules uses.

My code looks like this.

public void Init(HttpApplication app)
{
    if (CommunityAuthenticationIntegration.IsEnabled)
    {
        FormsAuthenticationModule formsAuthModule = (FormsAuthenticationModule) app.Modules["FormsAuthentication"];         

        formsAuthModule.Authenticate += new FormsAuthenticationEventHandler(this.OnAuthenticate);
    }
}

Here is an example how it is done in the RoleManagerModule from the .NET framework:

public void Init(HttpApplication app)
{
    if (Roles.Enabled)
    {
        app.PostAuthenticateRequest += new EventHandler(this.OnEnter);
        app.EndRequest += new EventHandler(this.OnLeave);
    }
}

Does anyone know what is going on?

(I just hope someone out there can tell me why this is happening and assure me that everything is perfectly fine) 🙂

UPDATE:

I have tried to narrow down the problem and so far I have found that the init method being called is always on a new object of my http module (contrary to what I thought before).

I seems that for the first request (when starting up the site) all of the HttpApplication objects being created and their modules are all trying to serve the first request and therefore all hit the eventhandler that is being added.
I can't really figure out why this is happening.

If I request another page all the HttpApplication's created (and their modules) will again try to serve the request causing it to hit the eventhandler multiple times.

But it also seems that if I then jump back to the first page (or another one) only one HttpApplication will start to take care of the request and everything is as expected – as long as I don't let it hang at a break point.

If I let it hang at a breakpoint it begins to create new HttpApplication's objects and starts adding HttpApplications (more than 1) to serve/handle the request (which is already in process of being served by the HttpApplication which is currently stopped at the breakpoint).

I guess or hope that it might be some intelligent "behind the scenes" way of helping to distribute and handle load and / or errors. But I have no clue.
I hope some out there can assure me that it is perfectly fine and how it is supposed to be?

Best Answer

It's normal for the Init() method to be called multiple times. When an application starts up, the ASP.NET Worker process will instantiate as many HttpApplication objects as it thinks it needs, then it'll pool them (e.g. reuse them for new requests, similar to database connection pooling).

Now for each HttpApplication object, it will also instantiate one copy of each IHttpModule that is registered and call the Init method that many times. So if 5 HttpApplication objects are created, 5 copies of your IHttpModule will be created, and your Init method called 5 times. Make sense?

Now why is it instantiating 5 HttpApplication objects say? Well maybe your ASPX page has links to other resources which your browser will try to download, css, javascript, WebResource.aspx, maybe an iframe somewhere. Or maybe the ASP.NET Worker Process 'is in the mood' for starting more than 1 HttpApplication object, that's really an internal detail/optimisation of the ASP.NET process running under IIS (or the VS built in webserver).

If you want code that's guaranteed to run just once (and don't want to use the Application_StartUp event in the Global.asax), you could try the following in your IHttpModule:

private static bool HasAppStarted = false;
private readonly static object _syncObject = new object();

public void Init(HttpApplication context)
{
    if (!HasAppStarted)
    {
        lock (_syncObject)
        {
            if (!HasAppStarted)
            {
                // Run application StartUp code here

                HasAppStarted = true;
            }
        }
    }
}

I've done something similar and it seems to work, though I'd welcome critiques of my work in case I've missed something.

Related Solutions

C# – How to find the method that called the current method

Try this:

using System.Diagnostics;
// Get call stack
StackTrace stackTrace = new StackTrace(); 
// Get calling method name
Console.WriteLine(stackTrace.GetFrame(1).GetMethod().Name);

one-liner:

(new System.Diagnostics.StackTrace()).GetFrame(1).GetMethod().Name

It is from Get Calling Method using Reflection [C#].

C# – Why is it important to override GetHashCode when Equals method is overridden

Yes, it is important if your item will be used as a key in a dictionary, or HashSet<T>, etc - since this is used (in the absence of a custom IEqualityComparer<T>) to group items into buckets. If the hash-code for two items does not match, they may never be considered equal (Equals will simply never be called).

The GetHashCode() method should reflect the Equals logic; the rules are:

if two things are equal (Equals(...) == true) then they must return the same value for GetHashCode()
if the GetHashCode() is equal, it is not necessary for them to be the same; this is a collision, and Equals will be called to see if it is a real equality or not.

In this case, it looks like "return FooId;" is a suitable GetHashCode() implementation. If you are testing multiple properties, it is common to combine them using code like below, to reduce diagonal collisions (i.e. so that new Foo(3,5) has a different hash-code to new Foo(5,3)):

unchecked // only needed if you're compiling with arithmetic checks enabled
{ // (the default compiler behaviour is *disabled*, so most folks won't need this)
    int hash = 13;
    hash = (hash * 7) + field1.GetHashCode();
    hash = (hash * 7) + field2.GetHashCode();
    ...
    return hash;
}

Oh - for convenience, you might also consider providing == and != operators when overriding Equals and GetHashCode.

A demonstration of what happens when you get this wrong is here.

Best Answer

Related Solutions

C# – How to find the method that called the current method

C# – Why is it important to override GetHashCode when Equals method is overridden

Related Topic