Is it sufficient to have System.Web.Configuration.HttpRuntimeSection.EnableHeaderChecking set to true (default) to fully prevent Http Header Injection attacks like Response Splitting etc.?
I'm asking because a white box penetration testing tool (fortify) reports exploitable http header injection issues with HttpResponse.Redirect and cookies but I haven't found a way to successfully perform an attack. (edit:..and we have EnableHeaderChecking turned on..)
EnableHeaderChecking is only for untrusted data. If you're passing data directly from a cookie into a Redirect, maybe the resulting headers are considered trusted and \r\n values aren't escaped.
AFAIK it's enough and it should be turned on by default.
I think Fortify is just thinking about defence in depth as if you change the configuration in the deployment etc.
I assume you did not strictly set it on in your configuration, if you have maybe Fortify is not that smart to figure that our.
Best way to confirm is exploiting it :)
I've been looking at this for some time now and draw the conclusion that setting EnableHeaderChecking to true is in fact good enough to prevent http header injection attacks.
Looking at 'reflected' ASP.NET code, I found that:
HttpResponseHeader (internal)HttpResponseHeader.MaybeEncodeHeader (for IIS7WorkerRequests)HttpResponseHeader instances are created before known headers like RedirectLocation or ContentType are sent (HttpResponse.GenerateResponseHeaders)HttpResponseHeader constructor checks the EnableheaderChecking setting and calls HttpResponseHeader.MaybeEncodeHeader when set to trueHttpResponseHeader.MaybeEncodeHeader correctly encodes newline characters which makes HTTP header injection attacks impossibleHere is a snippet to roughly demonstrate how I tested:
// simple http response splitting attack
Response.AddHeader("foo", "bar\n" +
// injected http response, bad if user provided
"HTTP/1.1 200 OK\n" +
"Content-Length: 19\n" +
"Content-Type: text/html\n\n" +
"<html>danger</html>"
);
The above only works if you explicitly turn EnableHeaderChecking off:
<httpRuntime enableHeaderChecking="false"/>Fortify simply doesn't take configuration into account (setting EnableHeaderChecking explicitly had no effect) and thus always reports these type of issues.
Josef, HttpResponse.AppendHeader() is not the only place where untrusted data can enter the HTTP Response Headers.
Any data from the attacker that ends up in Cookies or HTTP redirects can write new headers if the data contains a carriage return (or anything that is interpreted as a carriage return).
In general, it's a much better use of your time to validate your data than to sit around and try to work out exploits. Chances are, the hackers are going to be better at this than you or I are.