Hey Steve,

I though I'd update you and the list on what actually transpired.

On Wed, Dec 22, 2010 at 2:01 PM, Steve Vinoski <vinoski@ieee.org> wrote:
On Wed, Dec 22, 2010 at 11:48 AM, Michal D. <michal.dobrogost@gmail.com> wrote:
> Sure, that makes sense, but it seems to me that isReady() should eventually
> ...
> get a response back?

I was running this in Firefox with Firebug enabled and watching the
JavaScript console, and you can definitely see that it gets to the
point where if you hit the reset button at just the right time, the
request loop stops. I don't know if it's because the submit() function
sets onreadystatechange to an empty function or what, but that's what
I see.

The problem was that I was missing the 'var' keyword before defining my XMLHttpRequest objects. 

xmlhttp = NewXMLHttpRequest();
Should have had:
var xmlhttp = NewXMLHttpRequest();

Since I was creating them in two places I ended up clobbering one when I created the other and vice-versa.  I found it almost impossible to track this down - the major clue came when I made the server side 'submit' code return something and that showed up in the 'request' output in the client.  A strict hint that something funky was going on with the onreadystate callback.

The crash that was happening in the server itself never got tracked down.  It didn't seem to affect the state of any of the processes and so it remains a bit of a mystery.  With rewriting and adding stuff it eventually went away.

Thanks for your help Steve,