Re: Debugging Challenge - no source available

From: Ray Cassick \(Home\) (rcassickNOSPAM_at_enterprocity.com)
Date: 09/24/04 

Date: Thu, 23 Sep 2004 22:57:54 -0400

My guess might be that some judiciously placed debug.writeline statements
might be in order to at least help you collect some runtime data as to where
you were in the code at the time of the crash.

"markusszil" <markusszil@discussions.microsoft.com> wrote in message
news:5063FA9E-27B3-4DC3-A0C0-957E4464F404@microsoft.com...
> Hello all,
>
> I am working on an application which uses quite a few bitmaps as skins for
> custom controls.
>
> I am getting a very difficult bug to track - my application performs quite
a
> few bitmap operations which use a timer to help time-slice so the UI
remains
> responsive. At one point, I am setting a flag to indicate to the timer
that a
> particular control needs to have its background refreshed. This is placed
> into a list for future handling. This completes fine. Debugging in the
timer
> also doesn't reveal any problems, although it gets rather difficult to
follow
> the execution thread.
>
> At any rate, at some, not particularly repeatable point, I press F10 and
> control returns to .net/Windows and I get the error:
>
> "A first chance exception of type 'System.ArgumentException' occurred in
> system.drawing.dll Additional information: Invalid parameter used."
>
> The problem is that I don't have any clue as to where the error is
occurring
> - the debugger stops at the Application.Run() line, there's no call stack,
no
> objects I can evaluate, I'm kinda stuck.
>
> If I turn off exception debugging (i.e. set "When the exception is thrown"
> radio-button to "Continue"), then the error changes to:
>
> "An unhandled exception of type 'System.ArgumentException' occurred in
> system.windows.forms.dll Additional information: Invalid parameter
used."
>
> In both cases, the debugger catches execution at the same point, just
before
> shutting the program down.
>
> I'm looking for any pointers on steps I can take to at least cause
execution
> to pause at some useful point.
>
> Thanks,
> Markus Szillat

Loading