Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: Really clear Windows Error Messages:>)

[Subject Prev][Subject Next][Thread Prev][Thread Next]
There is also another problem called memory leaks in many progrgrams that are
designed with C & C++.  Memory leak is not a easy problem to solve.  It usually
happens after operating a program for long time.  And then you machine locks up.
I was told that the Microsoft's Word had a pretty bad memory leak one time.  A new
language--JAVA, supposely dosen't have this problem.

Usually, memory leak will not damage you machine.

Micorsoft declared yesterday that the new version of Words will be HTML based.
Would it be possible for RISA to have the option of output in HTML in the future?
We will be able to have fonts and color to be incorporated in to our reports
automatically!

Sam Chang, SE
Cupertino, CA

PS:  Don't even think about killing yourself untill Bill Gate did it first.



BRBATES(--nospam--at)aol.com wrote:

> Dennis Wish wrote:
>
> <<
>  If you run Dr. Watson with Windows it's suppose to help you track down the
>  cause of a system crash. Even Einstein couldn't figure this one out:>)
>
>  "The application overflowed its stack.  This is typically caused by
>  unrestricted recursion or an exception inside an exception handler."
>
>  Glad to see that Microsoft is improving product reliability - any bet's I
>  won't be able to find this one on their knowledgebase?
>   >>
>
> Dennis,
>
> What you have here is probably a legitimate program bug, unless the software
> advertises a "quick return to the operating system" feature ;-) .
>
> What's probably happening inside the program is that a subroutine is being
> called recursively, i.e. it's looping back on itself. For example, subroutine
> A calls subroutine B. Subroutine B calls subroutine C, and then subroutine C
> calls subroutine A. Each subroutine allocates stack space, which is
> automatically freed when the subroutine terminates, but because of the
> recursion, subroutine A's stack is never released and in fact is being
> allocated again and again until finally there's no stack left and the error is
> generated by the OS.
>
> This kind of error is not uncommon in beta releases during the software
> development process, but should not occur in a production release.
>
> This is not a user error, this is a program error. About the only thing you
> can do other than contacting the program vendor and requesting a fixed version
> is to identify the sequence of events that triggered the error and avoiding
> that sequence in the future.
>
> Regards,
>
> Bruce Bates
> RISA Technologies
>
> P.S. If you now tell me you got this error when running a RISA product, I'll
> have to kill myself.