Disabling popup window on application memory fault

[No Such Luck]

Hi All:

I have developed and application (.exe) that I am using on WinXP to
process data at night. A memory fault occurs on approximately .7% of
the data (not a problem, the application does not have to be perfect
right now). However, when a memory fault does occur, the annoying error
report window pops up, suspending the processing until a user clears
the window. For overnight processing jobs, this is unacceptable.

I have disabled error reporting for this application, but a memory
fault still causes an alert window to pop up, that must be cleared by a
user.

Aside from fixing the problem that causes the mem fault, any idea on
how to have a memory fault handled without this popup alert, and have
processing continue?

Any help is appreciated.

 

[Dave Patrick]

To disable the good Dr. Watson completely.

HKLM\Software\Microsoft\Window­s NT\CurrentVersion\AeDebug
set the string value
Auto="0"
and possibly change the string for
Debugger="drwtsn32 -p %ld -e %ld -g"
to
Debugger=""

If you later change your mind, from the "Run" box
drwtsn32 -i

Else if you just want no notification, from the run box;
drwtsn32
then uncheck the box for "Visual Notification"

HKEY_LOCAL_MACHINE\SOFTWARE\Mi­crosoft\DrWatson
"VisualNotification"=dword:000­00000

--
Regards,


Dave Patrick ....Please no email replies - reply in newsgroup.
Microsoft Certified Professional
Microsoft MVP [Windows]
http://www.microsoft.com/protect

:
| Hi All:
|
| I have developed and application (.exe) that I am using on WinXP to
| process data at night. A memory fault occurs on approximately .7% of
| the data (not a problem, the application does not have to be perfect
| right now). However, when a memory fault does occur, the annoying error
| report window pops up, suspending the processing until a user clears
| the window. For overnight processing jobs, this is unacceptable.
|
| I have disabled error reporting for this application, but a memory
| fault still causes an alert window to pop up, that must be cleared by a
| user.
|
| Aside from fixing the problem that causes the mem fault, any idea on
| how to have a memory fault handled without this popup alert, and have
| processing continue?
|
| Any help is appreciated.
|

 

[Sten Westerback]

Hi All:

I have developed and application (.exe) that I am using on WinXP to
process data at night. A memory fault occurs on approximately .7% of
the data (not a problem, the application does not have to be perfect
right now). However, when a memory fault does occur, the annoying error
report window pops up, suspending the processing until a user clears
the window. For overnight processing jobs, this is unacceptable.

I have disabled error reporting for this application, but a memory
fault still causes an alert window to pop up, that must be cleared by a
user.

Aside from fixing the problem that causes the mem fault, any idea on
how to have a memory fault handled without this popup alert, and have
processing continue?

By using Structured Exception Handling in Windows... but there isn't much
else you can do than to suspend the thread... and hope that other threads
can run without it without being blocked due to some mutex/critical section
or semaphore tokens the thread has aquired.

So best to fix the bugs....

- Sten

 

[Kellie Fitton]

Hi,

Well, you can use the win32 API SetErrorMode() function to inform the
system that
your application program will handle any serious errors on its own. You
need to
include the following flags within the function call, to prevent the
display of the GPF
messageBox, when the runTime system detect the memory fault error.

SEM_FAILCRITICALERRORS
SEM_NOGPFAULTERRORBOX

That said, you need to call this function again when the application
program had
stopped, to restore back the old error mode for the other applications.

http://msdn.microsoft.com/library/default.asp?url=/library/en-us/debug/base/seterrormode.asp


Hope these information helps.

Kellie.

 

[TC]

Um, an exe that memory-faults is not "imperfect" - it is "buggy"!

Surely, if you already know where the bug is, it shouldn't be all that
hard to fix it?

TC