fatal run time error: unknown source position

[Andy Wittmann]

Hi Brandon
thanks for your fast response.
Yes I deleted everything in the registry related to National Instruments, I was searching through the entire registry and looked for the keywords "National" and "CVI". I deleted allways the complete entry.
I spend the whole weekend from friday till sunday in killing and reinstalling CVI 6, the Runtime engines 6 and 7.1, GPIB software 1.7 and 2.4.
Additionally to the cleaning of the Registry I went through the Windows system32  and driver directories and deleted by hand every file related to National Instruments which I could find. But after reinstalling the GPIB and CVI I still got the error always and only when exiting the  programs. When stepping through in debug mode line by line to the very end of the program I could not see the problem, but when using the Run button then on the end of the programs the error was back again.
I guess when upgrading to version 8.1 the problem may disappears. It probably just comes up when downgrading from the Evaluation version 8.1 to the Full Development version 6. again ?
Somewhere there must be an entry or file left which is still active and I could not find it so far.
During the week I don't have much time to try around with the Laptop, I usually work on problems like this during the weekends (if my wife is not killing me)  🙂

Best regards
Andy

[menchar]

We've seen this same exception : 0xc0000005.   

I believe this access violation exeception may be most common on XP SP2.  Microsoft has added a service (called DEP) to XP SP2 that attempts to detect if data on a page marked as "non-execute" is trying to execute.  The intent is to catch malicious code:  worms / viruses commonly bring in malicious code as data (the virus' payload) and then try to execute the "data".

You can turn this feature off at boot time. (DEP function in WinXP SP2).

See this :

http://www.updatexp.com/0xC0000005.html

Menchar

[menchar]

A thought is that you have to be sure you uninstalled the 8.1 CVI RTE DLL's if you're trying to roll back to a previous CVI version.   Simply re-installing an older CVI version won't remove the 8.1 RTE - the installer will normally leave the newer version as is, since it's a newer version.

If I'm not mistaken, uninstalling CVI doesn't by default remove the RTE.

We had this problem trying to unhose a serial library problem we have with CVI 8.1.  We re-installed 7.1.1 and the problem didn't go away - we were still running the 8.1 RTE.

So if you're trying to dodge behavior you think might be particular to a specific version of CVI, take care to ensure you really have the CVI RTE you want.

Menchar

[Brandon_V]

---

@Andy Wittmann wrote:
Hi Brandon
thanks for your fast response.
Yes I deleted everything in the registry related to National Instruments, I was searching through the entire registry and looked for the keywords "National" and "CVI". I deleted allways the complete entry.
I spend the whole weekend from friday till sunday in killing and reinstalling CVI 6, the Runtime engines 6 and 7.1, GPIB software 1.7 and 2.4.
Additionally to the cleaning of the Registry I went through the Windows system32  and driver directories and deleted by hand every file related to National Instruments which I could find. But after reinstalling the GPIB and CVI I still got the error always and only when exiting the  programs. When stepping through in debug mode line by line to the very end of the program I could not see the problem, but when using the Run button then on the end of the programs the error was back again.
I guess when upgrading to version 8.1 the problem may disappears. It probably just comes up when downgrading from the Evaluation version 8.1 to the Full Development version 6. again ?
Somewhere there must be an entry or file left which is still active and I could not find it so far.
During the week I don't have much time to try around with the Laptop, I usually work on problems like this during the weekends (if my wife is not killing me)  🙂

Best regards
Andy

---

There has to be something still present in the registry for this error to occur. This error is a generic corrupt registry error and therefore there has to be something left behind. Be sure to look through all pieces of your registry and find anything related to NI, National Instruments, LabWindows, CVI, GPIB or any other drivers you had installed. Everything needs to be removed since something in there is corrupt.

[mtnbikeAZ]

I'm getting errors similar to what Hårek and Andy have seen...

Debug Mode: Unknown source position, thread id 0x00000EC0 / The program has caused a 'General Protection' fault at 001B:00EC1474

Release Mode: The instruction at "0x00ec1474" referenced memory at "0x00b704a4". The memory could not be "read". / Click on OK to terminate the program

I'm having a hard time believing this is a corruption issue, as it didn't seem to solve the problem for those who have already posted. Meanwhile, I'm dealing with two different computers, each running a LabWindows/CVI program developed by a different person. The computers are new and have fresh CVI 7.1 installations that haven't been used for any other program or purposes other than to run these specific programs. Yet both programs experience the same error on their respective computers.

I've been troubleshooting this issue (and another issue) for several days. I know this may sadden some true programmers, but is it possible to change a setting such that these errors are simply ignored? My error also pops up when exiting the program and has no real effect on the program. I just want the error to disappear!

[menchar]

One way would be to "wrap" the invocation of the CVI app in a C++ or C# try/catch block and catch and swallow the GPF exception.

We do this when it's not worth tracking down some CVI or OS weirdness, though we haven't seen this particular fault,

If you're running .net though (e.g.C#), the .net runtime can intervene and it will catch the exception out of the CVI app and do its own popup

You can defeat this though and avoid all popups on exceptions coming up out of a CVI application.  Unfortunately you have to use some other, external framework to do it, there's no way to catch an exception directly in a CVI application that I know of.

Menchar

[WesleyD]

I am also getting a fatal run-time error when using CVI 8.0.1. 

" The application failed to initialize properly (0xc0000005)"

This error appeared on a new NI PXI computer with CVI 8.0.1 installed. Everything was working fine and today I powered up and none of the LabWindow programs run.

The error appears when I reference any device in the PXI Chassis from within the labwindows C code.  The function panels still work find and I am able to control the instruments. The above error pops up when these function call are in the code being compiled and linked.

Anyone have suggestions on a solution ??

Thank you

WesleyD

[Brandon_V]

Wesley,

What have you tried for this? Has anything changed since you had this working? Is there any other machine you can test this on or that shows the same behavior?

[WesleyD]

Hi Brandon

The problem is now solved. ( I hope). The computer had a virus called called SVOHOST.Exe.   It must of either been on one of the NI CD's , NI Hard Drive or through the internet when I registered the NI product.

The virus was on both of my NI PXI Embedded Controllers. 

My other two Laptop computers don't have it. AVG virus scan found and repaired it.

The computer is a stand alone with no other media hooked up to it. Not sure why everthing was working then on 7/7/07 it hit ?

The problem is fixed for now. Hope this can help someone else.

Thanks

Wes

[Marikay]

I am also having a similar problem.

I have a CVI application (compiled on 8.0)  installed on a computer that was running fine a week ago but now when select the application - nothing happens, the program doesn't start and there is no error message.

CVI was not installed on this computer, just the run time engine that is installed with the distribution package.

It has Windows XP SP2.

This application runs on other computers just fine.

I installed CVI 8.0 to try to trouble shoot the problem. When I attempt to run in debug mode, I get FATAL RUN-TIME ERROR: Unknown source position, thread id 0x000006D4: The program has caused a 'General Protection' fault at ...

I copied and compiled the test program  from earlier in the thread and it works just fine.

I've tried removing the program, CVI 8.0, rebooting, reinstalling with no effect.

Symantic found no viruses.

MSIBlast found nothing to uninstall.

Tried removing the registry components - no effect

Has CVI 8.5 corrected this?

Marikay