Labview: Null Window ... stopped at unknown on the block diagram

[Jimmej]

Hi,

I have a fairly large program that was written in Labview 6.0. Recently we upgraded to Labview 9.0 and this program was exported to an executable with the new version.

When running the executable I sometimes get the following error as seemingly random places. I have not been able to make it crash when not using the executable. Labview: Null Window. The top-level VI ".vi" stopped at unknown on the block diagram of ".vi"

 

I apologise for having to blank out the vi names but i cannot give vi names, say what the software is used for or give over code. I can say that it uses NI USB 6212 IO cards, interfaces with Rhode and Swartz instruments and uses alot of file IO.

I was hoping that someone could give me a hint on how to debug such an error. I have noticed that When this error occurs, the memory use almost doubles.

Thankyou

James 

[MikeS81]

Hi James,

did you try it with the debug functionallity of applications? Do you call vi's dynamically?

Mike

[TCPlomp]

I think you open the front panel of a VI that has it's front panel removed during the build. Try to adjust the built settings of that specific VI to include the front panel.

Ton

[AdnanZ]

This link will help:

Why Do I Get a "LabVIEW: Null Window" Error When Running My Executable?

[Jimmej]

Thanks for the replies!

I have gone through and unticked all the boxes to remove the front panels. I thought we were on a winner but it is still giving this error.

It appears to happen at the same place. This program has a few different operations run from buttons on a front panel. Occasionally it crashes running one of these buttons. It is the first VI called that has a front panel enabled. Most of the time is runs fine.

Any more thoughts?

Thanks

James 

[Jimmej]

Ok, I have been looking at this code all day now. I have narrowed down what is happening.

As i said before there are two buttons. The first collects data and saves it to files. The second loads from these files and  displays the data. Read and Write Spreadsheet file is used for the file IO. Both buttons address the same VI. This VI either saves or loads depending on the input.

If data is collected (button 1) and then read (button 2), it will crash at the first VI with a front panel enabled giving the error stated above.

If data is collected (button 1) then the application is exited and restarted. Then the data is read (button 2), it will work without an error.

As stated earlier, when it crashes the memory in use doubles.

Thanks for any input. 

Regards

James 

[Ipshita_C]

Hi James,

Do you get an option to "investigate error" when you reopen LabVIEW? You can go through those steps so that the error log is sent to National Instruments and we can take a closer look at the reasons for LabVIEW crashing. Hope this helps!

Ipshita C.

[Jimmej]

Hi,

No there is no option to investigate error.

More experimentation has shown that it never crashes when run from the VI.

It is only crashing from the executable.

I believe I have made sure all front panels etc are included but now that i have narrowed it down to an executable problem, i will experiment with a few different options.

Cheers

James 

[Jimmej]

Is there something I am missing here? 

[Jimmej]

Hi,

More fidling.

If i enable Reentrant Execution on the VI that comes up with the error, the error moves on to the next VI with a Front Panel enabled.

So it looks like a partial solution, if i enable this on every VI with a front panel.

Is there something i am missing here? This shouldn't be required?