Non-Fatal Run-time Error "%%s"

[CVI-User]

This was shown yesterday by an non-debug program written with CVI 9.01 and run-time 9.1 (the current "428" build) on the target.

I was trying to fix "-12: Out of memory" popups (maybe a race condition inside ProcessSystemEvents/GetUserEvent?) and now this.

Any idea what this is?

[muks]

How do you prevent non-fatal run-time errors from being displayed?

[CVI-User]

Thanks. I know. But what is CVI trying to tell me in the first place?

[Daniel Coelho]

Hi.

Did you find an answer to your question?

I'm getting the same error on a DLL of mine.

I can't see what is wrong and I would appreciate some tips on this error.

Thanks.

[nickb]

Hey Daniel -

I've had others reporting this recently, but I've been unable to track it down or reproduce it.  Is there any way you could post some code that reproduces this error?  If you'd rather not post it to this forum, you can post it to our write-only ftp server at ftp://ftp.ni.com/incoming

NickB

National Instruments

[nickb]

I just wanted to give an update that I was able to reproduce the "%%s" dialog a little while ago.  You no longer need to try to put together something that reproduces it for us.  We're debugging the issue now, and I will post back when we have more information.

NickB

National Instruments

[nickb]

A couple updates:

From all indications, this dialog should only be shown if your binary (exe) is built in release mode.  If you build in debug mode, you should be able to see the error message that is trying to display.  Also, if our preliminary investigation is correct, the error should be in your release mode exe, and not any loaded dlls.

From all indications, this dialog should only be shown if your binary (exe) is built in release mode and loads dlls that are build in debug mode.  If you build your dependencies in release mode, no dialog should be displayed at all (because this is a non-fatal error).

Are you able to verify these statements?

Thanks -

NickB
National Instruments

[Daniel Coelho]

I verified that using TestStand to load a DLL in release and I get "%%s". The same happens with an EXE that loads the DLL in release.

If I compile the DLL in debug mode, I'm able to see the actual error.

[kplafa]

I've found that if everything is built in debug mode or everything is built in release mode, then the error message is displayed correctly. Hwoever it still stops on some non-fatal errors in release mode. At least I can see the error and fix it to prevent it happening again.