Fatal Assertion failed

[Kobi_K]

Hi all ,

I'm having a problem with my CVI software ,

I'm using CVI 9.1 under win XP with NI USB 6536.

The software basically generates 2 AO with 3 DO and at the same time Acquiring 8 AI simultaneously.

The software worked fine for the past 5 days but yesterday I had 5 fatal crashes , this is the msg from the windows event viewer :

The description for Event ID 0 from source CVI cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.
If the event originated on another computer, the display information had to be saved with the event.
The following information was included with the event: 
[17:08:20.770] [FATAL] [DoTrack] [..\lvsource\transact.c:1579] Assertion failed: lvData->doTrackNestingLevel <= 2 (Dangerous tracking nesting level; information lost.)

Maybe some one knows why it happens ?

[LuisG]

1. Can you let us know the exact size and version of c:\windows\system32\cvirte.dll

2. Can you check to see whether you have the following registry key, and if so, let us know what the various values under that key is:

HKEY_CURRENT_USER\Software\National Instruments\CVI Run-time Engine\9.1\Logger

Thanks,

Luis

[Tomsaw]

Hi there !

Are there any news regarding the above mentioned problem ? I've got an event log (windows application log) from a customer containig the same errors.

The application is running most of the time without any problems but sometimes it crashes. Many events appear in the log then (one case e.g. about 40). They all are similar to this:

Ereignistyp: Fehler Ereignisquelle: CVI Ereigniskategorie: Keine Ereigniskennung: 0 Datum:  03.04.2012 Zeit:  16:16:20 Benutzer:  Nicht zutreffend Computer: 612971-001 Beschreibung: Die Beschreibung der Ereigniskennung ( 0 ) in ( CVI ) wurde nicht gefunden. Der lokale Computer verfügt nicht über die zum Anzeigen der Meldungen von einem Remotecomputer erforderlichen Registrierungsinformationen oder DLL-Meldungsdateien. Möglicherweise müssen Sie das Flag /AUXSOURCE= zum Ermitteln der Beschreibung verwenden. Weitere Informationen stehen in Hilfe und Support. Ereignisinformationen: [16:16:20.421] [FATAL] [WSize] [..\mgsource\window2.c:2589] Assertion failed: 0 (Region creation failed.).

Only the last lines differ - here some examples:

Ereignisinformationen: [16:16:46.890] [FATAL] [IClose] [..\mgsource\image.c:8833] Assertion failed: WCurWind() (No wind in IClose).

...

Ereignisinformationen: [16:16:46.890] [FATAL] [DSelectNormPen] [..\mgsource\drawmgr.c:2856] Assertion failed: FALSE (bad Pen).

...

Ereignisinformationen: [16:16:46.890] [FATAL] [GetReady] [..\mgsource\drawmgr.c:2416] Assertion failed: FALSE (no HDC).

...

Ereignisinformationen: [16:16:46.890] [FATAL] [DSelectNormPen] [..\mgsource\drawmgr.c:2803] Assertion failed: SelectObject (mgrData->gPort.hdc, gStockPen) (selecting stock pen while printing).

It seems to occur when multiple panels and child windows are being loaded in the application.

The used CVI runtime on the customers machine is 9.1.1.450.

A cause might be that I mixed up runtime version - compiled the application using CVI 2010 on my dev. machine and applied it to the cutomers machine with CVI 9.1.1.450. - but I'm not sure. What do you think ?

[Kobi_K]

Hi,

The truth is that I really don't know until know what was the problem,

After 1 Restart the problem didn't come back so I didn't had any chance to examine it ,

I can tell you that in my program there was only 1 Panel.... So if it's your idea maybe you can check it but wait for some better advice before starting to "dig"

If any thing will come up I will update....

[Tomsaw]

There were other posts describing that the cause is probably the usage of Delay and multiple threads and that Delay has to be replaced by Sleep. But it was said the problem with Delay was fixed after CVI 9.0. As I'm runnning CVI 9.1.1.450 it seems to me the problem isn't fixed yet ...

[LuisG]

The problem with the Delay function was indeed fixed in version 2009 (9.1). So, if you're still having problems with 2009, then it must be something else.

If anyone has some code that can reproduce any of these problems, feel free to submit it and we'll take a look at it as soon as possible.

Luis

[Tomsaw]

Hi there !

Anyone ever heard of a similar issues again ?

I'm still facing this problem with a software running on my customers machines. All my attempts to reproduce the problem failed, but from time to time it happens at my customers'.

A slightly older version of the software is running with CVI 8.5 and there's no problem. On the machines having CVI 2009 SP1 sometimes appear lots of errors in windows event viewer.

[02:54:24.696] [FATAL] [DDestroyRgn] [..\mgsource\drawmgr.c:2112] Assertion failed: DeleteObject((HRGN)rgn)

[02:54:24.696] [FATAL] [DDestroyRgn] [..\mgsource\drawmgr.c:2107] Assertion failed: rgn

[02:54:24.586] [FATAL] [WVisibleRgn] [..\mgsource\window2.c:1761] Assertion failed: 0 (Region creation failed.)

I suppose it happens when drawing / deleting controls on the screen.

After a longer period of time - thinking the issue was solved (cleared some ressource leaks in my code before) it shows, the problem still persists.

[Alex+ra]

Hi Tomsaw,

Does the error show up in debug or the released executable?

Is it happening on multiple machines or just one?

Has the customer tried running the code on another machine?

Is the CVI 8.5 code on the same machine?

If there is one machine with CVI 8.5 and one with CVI 2009, what is the difference between the two machines in terms of age, operating system, hardware, etc?

What toolkits or module are you using in your code?

How long does the code run?

Is there any consistency to the errors, like time elapsed, certain function calls, first run after restart, etc?

Check out this forum with a similar issue: http://forums.ni.com/t5/LabWindows-CVI/Application-crash/td-p/1001743/page/3?view=by_date_ascending

Can you attach a copy of your code (with any depencies)? Also, can your customer generate a crash dump that we can look through? Crash dump instructions here: https://decibel.ni.com/content/people/nickb/blog/2009/06/10/generating-crash-dumps-for-analysis-by-n...

Other users with the same issue have had success with reverting code to a previous version of CVI. If that's possible, it may be the quickest fix, instead of trying to capture the intermittant error.

Regards,

Alexandra