Fatal Internal Error fpsane.cpp line 442 - can't find the insane object, even with Heep Peek window

[DavidZ.]

Starting from an old vi built in LV 8.5 i added a histogram, first in LV 2010 and now in LV 2010 SP1. In both I get in more and more increasing frequence the following fatal error message: Fatal Internal Error : "fpsane.cpp", line 442 . 

I already went through the forum and found these pages about the subject: fpsane.cpp, line 444 , read the labview wiki and knowledgebase article and found Darren's weekly nugget . So I already restarted from scratch the in LV 2010 added part of the vi, deleted and rebuilt the last objects after a crash, copied and pasted only the block diagram in a new vi, all without success. And now I try to use the Heap Peak.

More information about the vi: The newly added histogram is in a seperate pane, I use a splitter bar to be able the hide or make visible this new part, and keep the whole thing resizable. The vi is pretty big, so it's too much work to delete every old single element and find the insane object. So I tried out the Heap Peak window to find the object, which in the lvlog-file (file attached) is declared as "Insane object at FPHP+307C004, UID 10". But I just can't find this number in the upper right section of the Heap Peak window (file attached), as it is explained in the labview wiki article.

What am I doing wrong? And what can the information "line 442" in the error message help?

Thanks a lot for any help!

david

[HoussamK]

Hi

Did you try to delete the existing case structure from the old VI and replace it with a new one from the palette?

[benthescientist]

I can't offer any help, but I too have recently recieve this error when saving a subVI. The first time it usually saves fine. The second, it crashes.

Insane object at FPHP+23A3DC48, UID 10, in "Calc_Abb_Holo_Line_3.vi": {dsitem } (0x400): Panel (FPSC)

Like dazur, I too cannot find the corrupt item in Heap Peak. I did however, note that UID 10 seems to be the front panel itself (although I can not refind the section of Heap Peak where I saw this). Anyway, I decided that as i could not locate the problem, I would copy the block diagram over to a new blank VI and that maybe the copy would resolve the error. It didn't work, so I set about creating a new VI from scratch. Nothing but the text in a Matlab script node copied from the corrupt VI. It still didn't work. Even with a freshly created VI, the error still occurs.

I'm going to try recreating the main VIs controls that feed this subVI in case they are corrupting it somehow. I'll also change PC to see if the problem is limited to this installation.

[DavidZ.]

Hi hsm78, thanks for your idea. The old vi is huge, including plenty of case structures, subvis etc. So I was hoping to find an easier and faster solution with this heap peak window than completely replacing everything to find (the) insane object(s). But your idea reminds me, that in other posts about insane objects, all kind of elements in the block diagram could cause an insane object. I keep you up to date.   

[DavidZ.]

To Benthescientist, thanks a lot for your post! Good to know I'm not the only one with this kind of problem... 😞

In the meantime I keep getting this fatal internal errors fpsane.cpp, but I finally found now a suspect object, indicated this time with 307B2E4 (FPHP+307B2E4) in the lvlog-textfile (new actual lvlog-file attached below). This time I searched not only through the items listed in the top-right section when the FPHP of the main vi is selected in the top-left section, but I went down all the items in the top-left section of the heap peak window and at each item I searched then the listed elements in the top-right section for the element 307B2E4 . This way I found very surprised a vi of a labview library, called Details Display Dialog.vi .

I have no clue, what this means, but I don't think it's just coincidence, because heap peak showed me twice this same vi after two crashes. I wonder now, if this vi is causing some trouble, or if it's just an indicator for other problems.

I noticed also, automatic error handling was disabled in my main vi. So I enabled the automatic error handling, and then got several errors after starting the vi. I resolved the errors or disabled block panel elements where I couldn't see the cause (at connectors of the vis "Insert Menu Items" and "Set Menu Item Info"), and after this I could work a little bit longer on the main vi again. But after an hour it happens again, don't know why...

Any ideas?

[omas]

Hello

I get the error if I connect a void array to the X input of a General Histrogram VI

Hope it helps.

[csalwen]

I've been getting an insane object error on my vi and I couldn't narrow it down to a particular object.  I ended up figuring out that any time the vi was saved after having been run Labview would crash with an insane object error. This would happen even if the vi hadn't been changed at all.  It also would crash when it was being autosaved.

I spoke to support about it and they duplicated the problem and ended up telling me a couple of weeks later that it would be fixed in a future version.  I don't know if this has anything to do with other people's problems.  My work around is that I save and close a lot.  I make a change, save, close, open, run, close again without saving, etc.

[benthescientist]

That's exactly my problem too, csalwen.

At least they are going to fix it. Change, save, run, close, open, change is an incredible amount of effort to program!

[csalwen]

Not so bad once I figured it out.  I only crashed once today.    A fix would be nice, though.  I think it also depended on the Windows version.  If I remember right, when the guy was trying to duplicated it, he couldn't initially on Windows 7, but when he tried my vi on XP it crashed.  If you care I presume support could tell you.

[benthescientist]

My problem was on Windows 7 64 bit, although the program was first created on Windows XP 32 bit.