Internal error: Too many points in plot

[msavina]

Labview keeps crashing and upon relaunch I get a message saying "... an internal error or crash occurred at plotsupp.cpp, line 4049"

The failure log generated is:

####
#Date: Thu, Sep 13, 2007 3:37:07 PM
#OSName: Windows NT
#OSVers: 5.1
#AppName: LabVIEW
#Version: 8.5
#AppKind: FDS
#AppModDate: 07/25/2007 00:07 GMT

c:\builds\penguin\labview\branches\Jupiter\dev\source\panel\plotsupp.cpp(4049) : DWarn: too many points in plot
$Id: //labview/branches/Jupiter/dev/source/panel/plotsupp.cpp#52 $
0x007D3BDF - LabVIEW <unknown> + 0
0x007D60B2 - LabVIEW <unknown> + 0
0x0130124B - LabVIEW <unknown> + 0
0x011DE932 - LabVIEW <unknown> + 0
0x0123B148 - LabVIEW <unknown> + 0
0x011DE932 - LabVIEW <unknown> + 0
0x0067001A - LabVIEW <unknown> + 0
0x007C379B - LabVIEW <unknown> + 0
0x00B185C9 - LabVIEW <unknown> + 0
0x00B188DA - LabVIEW <unknown> + 0
0x00B18C4D - LabVIEW <unknown> + 0
0x005E1855 - LabVIEW <unknown> + 0
0x0049C139 - LabVIEW <unknown> + 0
0x0049A981 - LabVIEW <unknown> + 0
0x0049C1EC - LabVIEW <unknown> + 0
0x00405678 - LabVIEW <unknown> + 0
0x0040599B - LabVIEW <unknown> + 0
0x0045CFF6 - LabVIEW <unknown> + 0
0x00505D85 - LabVIEW <unknown> + 0
0x00505FAB - LabVIEW <unknown> + 0
0x00505FF0 - LabVIEW <unknown> + 0
0x0149415F - LabVIEW <unknown> + 0
0x7C816FD7 - kernel32 <unknown> + 0
0x00000000 - LabVIEW <unknown> + 0

[Yi Y]

Hi msavina,
Could you please attach the original LabVIEW crash log as a txt file? This doesn't seem like a reported bug yet.  Is there a specific task you do that reproduces this behavior? Maybe a specific function that's causing LabVIEW to crash? Any information that would help us reproduce this error would be greatly helpful.
Thank you for reporting this bug, and once I have the error log file, I will report it to R&D for further investigation.

Yi Y.
Applications Engineer
National Instruments
http://www.ni.com/support

[msavina]

The crash log is attached.

The attached  sZLS ECG6 simplified.vi  communicates over RS232 with ECG monitoring hardware. There are 3 tab views in the main VI, which by default displays the  Heartrate  tab upon startup. The crash seems to happen when attempting to view the  ECG Chart  tab. The  ecgeProcessPackets.vi  subVI is also attached - this is where the original  ECG Chart  ouput resides.

Also, I recently updated to LV 8.5 and I was able to run this VI in both LV 8.2 and LV 8.0 several times without any crashes.

[F._Schubert]

I had problems with plotting huge data sets in Tabs as well, see:
http://forums.ni.com/ni/board/message?board.id=170&message.id=271044&query.id=195534#M271044
it was 'just' performance issues (I got rid of them because I don't need that many points), but with the profile manager I found out, that switching from a tab takes twice the time than the draw operation alone. And half of it made of 'Tracking'. On an old PC I had a lot of performace issues using intensity charts in tabs as well as nested clusters. I guess, it is a draw/update problem, where the underlying ('overpainted') controls must be calculated as well...

Felix

[Nick R]

So does LabVIEW actually crash as in dissappear in the middle of your program running? This crash log is just for a dwarn, which is not nearly as significant as a dabort.

Also, there are still several VIs missing from your program, so there is no way I can try to reproduce this problem. I can submit this crash log and your VIs to R&D to look at, but without being able to reproduce the problem there is normally not much they can do.

[msavina]

Yes, while the program is running and consistently it seems to happen shortly after selecting the  "ECG Chart" tab, I get the Microsoft error message (Error1.png attached) and after acknowledging the error Labview goes bye bye.

Upon restarting Labview I get the "previous internal error" message (Error2.png attached) and the error log file is generated (lvlog10-04-07-13-48-17.txt attached).

I don't know if it will help since you probably can't run the VI's without the ECG monitoring hardware (a device manufactured by Active Corp. communicating with my PC serially via a COM port), but I will go ahead and attach all of the related subVI's in the ensuing post.

msavina

[msavina]

The main VI is "sZLS ECG6 simplified".

[Yi Y]

Hi msavina,
Thanks for the attached files.  Like you said, since I don't have the devices, I cannot test the VI.  I did some stress test plotting 10000000 data points onto two charts per iteration, and there doesn't seem to be any issue other than the slow-ness caused by large memory usage.  Have you tried to run this program on a different computer with LabVIEW 8.5 installed? Does the same crash happen?  Is there any way we can remove the hardware involved in the code and use simulated data to reproduce the same issue?
I can report the issue to R&D, but it's going to be more difficult for them to figure out the issue without it being reproducible.  I'm trying to gather as much information as possible so this can be adequately addressed.  The attached file is the stress test I used.  Does this cause a crash on your computer?

Thanks!

Yi Y.
Applications Engineer
National Instruments
http://www.ni.com/support

[JeffreyH]

Hi there,

I've got this same error using the attached test VI.. (On a side-note, this VI was originally created in 8.2.1 and didn't have the problem there.)
On my system, it will crash LV85 after two sweeps of the chart.

I hope this helps finding (and fixing) the problem..

Regards,
Jeffrey Habets

[Yi Y]

Hi Jeffrey,
Thank you for the VI submission, this should be a big help for reproducing the issue. I have attached the VI to the bug report, and this bug is being looked into by R&D. For your reference, the bug report ID is 4D5LG45D.

Yi Y.
Applications Engineer
National Instruments
http://www.ni.com/support