LabVIEW: End of file encountered. While saving a freshly opened 8.6.1 vi

[SigRob]

I recieved an error today while trying to migrate a LV 8.6.1 program into LV2009.  The program opens sucessfully, and shows in the title bar with an asterisk showing that there has been changes.  If I attempt to save the VI, the system hangs for a moment then gives the errors that I have attached below.  These errors do not occur in LV 8.6.1.

I cannot post the entire VI for analysis, but I believe I have found the culprit.  Attached, I also have a small demo VI that merely has 4 strip waveform charts.  The VI was created in LV version 8.6, and on every LV2009 computer I have available it will crash when trying to save.  The computer I have been using is a 3.2 Dual-core with 4GB of RAM, running windows XP SP3.

I realize the VI is quite large, and i'm not sure why that is.  These waveform charts were copied from my application VI, then I cleared the data on them and reset the history to 10 samples.

I'm eager to know if this can be reproduced, and if there is something I would be able to do for a workaround. Thanks.

 

[RavensFan]

Someone else had a similar problem in this thread.  Similar symptoms, but sounds like a different cause.

Having 15 MB on disk and 77MB of data in memory for 5 empty graphs is pretty suspicious.  When I deleted the graphs from the VI, the memory usage went almost to zero and I was able to save successfully.

Perhaps you can try deleting the graphs out of your 8.6 version of the VI, open in LV 2009, save it, then recreate the graphs from scratch.

[SigRob]

I was able to save when the graphs were deleted as well, that might get me by, but that solution doesn't really satisfy me.  Why are the graphs using the memory in the first place?  Is this an actual bug or is there something I have done incorrectly?

[RavensFan]

I don't know.  I would suspect some sort of bug.  I think someone at NI would have to look into the VI that you posted and figure that out be looking at the binary data stored in that 15 MB, (and in memory.)  Perhaps you can post the LV 8.6 version of that file as well so they have something to compare it to.

[SigRob]

That file is the 8.6 version, since i would have had to delete the graphs in order to save it in LV2009

[RavensFan]

---

SigRob wrote:
That file is the 8.6 version, since i would have had to delete the graphs in order to save it in LV2009

---

Of course.  You're right.  I was thinking LV2009 because that was what I opened it with to see the 77 MB of objects and to see the crash.  So I was think that was the "after" problem file, and I want to see the "before" file for comparison.

Of course the "before" is opening that file in LV 8.6 and seeing its properties, which I didn't do earlier.  Interestingly enough, I get the same sort of memory numbers.  77.7 MB of data.  And I was able to save it with LV 8.6.  It took a long time, but it worked.

So there are really two questions here.

1.  Why does a VI with very little in it take up 77 MB worth of data in memory?  What is it and how do you clear it out?

2.  Why is LV 8.6 able to save that file, but LV 2009 cannot?

When I delete all but the first graph, I get 3 MB on file and 15 MB of data in memory.  Pretty much 1/5 of the original in both cases.  But LV 2009 was able to save that.

[Vivek_N]

I just wanted to let you all know that I will look into this further and post on this thread again. Thank you for the test VI and details.

[SigRob]

Thanks Vivek.

[Vivek_N]

Hello,

In order to figure out why this is happening, I want to be able to re-create the issue on my end. I can see the problem with the VI that you have attached here, however I cannot recreate a VI with 5 waveform charts in LabVIEW 8.6.1 which behaves the same way when saved in LabVIEW 2009. 

Have you been able to do this with a newly created VI? If so, please let me know what steps you followed to create the 5 charts.  

[SigRob]

Vivek,

The 5 charts that i have supplied were copied + pasted into a new VI from a portion of code written in one of our applications.  In my own tinkering i found that those VI's were the culprit in not letting me save my application VI within LV 2009.  The application basically does SQ calculations and pushes the resulting value into one of those charts.  The charts were originally set up to display the past 1024 points each, and the VI was run within a LV 8.6.1 development system for long periods of time.

Hopefully this information is of some help.