LabVIEW 2010 crashes often when deleting multiple diagram objects

[altenbach]

LabVIEW 2010f2 has been reasonably stable for me so far, but I seem to get crashes to the desktop exactly under the following conditions:

• When deleting larger sections of code (=several objects selected with the mouse, typically including bad wires in the selection, followed by hitting the delete key).

When this happens, LabVIEW becomes unresponsive for about 10 seconds, then crashes to the desktop. This scenario is quite reproducible in a statistically significant way. While it is not 100% reproducible, I've had it happen a couple of times in a row on the same restored VI, doing basically the same delete procedure. It happened in quite a few unrelated VIs.

This is LabVIEW 2010f2 32bit on Windows 7 64bit. There is nothing in the failure log.

I wonder if anybody else has seen similar correlations.... Thanks!

[altenbach]

Maybe related, maybe not (!), just playing and trying to generate the above mentioned crashes a few more times.

This time I was editing the VI posted here.

Randomly grabbed and delete a few code sections and it did not crash (this time!), so I held down ctrl+z to undo to the beginning of the try. While undoing, LabVIEW crashed to the desktop and this time I got a failure log, see attached.

Things like:

DWarn: How could we have read more bytes than the size of the thrall map?

DWarn: FPDCO::LoadObjectData unexpected TDOffset tag in VI?

(just quoted here in case somebody searches for these strings later...)

What is the world is a thrall map???

[JackDunaway]

Cannot reproduce with Win 7 64bit, LV2010 64 bit without f2 patch.

[altenbach]

Thanks Jack,

Good to have positive and negative controls. 😄

Since this is happening to me on a regular basis, I'm hoping that it's not only me, but maybe it is.....

Yesterday, I was trying to record a jing movie of the crashing. Unfortunately, it did not crash while recording, but crashed nearly consistently when doing the same things while not recording screen activity..... grrrr..

[Matt_W1]

I've had the same crash several times as well.

[Ben_J1]

Altenbach,

   I'll be looking forward to a jing or reproducable code of the issue.

[JackDunaway]

---

Ben J wrote:

I'll be looking forward to a jing or reproducable code of the issue.

---

Have you tried the VI in altenbach's second comment: "This time I was editing the VI posted here."

[Ben_J1]

Jack,

   I apologize. I tried it and could not reproduce a crash. I have LV2010.f2 32-bit and Windows 7 32 bit.

[altenbach]

---

Ben J wrote:

   I'll be looking forward to a jing or reproducable code of the issue.

---

OK, got one!

Open the attached file in 2010f2. (from this discussion. All that padding and re-stripping of the 2D array is not necessary, so we try to clean up the code a bit. ;))

Jing file is here.

We delete a few things in the right, then on the left, then on the bottom left.

At 7 seconds, LabVIEW becomes unresponsive. At 12 seconds, it crashes to the desktop.

(I was able to record this two times in a row!)

[RavensFan]

I can confirm the behavior on this VI.  Differences for me were that I could get the matrix size function to the lower left to delete, then it locked up.  It never crashed to the desktop.  After several minutes, it is still sitting there dimmed out and "(Not Responding)".

The first attempt I tried, I accidentally deleted the Replace Array Subset to the center left that was definitely unresponsive in  Altenbach's, then went to the Matrix Size and was able to delete that.  The VI did not crash.

I watched the video again to be sure I deleted the same items in the same order.  Since it did not crash all the way to the desktop, I had to use the Windows dialog to force LabVIEW to exit.  ("Windows looking for solution", "blah blah blah", then to the desktop.  Vista by the way.)  When I restarted LabVIEW, I got the LabVIEW "there was an error message" and something about setstate, I didn't note the line number.  It saved this error log that is attached.

What is interesting, is that there are a lot of notes from several days ago in it (Oct 28) about VI's I've had open, but have been closed down a long time ago.  I probably have not shutdown and restarted the LabVIEW environment for several days.