Broken code with no broken arrow.

[Michael Aivaliotis]

Not sure what causes this but I've experienced this several times lately.

I run a toplevel VI several times and somewhere I get an error on the error cluster. So I decide to start debugging by opening up the subVI's. When I open one of the subVI's I notice that it has a broken run arrow and now I also notice that my toplevel VI has a broken run arrow. How is it possible for LabVIEW to run broken code? The only thing I can think of is that the broken code has not propogated to the current running application instance until I open it up. Not sure why however since I only have one instance. I have the project environment open too.

[jdunham]

Do you have a conditional disable structure on the diagram or in a subvi? If you have changed the conditional symbol definition in your project, then the same VI could definitely be working inside the project but broken outside of it.

[Michael Aivaliotis]

QUOTE(jdunham @ Nov 15 2007, 05:42 PM)

Do you have a conditional disable structure on the diagram or in a subvi? If you have changed the conditional symbol definition in your project, then the same VI could definitely be working inside the project but broken outside of it.

No, it's not that.

[silmaril]

Did you try Ctrl-Shift-Run on your top level VI?

Was the run arrow still unbroken after this?

[Michael Aivaliotis]

Well, it happened again. Check this out. There's a broken wire on the diagram but the arrow's intact.

Running the VI crashes LabVIEW. :thumbdown:

[Michael Aivaliotis]

Well, it happened again. Check this out. There's a broken wire on the diagram but the arrow's intact.

Running the VI crashes LabVIEW. :thumbdown:

[Val Brown]

Are you working in XP or Vista?

[Val Brown]

Are you working in XP or Vista?

[Michael Aivaliotis]

Here's a video that shows how you can have a broken VI on the diagram of a non-broken VI. No I'm not using a disable structure. If I click on the run arrow of the caller it breaks and shows the error list. The arrow then stays broken.

[Aristos Queue]

You have filed a CAR, right?

[LAVA 1.0 Content]

QUOTE (Aristos Queue @ Aug 20 2008, 02:42 AM)

You have filed a CAR, right?

Have you ever tried to file a CAr without being able to reproduce the issue on demand?

I have seen a couple of those myself, also working LV 8.5. I'm not using globals so if its the same bug I have seen it is not due them. THe other observation that I can add is that a really fuzzy general type observation. I would swear (if pressed) that code changes are not propagating to all of the VI's that are not open. I say this because I work with a tree.VI open and after amking changes (but before running) I save everything (from tree.vi save all) then close the tree.vi. Backup then re-open and test the code. WHen closing the tree.vi, I see VI's that want to be saved. As a result I ahve resorted to an old habit of opening and closing the tree a couple of times to make sure all of the changes get propogated.

Just reporting what I have ben seing.

Ben

[hfettig]

Hey Mike,

I think I found your Broken Arrow. Somehow it found its way into my VI.

I seem to be experiencing the opposite of your problem. I have a broken arrow and 0 errors and warnings listed in the error list.

Cheers,

Heiko

[Michael Aivaliotis]

QUOTE (neB @ Aug 20 2008, 05:36 AM)

Have you ever tried to file a CAr without being able to reproduce the issue on demand?

I agree. Reporting this to NI is a futile exercise. Not reproducible at all.

QUOTE (neB @ Aug 20 2008, 05:36 AM)

I would swear (if pressed) that code changes are not propagating to all of the VI's that are not open. I say this because I work with a tree.VI open and after amking changes (but before running) I save everything (from tree.vi save all) then close the tree.vi. Backup then re-open and test the code. WHen closing the tree.vi, I see VI's that want to be saved. As a result I ahve resorted to an old habit of opening and closing the tree a couple of times to make sure all of the changes get propogated.

Ben, there's a thread for this issue here.