Crash to desktop switching to manufacturing workspace

[BobVawter]

I have a design which reliably causes a crash-to-desktop on both Mac and Windows installations of Fusion due to what looks like a stack overflow when switching to the manufacture workspace.  It crashes so hard the normal error-reporting framework doesn't fire, and I'm left with the "Report to Apple" stack dump (attached).

I've tried exporting an f3z archive and re-importing, which also triggers the crash.

Is there a way through the scripting API to delete the setups or entirely reset the CAM workspace?  This particular file has several fixtures that I'm hoping to not to have to rebuild; the CAM operations would be straightforward to recreate from an earlier version of the design which does not crash.

Thank you.

[karina.harper]

Hi @BobVawter 

Can you attach the file here? There is a way to remove the CAM data if you send over the .f3z. How is your local memory? Do you have a good amount of free space?

Cheers,

Karina

[BobVawter]

@karina.harper I wasn't able to get permission from my customer to send the file out.  However, I spent some time poking at the f3z and enclosed f3d files to see that they're just zip archives.  I was able to drill in and remove the `Toolpath/` directory in the design, which achieved the desired effect of resetting the Manufacturing workspace.

In poking around some more, I've been able to create a minimal reproduction case:

* Create a Setup1

* Add a Manual NC Comment

* Add a 3d adaptive clearing operation

* Create a new Setup2, with "use stock from previous operation" and just reverse the Z axis

* Add a Manual NC Comment to Setup2

* Add a 3d Adaptive to Setup2

* Add a Manual NC Comment to Setup2

* It crashes here 2/3rds of the time.

* Otherwise, re-order the second manual NC above the adaptive clearing op

* It hasn't failed to crash here.

This reproduces the stack overflow in libIronCore10.dylib that's attached in my initial post.  See attached screenshot from right before a crash adding the third Manual NC Comment.

[karina.harper · Accepted solution]

Hi @BobVawter 

Yep, I was able to reproduce this crash. I'll log it with the team so they can have a look!

Thanks for doing the legwork here 🙂

Cheers,

Karina