trying to slice a duodecahedron and slicing reliably fails on the stl.

Ultimaker / Cura

3D printer / slicing GUI built on top of the Uranium framework

GNU Lesser General Public License v3.0

6.2k stars 2.08k forks source link

trying to slice a duodecahedron and slicing reliably fails on the stl. #19780

Closed interrupt80 closed 3 weeks ago

interrupt80 commented 4 weeks ago

Cura Version

5.7.1

Operating System

Windows 10 Pro/ NVIDIA GeForce GT 730

Printer

Creality ender-3 S1

Name abnormal settings

standard profile(TPU) work, moving to gyroid and 5 % infill works, wall to .4 from .8 causes crash

Describe model location

center

Describe your model

Add your .zip here ⬇️

CE3S1_dodecahedron_crash.zip original modeL: https://www.thingiverse.com/thing:198

GregValiant commented 3 weeks ago

Thanks for the report. I can't duplicate the problem. The project slices for me on my Win10Pro laptop.

Locate your Configuration Folder ("Help | Show Configuration Folder") and the cura.log file. In Cura, set up the slice again and immediately after it fails - zip the cural.log file and post it here. (It must be posted directly and not via email.)

interrupt80 commented 3 weeks ago

cura.zip

interrupt80 commented 3 weeks ago

I have a path forward, I just like getting bugs to go away.

GregValiant commented 3 weeks ago

From your log file. "Backend exited abnormally with return code 3221226505!"

There were several bug reports that mentioned that code. @HellAholic and the Cura team have attempted to fix this in newer releases. You can try running 5.7.1 "As Administrator" or maybe upgrading to Cura 5.8.x.

I'll mark this as a duplicate and close it as it appears this was fixed. If the problem still exists in 5.8.x then let us know and this can be revisited.

HellAholic commented 3 weeks ago

Thanks for the report and the heads up. With the sentry implementation we get the crash reports directly from engine anonymously if you're connected to internet when the crash happens and if you have the options selected in preferences to share the crash data, the code 3221226505 is just generic "engine said no" response. I'm pushing for some more focus on addressing the slice fail issues, 🤞

Also to clarify a bit about what I mean with the crash data, it tells us that this function had an issue, division by 0 or expected something that wasn't there, mostly developer language, nothing regarding the model or project file. So it is always nice to have the models that make the engine crash at hand and double check if the fix is effective.

Cheers!

GregValiant commented 3 weeks ago

Since this occurred with 5.7.1 is this an ongoing issue? If it is then I'll re-open this ticket.

HellAholic commented 3 weeks ago

Can't say for sure, since I can't reproduce it either, so we have no way of checking if it is resolved or not. Main part was getting the model, I'll add it to the automated slicing test so hopefully if there is a crash or regression on the model it will catch it. TLDR: It's fine!

interrupt80 commented 3 weeks ago

Can't say for sure, since I can't reproduce it either, so we have no way of checking if it is resolved or not. Main part was getting the model, I'll add it to the automated slicing test so hopefully if there is a crash or regression on the model it will catch it. TLDR: It's fine!

5.8.1 resolves it for me