"fmodstudioL.pdb not loaded" crashes with Unreal Engine 5.1 and FMOD Studio 2.02.16

Unreal Engine
1

Hello!

I’m working with a client using FMOD Studio 2.02.16, Unreal Engine 5.1 and Plastic SCM (this might be relevant) and we’re getting strange crashes in-editor, usually when the same sound gets retriggered a couple of times in a player combo. I haven’t been able to reproduce it on my side and it’s pretty rare to happen but we have very little information to go with except that “fmodstudioL.pdb is not loaded”.

image (1)
image (1)1393×874 37.2 KB

We’ve let all binaries out of the ignore file in Plastic but this seems to cause other annoyances and not solve the issue so I was wondering what files are safe to keep in the ignore for the FMOD Plugin in Unreal?

And, lastly, we’re missing the debug symbols for this version, could you send that for me?

Thank you in advance!

2

Hi Bruno,

I have uploaded the debug symbols archive for FMOD 2.02.16 for UE5.1 to the uploads section of your fmod.com user profile.

I’ll hand over to Connor to help you with your remaining issues.

Kind regards,
Morgan

1 Like
3

Thank you so much, I now have updated information regarding the crash, still didn’t gave us much light but might help Connor to know what might be happening

image
image2560×1400 281 KB

1 Like
4

Hi,

Thank you for the screen shot, you are right it doesn’t give us too much to go on.

Would it be possible to get some reproduction steps that I could test on my side to try to replicate the issue? Or would it be possible to get a stripped-out version of the UE and FMOD studio project uploaded to your FMOD profile?

Thanks

1 Like
5

Ok so, after checking in the debug information and thinking a bit about what might have caused it, I think I found it.
Basically, we have a system in place that can stop even instances when certain character moves get interrupted, I believe the crashes were being caused because the system would target an instance that had either been destroyed already or was already stopped.
We added a step to check the state of the instance before sending the stop command and that seems to do the trick, we’re not 100% sure though cause this crash was really hard to reproduce but our current method of testing it didn’t return anything.
We’ll keep testing here for a while to see if we get any new crashes and if not I’ll come back here to mark this as solution!

Thank you so much regardless, Connor, if anything happens I’ll sure to send the files to you!

1 Like
6

Hi,

Good to hear you might have found the solution. Would it be possible to get a crash dump? There may be systems that we could put in place to stop this from happening in the future but we would need to confirm it is an invalid instance issue.

Thank you for bringing this to our attention.