Awesome!

Bummer, because when mods do add their actions, it's impossible to anticipate what other mods add what type of actions. It's only possible to really anticipate this when staying within the same mod.

Can definitely reproduce and confirm freeze/desync/teleport issue when player changes stance while attached to vehicle (using LinkToLocalSpaceOf). I noticed that jumping/falling onto a vehicle while already crouched/prone does not cause the issue, it's only caused by changing stance.

Looks like it's fixed in 1.19 Experimental.

@Geez sorry for the ping, any update on this?

Work-around for comparison:

The problem is not infinite recursion.
The problem is that 1 < -2147483647 should NOT evaluate to true, ever (or any other positive non-zero number that's in-range for a 32-bit int).

Follow-up: Because all vanilla inherited cars call super in EOnPostSimulate, this also means they have double the fuel leakage than intended.

I've attached a minimal example modding a non-existent class. If this PBO is loaded with at least one other mod (this seems to be the important part), then the filename in the error message points to last line/EOF of some unrelated script.

I was half wrong actually, literally. There is only one NULL pointer key needed in the map for it to be not removable with map.Remove(NULL), as that would try to remove literal NULL from the map, not the key that originally pointed to an instance, but now points to NULL. Effect stays the same though.

And the underlying issue is this: https://feedback.bistudio.com/T166724

A temporary workaround until this is fixed properly is to override MissionServer::UpdateLogoutPlayers:

This bug can be fixed by removing ignorepointer=1 from alternative_bind and alternative_clear in gui\layouts\new_ui\options\keybindings_selectors\keybinding_option.layout, essentially reverting it to the (correctly functioning) 1.17 version.

IsInVehicle() started checking for parent transport in 1.16.

In T154093#2144684, @Helkhiana wrote:

With 1.10 update it was made actually again worse because I cannot override just a couple scripts, I have to override the whole ItemBase folder. This means any updates from DayZ in that folder I must update mine as well...

This should be enough for 1.10 and is what I'm currently using:

@Geez any ETA on when the current fix in Experimental will appear in stable?

No issue for me with the mods I run in Experimental. I would be very happy (well not only me I'm sure) if the current fix in Experimental would appear in stable soon.

@ukulelekid713 it's not related to VPP or any other single mod.

Latest Experimental looks like it did the trick. That's a huge relief. It allowed me to add the three mods back in that I previously had to deactivate, plus some more.

Thank you @Geez.

No change with latest Experimental for me unfortunately.

Moving code from 4_World into 3_Game definitely helps mitigate the problem, as does getting rid of any JsonFileLoader usage. What I also found helps is grouping all mods that use JsonFileLoader and load them last, after other mods. Hopefully the cause(s) for these freezes can be found and a proper fix be implemented.

You are wrong. Also, like I said, there's no mods making containers nestable to begin with.

Some mod objects was allowing someting like this: Container inside container attached to container inside container attched to container ...

The recursive weight calculation doesn't seem to work as intended. Fixable by modding: