Is there a more concrete thing we can check here that is actually related to the crash? E.g. is it actually that the gamerules ptr is invalid between maps and sdktools should be clearing it out?

Is there a more concrete thing we can check here that is actually related to the crash? E.g. is it actually that the gamerules ptr is invalid between maps and sdktools should be clearing it out?

Looking in IDA I think that the gamerules pointer doesnt get set until CWorld::Precache is called (CWorld::Precache -> InstallGameRules -> CreateGameRulesObject), which I think only happens if a map is loaded, unless I'm misunderstanding what CWorld is.

You should also add the errors to the docstring in tf2.inc

Is there a stack from a Linux box or similar for this one? Asher is right if the ptr is invalid at a global level we should correct that instead.

Is there a stack from a Linux box or similar for this one? Asher is right if the ptr is invalid at a global level we should correct that instead.

https://crash.limetech.org/ib3armhzana5

I let the server start up (on itemtest),

then did

killserver
map itemtest

and it crashed.

TF2 uses the GameRulesProxy way to grab the g_pGameRules pointer instead of using the other way of a pointer to that pointer which probably gets cleared by the game on mapchange.

https://github.com/alliedmodders/sourcemod/blob/master/extensions/sdktools/vglobals.cpp#L131-L152

That s_pGameRules isn't set to nullptr on LevelShutdown, so ISDKTools::GetGamerules() returns the old value in between.
I guess the correct fix here is to set s_pGameRules to nullptr in LevelShutdown.

Thanks for the input. I'll create a new PR that fixes this issue in a more concrete way.