thepeacockproject
Game Crash when loading into missions
Platform: Steam and Epic OS: Windows 11
Getting game crashes intermittently (about ~10% of the time) when loading or unloading a level. Analyzed MANY crash dumps and almost all of them are failing here
ExceptionAddress: 000000014138a96c (HITMAN3!cat::ClientEasyHandshake::ProcessAnswerWithIdentity+0x000000000031289c)
I get this same crash on both the Steam(no mods) and Epic(SMF+ZHM SDK) versions of WoA, on the current version 3.230. Installed the game completely fresh on Steam, with no mods, and a fresh install of peacock 8.1.0 and nothing else, and still get these intermittent crashes. When I don't use peacock I don't get any game crashes. There doesn't seem to be any pattern as to which maps will crash when loading. Looks like something in the client handshake response is causing an exception.
You're getting this on an unmodified version of the game, with Peacock?
I'd be very surprised if this is a Peacock-caused crash. Looks a lot more like Windows or the game itself.
Yeah, seems like this is an SSL error - Peacock doesn't come with any certificates so it's likely your system trust store is messed up somehow, which we can't really fix.
It's weird that if it's an SSL error it wouldn't crash every time. I thought it was crashing because of mods / SDK so I tried a completely clean install from Steam, and the latest release of peacock 8.1.0 with no plugins, still get crashes on ~10% of level loads. When I play on official servers I'm not able to replicate the crashes.
I'm thinking it might be a timing issue where the game isn't expecting to get a response back so quick (the server wasn't intended to be local) and maybe hitting some code before something gets initialized. I'll try inserting some delays in a few functions and see if that affects anything
Thousands of people have successfully used Peacock locally and you're the only one to ever encounter this crash, so I doubt it's a game issue.
I've recreated this issue now on 2 separate windows machines with clean installs of Hitman WoA (steam) and Peacock 8.1.0. Doesn't crash on official servers, does on peacock. I will figure out what the problem is and create a PR once I've fixed it :) sorry to have bothered you with it
I've been getting random crashes during loading for several months. It doesn't occur when I have a vanilla Steam install on IOI servers without the SDK. The crashes start again when I use the SDK to use the Peacock server. The crash_metrics.dat and piepieonline's CrashDebugHelper.dll didn't show a reason for the game crashing. So I originally thought it was due to something in the SDK itself.
I've seen several other mentions of random crashes during loading on the Glacier Modding Discord, usually in the #sdk channel. There was another report of it yesterday in #peacock-chat when I linked this Issue.
Yeah I'd be very surprised if nobody else is seeing it. I just let it slide for almost a year because of how intermittent it was, figuring it was just me configuring mods wrong and causing crashes. But I finally got annoyed enough to narrow it down and it's happening on completely clean Hitman installs, only with Peacock running and not on IOI servers. Seems like it's usually happening after some save operation, I'm gonna create a debug branch tonight and see if I can figure it out.
I wrongly blamed the SDK for the crashes at first because some of the functions it crashes in appear in the SDK as well, but I still get them even when remove the SDK and use only the patcher.
I have started to get this issue as well on my system. This wasn't happening before the latest update so I am assuming this was introduced recently
Same offset
Only happens to me when loading into Freelancer sometimes
I've checked my logs, I seem to be getting it in Freelancer too. I'm 95% sure I've had it happen on official too though.
I have started to get this issue as well on my system. This wasn't happening before the latest update so I am assuming this was introduced recently
I started getting this crash in 3.210 and didn't have any crashes in versions before that
The functions identified in the stack trace are almost certainly not the functions causing the crash. The addresses are unrelated to the functions listed.