Linux + SOG map + Domination

This is an exciting opportunity to squash that bug. :)

I haven't been able to craft a repro for this. I only see it when I run the full mission (Domination) on a dedicated server.

I tested the same PBO on Windows and Linux. The bug is easily reproduced on the Linux server. I can't reproduce the bug with Windows server.

Has anyone seen this bug running a Windows server? I have only seen it on Linux servers.

The hunchback bug also occurs with civilians. These are spawned in the mission by either 1) civilian module or 2) with createAgent. Very little scripting otherwise, just a couple of event handlers (killed, firedNear) attached to the civilian unit.

This occurs with both the 32 and 64 bit Linux server.

There are no server-side or client-side mods. The mission uses a revive module. Not sure if that is vanilla?

BIS needs to fix this. This breaks some very old missions.

This is the poison animation. This is the logged animationState when the bug occurs.

Newly created units are spawning in a "reaching down" animation and then staying in that position forever. They are able to fire their weapon but cannot move.

After the CDLC update I was able to run a large Domination mission overnight. Looks fixed?

I only tested with SOG Prairie Fire + Domination. The number of enemies are configurable but the crash occurs with any settings, even with "small" mission parameters.

I am not sure where crash logs can be found for a linux server. Here I am only seeing a 2GB core dump file.

possibly related to:

As dedmen called it, objects out of terrain bounds with an adverse affect on the server. Fixed in the mission. I would recommend closing this issue.

When the server is "overflowing" badly it can spam the logs with a million errors in a few hours and causes serious performance problems. Alternatively a lightly "overflowing" server sometimes self-corrects and the players don't even notice the issue.

Some object is way out of terrain bounds. Probably under the ground.

Just to clarify, I am able to see the "Overflow" error message in the server logs but the server does not crash in my current test environment. More testing at the moment.

It seems like something has changed since I last tested this in 2018. I'm not able to reproduce the issue on my current configuration. It could be due to newer Linux kernel, different hardware... not sure.

Hi @Wulf. Yes, I will provide logs and update this ticket.