FiveM server crashes when people connect - System issue

KalipayJ · 2019-05-10T17:27:48.043Z

I’ve installed FiveM server on Debian 8, Ubuntu 16.04, Ubuntu 18.04 and Windows 8.
All the installations have 2 things in common, which is that the FiveM server itself is running great, it shows up on my FiveM browser, but as soon as I connect to the server, my FiveM server crashes. And the second thing is that they were all installed on a VM.

I’m not 100% sure of the issue, which is why I’m asking here. Suspecting CPU incompatibility? Maybe someone has experienced the same, or maybe even has a solution, or can explain the reason for the crashes. I’ve attached a dump from Windows and some logs I made of what the console is showing me on Linux.

Hardware:
CPU: Xeon E3-1230 v3 @ 3.30Ghz
Motherboard: H97M-D3H
RAM: Tried between 1-8GB allocated memory.

The Hypervisor is XenServer 6.5

Console copy from Debian (with Essentialmode)
Debian.txt (6.5 KB)

Console copy from Ubuntu (bare bone install of FiveM)
Ubuntu.txt (5.3 KB)

Dump from Windows (bare bone install of FiveM)
6b422516-3610-4a04-a2b4-8b4310a372bb.dmp (932.7 KB)

The Hypervisor is running numerous VMs and none of them are failing, which makes me doubt my theory.
I hope someone can decode the logs and help me figure out what the problem is.

Thanks in advance,
Kali

barium · 2019-05-10T18:07:02.722Z

It seems your hypervisor claims it supports AVX in CPUID, when in reality this leads to a #UD fault when executing the vpxor instruction in the Botan crypto library.

A quick google shows this is related to Xen masking some XSAVE feature.

KalipayJ · 2019-05-10T18:37:58.706Z

Thank you, I believe this was the kind of brain I was looking for. I’ll look into that and see if there’s even a solution out there.!

Thanks again!

KalipayJ · 2019-05-11T12:14:15.842Z

From your information I searched about for a while, and came to the conclusion that an update of my XenServer might solve the problem. Updating from 6.5 to 6.5 SP1 did indeed solve the problem, and my FiveM server now runs flawlessly.

Thank you, once again!