Question Render server freezes up ?

[BeezKnezzMan]

Hello! I have a weird issue with a render server.
Specs:
MB: Asus Pro WS WRX80E-Sage SE WIFI, AMD WRX80
Proc: AMD Threadripper PRO 5955WX Socket sWRX8
M.2: Samsung 990 PRO 1TB M.2
PCU: 3 X Corsair HX1500i 1500W FM ATX3 PCIe5
Mem: 2X Corsair Vengeance RGB RT 128GB DDR4 3600MHz CL18 Quad
Channel
GPU: 8X MSI nVidia GeForce RTX 4090 GAMING X TRIO 24GB (attached to the MB with riser cables).
GDDR6X, 384bit

I used to have some problems about 8 months ago, but I was helped and found out I needed more RAM.
As well I had an issue with a riser cable, and I fixed it some time ago.

All the PC does get workload frames (Cinema4d or other like Blender), it downloads them, renders them and uploads them back.

What happens now is, I get a few bigger renders in a row and the system freezes while the GPU's remain in load. Happened last night and found it like this in the morning.

I researched it a bit and it could be a PSU, but does anyone know if I can figure it out in a way?

Thank you!

 

[Ralston18]

Key is to discover what may have happened via Window's view of things.

Look in Reliability History/Monitor and Event Viewer for error codes, warnings, or even informational events just before or at the time of the freezes.

And, if I follow correctly, "bigger renders", means more CPU/GPU processing, etc. which could be heating things up. And something begins to falter or fail. Or some resource hits a threshold limit and cannot continue.

Start with Reliability History/Monitor. Much more end user friendly and the time line format may help you see some pattern to the described events (freezes).

Event Viewer requires more time and effort to navigate and understand.

To help:

How To - How to use Windows 10 Event Viewer | Tom's Hardware Forum (tomshardware.com)

Take your time, no need to rush through any of it.

In both tools any given errors or entries will provide more details. The details may or may not be helpful.