Hi,

i need experts to help diagnose my system.

Last night i migrated to new motherboard and processor, re-using old parts such as RAM, hard drives, SSD, NIC, SAS card.

My Old Specs:

Asus z170 Sabertooth + Core i7 6700k

Corsair vengeance  DDR4 2666mhz1

Palit GTX1070 for Plex Transcode

Intel X550-T2 10G NIC

LSI 3008 24 Port SAS Card

Samsung 980 Pro 2TB as Cache

8x 16TB, 4x 10TB, All Toshiba Enterprise Capacity Drives

 

I only use it for Plex with GPU Passthrough for transcoding, qbittorrent, and i run a couple of CentOS and Windows VM.

 

System has been running well for 2 years. Until i decided to upgrade to Asus Crosshair VIII x570 + Ryzen 5900x last night.

Then as i was running parity check after bringing up the system, it rebooted abruptly and i get the Machine Check Event error for the first time.

Attached Diagnostics logs.

When i do a "mcelog" in terminal, i get the following error:

mcelog: ERROR: AMD Processor family 25: mcelog does not support this processor.  Please use the edac_mce_amd module instead.
CPU is unsupported

 

Just Disabled Global C-States in UEFI. Observing.

 

Thanks everyone.

nasty-diagnostics-20240512-2136.zip

Edited May 12, 20242 yr by sgrangers

19 hours ago, sgrangers said:

it rebooted abruptly

Server rebooting on its own is almost always a hardware issue, but can be caused by different components, since you are re-using the RAM, the next suspects after that would be PSU, board, CPU.

its been 20 hours. so far seem stable. *fingers cross*

lets hope its really the C states.

 

but regarding the mcelog. how do i verify? cuz it says this processor is not supported.

 

Update: Ah forget it. i realized i can't pipe the logs to the shares as the array is offline during bootup and shutdown process.

and mcelog probably gets reset during a reboot. so i will just pray that my system is stable now.

 

Edited May 13, 20242 yr by sgrangers