New unRAID Crashing When Under Load

[SeanP]

Hey guys. I've been running this unRAID server for 5-6 days at this point, and it seems to be really crashy. Going through the steps I came across this error, but can't really decipher it. I don't believe my CPU is overheating (Ryzen 5 1500X), based on the temperature of the disks and the monitors I have access to. Let me know what you think, I have a Ryzen 3 1200 CPU I could swap in as well. I will say that I haven't had the opportunity to try and visualize the output during a crash, I'll just be doing something and get a popup that my share disconnected and the server will not be available for 5-10 minutes.

 

Feb 22 08:16:32 Oathbreaker kernel: mce: [Hardware Error]: Machine check events logged

Feb 22 08:16:32 Oathbreaker kernel: mce: [Hardware Error]: CPU 5: Machine Check: 0 Bank 5: bea0000000000108

Feb 22 08:16:32 Oathbreaker kernel: mce: [Hardware Error]: TSC 0 ADDR 1ffff8168ab3e MISC d012000100000000 SYND 4d000000 IPID 500b000000000

Feb 22 08:16:32 Oathbreaker kernel: mce: [Hardware Error]: PROCESSOR 2:800f11 TIME 1582388171 SOCKET 0 APIC 3 microcode 8001138

[trurl]

See the Unraid FAQ pinned near the top of this same subforum for Ryzen tweaks. 

[Hoopster]

24 minutes ago, SeanP said:

I've been running this unRAID server for 5-6 days at this point, and it seems to be really crashy.

First generation Ryzen CPUs are well know for crashing in Linux.  There are some things you need to adjust in the BIOS such as disabling C-States and adjusting power supply idle control.

 

See the thread Trurl mentioned.

 

 

[SeanP]

Awesome, great info! I was scrolling down through the logs and noticed this as well:

 

Feb 22 08:32:38 Oathbreaker root: Fix Common Problems: Error: Machine Check Events detected on your server

Feb 22 08:32:38 Oathbreaker root: mcelog: ERROR: AMD Processor family 23: mcelog does not support this processor. Please use the edac_mce_amd module instead.

 

I'm going to see about using that other module instead. Thanks for the info!