ccp 0000:04:00.2: SEV: failed to INIT error 0x8003

Jarvar

Active Member
Aug 27, 2019
317
10
38
Good morning everybody.
Since last week, whenever I reboot or shutdown then power on my Proxmox server I get this error.
Can anybody help me decipher it?
Not entirely sure what it is, but last week I upgrade the BIOS on my Supermicro Motherboard M11SDV-8C-LN4F
AMD Epyc 3251 SOC with 2 dimms populated of 32 GB ECC Rdimm.
However, I think I updated the BIOS on th 26th and I happened to be going through the logs after seeing the Proxmox boot error.
Proxmox still runs and we've been using it all week, but I am trying to sort out this error.

Last week, on July 25th, I also logged a Uncorrectable ECC / other uncorrectable memory error @DIMMA2 - A
in the Systems Events log.
I have been trying to run the Memtest86+ 5.01 bundled with Proxmox during the night when the their operations is closed but so far can only manage 2 passes and up to test 10 at the 26 hour mark.

When it happened last weekend, I actually got
Proxmox Hypervisor showing SEV Uncorrectable ECC Error.

I saw somewhere there it could be a BIOS setting related as well.

Any insight would be much appreciated.
Thank you so much.
 
Last edited:
hi, have you found a solution to this, just saw the exact same message in my dmesg protocol with proxmox and the exact same cpu, 2 memory ECC sticks
 
@Machtl
Sorry I never fully figured it out. To be safe, I got two separate stick of memory to test out. However I've chalked it up to be either a software issue or incompatibility. I guess I could have sent off the unit under warranty, but everything else seemed to be working okay.

I'm not sure, I was hoping that it was a special case scenario. I happened to have upgraded the bios of the motherboard around the same time which I think could have contributed to the error. I did eventually replace the memory or at least 2 sticks extra a backup for it. I tested both sets of memory using both Memtett86 and Memtest86+. The former would pass, but the later program kept getting errors. I did find somewhere at the time that the errors are probably related to the memory testing software and not necessarily the ECC Ram itself. Memtest86 was clear, although it maxes out with 4 tests. I think that's what happened.
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!