Used EPYC 7601 post code 78

justme

New Member
Aug 14, 2020
6
0
1
Hey,

So I've got a supermicro H11DSI-NT, got two different epyc 7601s from ebay from 2 different vendors that looked somewhat reputable. Both are giving me post code 78. Fans spinning, but nothing booting, no video, IPMI not even showing correct hardware info. Exact same behavior for both CPUs.

Eventually got a retail epyc 7251 and everything's working perfectly fine!

I'm trying to figure out if I've been scammed, am super unlucky or perhaps I'm missing something.
Perhaps something about the 7601 that's not making them boot? Anything I should try before giving up on them?
 

ari2asem

Active Member
Dec 26, 2018
547
87
28
The Netherlands, Groningen
so, if i understand you correctly....you have 2* 7601 from 2 different sellers ? do they have the same part-number ?? it begins with 100-*****.
are those 7601 ES or QS ??

and your 7251 is 1 cpu on your board and it boots fine ??
 

justme

New Member
Aug 14, 2020
6
0
1
so, if i understand you correctly....you have 2* 7601 from 2 different sellers ? do they have the same part-number ?? it begins with 100-*****.
are those 7601 ES or QS ??

and your 7251 is 1 cpu on your board and it boots fine ??
Correct, the 1 7251 boots fine. Neither of the 7601s boot.

Not sure how I'd tell if it's ES or QS. Both vendors seem to be involved in decommissioning server hardware so I assume they should be just used cpus.

Attaching images of all 3 I've got. (PS7601BDVIHAF is the part number for the 7601s)
 

Attachments

Last edited:

ari2asem

Active Member
Dec 26, 2018
547
87
28
The Netherlands, Groningen
7601 have indeed the same part numbers. they are retail cpu's. no ES, no QS.
but i see they hav different UA-numbers. UA-numbers tell something when and where the cpu are produced
https://www.reddit.com/r/Amd/comments/6scnlg this difference shouldn;t be the reason for 79-error.

did you tried single 7601-cpu to boot, like with single 7251 ?

code 78 ==>> ACPI initiliazation

or thie one


try to downlod the same bios as your board now has and flash t0 same bios version as it is now...and see what hapens
 
Last edited:

justme

New Member
Aug 14, 2020
6
0
1
7601 have indeed the same part numbers. but i see they hav different UA-numbers. UA-numbers tell something when and where the cpu are produced
https://www.reddit.com/r/Amd/comments/6scnlg
did you tried single 7601-cpu to boot, like with single 7251 ?
Yea, mostly only trying this way in fact. Both 7251 and 7601 are rated as 1P/2P CPUs.
Different RAM seating configurations don't help.
Other than that not sure what other variable could be involved.

the post sequence for 7601s goes like this as far as I can tell from IPMI
3e d1 d3 b5 c8 b7 78
 

ari2asem

Active Member
Dec 26, 2018
547
87
28
The Netherlands, Groningen
try to reset your bios.

- first install both 7601
- turn on your system and get 78-error
-completely shutdown your system
- pull off PSU totally from your mainboard (disconnect 24pin atx cable and all cpu 8pin cables)

- pull out cmos battery for4-5minutes
- short connect cmos reset pins on your board for around 4 minutes

- connect psu cables back...and see what happens...

if still 78-error, then try to update your bios.
do this update step by step (first bios version 2.0b, then newer version, then more newer version and then latest version.)

and try with minimum memory sticks, according the manual

you have to figure out how to get past that error....it is trial and error learning way...
 

justme

New Member
Aug 14, 2020
6
0
1
try to reset your bios.

- first install both 7601
- turn on your system and get 78-error
-completely shutdown your system
- pull off PSU totally from your mainboard (disconnect 24pin atx cable and all cpu 8pin cables)

- pull out cmos battery for4-5minutes
- short connect cmos reset pins on your board for around 4 minutes

- connect psu cables back...and see what happens...

if still 78-error, then try to update your bios.
do this update step by step (first bios version 2.0b, then newer version, then more newer version and then latest version.)

and try with minimum memory sticks, according the manual

you have to figure out how to get past that error....it is trial and error learning way...
Oh, I've tried all kinds of sequences and combinations. On the newest bios now. Nothing makes any difference.
And the retail 7251 always boots perfectly on the first try regardless of the previous configuration.

If there's no trick here or something specific to the 7601 pretty sure there's nothing else I can do.

A few threads out there with people getting stuck on 78 with ebay epycs. Seems like everybody ends up giving up.

It's just so weird, vendors really looked reputable. Have tons of feedback, proper websites and addresses.
One listing explicitly claimed it has been tested. Was shipped to me in the EU from the US.
Packed pretty well too so doubt it's shipping damage (plus that's kind of unlikely in the first place with CPUs).
Trying to figure out if I should even bother trying to play this ebay game again. Seems very high risk to me now.
 
Last edited:

pixelBit

New Member
Jul 16, 2019
6
0
1
Must the BIOS be signed?

Perhaps editing the DMI strings of the motherboard's BIOS to reflect a Dell system that runs EPYC could do the trick. DMI strings such as the "Motherboard Manufacturer Name", "System Manufacturer", and "System Product Name".
 

gv64

New Member
Aug 29, 2020
5
1
3
Must the BIOS be signed?
Unfortunately I don't think simply replacing the DMI strings would work, as that would make it way too easy for the BIOS firmware to be "compromised". Here are the two relevant paragraphs from the security document I linked (emphasis in bold):
The first generation of the AMD EPYC processors have the AMD Secure Processor –an independent processor core integrated in the CPU package alongside the main CPU cores. On system power-on or reset, the AMD Secure Processor executes its firmware while the main CPU cores are held in reset. One of the AMD Secure Processor’s tasks is to provide a secure hardware root-of-trust by authenticating the initial PowerEdge BIOS firmware. If the initial PowerEdge BIOS is corrupted or compromised, the AMD Secure Processor will halt the system and prevent OS boot. If no corruption, the AMD Secure Processor starts the main CPU cores, and initial BIOS execution begins.

The very first time a CPU is powered on (typically in the Dell EMC factory) the AMD Secure Processor permanently stores a unique Dell EMC ID inside the CPU. This is also the case when a new off-the-shelf CPU is installed in a Dell EMC server. The unique Dell EMC ID inside the CPU binds the CPU to the Dell EMC server. Consequently, the AMD Secure Processor may not allow a PowerEdge server to boot if a CPU is transferred from a non-Dell EMC server (and CPU transferred from a Dell EMC server to a non-Dell EMC server may not boot).
The "secure hardware root-of-trust" and mentioning of authenticating the BIOS firmware implies that there is some sort of cryptographic verification happening, with a public key that's permanently burned into OTP e-fuses on the EPYC CPU the first time such a CPU is booted on a Dell system.

I returned the CPU to the eBay seller, they re-listed the CPU as a "Dell-only" unit, and it hasn't sold since. I absolutely hate that Dell is doing this (and that AMD is allowing it to happen), wasting perfectly good CPUs in the process.
 

Patrick

Administrator
Staff member
Dec 21, 2010
11,969
4,931
113
I had the same issue: code 78 on an Asrock Rack EPYCD8-2T with a 7551 from eBay. Turns out the issue was that the CPU was pulled from a Dell system, which burn a key into the CPU such that it will only boot a Dell-signed BIOS. This makes the CPU unusable in any non-Dell system.

More details here: https://downloads.dell.com/manuals/common/security_poweredge_amd_epyc_gen2.pdf
This is what we covered in the PowerEdge C6525 video/ review. Doing a full piece on this.

Just so we are clear, Dell is not the only vendor doing this.
 

gv64

New Member
Aug 29, 2020
5
1
3
This is what we covered in the PowerEdge C6525 video/ review. Doing a full piece on this.

Just so we are clear, Dell is not the only vendor doing this.
Thanks Patrick! It was from that video that I put 2 and 2 together and figured out why the CPU I ordered a while back wouldn't work.

I could find barely any information on this on the Internet, and I wanted to implicate POST code 78 (at least on Asrock and other AMI Aptio V based boards) to this exact problem to help others who purchase pulled EPYC CPUs. Looking forward to your video to learn more about this.