troubleshooting ECC error on ASUS Pro WS C621-64L SAGE/10G

[goretsky]

Hello,

I have an ASUS Pro WS C621-64L SAGE/10G, populated with 12 SK hynix 64GB DDR4-2666MHz ECC rDIMMs, model HMAA8GL7AMR4N-VK.

Recently, the system started to have sudden power-offs, and after swapping out the motherboard, CPU, and PSU with spares, bypassing the UPS, and have moved on to troubleshooting by testing the RAM.

I initially populated the motherboard with a pair of DIMMs, and increased that by a pair after each test cycle (i.e., 2, 4, 6, 8, 10, and finally 12 DIMMs installed). I am using PassMark MemTest86 v11.7 (the free edition) to test the memory. With all 12 slots populated, I get the following ECC error:

> [ECC ERRORS] Test: 5 Channel-Slot: 3-0
> [ECC ERRORS] Test: 6 Channel-Slot: 3-1

What I am trying to to confirm is whether Channel 3, Slot 0 and Channel 3, Slot 1 correspond to DIMM_C1 and DIMM_C2 on the motherboard or to DIMM_C1 and DIMM_F1 on it, as those were the slots I put the third pair of DIMMs into, per the motherboard's manual. Thanks if anyone can help confirm the location.

Oh, and if you have any other ideas about troubleshooting, please let me know.

Regards,

Aryeh Goretsky

 

[RolloZ170]

0 A1/A2
1 B1/B2
2 C1/C2
3 D1/D2 <<<
4 E1/E2
5 F1/F2
but can be overheating issue because RAM slots are not orientated to normal Air flow from front to back...

 

[goretsky]

Hello,

Thank you very much, @RolloZ170!

Strangely enough, problems have moved to different DIMMs after replacing those and retesting.

Will keep everyone informed of progression.

Regards,

Aryeh Goretsky

 

[RolloZ170]

Strangely enough, problems have moved to different DIMMs after replacing those and retesting.

do you run 2DPC with same speed than 1DPC ? maybe your system can't hold speed.
enable enforce POR.

 

[goretsky]

Hello,

I have been running 2DPC, fully-populated (12 DIMMs, all same model and part number).

Turns out it was not an ECC, frequency or voltage error at all.

It was an overheat issue, caused by the fan settings reverting to their default values after a BIOS (UEFI) firmware update. In a sense, that was a memory issue on my part—I had forgotten I had changed those ~3 years ago.

I've written it up here. Definitely a learning experience.

Thank you for taking the time to answer my questions.

Regards,

Aryeh Goretsky