DDR4 ECC Error/Bug

[tempt8]

Hi all,

First time poster. I have run into the following issue where my system seems to pass the tests but reports ECC errors after every test. I am unsure if this caused by a bad CPU/RAM/MOBO on my system or if it is some incompatibility between my BIOS and MemTest86 itself. I have even reset my BIOS settings to default but the same thing keeps happening. I would like to do a full sweep of MemTest86 assuring that the ECC functionality on my system is working as expected; do I need to worry about these errors?

I have attached screenshots of everything that might be useful for troubleshooting, if anything else is required let me know

Thanks in advance!

[David (PassMark)]

seems to pass the tests but reports ECC errors

The job of ECC RAM is to correct single bit errors and detect 2 bit errors.

So in your report above, you had a bunch of RAM errors, but they were all corrected by the ECC function. As there was no memory corruption or data loss, the test passed.

assuring that the ECC functionality on my system is working as expected

Yep, it is working.

You could argue that it would be better if it didn't have to work so hard and the somewhat bad RAM should be replaced, just to be on the safe side. That is a financial / risk decision you need to make.

[tempt8]

Thanks for the reply David! One funny observation: I pulled out all the DIMM sticks put them in a different motherboard (not SuperMicro this time) and as you said the memory still passed all the test but this time I didn't get any of the ECC Corrected errors. I looked around in the MemTest86 Technical Docs and found the following:



source: https://www.memtest86.com/ecc.htm

I looked around in my BIOS settings and couldn't find anything related to Quick Boot. Do you think this might have to do with an incompatibility from the BIOS side or that there might be something wrong with the Motherboard itself? As a good measure I have attached the log of the memtest run just in case it might help your team with debugging! MemTest86-20240108-234641.log

[David (PassMark)]

As your ECC errors occurred a long way into the testing I don't think the problem is the Quick boot issue.

Isn't possible to be 100% sure who is to blame without some very high end test equipment (oscilloscopes, etc..) or by doing a lot of experimentation.
e,g, small changes in voltages or timings might fix the problem on the supermicro board.