ECC errors on 3 new DL360/380

[pirx]

Something in the environment probably changed. e.g temperature. EMI, etc..
It isn't very likely the sticks fixed themselves.

If HP claim to have ECC error logging that works, then I would monitor the logs for a few weeks and see if you get any corrected errors in real usage.

At least on one server I see the same error again. Still HPE claims that they see nothing in the ILO logs. It's hard to force them to replace the DIMM when without any errors in their tools.

[David (PassMark)]

Something in the environment probably changed. e.g temperature. EMI, etc..
It isn't very likely the sticks fixed themselves.

If HP claim to have ECC error logging that works, then I would monitor the logs for a few weeks and see if you get any corrected errors in real usage.

[pirx]

Hi,

we got a couple of new HPE Proliant DL360Gen10/Dl380Gen10 and I ran memtest86 on them. 3 of them had ECC errors. I opened a ticket at HPE but they did not find anything in ILO event logs or generated AHS logs. I rerun the test now and it's already running 1-2h longer than before, without any errors. So are these ECC errors something to worry about?