Bad memory on node MOC-R4PAC10U13-S3

[larsks]

ESI node MOC-R4PAC10U13-S3 (service tag J16N3Y2) failed earlier today with:

	Tue Sep 24 2024 12:55:45	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:45	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:45	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:45	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:45	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:45	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:45	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:44	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:44	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.
		Tue Sep 24 2024 12:55:44	Multi-bit memory errors detected on a memory device at location(s) DIMM_B2.

The errors did not repeat after a cold boot, but perhaps we should have the memory replaced.

[schwesig]

from a call today
Yeah probably the ECC memory corrector is unable to fix things anymore. That might be why on the reboot, it clears out that counter.

[hakasapl]

@larsks is this node currently in use? I see it as active in ESI

[larsks]

@hakasapl I thought we pulled it from the cluster, but @tssala23 is probably the best person to ask.

[tssala23]

@larsks @hakasapl We didn't pull it from the cluster, when it had the memory issue it wouldn’t boot without someone telling it, it was “okay” to boot. Lars you went in a did that and it was fine in terms of working with the cluster. So it is still active in the beta-test cluster