2.0->2.2(5) produces lots of P0V75_DDR3_T_P1A:Voltage Threshold Crossed

[hostasaurus]

Hi all, just completed an upgrade from 2.0(2q) to 2.2(5).  After upgrading, every few days, across all the blades in our cluster, we're seeing errors like this:

Health LED of server 1/3 shows error. Reason: P0V75_DDR3_T_P1A:Voltage Threshold Crossed

which will then self-recover a minute or two later.  Obviously memory errors make one nervous, but then when it began occurring across all the servers, I figured it's a cosmetic bug.  Any ideas if that's the case?

[Keny Perez]

you may be hitting this https://tools.cisco.com/bugsearch/bug/CSCul70994/?reffering_site=dumpcr

-Kenny

[hostasaurus]

Unfortunately I don't seem to have permission to view that bug.

[Keny Perez]

Yeah, sorry... can you at least see the tittle of the bug?

-Kenny

[hostasaurus]

Nope :-)  Just tells me no permission.

I did some other searches for UCS-related bugs with that text and only got four, none of which appear to apply to 2.2 series software.

All blades are B440 M2's in case that is useful. (4) 2.4 GHz cpu's, 256 GB mem fully populated, so 8 GB dimm's.

[Keny Perez]

Ok that bug talks about B230-M2's... open a case saying you got these error messages and that you suspect from CSCul70994 but not sure if it applies to your platform.

-Kenny