Strange point about temperature sensors of gpu and memory test in ML110 G10

[DavidHur]

Hi, I have two kinds of question as below;
1. Tried to check the temperature, there have GPU 1 and GPU4 sensor in IO board, there is temperature is about 78 celsius. I think it is abnormal status because it doesn't working anything only power up status.
What is problem in this case? I searched any firmware of it but I couldn't find.
If 78 celsius is really correct, it should feel very hot when it touched to the GPU core but it doen't.

2. I tried to  memory test diagnostics self but it doesn't working correctly. I tested memory to the other server and it tested as well but have problem memory diagnostic problem in ML110 Gen10.
Do you know what is the problem?
FAILURE ID : 90BR79-000AA0-X8622J-403903
PRODUCT ID : P03685-425

[ksram]

Hi @DavidHur ,

Hi 

Please try updating the ILO and BIOS firmware related to the Temperature issue :

ILO : https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_950d3dfe4d854854a0be9ec6a0#tab-history
BIOS : https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_ce335c0364fa46aabd7f326ed0#tab-history

For the hardware test, we would recommend to update the HPE Intelligent Provisioning version and then try the test.

IP : https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_58d0df2101e94ba1b8ada2344d

Thank you

Ram

[DavidHur]

No, it wasn't solved after firmware updated you recommended.
Temperature sensors and Memory test doesn't solved anything.

[ksram]

Hi @DavidHur,

Please refer : https://support.hpe.com/hpesc/public/docDisplay?docId=a00091776en_us&docLocale=en_US regarding the Memory Test.

Please share the GPU details, lets check if there are any Firmware / Driver fixes for the same.

Also share the OS details.

Thank you

Ram