URL:
  <https://savannah.gnu.org/support/?111312>

                 Summary: freeipmi is reading certain sensors incorrectly on
Nvidia GB200 platform
                   Group: GNU FreeIPMI
               Submitter: None
               Submitted: Mon 08 Sep 2025 09:10:41 PM UTC
                Category: ipmi-sensors
                Priority: 5 - Normal
                Severity: 4 - Important
                  Status: None
                 Privacy: Public
             Assigned to: None
        Originator Email: [email protected]
             Open/Closed: Open
         Discussion Lock: Any
        Operating System: GNU/Linux


    _______________________________________________________

Follow-up Comments:


-------------------------------------------------------
Date: Mon 08 Sep 2025 09:10:41 PM UTC By: Anonymous
Hi,

On the latest version of freeipmi (1.6.15), certain sensors are read and
reported incorrectly. Specifically, sensors related to temperature of ports on
the CX7 NICs are reported in the range of 190C+, which is clearly incorrect.
Below is the partial output of ipmi-sensors which shows this issue:

77  | IOBo0CX70Por0Tem | Temperature              | Nominal  | 198.43     | C
   | 'OK'
78  | IOBoard0CX70Temp | Temperature              | Nominal  | 40.30      | C
   | 'OK'
79  | IOBo0CX71Por0Tem | Temperature              | Nominal  | 199.06     | C
   | 'OK'
80  | IOBoard0CX71Temp | Temperature              | Nominal  | 42.82      | C
   | 'OK'
81  | IOBo1CX70Por0Tem | Temperature              | Nominal  | 196.54     | C
   | 'OK'
82  | IOBoard1CX70Temp | Temperature              | Nominal  | 44.08      | C
   | 'OK'
83  | IOBo1CX71Por0Tem | Temperature              | Nominal  | 196.54     | C
   | 'OK'
84  | IOBoard1CX71Temp | Temperature              | Nominal  | 44.08      | C
   | 'OK'

The same sensors are read correctly using ipmitool:

IOBo0CX70Por0Tem | 4Dh | ok  |  0.1 | 37.78 degrees C
IOBoard0CX70Temp | 4Eh | ok  |  0.1 | 40.93 degrees C
IOBo0CX71Por0Tem | 4Fh | ok  |  0.1 | 37.78 degrees C
IOBoard0CX71Temp | 50h | ok  |  0.1 | 42.82 degrees C
IOBo1CX70Por0Tem | 51h | ok  |  0.1 | 35.26 degrees C
IOBoard1CX70Temp | 52h | ok  |  0.1 | 44.08 degrees C
IOBo1CX71Por0Tem | 53h | ok  |  0.1 | 35.26 degrees C
IOBoard1CX71Temp | 54h | ok  |  0.1 | 44.08 degrees C

Since ipmitool report the sensor values correctly, we believe this is a
software bug in freeipmi, and we were hoping a patch/fix could be provided to
address the issue. Please let me know if I can provide any extra information
to help debug and resolve this issue.







    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/support/?111312>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/

Attachment: signature.asc
Description: PGP signature

Reply via email to