When I open my workstation, 'mst status -v' can find clear Bluefield at first. However, few minutes later, disappeared with dmesg error

[ziji.chen]

And then

dmesg log:

[ziji.chen]

Ask for help!!! I’m totally desperate. I cannot graduate without using this card, long story, lots of effort and 2 days to locate the problem. Should I add more fans to my workstation? Anyway based on the errors.

[samerka]

Hi ,

Please refer to the Bluefield Hardware Spec , as it looks like it getting high temperature and as a safety step the NIC DPU is shutting down .

Kindly check the cooling system on the server and the adapter if it follows the spec
https://docs.nvidia.com/networking/display/bf3dpu/specifications
Environmental Temperature
**|Operational|0°C to 55°C|
|Non-operational|-40°C to 70°C (b)|

Also you can review supported servers for Bluefield 3
https://docs.nvidia.com/networking/display/bf3dpu/supported+servers+and+power+cords

Thanks
Samer

[ziji.chen]

I’m using Bluefield2.

[samerka]

Hi ,

Please see below
https://docs.nvidia.com/networking/display/bluefield2dpuvpi/supported+servers+and+power+cords

https://docs.nvidia.com/networking/display/bluefield2dpuvpi/specifications

Temperature	Operational	0°C to 55°C
Non-operational	-40°C to 70°C(c)

Thanks
Samer

[ziji.chen]

Hi, thank you. But I can’t find NVIDIA BlueField-2 DPUs Power and Airflow Specifications document. Could you please send me the link directly?

“Power and airflow specifications are provided in NVIDIA BlueField-2 DPUs Power and Airflow Specifications document, which is available at NVOnline following login.”

[samerka]

Hi ,

It’s a confidential file that requests an NDA signed.
Please make sure you have a valid support entitlement and send a request to
enterprisesupport@nvidia.com

Thanks
Samer

[cs.lev]

I know it might be late, but I only checked in today.
You can check if the temperature is really a problem by actually getting the temperature of the SmartNIC.
Check this thread to do so:

How can I check the current temperature and set threshold of MLX_5(Connectx-6)?

Dear Mellanox Support team, The following message is recorded in /var/log/messages. Dec 3 18:47:13 hgpu009 kernel: mlx5_core 0000: 3b: 00.0: mlx5_temp_warning_event: 564: (pid 0): High temperature on sensors with bit set 0 0 ?It seems that the following times have been recorded in the past month. [root @ hbcm01 log] # grep “High temperature” messages * | grep gpu | wc -l 77 [root @ hbcm01 log] # grep “High temperature” messages * | grep cpu | wc -l 4 ? How can I check the current temper…

[ashfaq1]

I faced similar issue with my BF2. Installing a cooling fan on top of BF2 solved the issue. I am using a fan which can be attached in the expansion slots.

[ziji.chen]

Hi, thank you! I have solved my problem by adding an extra cooling fan. Thanks again.

[ziji.chen]

Thank you! BTW, dmesg can also provided many messages.