Error setting watches. Result: -33: This request is serviced by a module of DCGM that is not currently loaded

[mintchocohoco]

hello, I am using p100 gpu, and there is a problem that more than 1000 features(1002,1003,1004,1005....) are not work with this error code

Error setting watches. Result: -33: This request is serviced by a module of DCGM that is not currently loaded

• Features less than 1000(e.g. 82,83....) work fine
• p100 is using an AC1 instance of IBM CLOUD (ubuntu18.04,cuda11.4,python3.7)
• dcgm was downloaded with the following command.

  wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/cuda-keyring_1.0-1_all.deb sudo dpkg -i cuda-keyring_1.0-1_all.deb sudo add-apt-repository "deb http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/ /" sudo apt-get update && sudo apt-get install -y datacenter-gpu-manager sudo systemctl --now enable nvidia-dcgm

• When I type dcgmi in the terminal I get:

[nikkon-dev]

@mintchocohoco,

The DCP metrics (1001...) are supported starting from Turing architecture. Pascal is not supported. For some metrics (1013,1014) you would need at least an Ampere GA100 chip.