I am trying to run this on my instance which has 8 NVIDIA H100 GPUs but my docker container seems to be exiting due to 137. I tried setting the memory to be at 1GB and still happening with the same result. Is anyone else having this issue?
NVIDIA-SMI 535.86.10 Driver Version: 535.86.10 CUDA Version: 12.2
Docker version 20.10.21, build baeda1f
docker run -d --gpus all --m 1g -p 9400:9400 nvcr.io/nvidia/k8s/dcgm-exporter:3.2.5-3.1.8-ubuntu20.04
Hello,
I am trying to run this on my instance which has 8 NVIDIA H100 GPUs but my docker container seems to be exiting due to 137. I tried setting the memory to be at 1GB and still happening with the same result. Is anyone else having this issue?
NVIDIA-SMI 535.86.10 Driver Version: 535.86.10 CUDA Version: 12.2
Docker version 20.10.21, build baeda1f
docker run -d --gpus all --m 1g -p 9400:9400 nvcr.io/nvidia/k8s/dcgm-exporter:3.2.5-3.1.8-ubuntu20.04