Hi there! I am preparing a Power8 rhel 7.7 system with a Tesla P100-SXM2-16GB to be used by an containerized application but unfortunately I have been getting version conflicts.
The last CUDA version for a Power8 rhel 7.7 is the 10.2 and the Nvidia diver is the 440.95.01. But for some how the nvdia-container asks for the CUDA 11.0 or a earlier container version.
I used the command yum install --setopt=obsoletes=0 nvidia-container-runtime-hook-1.4.0-2 to keep the nvidia-container at a supported version but didn’t work.
At the container run the docker pulled a newer version and I got the conflict.
error message when running the container
docker run --rm nvidia/cuda-ppc64le nvidia-smi
*Status: Downloaded newer image for docker.io/nvidia/cuda-ppc64le:latest
container_linux.go:235: starting container process caused “process_linux.go:339: running prestart hook 1 caused "error running hook: exit status 1, stdout: , stderr: exec command: [/usr/bin/nvidia-container-cli --load-kmods configure --ldconfig=@/sbin/ldconfig --device=all --compute --utility --require=cuda>=11.0 --pid=7799 /var/lib/docker/overlay2/e6e2586c37aabece26ac39b140bce0b9b5ed6ec881abba8096396a21a9935ace/merged]\nnvidia-container-cli: requirement error: unsatisfied condition: cuda>=11.0, please update your driver to a newer version, or use an earlier cuda container\n"”
*
I would appreciate any help out there
Thanks