![]() ![]() I've modified the Dockerfile like this: FROM ubuntu:18. It looks like my docker image needs an nvidia driver. ![]() Make sure that the latest NVIDIA driver is installed and running. nvidia-smi NVIDIA-SMI has failed because it couldn’t communicate with the NVIDIA driver. To check the version of the Nvidia driver installed on your Ubuntu system from the command line, first open a terminal window. It is also possible that the driver is not available in your systems driver database. ![]() You can run the following command to check if your system is running on the open-source driver nouveau. When I run the docker image through nvidia-docker2 or kubernetes it gives this error: Cuda failure addarrays.cu:9: 'CUDA driver version is insufficient for CUDA runtime version'Ĭuda failure addarrays.cu:9: 'CUDA driver version is insufficient for CUDA runtime version'Īddarrays: addarrays.cu:62: int main(): Assertion `hostArrayTmp = hostArrayDest' failed. NVIDIA tools indicate that the driver is not running, though. This is my terminal output of Nvidia nvidia-smi If the output is negative for nouveau, then all is well with your installation. I've set up the nodes to use the kubernetes nvidia device plugin with nvidia-docker2. I need to run a cuda binary on kubernetes. There are several NVIDIA drivers (both open-source and proprietary) that you can use. ![]()
0 Comments
Leave a Reply. |