ansible-role-nvidia-driver
ansible-role-nvidia-driver copied to clipboard
NVIDIA
The role was not installing latest driver version by default on RHEL systems. With this change the latest driver is installed unless a specific version is selected using var `nvidia_driver_package_version`...
is there any reason why this role doesn't include a task like this? ``` - name: Blacklist the nouveau driver module community.general.kernel_blacklist: name: nouveau state: present ```
First, thanks for this very useful role. I'm having an issue when trying to install the drivers on an Ubuntu 18.04 VM with PCIe passthrough (on plain KVM). I ran...
I've got a problem to install specific NVIDIA Driver on RHEL8 After PR [#53](https://github.com/NVIDIA/ansible-role-nvidia-driver/pull/53) was able to install the branch version of the driver. But I need to install an...
The logic in line: https://github.com/NVIDIA/ansible-role-nvidia-driver/blob/8fa78e47b2e974e1a6dde961ea3622f47c2647e5/tasks/install-redhat.yml#L68 hasn't been working for me. I've specified: ``` vars: - nvidia_driver_package_version: '465.19.01-1' - nvidia_driver_branch: '465' ``` and yet I see this error: `No package matching...
I'm currently running into [this longstanding issue](https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1853977) on Ubuntu 20.04 with the following configuration: ``` - role: nvidia.nvidia_driver nvidia_driver_ubuntu_branch: 340 nvidia_driver_ubuntu_packages: [ 'nvidia-340' ] ``` I don't have this issue...
**Ansible role fails with :** > TASK [nvidia.nvidia_driver : add key] > ok: [gpu1] > > TASK [nvidia.nvidia_driver : add repo] > fatal: [gpu1]: FAILED! => { > "changed": false,...
Please check the base url of the arm64 server https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/sbsa/ There should be a patch like: ``` diff --git a/vars/main.yml b/vars/main.yml index c994f5b..c01187d 100644 --- a/vars/main.yml +++ b/vars/main.yml @@ -1,2...
I am trying to set up a headless render node for Davinci Resolve running on CentOS 7. The ansible playbook *claims* that everything ran successfully, but following a reboot I...
