From 4e49a623a92dde5be534214ad0b89bce82ffd49b Mon Sep 17 00:00:00 2001 From: "Barry.Cheng" Date: Tue, 4 Feb 2025 14:11:36 +0800 Subject: [PATCH] docs: Add note for nvidia-smi execution in example deployment --- README.md | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 9a43d590..1501c5d6 100644 --- a/README.md +++ b/README.md @@ -49,13 +49,21 @@ resources: Verify that it has been scheduled on one of the __CPU__ nodes. -You can also test by running the example deployment YAML under the [example](./example) folder +You can also test by running the example deployment YAML under the [example](./example) folder. + +**Note:** If you want to execute `nvidia-smi` in the example deployment, you need to add the following snippet to the `deployment.yml` file. Replace `` with the node name you labeled during installation: + +```yaml +env: + - name: NODE_NAME + value: +``` ## Troubleshooting [Pod Security Admission](https://kubernetes.io/docs/concepts/security/pod-security-admission/) should be disabled on the gpu-operator namespace -``` +```sh kubectl label ns gpu-operator pod-security.kubernetes.io/enforce=privileged ```