I'm trying to deploy the gpu-feature-discovery using the provided helm chart but due to the pod priority admission plugin in EKS the pods with system-node-critical cannot be scheduled outside of kube-system namespace. I've found some details about this here .
I thought I'll create this PR maybe it can be approved.
Thanks
I'm trying to deploy the gpu-feature-discovery using the provided helm chart but due to the pod priority admission plugin in EKS the pods with
system-node-critical
cannot be scheduled outside ofkube-system
namespace. I've found some details about this here . I thought I'll create this PR maybe it can be approved. Thanks