Hi All I've deployed k3s on openSUSE Leap 15.5 along with the Nvidia gpu-operator which is all working fine. I've deployed open-webui and it is all working fine until a reboot, when I get;

Reason:  UnexpectedAdmissionError
Message: Pod was rejected: Allocate failed due to no healthy devices present; cannot allocate unhealthy devices <http://nvidia.com/gpu|nvidia.com/gpu>, which is unexpected

How do I set a delay on the open-webui-ollama pod or open-webui namespace to wait for the gpu-operator to be up and running before starting?