The GPU partitioning lab instructions assume that the example YAML manifests are available on the same machine where kubectl is being executed.
This works when the HAMi repository is cloned directly on the GCP VM, but it does not work in the AWS EC2 setup, where the repository is typically on the user's local machine and kubectl runs on the remote EC2 instance.
As a result, commands using relative paths such as:
kubectl apply -f tutorials/labs/examples/03-gpu-partitioning/<manifest>.yaml fail because the referenced files are not present on the EC2 instance.
The GPU partitioning lab instructions assume that the example YAML manifests are available on the same machine where
kubectlis being executed.This works when the HAMi repository is cloned directly on the GCP VM, but it does not work in the AWS EC2 setup, where the repository is typically on the user's local machine and
kubectlruns on the remote EC2 instance.As a result, commands using relative paths such as:
kubectl apply -f tutorials/labs/examples/03-gpu-partitioning/<manifest>.yamlfail because the referenced files are not present on the EC2 instance.