Is there a way to automatically taint a few node pools to allow gpu access upon creation? Like this but through the values.yaml when helm installing? Also what should the key and value be for the resource tolerations?

can you elaborate a bit?

So I can set the tolerations on the node pools by doing this in values.yaml

k8s:
    plugins:
      # -- Configuration section for all K8s specific plugins [Configuration structure](<https://pkg.go.dev/github.com/lyft/flyteplugins/go/tasks/pluginmachinery/flytek8s/config>)
      k8s:
        default-env-vars: []
        #  DEFAULT_ENV_VAR: VALUE
        default-cpus: 100m
        default-memory: 100Mi
        resource-tolerations:
          - <http://nvidia.com/gpu|nvidia.com/gpu>:
            - key: "flyte/gpu"
              operator: "Equal"
              value: "dedicated"
              effect: "NoSchedule"

But in order for a node pool to have a taint that matches these tolerations, I would need to taint them after helm installation. Is there a way to avoid doing the manual step of looking up node pools and setting the taint? It would be ideal if the node pool creator would apply taints to a few node pools.

how are you creating the node pools?

okay i understand now. thank you. I thought flyte created node pools since there are some default nodes under the development namespace.