Curtis Vogt

> Feels like you'd pretty much always want this I agree with this. I'd roll this into the `K8sClusterManager` constructor. We already retrieve the manager pod information there.

Looks like we should hold off on this for a while: ``` ! Specified Kubernetes version 1.27.0 is newer than the newest supported version: v1.27.0-rc.0. Use `minikube config defaults kubernetes-version`...

Just checked the minikube repo for an issue on this and I didn't find one. However, there is a bot that attempted to do this update and ran into some...

The nested task error: `Unable to determine the pod name from: ""` is from `create_pod` and shows that the external command call resulted in no stdout (the empty string reported)...

Code is currently not resulting in any layer cache reuse. Something isn't working here

Definitely should be added. Should be pretty straight forward to add. We'll just need to [pass `params [:env]` into `worker_pod_spec`](https://github.com/beacon-biosignals/K8sClusterManagers.jl/blob/905c995d0ca6300a188e2e8dc7ef5c87a7548295/src/native_driver.jl#L91-L96) and then [update `worker_pod_spec` to add the environmental variables into...

> Maybe even set a label by default, maybe `name: `? Some default labels would be nice. We should probably have a label that is used for the manager and...

Workers now automatically have a `worker-id` label which is the same as there worker ID (`myid`) as of #48

Related to https://github.com/beacon-biosignals/K8sClusterManagers.jl/issues/54 but slightly different.

It should also be mentioned that k8s jobs have a [`.spec.ttlSecondsAfterFinished`](https://kubernetes.io/docs/concepts/workloads/controllers/ttlafterfinished/) to automatically delete themselves after they have completed. However, this feature is only available for jobs.