[DPE-1423] Update ocean-kubernetes-controller version to 0.1.59

[BryanFauble]

Problem:

• There is an updated helm chart available for the ocean kuberenetes controller
• New nodes that were coming up were not joining the EKS cluster

Solution:

• Use the following to find the new chart version, and generate a new values.yaml file:

-- Update all repos I have added:
helm repo update


-- Lists the repos I have added
helm repo list

-- Find the updated chart number
helm search repo spot

-- Generate a new values file I can copy/paste into the existing values.yaml
helm show values spot/ocean-kubernetes-controller > new_values.yaml

• Update the user_data section of the spot terraform module to ensure that it will bootstrap correctly

https://awslabs.github.io/amazon-eks-ami/nodeadm/
https://registry.terraform.io/providers/spotinst/spotinst/latest/docs/resources/ocean_aws#user_data-1

Testing:

• This has deployed to the sandbox k8s cluster
• Nodes are joining the cluster updated to 1.33

[linglp]

Why do we need to set command: []? Does it mean to use the default entry point?

[BryanFauble]

I'm not exactly sure. Many of these changes are what the developers have as defaults, so I decided to leave them in.

In helm when you run

helm show values spot/ocean-kubernetes-controller

It gives you what their default values file is for the version you are using. The changes you see in the values file are what their new suggested defaults are, and I kept a bit of our specific configuration as well that isn't the same as the default.

[linglp]

Ah got it. I saw it on the UI too:

[linglp]

Do we usually just copy the template that they provide? I can find template here: https://artifacthub.io/packages/helm/spot/ocean-kubernetes-controller

[BryanFauble]

Yeah exactly! I do it through helm commands, but there are a bunch of places that we can get their template.

Copying the default template and adjusting it is how I have done all of the helm based deployments in the repo

[linglp]

Based on my understanding, both 10001 and 1000690000 are both non-root users. Why do we need to switch from one non-root user to another?

[linglp]

oh I undestand why now! For some reasons, that was changed in their template:

[thomasyu888]

🔥 LGTM! Thanks for doing this update!