Storage node example in kube.tf

[loxy]

In the kube.tf example there is an example of a storage node

{
      name        = "storage",
      server_type = "cpx21",
      location    = "fsn1",
      # Fully optional, just a demo.
      labels      = [
        "node.kubernetes.io/server-usage=storage"
      ],
      taints      = [],
      count       = 1
}

Can someone please explain for what I need that? Is it only relevant when using for Lonhorn?

[WebSpider]

No, labels can be used for scheduling workloads onto nodes, and work together with taints and tolerations.
There is a lot out there on the subject, and ChatGPT can probably help you as well, but in short:

Labels:

• Are applied to a node
• Give hints as to what this node can be used for
  Taints:
• Are applied to a node
• Can cause a pod not to be scheduled on a node, even if it has a label would cause it to normally be selected
  Tolerations:
• Are applied to a pod
• Can make scheduling that would fail because of a Taint still happen

In practise, this is used to make sure the right node gets the right workload, so it could apply to Longhorn storage nodes as well, but doesn't need to.

[loxy]

Thank you for you answer! But I already asked ChatGPT a lot ;-)

I know what labels and taints are. My question was more in the direction of reason to create/user such a node config in general. It is "enabled" by default in kube.tf and creates a dedicated node (for storage as the name suggests). But I don't need that when using cloud volumes, right?

That's why I wanted to understand why someone would use that node (because agents where already defined before). So, I guess no pods should be scheduled to it?

When it is not only used for Longhorn, can you give other examples where this setup is useful and I should use a "storage" node?

Sorry for my maybe dumb questions...