Consider returning warnings in validation for usage of deprecated fields

[chrischdi]

What would you like to be added (User Story)?

As a user I would like to get a warning when I make use of deprecated fields.

Detailed Description

We already have some deprecated fields:

• Cluster
  • spec.topology.rolloutAfter
  • spec.topology.variables[].definitionFrom
  • spec.topology.controlPlane.variables.overrides[].definitionFrom
  • spec.topology.workers.machineDeployments[].variables.overrides[].definitionFrom
  • spec.topology.workers.machinePools[].variables.overrides[].definitionFrom
• ClusterClass
  • spec.variables[].metadata
• MachineDeployment
  • spec.revisionHistoryLimit
• MachineHealthCheck
  • spec.maxUnhealthy
  • spec.unhealthyRange
• KubeadmConfig (also KubeadmConfigTemplate, KubeadmControlPlane, KubeadmControlPlaneTemplate)
  • spec.useExperimentalRetryJoin
• DockerMachine
  • .spec.bootstrapped

and considering e.g.:

• Implement support for kubeadm v1beta4 API #10708

we may introduce more in future.

For providing a user the information that it makes use of a deprecated field, we could use warnings in validation to bubble that information up.

Anything else you would like to add?

No response

Label(s) to be applied

/kind feature
One or more /area label. See https://github.com/kubernetes-sigs/cluster-api/labels?q=area for the list of labels.

[sbueringer]

Yup also

• Deprecate revision management in MachineDeployments #10479
• Deprecate definitionFrom in cluster variables #10666
• ✨ Add metadata for top-level and nested variables & deprecate previous metadata field #10778
• experimentalRetryJoin

I think this is a great idea, and we should do this for all fields now & going forward

[fabriziopandini]

/triage accepted
/help

we can evaluate case by case on follow up PRs

[k8s-ci-robot]

@fabriziopandini:
This request has been marked as needing help from a contributor.

Guidelines

Please ensure that the issue body includes answers to the following questions:

• Why are we solving this issue?
• To address this issue, are there any code changes? If there are code changes, what needs to be done in the code and what places can the assignee treat as reference points?
• Does this issue have zero to low barrier of entry?
• How can the assignee reach out to you for help?

For more details on the requirements of such an issue, please see here and ensure that they are met.

If this request no longer meets these requirements, the label can be removed
by commenting with the /remove-help command.

In response to this:

/triage accepted
/help

we can evaluate case by case on follow up PRs

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[chrischdi]

Yup also

• Consider deprecation of revision management in MachineDeployments #10479
• Deprecated definitionFrom in cluster variables #10666
• ✨ Add metadata for top-level and nested variables & deprecate previous metadata field #10778
• experimentalRetryJoin

I think this is a great idea, and we should do this for all fields now & going forward

moved these up to the list

[sbueringer]

Updated the list in the issue description & structured by CRD

[sbueringer]

A few general comments from my side:

• I think we should be careful with fields that are set per default. I think it's a bad idea to emit warnings for fields that we default, because then it's not actually the user/controller/client setting them and every single user/controller/client will get the warnings (they will probably (afaik) also show up in our own core CAPI controller logs)
• I think in any case it would be good to have a flag to opt-out from this feature. I think if some folks are using Cluster API within their products it's totally fine if they are not forced to emit these warnings to their end users. So there should be away to avoid spaming everyone with warnings if folks are aware of the deprecations