controller-tools +kubebuilder:default={} does not use nested defaults

So I believe I exposed a bug in kubebuilder...

when I define a +kubebuilder:default={} as a pointer to another type with +kubebuilder:default values I assume those values should propagate to the parent object.

For example

// policyAuditConfig is the configuration for network policy audit events. If unset,
// reported defaults are used.
// +kubebuilder:default={}
// +optional
PolicyAuditConfig *PolicyAuditConfig `json:"policyAuditConfig,omitempty"`

Where the PolicyAuditConfig is defined in the same package

type PolicyAuditConfig struct {
	// rateLimit is the approximate maximum number of messages to generate per-second per-node. If
	// unset the default of 20 msg/sec is used.
	// +kubebuilder:default=20
	// +kubebuilder:validation:Minimum=1
	// +optional
	RateLimit *uint32 `json:"rateLimit,omitempty"`

	// maxFilesSize is the max size an ACL_audit log file is allowed to reach before rotation occurs
	// Units are in MB and the Default is 50MB
	// +kubebuilder:default=50
	// +kubebuilder:validation:Minimum=1
	// +optional
	MaxFileSize *uint32 `json:"maxFileSize,omitempty"`

	// destination is the location for policy log messages.
	// Regardless of this config, persistent logs will always be dumped to the host
	// at /var/log/ovn/ however
	// Additionally syslog output may be configured as follows.
	// Valid values are:
	// - "libc" -> to use the libc syslog() function of the host node's journdald process
	// - "udp:host:port" -> for sending syslog over UDP
	// - "unix:file" -> for using the UNIX domain socket directly
	// - "null" -> to discard all messages logged to syslog
	// The default is "null"
	// +kubebuilder:default=null
	// +kubebuilder:pattern='^libc$|^null$|^udp:(([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5])\.){3}([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5]):([0-9]){0,5}$|^unix:(\/[^\/ ]*)+([^\/\s])$'
	// +optional
	Destination string `json:"destination,omitempty"`

	// syslogFacility the RFC5424 facility for generated messages, e.g. "kern". Default is "local0"
	// +kubebuilder:default=local0
	// +kubebuilder:Enum=kern;user;mail;daemon;auth;syslog;lpr;news;uucp;clock;ftp;ntp;audit;alert;clock2;local0;local1;local2;local3;local4;local5;local6;local7
	// +optional
	SyslogFacility string `json:"syslogFacility,omitempty"`
}

However rather than rendering the default correctly I see the following in the generated yaml

policyAuditConfig:
  description: policyAuditConfig is the configuration for network policy audit events. If unset, reported defaults are used.
  type: object
  default: null
  properties:
    destination:
      description: 'destination is the location for policy log messages. Regardless of this config, persistent logs will always be dumped to the host at /var/log/ovn/ however Additionally syslog output may be configured as follows. Valid values are: - "libc" -> to use the libc syslog() function of the host node''s journdald process - "udp:host:port" -> for sending syslog over UDP - "unix:file" -> for using the UNIX domain socket directly - "null" -> to discard all messages logged to syslog The default is "null"'
      type: string
      default: "null"
    maxFileSize:
      description: maxFilesSize is the max size an ACL_audit log file is allowed to reach before rotation occurs Units are in MB and the Default is 50MB
      type: integer
      format: int32
      default: 50
      minimum: 1
    rateLimit:
      description: rateLimit is the approximate maximum number of messages to generate per-second per-node. If unset the default of 20 msg/sec is used.
      type: integer
      format: int32
      default: 20
      minimum: 1
    syslogFacility:
      description: syslogFacility the RFC5424 facility for generated messages, e.g. "kern". Default is "local0"
      type: string
      default: local0

Specifically the default value is null where it should be populated automatically by the defined nested defaults.

I am happy to help propose a fix here, however after spending some time playing with the codebase I'm realizing I will probably need some assistance from someone more experienced with kubebuilder

I already reached out on slack and got no response, which prompted me to open this issue.

Thanks, Andrew

Sep 21 '21 21:09 astoycos

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Mark this issue or PR as fresh with /remove-lifecycle stale
Mark this issue or PR as rotten with /lifecycle rotten
Close this issue or PR with /close
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

Dec 20 '21 22:12 k8s-triage-robot

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Mark this issue or PR as fresh with /remove-lifecycle rotten
Close this issue or PR with /close
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

Jan 19 '22 22:01 k8s-triage-robot

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue or PR with /reopen
Mark this issue or PR as fresh with /remove-lifecycle rotten
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

Feb 18 '22 22:02 k8s-triage-robot

@k8s-triage-robot: Closing this issue.

In response to this:

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied

After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied

After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue or PR with /reopen

Mark this issue or PR as fresh with /remove-lifecycle rotten

Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

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/test-infra repository.

Feb 18 '22 22:02 k8s-ci-robot

/reopen

Feb 23 '22 00:02 astoycos

@astoycos: Reopened this issue.

In response to this:

/reopen

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/test-infra repository.

Feb 23 '22 00:02 k8s-ci-robot

/bug

Feb 23 '22 00:02 astoycos

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue or PR with /reopen
Mark this issue or PR as fresh with /remove-lifecycle rotten
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

Mar 25 '22 00:03 k8s-triage-robot

@k8s-triage-robot: Closing this issue.

In response to this:

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied

After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied

After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue or PR with /reopen

Mark this issue or PR as fresh with /remove-lifecycle rotten

Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

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/test-infra repository.

Mar 25 '22 00:03 k8s-ci-robot

/reopen

Apr 30 '22 16:04 therealak12

@therealak12: You can't reopen an issue/PR unless you authored it or you are a collaborator.

In response to this:

/reopen

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/test-infra repository.

Apr 30 '22 16:04 k8s-ci-robot

/reopen

May 01 '22 16:05 astoycos

@astoycos: Reopened this issue.

In response to this:

/reopen

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/test-infra repository.

May 01 '22 16:05 k8s-ci-robot

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue or PR with /reopen
Mark this issue or PR as fresh with /remove-lifecycle rotten
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

May 31 '22 16:05 k8s-triage-robot

@k8s-triage-robot: Closing this issue.

In response to this:

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied

After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied

After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue or PR with /reopen

Mark this issue or PR as fresh with /remove-lifecycle rotten

Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

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/test-infra repository.

May 31 '22 16:05 k8s-ci-robot

/reopen

Jul 08 '22 19:07 astoycos

@astoycos: Reopened this issue.

In response to this:

/reopen

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/test-infra repository.

Jul 08 '22 19:07 k8s-ci-robot

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue or PR with /reopen
Mark this issue or PR as fresh with /remove-lifecycle rotten
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

Aug 07 '22 20:08 k8s-triage-robot

@k8s-triage-robot: Closing this issue.

In response to this:

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied

After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied

After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue or PR with /reopen

Mark this issue or PR as fresh with /remove-lifecycle rotten

Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close

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/test-infra repository.

Aug 07 '22 20:08 k8s-ci-robot

This issue seems not fixed?

Dec 06 '22 09:12 liubog2008

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue with /reopen
Mark this issue as fresh with /remove-lifecycle rotten
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close not-planned

Jan 05 '23 10:01 k8s-triage-robot

@k8s-triage-robot: Closing this issue, marking it as "Not Planned".

In response to this:

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues according to the following rules:

After 90d of inactivity, lifecycle/stale is applied

After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied

After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue with /reopen

Mark this issue as fresh with /remove-lifecycle rotten

Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close not-planned

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/test-infra repository.

Jan 05 '23 10:01 k8s-ci-robot

This is still an issue. What "works" as a sort of workaround is specifying the default verbatim in every +kubebuilder:default marker, for instance:

type XStrategy struct {
	// +kubebuilder:validation:Optional
	// +kubebuilder:default={"plan":{"deployment":"ScaleDownOnPromotionScaleUpOnRollout"}}
	Resources RolloutResources `json:"resources,omitempty"`
}

type RolloutResources struct {
	// +kubebuilder:validation:Optional
	// +kubebuilder:default={"deployment":"ScaleDownOnPromotionScaleUpOnRollout"}
	Plan RolloutResourcePlans `json:"plan,omitempty"`
}


type RolloutResourcePlans struct {
	// +kubebuilder:validation:Enum=ScaleDownOnPromotionScaleUpOnRollout;DeleteOnPromotionRecreateOnRollout
	// +kubebuilder:default=ScaleDownOnPromotionScaleUpOnRollout
	Deployment RolloutResourcePlanDeployment `json:"deployment,omitempty"`
}

which is error prone and counterproductive. Field based defaults should be enough.

/reopen

Aug 23 '23 15:08 pmalek

@pmalek: You can't reopen an issue/PR unless you authored it or you are a collaborator.

In response to this:

This is still an issue. What "works" as a sort of workaround is specifying the default verbatim in every +kubebuilder:default marker, for instance:

type XStrategy struct {
  // +kubebuilder:validation:Optional
  // +kubebuilder:default={"plan":{"deployment":"ScaleDownOnPromotionScaleUpOnRollout"}}
  Resources RolloutResources `json:"resources,omitempty"`
}

type RolloutResources struct {
  // +kubebuilder:validation:Optional
  // +kubebuilder:default={"deployment":"ScaleDownOnPromotionScaleUpOnRollout"}
  Plan RolloutResourcePlans `json:"plan,omitempty"`
}


type RolloutResourcePlans struct {
  // +kubebuilder:validation:Enum=ScaleDownOnPromotionScaleUpOnRollout;DeleteOnPromotionRecreateOnRollout
  // +kubebuilder:default=ScaleDownOnPromotionScaleUpOnRollout
  Deployment RolloutResourcePlanDeployment `json:"deployment,omitempty"`
}

which is error prone and counterproductive. Field based defaults should be enough.

/reopen

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/test-infra repository.

Aug 23 '23 15:08 k8s-ci-robot

/reopen

Aug 23 '23 15:08 astoycos

@astoycos: Reopened this issue.

In response to this:

/reopen

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/test-infra repository.

Aug 23 '23 15:08 k8s-ci-robot

/remove-lifecycle rotten

Aug 23 '23 15:08 astoycos

This issue has been here for over two years now 😵 Does anyone plan to work on it?

Oct 04 '23 12:10 franpog859

I just experienced the same behavior. I've used https://github.com/kubernetes-sigs/controller-tools/issues/622#issuecomment-1690128862 as a workaround.

Dec 18 '23 10:12 morucci

/lifecycle frozen /kind bug

Jan 12 '24 18:01 ahmetb