[EKS Controller] constantly spamming in logs

[gecube]

with

{"level":"info","ts":"2025-04-29T08:28:29.342Z","logger":"ackrt","msg":"desired resource state has changed","kind":"Nodegroup","namespace":"infra-production","name":"production-fix-ng-20250413-c","account":"*****","role":"arn:aws:iam::*****:role/ack-ec2-controller-tooling","region":"eu-west-2","is_adopted":false,"generation":2,"diff":[{"Path":{"Parts":["Spec","Taints"]},"A":[{"effect":"NO_SCHEDULE","key":"node-role.kubernetes.io/fix","value":""}],"B":[{"effect":"NO_SCHEDULE","key":"node-role.kubernetes.io/fix"}]}]}

the source manifest:

apiVersion: eks.services.k8s.aws/v1alpha1
kind: Nodegroup
metadata:
  name: production-fix-ng-20250413-c
  namespace: infra-production
spec:
  name: production-fix-ng-20250413-c
  clusterName: production-new
  diskSize: 100
  subnetRefs:
    - from:
        name: production-private-eu-west-2c
  nodeRole: arn:aws:iam::*****:role/eks-production-nodegroup-role
  scalingConfig:
    minSize: 1
    maxSize: 1
    desiredSize: 1
  instanceTypes:
    - m5.4xlarge
  taints:
    - key: "node-role.kubernetes.io/fix"
      value: ""
      effect: "NO_SCHEDULE"
  labels: {}
  amiType: BOTTLEROCKET_x86_64

[github-actions]

Hello @gecube 👋 Thank you for opening an issue in ACK! A maintainer will triage this issue soon.

We encourage community contributions, so if you're interested in tackling this yourself or suggesting a solution, please check out our Contribution and Code of Conduct guidelines.

You can find more information about ACK on our website.

[gecube]

looks like fixed with changing the manifest to:

apiVersion: eks.services.k8s.aws/v1alpha1
kind: Nodegroup
metadata:
  name: production-fix-ng-20250413-c
  namespace: infra-production
spec:
  name: production-fix-ng-20250413-c
  clusterName: production-new
  diskSize: 100
  subnetRefs:
    - from:
        name: production-private-eu-west-2c
  nodeRole: arn:aws:iam::*****:role/eks-production-nodegroup-role
  scalingConfig:
    minSize: 1
    maxSize: 1
    desiredSize: 1
  instanceTypes:
    - m5.4xlarge
  taints:
    - key: "node-role.kubernetes.io/fix"
#######      value: "" ##### removed
      effect: "NO_SCHEDULE"
  labels: {}
  amiType: BOTTLEROCKET_x86_64

[michaelhtm]

Hi @gecube
What would you like the desired outcome to be here?

[gecube]

@michaelhtm Hi! Thanks for asking. I see the next options which could be combined:

• make a proper documentation which parameters are mandatory and which are optional and provide an examples how to make a NG with a taints.
• allow empty value in value: "" effectively meaning none value
• disallow empty value in value: "" with admission controller (as it makes no sense and it could be clearly seen from the example above)
• somehow ignore the check (because the controller spams as it tries to apply something against cluster and there is a drift) or make some analogue of dry-run which would properly handle such a things...

[ack-bot]

Issues go stale after 180d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 60d of inactivity and eventually close.
If this issue is safe to close now please do so with /close.
Provide feedback via https://github.com/aws-controllers-k8s/community.
/lifecycle stale

[gecube]

/remove-lifecycle stale