Request for Stable 1.0.0 Release of controller-runtime

[camilamacedo86]

We deeply appreciate the maintainers' and community's excellent work and contributions behind the controller-runtime project. This library is a critical part of the Kubernetes ecosystem and has proven invaluable for building reliable controllers and operators.

However, the lack of a stable 1.0.0 release introduces challenges for projects relying on controller-runtime. A stable release would signal API stability, foster greater confidence, and encourage broader adoption.

Why a Stable 1.0.0 Release?

• Maturity: The library has undergone significant improvements and has proven reliable across diverse production use cases.
  Adoption: Kubebuilder and other frameworks widely use it, and its importance in the Kubernetes ecosystem is evident.
• API Stability: A 1.0.0 release would signal a commitment to backward compatibility and long-term support. It would help reduce the tool in the users and projects that relay on it.

Therefore, we would like to ask the maintainers to consider this RFE to help solidify controller-runtime as a trusted foundation for building ecosystem solutions.

Thank you for your dedication and for considering this request.
We are excited to participate in further talks with the community.

[camilamacedo86]

Dear folks,

I remember that this was discussed around 3 years ago, but it did not fly at the time.
Would it be possible to reconsider this effort?
Have we tracked everything we would like to address before making a stable release?
Could we organize all the required changes under a milestone to see if the community could help out with it?

WDYT?

[JoelSpeed]

I agree that there would be benefit to the community to start formalising a stable version. I'd be keen to see if we can enumerate all of the outstanding pain points that we would like to work on, and identify which require API changes, and therefore which we would need to solidify prior to that v1 release

[sbueringer]

Please see the previous discussion in #2327

I think one huge point is that we essentially need a stable release of client-go and the other k8s.io/* libs that we depend on. As long as that's not the case I don't see how we can guarantee stability.

[camilamacedo86]

Hi @sbueringer

Thank you. So, I understand that convey here is: #2327 (comment)

As you point out above. Since we have #2327, and you provided the answer above, I think it is OK to close this one. Thank you for your time and attention.