Bringing kube-bind under SIG-MC

[Stephen Kitt]

Hi everyone,

As discussed last week, there is some interest in bringing kube-bind
(<https://kube-bind.io/>) under SIG-Multicluster, providing a home
both for the project and ultimately for the relevant spec (if one
should emerge).

There doesn’t seem to be any other similar project; kro
(<https://kro.run>) was mentioned but doesn’t seem to target multiple
clusters, and has different use cases than kube-bind. Any spec
emerging from kube-bind would also be likely to fit naturally into
SIG-MC, and be amenable to implementation by multicluster controllers
(not just kube-bind).

See also the slides shared during the call
(<https://docs.google.com/presentation/d/1a5QqiaJNpu0UgyL50T0kLGg9-gaeF-1yqW3FZThlhnU/edit>)
and the GitHub project (<https://github.com/kube-bind/kube-bind>).

Any thoughts or concerns?

Regards,

--
Stephen Kitt
Senior Principal Software Engineer
Red Hat OpenShift Networking

[Tim Hockin]

I didn't get to see the demo.  I think I get what it is doing from the slides but it's not exactly clear why MC is the right home?

--
You received this message because you are subscribed to the Google Groups "kubernetes-sig-multicluster" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kubernetes-sig-mult...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/kubernetes-sig-multicluster/aHZ1K_7OkPMm7Onn%40redhat.com.

[ryan zhang]

I think I would like to clarify that I would not think hosting the kube-binding in the MC mean that the sig endorse either the API nor the implementation of it. It is different from the rest of the projects hosted in the sig-MC org. I think the real question is what is the sig’s policy in hosting projects like this.

[Hongcai Ren]

I have the same question as Tim, Why do you think SIG-Multicluster is the right place?

In addition, I remember the SIG-Multicluster clearly clarified that it would no longer develop any sub-projects other than building blocks/APIs. Are we going to change that?

To view this discussion visit https://groups.google.com/d/msgid/kubernetes-sig-multicluster/D7775638-17AD-4DA3-BA28-0317FADF49CB%40gmail.com.

[Marvin Beckers]

Hi folks,

We brought kube-bind to SIG Multicluster since kube-bind is by design a multi-cluster project, sharing APIs across cluster boundaries (with one cluster being sort of an CRD-based API "provider" and the other being the API "consumer"). While not exactly the same, we saw some (very high-level) similarities with MCS (with MCS being about Service consumption across clusters and kube-bind being about API consumption across clusters). While kube-bind is more "mesh-y" than other projects under SIG Multicluster we thought it might be aligned with the SIG and added to its toolbox of using/managing/connecting multiple clusters.

While kube-bind has a reference implementation in its repo, the focus for the project is also on getting API definitions right. Eventually, we expect other implementations to crop up, but the project built an initial implementation to be more tangible.

Thanks,

Marvin

To view this discussion visit https://groups.google.com/d/msgid/kubernetes-sig-multicluster/CAJ473nGoJBPcTNYy0Jt1a0Ny0E%2BhnT3eAY6S5vVbh%2BVhkpmU9Q%40mail.gmail.com.