Nick Young
Nick Young
Sorry about the delay @mghlaiel, if you could write something up, that would be great! One of us should be able to assist you with putting the guide into the...
The change basically lgtm, but you'll need to remove the "fixes ..." from the commit message @danehans, Prow doesn't let us do that in commit messages.
The docs don't say that the GatewayClass needs to not be accepted, just that the unsupportedVersion Condition needs to be added. The intent here is to give implementations a way...
The idea behind all of this is to _teach_ users to check the GatewayClass to find out what their controller supports (this is the point of the `status.supportedFeatures` field after...
@howardjohn, I know you don't like GatewayClass because it's cluster scoped, but the _intent_ behind GatewayClass was always: * provide a way for implementations to own some subset of Gateways...
I had a chat with some folks in #sig-apimachinery about this (see https://kubernetes.slack.com/archives/C0EG7JC6T/p1733285894738669), and the consensus was that, while it's important to be _careful_ about watching CRDs, it's okay in...
This is not a Cilium bug - Envoy does not allow this header to be changed using header editing operations. In order to remove this, we would need to add...
Hmm, I had forgotten about this. The implementer's guide has now been, well, implemented, and is ready for someone to take the table that's currently in GEP-1911 (at https://gateway-api.sigs.k8s.io/geps/gep-1911/#supporting-protocols) and...
LGTM, thanks @dprotaso /approve