gateway-api
gateway-api copied to clipboard
BackendObjectReference port should be compatible with named port
What would you like to be added: BackendObjectReference port should be compatible with named port
Why this is needed:
As user of the Gateway API, I'm forced to use the port number defined by the backend service. However, this port may change and I have to sync this manually and/or automate a check for it.
In Ingress API, we have a specific API, where the backend.service.port.name can be used to avoid a direct relationship to a port.
This port.name is a useful indirection allowing to separate the implementation (port.number) from the purpose (port.name).
With the GatewayAPI, we don't have this concept anymore.
The Kubernetes project currently lacks enough contributors to adequately respond to all issues.
This bot triages un-triaged issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Mark this issue as fresh with
/remove-lifecycle stale - Close this issue with
/close - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale
/unstale 😇
The Kubernetes project currently lacks enough active contributors to adequately respond to all issues.
This bot triages un-triaged issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Mark this issue as fresh with
/remove-lifecycle rotten - Close this issue with
/close - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle rotten
@shaneutt, is it possible to remove the lifecycle for it, please?
Yes, albeit nobody has stepped up to drive it forward, it's still technically in triage so we need to check on it still.
For the record, this was an intentional omission of the API. Supporting named ports adds more complexity for both implementers and users and generally just introduces more ways that things can break. For anyone interested in pushing this forward, you would need to show that the value of providing this option outweighed the complexity it introduces.
I worked on that for few weeks and I understand the reasoning.
Feel free to close it, at least we have an issue answering this question for future people doing the same migration as me… and using a lot named port.
The Kubernetes project currently lacks enough contributors to adequately respond to all issues.
This bot triages un-triaged issues according to the following rules:
- After 90d of inactivity,
lifecycle/staleis applied - After 30d of inactivity since
lifecycle/stalewas applied,lifecycle/rottenis applied - After 30d of inactivity since
lifecycle/rottenwas applied, the issue is closed
You can:
- Mark this issue as fresh with
/remove-lifecycle stale - Close this issue with
/close - Offer to help out with Issue Triage
Please send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale