Support Headless Services for Deployments as well

[lorenzo-biava]

Please describe the Improvement and/or Feature Request

Support for K8s Headless services has been introduced with #3477

However the same kind of support would be helpful for K8s Deployments as well.

To avoid overloading the generated Envoy config and increase resource consumption (most likely it needs to be a Listener per backend Pod?), the feature should be made opt-in per K8s Service, probably leveraging an annotation.

Scope (please mark with X where applicable)

• New Functionality [x]
• Install [ ]
• SMI Traffic Access Policy [ ]
• SMI Traffic Specs Policy [ ]
• SMI Traffic Split Policy [ ]
• Permissive Traffic Policy [ ]
• Ingress [ ]
• Egress [ ]
• Envoy Control Plane [ ]
• CLI Tool [ ]
• Metrics [ ]
• Certificate Management [ ]
• Sidecar Injection [ ]
• Logging [ ]
• Debugging [ ]
• Tests [ ]
• CI System [ ]
• Demo [ ]
• Project Release [ ]

Possible use cases

There are some scenarios where an application requires custom routing of requests to downstream Pods, without relying on the load balancing offered by the Mesh (e.g. when using extra contextual information, and only leveraging the Mesh for mTLS and observability features).

[lorenzo-biava]

@keithmattix FYI

[lorenzo-biava]

Looks like it's been implemented by #5250

[steeling]

marking this closed by the above comment