Maximum set Service IDs being increased

[hox]

I am running some large clusters with 65,535+ service IDs across all services and ports. New services will not get added to Cilium's LB map due to a hard limit set here: https://github.com/cilium/cilium/blob/main/pkg/service/const.go#L12

The error shown in agent logs:

level=error msg="Error while inserting service in LB map" error="Unable to allocate service ID 0 for {{xxx.xxx.xxx.xxx {TCP 20261} 0} 0}: no service ID available" k8sNamespace==****** k8sSvcName=****** subsys=k8s-watcher.

Ideally, the limit for the service IDs should be increased past the uint16 maximum, and the code is switched to use uint32 types to increase the limits on how many ports you can have across a cluster.

This would allow for larger clusters being able to support 16,777,215 service IDs, instead of only 65,535.

[github-actions]

This issue has been automatically marked as stale because it has not
had recent activity. It will be closed if no further activity occurs.

[hox]

bump

[github-actions]

This issue has been automatically marked as stale because it has not
had recent activity. It will be closed if no further activity occurs.

[github-actions]

This issue has not seen any activity since it was marked stale.
Closing.