[Feature]: Allow to configure a different port for service in Helm chart

[hpedrorodrigues]

The Feature

Allow configuration of a port in the service that differs from the port in the deployment in the Helm chart.

At the moment, the same configuration is used to set the port in both the service and deployment.

litellm/deploy/charts/litellm-helm/values.yaml

Lines 56 to 58 in 61d7704

	service:
	type: ClusterIP
	port: 4000

litellm/deploy/charts/litellm-helm/templates/service.yaml

Lines 13 to 14 in 61d7704

	ports:
	- port: {{ .Values.service.port }}

litellm/deploy/charts/litellm-helm/templates/deployment.yaml

Lines 43 to 44 in 61d7704

	- name: PORT
	value: {{ .Values.service.port | quote}}

litellm/deploy/charts/litellm-helm/templates/deployment.yaml

Lines 115 to 117 in 61d7704

	ports:
	- name: http
	containerPort: {{ .Values.service.port }}

Motivation, pitch

With this change, port 80 for the service can be set, while any desired port can be set in the deployment. This will make the configuration for clients easier since they won't need to worry about the port in the URI.

Are you a ML Ops Team?

Yes

Twitter / LinkedIn details

No response

[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.

[hpedrorodrigues]

I can work on it if it makes sense for the project. Please let me know your thoughts.

[wkbaran]

I'd appreciate this change. It's not core to the functionality and improves interoperability in different environments.

[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.

[hpedrorodrigues]

(adding this comment to remove the stale label)