Render service

Basic structure

To render our Kubernetes services, Prefapp-helm uses ph.service.render, which expects the following object structure:

name: <name_of_the_deployment>
selector:  <selectors>

ports: 
  - service: 80 #exposed port
    pod: 80 #target

Prefapp-helm-render uses the same nomenclature as the standard k8s artifact, with the exception of the ports, this allows us to use the same Values declaration for both the service and the deployment.

Given this values.yaml:

#values.yaml
frontend_ports:
  - service: 80 #exposed port
    pod: 80  #target

We can have the same line in the service:

#service.yaml
name: service-{{ .Release.Name }}-php
selector:  {{ .Values.selector_meiga | toYaml | nindent 2}}
  
ports: {{ .Values.frontend_ports | toYaml | nindent 6 }}

As well as the deployment:

#deploy.yaml
name: deploy-{{ .Release.Name }}-php
selector: {{ .Values.selector_meiga | toYaml | nindent 2}}

containers:
- name: app-php
  ...
  ports: {{ .Values.frontend_ports | toYaml | nindent 6 }}

Example of ph.service.render

{{ define "frontend.service.data"  }}
name: servizo-{{ .Release.Name }}-php
selector:  {{ .Values.selector_meiga | toYaml | nindent 2}}
  
ports: {{ .Values.ports | toYaml | nindent 6 }}

{{- end }}
{{ include "ph.service.render" (include "frontend.service.data" . | fromYaml )  }}