GKE Nginx Value Issue

[otavioss]

The ingress generated by GKE recommended values does not work, and that server-snippet is too complex and causes a lot of intermittent 404:

boldbi-kubernetes/helm/boldbi/templates/ingress.yaml

Lines 32 to 40 in 26c4920

	nginx.ingress.kubernetes.io/server-snippet: |
	client_max_body_size 200m;
	{{- if eq .Release.Namespace "default" }}
	set $bold_namespace "{{ .Values.namespace }}";
	{{- else }}
	set $bold_namespace "{{ .Release.Namespace }}";
	{{- end }}
	location ~* "^/bi/api(/|$)(.*)" {

The helm template does not allow changing the ingress class:

boldbi-kubernetes/helm/bold-common/templates/ingress.yaml

Line 16 in 7539a7a

kubernetes.io/ingress.class: "nginx"

boldbi-kubernetes/helm/boldbi/templates/ingress.yaml

Lines 510 to 512 in 26c4920

	{{- if and $ingressSupportsIngressClassName .Values.loadBalancer.ingressClassName }}
	ingressClassName: {{ .Values.loadBalancer.ingressClassName }}
	{{- end }}

Here is a sample of a ingress that works

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: bold-ingress-
  namespace: bold-bi
  annotations:
    nginx.ingress.kubernetes.io/affinity: cookie
    nginx.ingress.kubernetes.io/session-cookie-hash: sha1
    nginx.ingress.kubernetes.io/session-cookie-name: "bold.k8s.pod.id"
    nginx.ingress.kubernetes.io/session-cookie-max-age: '600'
    nginx.ingress.kubernetes.io/use-regex: 'true'
    nginx.ingress.kubernetes.io/proxy-buffer-size: "128k"
    nginx.ingress.kubernetes.io/proxy-buffers: "256k"
    nginx.ingress.kubernetes.io/proxy-busy-buffers-size: "256k"
    nginx.ingress.kubernetes.io/large-client-header-buffers: "16k"
    nginx.ingress.kubernetes.io/fastcgi-buffers: "16k"
    nginx.ingress.kubernetes.io/fastcgi-buffer-size: "32k"
    nginx.ingress.kubernetes.io/proxy-connect-timeout: "300"
    nginx.ingress.kubernetes.io/proxy-read-timeout: "300"
    nginx.ingress.kubernetes.io/proxy-send-timeout: "300"
spec:
  ingressClassName: example
  tls:
  - hosts:
    - bold-bi.example.com
    - www.bold-bi.example.com
    secretName: bold-bi-tls
  rules:
  - host: bold-bi.example.com
    http:
      paths:
      - path: "/bi/api/*"
        pathType: ImplementationSpecific
        backend:
          service:
            name: bi-api-service
            port:
              number: 6005
      - path: "/bi/jobs/*"
        pathType: ImplementationSpecific
        backend:
          service:
            name: bi-jobs-service
            port:
              number: 6006
      - path: "/bi/designer/*"
        pathType: ImplementationSpecific
        backend:
          service:
            name: bi-dataservice-service
            port:
              number: 6007
      - path: "/bi/*"
        pathType: ImplementationSpecific
        backend:
          service:
            name: bi-web-service
            port:
              number: 6004
      - path: "/api/*"
        pathType: ImplementationSpecific
        backend:
          service:
            name: id-api-service
            port:
              number: 6001
      - path: "/ums/*"
        pathType: ImplementationSpecific
        backend:
          service:
            name: id-ums-service
            port:
              number: 6002
      - path: "/*"
        pathType: ImplementationSpecific
        backend:
          service:
            name: id-web-service
            port:
              number: 6000

The ETL service needs to be created separately due to the rewrite target, if added together the application will return 502:

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
 name: bold-ingress-etl
 namespace: bold-bi
 annotations:
   nginx.ingress.kubernetes.io/affinity: cookie
   nginx.ingress.kubernetes.io/session-cookie-hash: sha1
   nginx.ingress.kubernetes.io/session-cookie-name: "bold.k8s.pod.id"
   nginx.ingress.kubernetes.io/session-cookie-max-age: '600'
   nginx.ingress.kubernetes.io/use-regex: 'true'
   nginx.ingress.kubernetes.io/proxy-buffer-size: "128k"
   nginx.ingress.kubernetes.io/proxy-buffers: "256k"
   nginx.ingress.kubernetes.io/proxy-busy-buffers-size: "256k"
   nginx.ingress.kubernetes.io/large-client-header-buffers: "16k"
   nginx.ingress.kubernetes.io/fastcgi-buffers: "16k"
   nginx.ingress.kubernetes.io/fastcgi-buffer-size: "32k"
   nginx.ingress.kubernetes.io/proxy-connect-timeout: "300"
   nginx.ingress.kubernetes.io/proxy-read-timeout: "300"
   nginx.ingress.kubernetes.io/proxy-send-timeout: "300"
   nginx.ingress.kubernetes.io/rewrite-target: /$2
spec:
 ingressClassName: example
 tls:
 - hosts:
   - bold-bi.example.com
   - www.bold-bi.example.com
   secretName: bold-bi-tls
 rules:
 - host: bold-bi.example.com
   http:
     paths:
     - path: "/etlservice(/|$)(.*)"
       pathType: ImplementationSpecific
       backend:
         service:
           name: bold-etl-service
           port:
             number: 6009

[sivakumar-devops]

Hi @otavioss,
Currently, we are addressing the concerns regarding the nginx Ingress file. We will update you once we have completed the changes.