[sophora-ugc] [sophora-ugc-proxy] add ServiceMonitors (#304)

* [ugc] add ServiceMonitor objects for UGC and UGC multimedia

* [ugc-proxy] add ServiceMonitor

* [ugc] add per service config for serviceMonitor and remove management port

* [ugc] rename "jolokia" service to "management" across chart files

* [ugc] allow monitoring via management service

* [ugc] deploy ServiceMonitor only if UGC multimedia is enabled

* [ugc] add health probes configuration

* [ugc] make management service optional and configure port

* [ugc] move gcp config to ugc and ugcMultimedia sections and make it adjustable per service

* [ugc] refactor ServiceMonitor target selection

* [ugc] adjust gcp backend policy test

* [ugc] revert test changes to gcp health

* [ugc] disable ServiceMonitor by default

* [ugc] move management service configuration to toplevel

* [ugc] improve docs

* [ugc] add changelog

* Update explaining comment

Co-authored-by: Jonas Pruditsch <6641029+kunstkomputer@users.noreply.github.com>

* Update explaining comment

Co-authored-by: Jonas Pruditsch <6641029+kunstkomputer@users.noreply.github.com>

* Update explaining comment

Co-authored-by: Jonas Pruditsch <6641029+kunstkomputer@users.noreply.github.com>

* Update explaining comment

Co-authored-by: Jonas Pruditsch <6641029+kunstkomputer@users.noreply.github.com>

---------

Co-authored-by: Jan Stubenrauch <129169967+js-sub@users.noreply.github.com>
Co-authored-by: Jonas Pruditsch <6641029+kunstkomputer@users.noreply.github.com>

Author: 3 people

charts/sophora-ugc-proxy/Chart.yaml

@@ -4,11 +4,11 @@ description: A Helm chart for Kubernetes to install Sophora UGC Proxy connected
 
 type: application
 
-version: 1.6.0
+version: 1.7.0
 annotations:
   artifacthub.io/changes: |
     - kind: added
-      description: "Add support for supplying API basic-auth credentials from a Kubernetes Secret"
+      description: "ServiceMonitor object for UGC proxy"
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to

charts/sophora-ugc-proxy/templates/service.yaml

@@ -13,6 +13,7 @@ spec:
     {{- include "sophora-ugc-proxy.selectorLabels" . | nindent 4 }}
   type: {{ .Values.service.type }}
   ports:
-    - protocol: TCP
+    - name: http
+      protocol: TCP
       targetPort: proxy
       port: {{ .Values.service.port }}

charts/sophora-ugc-proxy/templates/servicemonitor.yaml

@@ -0,0 +1,15 @@
+{{- if .Values.serviceMonitor.enabled -}}
+{{- $fullName := include "sophora-ugc-proxy.fullname" . -}}
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+    name: {{ $fullName }}
+    labels: {{- include "sophora-ugc-proxy.labels" . | nindent 8 }}
+spec:
+    selector:
+        matchLabels: {{- include "sophora-ugc-proxy.selectorLabels" . | nindent 12 }}
+    endpoints:
+        - interval: {{ .Values.serviceMonitor.interval }}
+          port: http
+          path: {{ .Values.serviceMonitor.path }}
+{{- end }}

charts/sophora-ugc-proxy/values.yaml

@@ -227,3 +227,12 @@ gcp:
 
 # extraDeploy is a list of full Kubernetes manifests that will be rendered additionally by this chart
 extraDeploy: []
+
+
+# serviceMonitor.enabled Whether the serviceMonitor resource should be deployed
+# serviceMonitor.interval Prometheus scrape interval
+# serviceMonitor.path HTTP path prometheus should use to scrape the application's metrics
+serviceMonitor:
+  enabled: false
+  interval: 10s
+  path: "/actuator/prometheus"

charts/sophora-ugc/Chart.yaml

@@ -15,11 +15,17 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 2.10.4
+version: 2.11.0
 annotations:
   artifacthub.io/changes: |
-    - kind: fixed
-      description: "fix linting issue on enabled GCPHealthChecks and GCPBackendPolicies"
+    - kind: added
+      description: "ServiceMonitor objects for UGC and UGC multimedia"
+    - kind: changed
+      description: "The parameter `service.jolokia.clusterIP` is now `service.management.clusterIP`. The deployed service and the corresponding container port are now named `management` instead of `jolokia`."
+    - kind: changed
+      description: "The `gcp` parameters can now be configured for UGC and UGC multimedia independently at `ugc.gcp` and `ugcMultimedia.gcp`."
+    
+
   artifacthub.io/links: |
     - name: Documentation for Sophora UGC
       url: https://subshell.com/docs/ugc/

charts/sophora-ugc/README.md

@@ -22,10 +22,18 @@ For details, check the [UGC Multimedia Service documentation page](https://subsh
 
 ```yaml
 service:
-  jolokia:
-    clusterIP: None
+ # Options for the services exposing UGC Webapp and UGC Multimedia
   webapp:
     type: LoadBalancer
+  # Options for the service exposing the UGC management port
+  management:
+    # Whether to deploy a management service
+    enabled: true
+    #  Port on which the management service should listen
+    # Must match Spring's management.server.port property in ugc.config.
+    port: 1694
+    # ClusterIP for the management service
+    clusterIP: None
 
 ingress:
   enabled: false
@@ -59,9 +67,6 @@ ugc:
     pullPolicy: IfNotPresent
     # Overrides the image tag whose default is the chart appVersion.
     tag: "latest"
-  binariesStorage:
-    size: 1G
-    storageClass: standard
 
   logback: |
     <?xml version="1.0" encoding="UTF-8"?>
@@ -123,6 +128,10 @@ ugc:
     rating:
       primaryTypes: ["sophora-content-nt:story"]
 
+  healthCheck:
+    # Which port to use for Kubernetes liveness/readiness probes at the ugc container, "webapp" or "management"
+    port: webapp
+
 ugcMultimedia:
   enabled: false # enable to deploy UGC Multimedia Service
   image:

charts/sophora-ugc/templates/deployment.yaml

@@ -93,15 +93,47 @@ spec:
           ports:
             - containerPort: {{ default 8080 (.Values.ugc.config.server).port }}
               name: webapp
-            - containerPort: 1694
-              name: jolokia
+            {{- with .Values.service.management.port }}
+            - containerPort: {{ . }}
+              name: management
+            {{- end }}
           volumeMounts:
             - name: logback-xml
               mountPath: /config/logback
             - name: application-yml
               mountPath: /config/application
           resources:
             {{- toYaml .Values.resources | nindent 12 }}
+          {{- with .Values.startupProbe }}
+          startupProbe:
+            httpGet:
+              port: {{ $.Values.ugc.healthCheck.port }}
+              path: /actuator/health/liveness
+            failureThreshold: {{ .failureThreshold }}
+            initialDelaySeconds: {{ .initialDelaySeconds }}
+            periodSeconds: {{ .periodSeconds }}
+            timeoutSeconds: {{ .timeoutSeconds }}
+          {{- end }}
+          {{- with .Values.livenessProbe }}
+          livenessProbe:
+            httpGet:
+              port: {{ $.Values.ugc.healthCheck.port }}
+              path: /actuator/health/liveness
+            failureThreshold: {{ .failureThreshold }}
+            initialDelaySeconds: {{ .initialDelaySeconds }}
+            periodSeconds: {{ .periodSeconds }}
+            timeoutSeconds: {{ .timeoutSeconds }}
+           {{- end }}
+           {{- with .Values.readinessProbe }}
+          readinessProbe:
+            httpGet:
+              port: {{ $.Values.ugc.healthCheck.port }}
+              path: /actuator/health/readiness
+            failureThreshold: {{ .failureThreshold }}
+            initialDelaySeconds: {{ .initialDelaySeconds }}
+            periodSeconds: {{ .periodSeconds }}
+            timeoutSeconds: {{ .timeoutSeconds }}
+            {{- end }}
         {{- if .Values.ugc.extraContainers }}
           {{ include "common.tplvalues.render" (dict "value" .Values.ugc.extraContainers "context" $) | nindent 8 }}
         {{- end }}

charts/sophora-ugc/templates/gcpBackendPolicy.yaml

@@ -1,4 +1,4 @@
-{{- if and .Values.gcp.enabled .Values.gcp.backendPolicy.enabled }}
+{{- if and .Values.ugc.gcp.enabled .Values.ugc.gcp.backendPolicy.enabled }}
 {{- $fullName := printf "%s-webapp" (include "sophora-ugc.fullname" .) -}}
 {{- $labels := include "sophora-ugc.labels" . -}}
 apiVersion: networking.gke.io/v1
@@ -9,27 +9,27 @@ metadata:
     {{- $labels | nindent 4 }}
 spec:
   default:
-    {{- with .Values.gcp.backendPolicy.logging }}
+    {{- with .Values.ugc.gcp.backendPolicy.logging }}
     logging:
       enabled: {{ .enabled | default false }}
       {{- if .enabled }}
       sampleRate: {{ .sampleRate | default 1000000 }}
       {{- end }}
     {{- end }}
-    {{- with .Values.gcp.backendPolicy.sessionAffinity }}
+    {{- with .Values.ugc.gcp.backendPolicy.sessionAffinity }}
     sessionAffinity:
       type: {{ .type | default "NONE" }}
       {{- if or (eq .type "GENERATED_COOKIE") (eq .type "HTTP_COOKIE") }}
       cookieTtlSec: {{ .cookieTtlSec | default 0 }}
       {{- end }}
     {{- end }}
     connectionDraining:
-      drainingTimeoutSec: {{ dig "drainingTimeoutSec" 0 (.Values.gcp.backendPolicy.connectionDraining | default dict) }}
-    timeoutSec: {{ .Values.gcp.backendPolicy.timeoutSec | default 30 }}
-    {{- with .Values.gcp.backendPolicy.securityPolicy }}
+      drainingTimeoutSec: {{ dig "drainingTimeoutSec" 0 (.Values.ugc.gcp.backendPolicy.connectionDraining | default dict) }}
+    timeoutSec: {{ .Values.ugc.gcp.backendPolicy.timeoutSec | default 30 }}
+    {{- with .Values.ugc.gcp.backendPolicy.securityPolicy }}
     securityPolicy: {{ . | quote }}
     {{- end }}
-    {{- with .Values.gcp.backendPolicy.iap }}
+    {{- with .Values.ugc.gcp.backendPolicy.iap }}
     iap:
       enabled: {{ .enabled | default false }}
       {{- if .enabled }}

charts/sophora-ugc/templates/gcpHealthCheck.yaml

@@ -1,4 +1,4 @@
-{{- if and .Values.gcp.enabled .Values.gcp.healthCheck.enabled }}
+{{- if and .Values.ugc.gcp.enabled .Values.ugc.gcp.healthCheck.enabled }}
 {{- $fullName := include "sophora-ugc.fullname" . -}}
 {{- $labels := include "sophora-ugc.labels" . -}}
 apiVersion: networking.gke.io/v1
@@ -12,15 +12,15 @@ spec:
     config:
       type: HTTP
       httpHealthCheck:
-        {{- with .Values.gcp.healthCheck.port }}
+        {{- with .Values.ugc.gcp.healthCheck.port }}
         port: {{ . }}
         {{- end }}
-        {{- with .Values.gcp.healthCheck.portName }}
+        {{- with .Values.ugc.gcp.healthCheck.portName }}
         portName: {{ . | quote }}
         {{- end }}
-        requestPath: {{ .Values.gcp.healthCheck.path | quote }}
+        requestPath: {{ .Values.ugc.gcp.healthCheck.path | quote }}
     logConfig:
-      enabled: {{ .Values.gcp.healthCheck.logging }}
+      enabled: {{ .Values.ugc.gcp.healthCheck.logging }}
   targetRef:
     group: ""
     kind: Service