feat(metrics): add prometheus rules (and update exporter)

Signed-off-by: WrenIX <[email protected]>

Author: wrenix

charts/nextcloud/Chart.yaml

@@ -1,6 +1,6 @@
 apiVersion: v2
 name: nextcloud
-version: 6.6.10
+version: 6.7.0
 # renovate: image=docker.io/library/nextcloud
 appVersion: 30.0.10
 description: A file sharing server that puts the control and security of your own data back into your hands.

charts/nextcloud/templates/metrics/deployment.yaml

@@ -64,6 +64,8 @@ spec:
               value: {{ .Values.metrics.tlsSkipVerify | quote }}
             - name: NEXTCLOUD_INFO_APPS
               value: {{ .Values.metrics.info.apps | quote }}
+            - name: NEXTCLOUD_INFO_UPDATE
+              value: {{ .Values.metrics.info.update | quote }}
           ports:
             - name: metrics
               containerPort: 9205

charts/nextcloud/templates/metrics/prometheus-rules.yaml

@@ -0,0 +1,61 @@
+{{- if .Values.metrics.rules.enabled }}
+{{- $fullname := include "nextcloud.fullname" . }}
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: {{ $fullname }}
+  labels:
+    {{- include "nextcloud.labels"  ( dict "rootContext" $ ) | nindent 4 }}
+    {{- with .Values.metrics.rules.labels }}
+    {{- toYaml . | nindent 4 }}
+    {{- end }}
+spec:
+  groups:
+    {{- with .Values.metrics.rules.defaults }}
+    {{- if .enabled }}
+    {{- $filter := .filter | default (printf `namespace="%s",job="%s"` $.Release.Namespace $fullname) }}
+    - name: {{ $fullname }}-Defaults
+      rules:
+        - alert: "nextcloud: not reachable"
+          expr: 'avg(nextcloud_up{ {{ $filter }} }) without(endpoint,container,pod,instance) < 1'
+          for: 5m
+          labels:
+            severity: "critical"
+            {{- with .labels }}
+            {{- toYaml . | nindent 12 }}
+            {{- end }}
+          annotations:
+            {{`
+            summary: "Nextcloud in {{ $labels.namespace }} is not reachable by exporter"
+            `}}
+        - alert: "nextcloud: outdated version"
+          expr: 'sum(nextcloud_system_update_available{ {{ $filter }} }) without(endpoint,container,pod,instance) > 0'
+          labels:
+            severity: "warning"
+            {{- with .labels }}
+            {{- toYaml . | nindent 12 }}
+            {{- end }}
+          annotations:
+            {{`
+            summary: "Nextcloud in {{ $labels.namespace }} is outdated"
+            `}}
+        - alert: "nextcloud: outdated apps"
+          expr: 'sum(nextcloud_apps_updates_available_total{ {{ $filter }} }) without(endpoint,container,pod,instance) > 0'
+          labels:
+            severity: "warning"
+            {{- with .labels }}
+            {{- toYaml . | nindent 12 }}
+            {{- end }}
+          annotations:
+            {{`
+            summary: "Nextcloud in {{ $labels.namespace }} has {{ $value }} outdated Apps"
+            `}}
+    {{- end }}
+    {{- end }}{{/* end-with prometheus.rules.default */}}
+    {{- with .Values.metrics.rules.additionalRules }}
+    - name: {{ $fullname }}-Additional
+      rules:
+        {{- toYaml . | nindent 8 }}
+    {{- end }}
+{{- end }}
+

charts/nextcloud/values.yaml

@@ -778,10 +778,11 @@ metrics:
     # Optional: becomes NEXTCLOUD_INFO_APPS env var in the nextcloud-exporter container.
     # Enables gathering of apps-related metrics. Defaults to false
     apps: false
+    update: false
 
   image:
     repository: xperimental/nextcloud-exporter
-    tag: 0.6.2
+    tag: 0.8.0
     pullPolicy: IfNotPresent
     # pullSecrets:
     #   - myRegistrKeySecretName
@@ -863,6 +864,27 @@ metrics:
     ##
     labels: {}
 
+  rules:
+    # -- Deploy Prometheus Rules (Alerts) for the exporter
+    # @section -- Metrics
+    enabled: false
+    # -- Label on Prometheus Rules CRD Manifest
+    # @section -- Metrics
+    labels: {}
+    defaults:
+      # -- Add Default Rules
+      # @section -- Metrics
+      enabled: true
+      # -- Label on the rules (the severity is already set)
+      # @section -- Metrics
+      labels: {}
+      # -- Filter on metrics on alerts (default just for this helm-chart)
+      # @section -- Metrics
+      filter: ""
+    # -- Add own Rules to Prometheus Rules
+    # @section -- Metrics
+    additionalRules: []
+
 
 rbac:
   enabled: false