[WIP] thanos ruler compoment

thanos/charts/templates/ruler-svc.yaml

@@ -0,0 +1,33 @@
+# SPDX-FileCopyrightText: 2024 SAP SE or an SAP affiliate company and Greenhouse contributors
+# SPDX-License-Identifier: Apache-2.0
+
+{{ if .Values.thanos.ruler.enabled }}
+apiVersion: v1
+kind: Service
+metadata:
+  {{- if .Values.thanos.query.annotations }}
+  annotations:
+    {{ toYaml .Values.thanos.query.annotations | nindent 8 }}
+  {{- end }}
+  labels:
+    {{- include "plugin.labels" . | nindent 4 }}
+    {{- include "thanos.labels" . | nindent 4 }}
+    {{- if .Values.thanos.ruler.serviceLabels }}
+    {{ toYaml .Values.thanos.ruler.serviceLabels | nindent 4 }}
+    {{- end }}
+  name: {{ include "release.name" . }}-ruler
+spec:
+  ports:
+  - name: grpc
+    port: 10901
+    protocol: TCP
+    targetPort: grpc
+  - name: http
+    port: 10902
+    protocol: TCP
+    targetPort: http
+  selector:
+    app.kubernetes.io/managed-by: {{ include "release.name" . }}
+    app.kubernetes.io/name: ruler
+
+{{- end }}

thanos/charts/templates/ruler.yaml

@@ -0,0 +1,98 @@
+# {{ if .Values.thanos.ruler.enabled }}
+apiVersion: apps/v1
+kind: StatefulSet
+metadata:
+  labels:
+    {{- include "plugin.labels" . | nindent 4 }}
+    {{- include "thanos.labels" . | nindent 4 }}
+    {{- if .Values.thanos.ruler.deploymentLabels }}
+    {{ toYaml .Values.thanos.ruler.deploymentLabels | nindent 4 }}
+    {{- end }}
+  name: {{ include "release.name" . }}-ruler
+spec:
+  replicas: {{ default 1 .Values.thanos.ruler.replicas }}
+  updateStrategy:
+    rollingUpdate:
+      maxSurge: 25%
+      maxUnavailable: 25%
+    type: RollingUpdate
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: {{ include "release.name" . }}-ruler
+      app.kubernetes.io/instance: {{ .Release.Name }}
+  serviceName: {{ include "release.name" . }}-ruler
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/name: {{ include "release.name" . }}-ruler
+        app.kubernetes.io/instance: {{ .Release.Name }}
+        prometheus-thanos-peer: "true"
+        {{- with .Values.thanos.ruler.additionalLabels }}
+          {{- toYaml . | nindent 8 }}
+        {{- end }}
+        region: {{ .Values.global.region }}
+        cluster_type: {{ .Values.global.clusterType }}
+      annotations:
+        prometheus.io/scrape: "true"
+        prometheus.io/port: "10902"
+    spec:
+      serviceAccount: {{ include "release.name" . }}-ruler
+      containers:
+        - name: {{ .Chart.Name }}-ruler
+          imagePullPolicy: {{ default "IfNotPresent" .Values.thanos.image.pullPolicy }}
+          image: "{{ .Values.thanos.image.repository }}:{{ .Values.thanos.image.tag | default .Chart.AppVersion }}"
+          args:
+          - rule
+          - --data-dir=/thanos/data
+          - --log.level={{ .Values.thanos.ruler.logLevel }}
+          - --eval-interval={{ .Values.thanos.ruler.evalInterval }}
+          - --tsdb.retention=24h
+          - --rule-file=/etc/thanos/rules/*/*.yaml
+          - --query={{ include "release.name" . }}-query:10902
+          - --label=thanos_ruler_replica="$(POD_NAME)"
+          - --label=cluster="qa-de-1"
+          - --label=cluster_type="metal"
+          - --label=region="qa-de-1"
+          - --alert.label-drop=thanos_ruler_replica
+          - --alertmanagers.config-file=/etc/thanos/config/alertmanager-config/alertManagerConfig.yaml
+          - --web.external-prefix=/ruler
+          - --web.route-prefix=/ruler
+          - --alert.query-url=https://metrics.metal.qa-de-1.cloud.sap
+          - --http.config=/etc/thanos/web_config/web-config.yaml
+          ports:
+            - name: http
+              containerPort: 10902
+              protocol: TCP
+            - name: grpc
+              containerPort: 10901
+              protocol: TCP
+          volumeMounts:
+          - mountPath: /etc/thanos/rules/{{ include "release.name" . }}-rulefiles
+            name: {{ include "release.name" . }}-rulefiles
+          - mountPath: /etc/thanos/config/alertmanager-config
+            name: alertmanager-config
+          - mountPath: /etc/thanos/web_config/web-config.yaml
+            name: web-config
+          - mountPath: /etc/thanos/secrets/thanos-metal-ruler-alertmanager-sso-cert
+            name: {{ include "release.name" . }}-alertmanager-sso-cert
+          - mountPath: /thanos/data
+          name: data
+      volumes:
+      - name: alertmanager-config
+        secret:
+          defaultMode: 420
+          secretName: {{ include "release.name" . }}-alertmanager-config
+      - name: web-config
+        secret:
+          defaultMode: 420
+          secretName: {{ include "release.name" . }}-web-config
+      - name: {{ include "release.name" . }}-alertmanager-sso-cert
+        secret:
+          defaultMode: 420
+          secretName: {{ include "release.name" . }}-ruler-alertmanager-sso-cert
+      - name: {{ include "release.name" . }}-rulefiles #what if multiple replicas ?
+        configMap:
+            name: {{ include "release.name" . }}-rulerrulefiles
+      - name: data
+        emptyDir: {}
+# {{- end }}

thanos/charts/templates/tests/test-thanos-config.yaml

@@ -26,12 +26,15 @@ data:
         verify "there is 1 deployment named '{{ .Release.Name }}-compactor'"
         verify "there is 1 deployment named '{{ .Release.Name }}-query'"
         verify "there is 1 deployment named '{{ .Release.Name }}-store'"
+        verify "there is 1 statefulset named '{{ .Release.Name }}-ruler'"
         verify "there is 1 service named '{{ .Release.Name }}-compactor'"
         verify "there is 1 service named '{{ .Release.Name }}-query'"
         verify "there is 1 service named '{{ .Release.Name }}-store'"
+        verify "there is 1 service named '{{ .Release.Name }}-ruler'"
         try "at most 2 times every 5s to get pods named '{{ .Release.Name }}-compactor' and verify that '.status.phase' is 'running'"
         try "at most 4 times every 5s to get pods named '{{ .Release.Name }}-query' and verify that '.status.phase' is 'running'"
         try "at most 4 times every 5s to get pods named '{{ .Release.Name }}-store' and verify that '.status.phase' is 'running'"
+        try "at most 4 times every 5s to get pods named '{{ .Release.Name }}-ruler' and verify that '.status.phase' is 'running'"
     }
 
     @test "Verify successful creation and bound status of {{ .Release.Name }} persistent volume claims" {

thanos/charts/values.yaml

@@ -2,11 +2,23 @@
 # SPDX-License-Identifier: Apache-2.0
 
 global:
+  clusterType:
 
+  region:
   ## Labels to apply to all resources
   ##
   commonLabels: {}
 
+# Create RBAC resources.
+rbac:
+  create: true
+
+# ServiceAccount to use for the Thanos.
+# Note that a ServiceAccount with name `default` cannot be created.
+# Instead the generated name will be used.
+serviceAccount:
+  create: true
+
 thanos:
   image:
     repository: quay.io/thanos/thanos
@@ -61,6 +73,12 @@ thanos:
 
     annotations:
 
+    chunkPoolSize:
+
+    indexCacheSize:
+
+    logLevel:
+
   compactor:
     # if this is a standalone Thanos Query, we wouldn't want a compactor
     enabled: true
@@ -102,6 +120,45 @@ thanos:
       # GiB
       size:
       labels: []
+
+  ruler:
+    enabled: true
+
+    replicas:
+
+    annotations:
+
+    deploymentLabels: {} 
+
+    serviceLabels:
+
+    paused:
+
+    logLevel:
+
+    externalPrefix:
+
+    evalInterval: 30s
+
+    externalClusterLabels:
+      enabled:
+
+    alertmanagers:
+      hosts:
+      authentication:
+        enabled:
+
+    queryEndpoints:
+
+    queries:
+
+    additionalFlags:
+
+    additionalLabels:
+
+    externalUrl: "https://"
+
+    retention: true
 
 testFramework:
   enabled: true