prometheus-rules, default values

charts/hermes/files/config.toml

@@ -8,7 +8,7 @@ enabled = {{ .Values.rest.enabled }}
 
 # Specify the IPv4/6 host over which the built-in HTTP server will serve the RESTful
 # API requests. Default: 127.0.0.1
-host = '127.0.0.1'
+host = '0.0.0.0'
 
 # Specify the port over which the built-in HTTP server will serve the restful API
 # requests. Default: 3000
@@ -35,7 +35,7 @@ tx_confirmation = {{ .Values.mode.packets.txConfirmation }}
 
 [telemetry]
 enabled = {{ .Values.telemetry.enabled }}
-host = '127.0.0.1'
+host = '0.0.0.0'
 port = {{ .Values.ports.telemetry }}
 
 [telemetry.buckets]

charts/hermes/templates/_helpers.tpl

@@ -22,6 +22,20 @@ We truncate at 63 chars because some Kubernetes name fields are limited to this
 {{- end -}}
 {{- end -}}
 
+{{/*
+Common labels
+*/}}
+{{- define "hermes.labels" -}}
+{{ include "hermes.selectorLabels" . }}
+{{- end }}
+
+{{/*
+Selector labels
+*/}}
+{{- define "hermes.selectorLabels" -}}
+app: {{ include "hermes.fullname" . }}
+{{- end }}
+
 {{/*
 Return if ingress is stable.
 */}}

charts/hermes/templates/prometheusrule.yaml

@@ -0,0 +1,20 @@
+{{- if .Values.alerting.enabled -}}
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: {{ template "hermes.fullname" . }}
+{{- if .Values.alerting.prometheusRule.namespace }}
+  namespace: {{ .Values.alerting.prometheusRule.namespace | quote }}
+{{- end }}
+  labels:
+    {{- include "hermes.labels" . | nindent 4 }}
+  {{- if .Values.alerting.prometheusRule.additionalLabels }}
+    {{- toYaml .Values.alerting.prometheusRule.additionalLabels | nindent 4 }}
+  {{- end }}
+spec:
+{{- if .Values.alerting.prometheusRule.rules }}
+  groups:
+  - name: {{ template "hermes.fullname" . }}
+    rules: {{- toYaml .Values.alerting.prometheusRule.rules | nindent 4 }}
+{{- end }}
+{{- end }}

charts/hermes/templates/service.yaml

@@ -16,7 +16,7 @@ spec:
       targetPort: rest
     {{- end }}
     {{- if .Values.telemetry.enabled }}
-    - name: telemetry-svc
+    - name: telemetry
       port: {{ .Values.ports.telemetry }}
       targetPort: telemetry
     {{- end }}

charts/hermes/values.yaml

@@ -3,7 +3,7 @@ global:
   replicaCount: 1
   logLevel: debug
 
-image: ghcr.io/penumbra-zone/hermes:main
+image: ghcr.io/astriaorg/hermes:sha-450f848
 imagePullPolicy: IfNotPresent
 
 fullnameOverride: ""
@@ -195,3 +195,26 @@ ingress:
       #  - secretName: chart-example-tls
       #    hosts:
       #      - chart-example.local
+
+alerting:
+  enabled: false
+  interval: ""
+  additionalLabels:
+    release: kube-prometheus-stack
+  annotations: {}
+  # scrapeTimeout: 10s
+  # path: /metrics
+  prometheusRule:
+    enabled: true
+    additionalLabels:
+      release: kube-prometheus-stack
+    namespace: monitoring
+    rules:
+    - alert: Chain_Node_Down
+      expr: up{container="cometbft"} == 0  # Insert your query Expression
+      for: 1m  # Rough number but should be enough to init warn
+      labels:
+        severity: critical
+      annotations:
+        summary: Chain Node is Down (instance {{ $labels.instance }})
+        description: "chain node '{{ $labels.namespace }}' has disappeared from Prometheus target discovery.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"