From 2ed35497aad06f802eb82793f405f4b4ea683849 Mon Sep 17 00:00:00 2001
From: "@tanya_r" <tiana.rogozina@gmail.com>
Date: Fri, 19 Jun 2026 02:04:18 -0300
Subject: [PATCH] feat(observability): add prometheus operator servicemonitor
 to the helm charts

Add an opt-in ServiceMonitor template to the ledger and payments charts
so a cluster running the Prometheus Operator scrapes each service's
/actuator/prometheus. The template is gated behind serviceMonitor.enabled
(default false), so the default render stays CRD-free and the existing
kubeconform-strict check is unchanged; a second CI step renders it with
the flag on and validates everything but the operator CRD.

Closes #249
---
 .github/workflows/helm-test.yml                  |  7 +++++++
 deploy/helm/ledger/templates/servicemonitor.yaml | 16 ++++++++++++++++
 deploy/helm/ledger/values.yaml                   |  6 ++++++
 .../helm/payments/templates/servicemonitor.yaml  | 16 ++++++++++++++++
 deploy/helm/payments/values.yaml                 |  6 ++++++
 5 files changed, 51 insertions(+)
 create mode 100644 deploy/helm/ledger/templates/servicemonitor.yaml
 create mode 100644 deploy/helm/payments/templates/servicemonitor.yaml

diff --git a/.github/workflows/helm-test.yml b/.github/workflows/helm-test.yml
index 18a52da..16eaa73 100644
--- a/.github/workflows/helm-test.yml
+++ b/.github/workflows/helm-test.yml
@@ -55,3 +55,10 @@ jobs:
         run: |
           helm template ${{ matrix.chart }} deploy/helm/${{ matrix.chart }} --set image.tag=test \
             | kubeconform -strict -summary -kubernetes-version "$KUBERNETES_VERSION"
+
+      - name: Validate the ServiceMonitor renders (operator CRD skipped)
+        run: |
+          rendered=$(helm template ${{ matrix.chart }} deploy/helm/${{ matrix.chart }} \
+            --set image.tag=test --set serviceMonitor.enabled=true)
+          echo "$rendered" | grep -q 'kind: ServiceMonitor'
+          echo "$rendered" | kubeconform -strict -summary -skip ServiceMonitor -kubernetes-version "$KUBERNETES_VERSION"
diff --git a/deploy/helm/ledger/templates/servicemonitor.yaml b/deploy/helm/ledger/templates/servicemonitor.yaml
new file mode 100644
index 0000000..31463e5
--- /dev/null
+++ b/deploy/helm/ledger/templates/servicemonitor.yaml
@@ -0,0 +1,16 @@
+{{- if .Values.serviceMonitor.enabled }}
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  name: {{ include "ledger.fullname" . }}
+  labels:
+    {{- include "ledger.labels" . | nindent 4 }}
+spec:
+  selector:
+    matchLabels:
+      {{- include "ledger.selectorLabels" . | nindent 6 }}
+  endpoints:
+    - port: http
+      path: {{ .Values.serviceMonitor.path }}
+      interval: {{ .Values.serviceMonitor.interval }}
+{{- end }}
diff --git a/deploy/helm/ledger/values.yaml b/deploy/helm/ledger/values.yaml
index 69b67df..beb310f 100644
--- a/deploy/helm/ledger/values.yaml
+++ b/deploy/helm/ledger/values.yaml
@@ -19,6 +19,12 @@ service:
   type: ClusterIP
   port: 8080
 
+# Prometheus Operator ServiceMonitor. Off by default; enable on a cluster running the operator.
+serviceMonitor:
+  enabled: false
+  interval: 30s
+  path: /actuator/prometheus
+
 resources:
   requests:
     cpu: 250m
diff --git a/deploy/helm/payments/templates/servicemonitor.yaml b/deploy/helm/payments/templates/servicemonitor.yaml
new file mode 100644
index 0000000..6d5f87d
--- /dev/null
+++ b/deploy/helm/payments/templates/servicemonitor.yaml
@@ -0,0 +1,16 @@
+{{- if .Values.serviceMonitor.enabled }}
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  name: {{ include "payments.fullname" . }}
+  labels:
+    {{- include "payments.labels" . | nindent 4 }}
+spec:
+  selector:
+    matchLabels:
+      {{- include "payments.selectorLabels" . | nindent 6 }}
+  endpoints:
+    - port: http
+      path: {{ .Values.serviceMonitor.path }}
+      interval: {{ .Values.serviceMonitor.interval }}
+{{- end }}
diff --git a/deploy/helm/payments/values.yaml b/deploy/helm/payments/values.yaml
index a3c416d..93363af 100644
--- a/deploy/helm/payments/values.yaml
+++ b/deploy/helm/payments/values.yaml
@@ -19,6 +19,12 @@ service:
   type: ClusterIP
   port: 8080
 
+# Prometheus Operator ServiceMonitor. Off by default; enable on a cluster running the operator.
+serviceMonitor:
+  enabled: false
+  interval: 30s
+  path: /actuator/prometheus
+
 resources:
   requests:
     cpu: 250m