COSI-46 expose metrics service in kubernetes and E2E test

.github/scripts/e2e_tests_metrics_service.sh

@@ -0,0 +1,59 @@
+#!/bin/bash
+set -e
+
+LOG_FILE=".github/e2e_tests/artifacts/logs/e2e_tests/metrics_service.log"
+mkdir -p "$(dirname "$LOG_FILE")"
+
+NAMESPACE="scality-object-storage"
+SERVICE="scality-cosi-metrics"
+LOCAL_PORT=8080
+TARGET_PORT=8080
+
+# Error handling function
+error_handler() {
+  echo "An error occurred during bucket creation tests. Check the log file for details." | tee -a "$LOG_FILE"
+  echo "Failed command: $BASH_COMMAND" | tee -a "$LOG_FILE"
+  exit 1
+}
+
+# Trap errors and call the error handler
+trap 'error_handler' ERR
+
+# Log command execution to the log file for debugging
+log_and_run() {
+  "$@" 2>&1 | tee -a "$LOG_FILE"
+}
+
+log_and_run kubectl describe svc scality-cosi-driver-metrics -n scality-object-storage
+
+log_and_run kubectl port-forward svc/scality-cosi-driver-metrics -n scality-object-storage 8080:8080 &
+PORT_FORWARD_PID=$!
+
+if ps -p $PORT_FORWARD_PID > /dev/null; then
+    log_and_run echo "Port-forwarding established. Querying metrics..."
+
+    log_and_run curl -f http://localhost:$LOCAL_PORT/metrics > metrics_output.txt
+
+    if [ $? -eq 0 ]; then
+        log_and_run  echo "Metrics fetched successfully. Metrics output:"
+        log_and_run cat metrics_output.txt
+        METRICS_WORKING=true
+    else
+        log_and_run echo "Failed to fetch metrics from http://localhost:$LOCAL_PORT/metrics"
+        METRICS_WORKING=false
+    fi
+
+    log_and_run kill $PORT_FORWARD_PID
+else
+    log_and_run echo "Port-forwarding failed to establish."
+    METRICS_WORKING=false
+fi
+
+if [ "$METRICS_WORKING" = true ]; then
+    log_and_run echo "Metrics service is working as expected."
+    exit 0
+else
+    log_and_run echo "Metrics service is not working as expected."
+    exit 1
+fi
+

.github/workflows/ci-e2e-tests.yml

@@ -104,6 +104,10 @@ jobs:
       run: |
         .github/scripts/e2e_tests_brownfield_use_case.sh
 
+    - name: E2E tests for metrics service using kustomize
+      run: |
+        .github/scripts/e2e_tests_metrics_service.sh
+
     - name: "Delay completion"
       if: ${{ github.event_name == 'workflow_dispatch' && inputs.debug_enabled }}
       uses: scality/actions/[email protected]

helm/scality-cosi-driver/templates/deployment.yaml

@@ -7,6 +7,10 @@ metadata:
     app.kubernetes.io/instance: {{ .Release.Name }}
     app.kubernetes.io/part-of: container-object-storage-interface
     app.kubernetes.io/managed-by: {{ .Release.Service }}
+  annotations:
+      prometheus.io/scrape: "true"
+      prometheus.io/port: "8080"
+      prometheus.io/path: "/metrics"
 
 spec:
   replicas: {{ .Values.replicaCount }}
@@ -27,6 +31,7 @@ spec:
           args:
             - "--driver-prefix=cosi"
             - "--v={{ .Values.logLevels.driver }}"
+            - "--metrics-address=0.0.0.0:8080"
           resources:
             limits:
               cpu: {{ .Values.resources.limits.cpu }}

helm/scality-cosi-driver/templates/service.yaml

@@ -0,0 +1,16 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ include "scality-cosi-driver.fullname" . }}-metrics
+  labels:
+    app.kubernetes.io/name: {{ include "scality-cosi-driver.name" . }}
+    app.kubernetes.io/instance: {{ .Release.Name }}
+    app.kubernetes.io/part-of: container-object-storage-interface
+    app.kubernetes.io/managed-by: {{ .Release.Service }}
+spec:
+  selector:
+    app.kubernetes.io/name: {{ include "scality-cosi-driver.name" . }}
+  ports:
+    - protocol: TCP
+      port: {{ .Values.metrics.port }}
+      targetPort: {{ .Values.metrics.port }}

helm/scality-cosi-driver/values.yaml

@@ -26,6 +26,10 @@ serviceAccount:
   name: scality-object-storage-provisioner
   create: true
 
+metrics:
+  enabled: true
+  port: 8080
+
 
 resources:
   limits:

kustomize/base/deployment.yaml

@@ -15,6 +15,10 @@ spec:
         app.kubernetes.io/component: driver
         app.kubernetes.io/version: main
         app.kubernetes.io/managed-by: kustomize
+      annotations:
+        prometheus.io/scrape: "true"
+        prometheus.io/port: "8080"
+        prometheus.io/path: "/metrics"
     spec:
       serviceAccountName: scality-object-storage-provisioner
       containers:
@@ -24,6 +28,7 @@ spec:
           args:
             - "--driver-prefix=cosi"
             - "--v=$(COSI_DRIVER_LOG_LEVEL)"
+            - "--metrics-address=0.0.0.0:8080"
           volumeMounts:
             - mountPath: /var/lib/cosi
               name: socket

kustomize/base/kustomization.yaml

@@ -8,3 +8,4 @@ resources:
   - serviceaccount.yaml
   - rbac.yaml
   - deployment.yaml
+  - metrics-service.yaml

kustomize/base/metrics-service.yaml

@@ -0,0 +1,14 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: scality-cosi-metrics
+  labels:
+    app.kubernetes.io/name: scality-cosi-driver
+    app.kubernetes.io/part-of: container-object-storage-interface
+spec:
+  selector:
+    app.kubernetes.io/name: scality-cosi-driver
+  ports:
+    - protocol: TCP
+      port: 8080  # The port Prometheus will scrape
+      targetPort: 8080  # The port exposed by the container

kustomize/base/rbac.yaml

@@ -12,6 +12,9 @@ rules:
   - apiGroups: [""]
     resources: ["secrets", "events"]
     verbs: ["get", "delete", "update", "create"]
+  - apiGroups: [""]
+    resources: ["services", "endpoints"]
+    verbs: ["get", "list", "watch"]
 
 ---
 apiVersion: rbac.authorization.k8s.io/v1

kustomize/overlays/dev/kustomization.yaml

@@ -15,6 +15,7 @@ resources:
   - ../../base/namespace.yaml
   - ../../base/serviceaccount.yaml
   - ../../base/rbac.yaml
+  - ../../metrics-service.yaml
 
 commonLabels:
   app.kubernetes.io/version: main