chore: optimize metrics creation and add documentation

.github/workflows/go-generate.yml

@@ -0,0 +1,25 @@
+# Terraform Provider testing workflow.
+name: go-generate
+
+# This GitHub action runs your tests for each pull request and push.
+# Optionally, you can turn it on using a schedule for regular testing.
+on:
+  pull_request:
+    paths:
+      - 'docs/**'
+      - 'tools/**'
+
+jobs:
+  generate:
+    name: Generate
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4 # v3.5.0
+      - uses: actions/setup-go@v5 # v4.0.0
+        with:
+          go-version-file: 'go.mod'
+      - run: go generate ./...
+      - name: git diff
+        run: |
+          git diff --compact-summary --exit-code || \
+            (echo; echo "Unexpected difference in directories after code generation. Run 'go generate ./...' command and commit."; exit 1)

.github/workflows/goproxy.yml → .github/workflows/go-proxy.yml

@@ -1,10 +1,8 @@
 name: 'Force pkg.go.dev release sync'
 
 on:
-    push:
-        tags:
-            - 'v[0-9]+.[0-9]+.[0-9]+'
-            - '**/v[0-9]+.[0-9]+.[0-9]+'
+    release:
+        types: [published]
 
 jobs:
     build:

.github/workflows/goreport.yml → .github/workflows/go-report.yml

@@ -9,4 +9,4 @@ jobs:
     runs-on:
       group: Default
     steps:
-      - uses: creekorful/goreportcard-action@v1.0
+      - uses: creekorful/goreportcard-action@v1.0

.github/workflows/gotest.yml → .github/workflows/go-test.yml

@@ -1,7 +1,7 @@
 name: Unit tests
 
 on:
-  pull_request_target:
+  pull_request:
   workflow_dispatch:
 
 permissions:

.github/workflows/new-release.yaml

@@ -83,10 +83,23 @@ jobs:
     - name: Run Go unit tests
       run: |
         go test ./...
-
+  generate:
+    needs: [pre-check]
+    name: Generate
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4 # v3.5.0
+      - uses: actions/setup-go@v5 # v4.0.0
+        with:
+          go-version-file: 'go.mod'
+      - run: go generate ./...
+      - name: git diff
+        run: |
+          git diff --compact-summary --exit-code || \
+            (echo; echo "Unexpected difference in directories after code generation. Run 'go generate ./...' command and commit."; exit 1)
   # * Step 2: Create a new tag
   tag:
-    needs: [golangci-lint, pre-check, tag-already-exist, testsunit]
+    needs: [golangci-lint, pre-check, tag-already-exist, testsunit, generate]
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4

cmd/admission-controller/main.go

@@ -17,6 +17,7 @@ import (
 	"github.com/orange-cloudavenue/kube-image-updater/internal/httpserver"
 	client "github.com/orange-cloudavenue/kube-image-updater/internal/kubeclient"
 	"github.com/orange-cloudavenue/kube-image-updater/internal/log"
+	"github.com/orange-cloudavenue/kube-image-updater/internal/metrics"
 )
 
 var (
@@ -38,6 +39,9 @@ var (
 )
 
 func init() {
+	// Init Metrics
+	metrics.AdmissionController()
+
 	// webhook server running namespace (default to "default")
 	if os.Getenv("POD_NAMESPACE") != "" {
 		webhookNamespace = os.Getenv("POD_NAMESPACE")

cmd/admission-controller/webhook.go

@@ -22,8 +22,8 @@ import (
 // func serveHandler
 func ServeHandler(w http.ResponseWriter, r *http.Request) {
 	// Prometheus metrics
-	metrics.AdmissionController().Total().Inc()
-	timeAC := metrics.AdmissionController().Duration()
+	metrics.AdmissionController().Total.Inc()
+	timeAC := metrics.AdmissionController().Duration.NewTimer()
 	defer timeAC.ObserveDuration()
 
 	var body []byte
@@ -34,7 +34,7 @@ func ServeHandler(w http.ResponseWriter, r *http.Request) {
 	}
 	if len(body) == 0 {
 		// increment the total number of errors
-		metrics.AdmissionController().TotalErr().Inc()
+		metrics.AdmissionController().ErrTotal.Inc()
 
 		log.Error("empty body")
 		http.Error(w, "empty body", http.StatusBadRequest)
@@ -45,7 +45,7 @@ func ServeHandler(w http.ResponseWriter, r *http.Request) {
 	contentType := r.Header.Get("Content-Type")
 	if contentType != "application/json" {
 		// increment the total number of errors
-		metrics.AdmissionController().TotalErr().Inc()
+		metrics.AdmissionController().ErrTotal.Inc()
 
 		http.Error(w, "invalid Content-Type, expect `application/json`", http.StatusUnsupportedMediaType)
 		return
@@ -55,7 +55,7 @@ func ServeHandler(w http.ResponseWriter, r *http.Request) {
 	ar := admissionv1.AdmissionReview{}
 	if _, _, err := deserializer.Decode(body, nil, &ar); err != nil {
 		// increment the total number of errors
-		metrics.AdmissionController().TotalErr().Inc()
+		metrics.AdmissionController().ErrTotal.Inc()
 
 		log.WithError(err).Warn("Can't decode body")
 		admissionResponse = &admissionv1.AdmissionResponse{
@@ -83,13 +83,13 @@ func ServeHandler(w http.ResponseWriter, r *http.Request) {
 	resp, err := json.Marshal(admissionReview)
 	if err != nil {
 		// increment the total number of errors
-		metrics.AdmissionController().TotalErr().Inc()
+		metrics.AdmissionController().ErrTotal.Inc()
 
 		http.Error(w, fmt.Sprintf("could not encode response: %v", err), http.StatusInternalServerError)
 	}
 	if _, err := w.Write(resp); err != nil {
 		// increment the total number of errors
-		metrics.AdmissionController().TotalErr().Inc()
+		metrics.AdmissionController().ErrTotal.Inc()
 
 		http.Error(w, fmt.Sprintf("could not write response: %v", err), http.StatusInternalServerError)
 	}
@@ -138,16 +138,16 @@ func mutate(ctx context.Context, ar *admissionv1.AdmissionReview) *admissionv1.A
 // create mutation patch for pod.
 func createPatch(ctx context.Context, pod *corev1.Pod) ([]byte, error) {
 	// Metrics - increment the total number of patch
-	metrics.AdmissionControllerPatch().Total().Inc()
-	timePatch := metrics.AdmissionControllerPatch().Duration()
+	metrics.AdmissionController().PatchTotal.Inc()
+	timePatch := metrics.AdmissionController().PatchDuration.NewTimer()
 	defer timePatch.ObserveDuration()
 
 	var err error
 	// find annotation enabled
 	an := annotations.New(ctx, pod)
 	if !an.Enabled().Get() {
 		// increment the total number of errors
-		metrics.AdmissionControllerPatch().TotalErr().Inc()
+		metrics.AdmissionController().PatchErrTotal.Inc()
 
 		return nil, fmt.Errorf("annotation not enabled")
 	}
@@ -172,7 +172,7 @@ func createPatch(ctx context.Context, pod *corev1.Pod) ([]byte, error) {
 			image, err = kubeClient.Image().Find(ctx, pod.Namespace, container.Image)
 			if err != nil {
 				// increment the total number of errors
-				metrics.AdmissionControllerPatch().TotalErr().Inc()
+				metrics.AdmissionController().PatchErrTotal.Inc()
 
 				log.
 					WithFields(logrus.Fields{
@@ -187,7 +187,7 @@ func createPatch(ctx context.Context, pod *corev1.Pod) ([]byte, error) {
 			image, err = kubeClient.Image().Get(ctx, pod.Namespace, crdName)
 			if err != nil {
 				// increment the total number of errors
-				metrics.AdmissionControllerPatch().TotalErr().Inc()
+				metrics.AdmissionController().PatchErrTotal.Inc()
 
 				log.
 					WithFields(logrus.Fields{

cmd/kimup/main.go

@@ -13,6 +13,7 @@ import (
 	"github.com/orange-cloudavenue/kube-image-updater/internal/httpserver"
 	"github.com/orange-cloudavenue/kube-image-updater/internal/kubeclient"
 	"github.com/orange-cloudavenue/kube-image-updater/internal/log"
+	"github.com/orange-cloudavenue/kube-image-updater/internal/metrics"
 	"github.com/orange-cloudavenue/kube-image-updater/internal/models"
 	"github.com/orange-cloudavenue/kube-image-updater/internal/triggers"
 )
@@ -24,6 +25,13 @@ var (
 )
 
 func init() {
+	// Initialize the metrics
+	metrics.Tags()
+	metrics.Events()
+	metrics.Actions()
+	metrics.Rules()
+	metrics.Registry()
+
 	// TODO add namespace scope
 	// Flag "loglevel" is set in log package
 	flag.Parse()

cmd/kimup/scheduler.go

@@ -30,9 +30,9 @@ func initScheduler(ctx context.Context, k kubeclient.Interface) {
 	// Add event lock
 	event.On(triggers.RefreshImage.String(), event.ListenerFunc(func(e event.Event) (err error) {
 		// Increment the counter for the events
-		metrics.Events().Total().Inc()
+		metrics.Events().Total.Inc()
 		// Start the timer for the event execution
-		timerEvents := metrics.Events().Duration()
+		timerEvents := metrics.Events().Duration.NewTimer()
 		defer timerEvents.ObserveDuration()
 
 		if l[e.Data()["namespace"].(string)+"/"+e.Data()["image"].(string)] == nil {
@@ -72,8 +72,8 @@ func initScheduler(ctx context.Context, k kubeclient.Interface) {
 			i := utils.ImageParser(image.Spec.Image)
 
 			// Prometheus metrics - Increment the counter for the registry
-			metrics.Registry().Total().Inc()
-			timerRegistry := metrics.Registry().Duration()
+			metrics.Registry().Total.Inc()
+			timerRegistry := metrics.Registry().Duration.NewTimer()
 
 			re, err := registry.New(ctx, image.Spec.Image, registry.Settings{
 				InsecureTLS: image.Spec.InsecureSkipTLSVerify,
@@ -93,20 +93,20 @@ func initScheduler(ctx context.Context, k kubeclient.Interface) {
 			timerRegistry.ObserveDuration()
 			if err != nil {
 				// Prometheus metrics - Increment the counter for the registry with error
-				metrics.Registry().TotalErr().Inc()
+				metrics.Registry().ErrTotal.Inc()
 
 				return err
 			}
 
 			// Prometheus metrics - Increment the counter for the tags
-			metrics.Tags().Total().Inc()
-			timerTags := metrics.Tags().Duration()
+			metrics.Tags().Total.Inc()
+			timerTags := metrics.Tags().Duration.NewTimer()
 
 			tagsAvailable, err := re.Tags()
 			timerTags.ObserveDuration()
 			if err != nil {
 				// Prometheus metrics - Increment the counter for the tags with error
-				metrics.Tags().TotalErr().Inc()
+				metrics.Tags().TotalErr.Inc()
 
 				return err
 			}
@@ -128,8 +128,8 @@ func initScheduler(ctx context.Context, k kubeclient.Interface) {
 				r.Init(tag, tagsAvailable, rule.Value)
 
 				// Prometheus metrics - Increment the counter for the rules
-				metrics.Rules().Total().Inc()
-				timerRules := metrics.Rules().Duration()
+				metrics.Rules().Total.Inc()
+				timerRules := metrics.Rules().Duration.NewTimer()
 
 				match, newTag, err := r.Evaluate()
 
@@ -138,7 +138,7 @@ func initScheduler(ctx context.Context, k kubeclient.Interface) {
 
 				if err != nil {
 					// Prometheus metrics - Increment the counter for the evaluated rule with error
-					metrics.Rules().TotalErr().Inc()
+					metrics.Rules().ErrTotal.Inc()
 
 					log.Errorf("Error evaluating rule: %v", err)
 					continue
@@ -159,8 +159,8 @@ func initScheduler(ctx context.Context, k kubeclient.Interface) {
 						}, &image, action.Data)
 
 						// Prometheus metrics - Increment the counter for the actions
-						metrics.Actions().Total().Inc()
-						timerActions := metrics.Actions().Duration()
+						metrics.Actions().Total.Inc()
+						timerActions := metrics.Actions().Duration.NewTimer()
 
 						err = a.Execute(ctx)
 
@@ -169,7 +169,7 @@ func initScheduler(ctx context.Context, k kubeclient.Interface) {
 
 						if err != nil {
 							// Prometheus metrics - Increment the counter for the executed action with error
-							metrics.Actions().TotalErr().Inc()
+							metrics.Actions().ErrTotal.Inc()
 
 							log.Errorf("Error executing action(%s): %v", action.Type, err)
 							continue
@@ -183,7 +183,7 @@ func initScheduler(ctx context.Context, k kubeclient.Interface) {
 		})
 
 		// Prometheus metrics - Increment the counter for the events evaluated with error
-		metrics.Events().TotalErr().Inc()
+		metrics.Events().ErrTotal.Inc()
 		return retryErr
 	}), event.Normal)
 }

docs/advanced/metrics.md

@@ -0,0 +1,47 @@
+---
+hide:
+  - toc
+---
+
+# Metrics
+
+kimup exposes metrics to monitor the performance. The metrics are exposed in the Prometheus format and can be scraped by Prometheus or any other monitoring tool that can scrape Prometheus.
+
+## Settings
+
+The following arguments can be used to configure the metrics *(Available in kimup-operator, kimup-controller and kimup-admission-controller)*:
+
+| Flag           | Default  | Description               |
+| -------------- | -------- | ------------------------- |
+| --metrics      | false    | Enable metrics collection |
+| --metrics-port | :9080    | Port to expose metrics on |
+| --metrics-path | /metrics | Path to expose metrics on |
+
+
+## Metrics
+
+The following metrics are exposed:
+
+| Metrics                                    | Description                                             |
+| ------------------------------------------ | ------------------------------------------------------- |
+| kimup_actions_duration                     | The duration in seconds of action performed.            |
+| kimup_actions_err_total                    | The total number of action performed with error.        |
+| kimup_actions_total                        | The total number of action performed.                   |
+| kimup_admission_controller_duration        | The duration in seconds of action performed.            |
+| kimup_admission_controller_err_total       | The total number of action performed with error.        |
+| kimup_admission_controller_patch_duration  | The duration in seconds of patch action performed.      |
+| kimup_admission_controller_patch_err_total | The total number of patch action performed with error.  |
+| kimup_admission_controller_patch_total     | The total number of patch action performed.             |
+| kimup_admission_controller_total           | The total number of action performed.                   |
+| kimup_events_duration                      | The duration in seconds of events performed.            |
+| kimup_events_err_total                     | The total number of events performed with error.        |
+| kimup_events_total                         | The total number of events performed.                   |
+| kimup_registry_duration                    | The duration in seconds of registry evaluated.          |
+| kimup_registry_err_total                   | The total number of registry evaluated with error.      |
+| kimup_registry_total                       | The total number of registry evaluated.                 |
+| kimup_rules_duration                       | The duration in seconds of rules evaluated.             |
+| kimup_rules_err_total                      | The total number of rules evaluated with error.         |
+| kimup_rules_total                          | The total number of rules evaluated.                    |
+| kimup_tags_duration                        | The duration in seconds for func tags to list the tags. |
+| kimup_tags_total                           | The total number of func tags is called to list tags.   |
+| kimup_tags_total_err                       | The total number return by the func tags with error.    |

docs/advanced/metrics.md.tmpl

@@ -0,0 +1,20 @@
+---
+hide:
+  - toc
+---
+
+# Metrics
+
+kimup exposes metrics to monitor the performance. The metrics are exposed in the Prometheus format and can be scraped by Prometheus or any other monitoring tool that can scrape Prometheus.
+
+## Settings
+
+The following arguments can be used to configure the metrics *(Available in kimup-operator, kimup-controller and kimup-admission-controller)*:
+
+{{ tableSettings }}
+
+## Metrics
+
+The following metrics are exposed:
+
+{{ tableMetrics }}

go.mod

@@ -12,8 +12,10 @@ require (
 	github.com/containers/image/v5 v5.32.2
 	github.com/containrrr/shoutrrr v0.8.0
 	github.com/crazy-max/diun/v4 v4.28.0
+	github.com/fbiville/markdown-table-formatter v0.3.0
 	github.com/go-chi/chi/v5 v5.1.0
 	github.com/gookit/event v1.1.2
+	github.com/iancoleman/strcase v0.3.0
 	github.com/onsi/ginkgo/v2 v2.20.2
 	github.com/ory/dockertest/v3 v3.11.0
 	github.com/prometheus/client_golang v1.20.4