Reduce default metric label cardinality

README.md

@@ -6,25 +6,26 @@ The `promgithub` service is a lightweight service designed to receive and proces
 
 The `promgithub` service exports the following Prometheus metrics:
 
-| Name                               | Type      | Labels                                                                                         | Description                               |
-|------------------------------------|-----------|------------------------------------------------------------------------------------------------|-------------------------------------------|
-| `promgithub_workflow_status`       | Counter   | `repository`, `branch`, `workflow_name`, `workflow_status`, `conclusion`                       | Total number of workflow runs with status |
-| `promgithub_workflow_duration`     | Histogram | `repository`, `branch`, `workflow_name`, `workflow_status`, `conclusion`                       | Duration of workflow runs                 |
-| `promgithub_workflow_queued`       | Gauge     | `repository`, `branch`, `workflow_name`                                                        | Number of workflow runs queued            |
-| `promgithub_workflow_in_progress`  | Gauge     | `repository`, `branch`, `workflow_name`                                                        | Number of workflow runs in progress       |
-| `promgithub_workflow_completed`    | Gauge     | `repository`, `branch`, `workflow_conclusion`,`workflow_name`                                  | Number of workflow runs completed         |
-| `promgithub_job_status`            | Counter   | `runner`, `repository`, `branch`, `workflow_name`, `job_name`, `job_status`, `job_conclusion`, | Total number of jobs with status          |
-| `promgithub_job_duration`          | Histogram | `runner`, `repository`, `branch`, `workflow_name`, `job_name`, `job_status`, `job_conclusion`  | Duration of jobs runs in seconds          |
-| `promgithub_job_queued`            | Gauge     | `runner`, `repository`, `branch`, `workflow_name`, `job_name`                                  | Number of jobs queued                     |
-| `promgithub_job_in_progress`       | Gauge     | `runner`, `repository`, `branch`, `workflow_name`, `job_name`                                  | Number of jobs in progress                |
-| `promgithub_job_completed`         | Gauge     | `runner`, `repository`, `branch`, `job_conclusion`, `workflow_name`, `job_name`                | Number of jobs completed                  |
-| `promgithub_commit_pushed`         | Counter   | `repository`, `branch`, `commit_author`, `commit_author_email`                                 | Total number of commits pushed            |
-| `promgithub_pull_request`          | Counter   | `repository`, `base_branch`, `pull_request_author`, `pull_request_status`                      | Total number of pull requests             |
+| Name                               | Type      | Labels                                                        | Description                               |
+|------------------------------------|-----------|---------------------------------------------------------------|-------------------------------------------|
+| `promgithub_workflow_status`       | Counter   | `repository`, `branch`, `workflow_name`, `workflow_status`, `conclusion`| Total number of workflow runs with status |
+| `promgithub_workflow_duration`     | Histogram | `repository`, `branch`, `workflow_name`, `workflow_status`, `conclusion`| Duration of workflow runs                 |
+| `promgithub_workflow_queued`       | Gauge     | `repository`, `branch`, `workflow_name`                       | Number of workflow runs queued            |
+| `promgithub_workflow_in_progress`  | Gauge     | `repository`, `branch`, `workflow_name`                       | Number of workflow runs in progress       |
+| `promgithub_workflow_completed`    | Gauge     | `repository`, `branch`, `workflow_conclusion`,`workflow_name` | Number of workflow runs completed         |
+| `promgithub_job_status`            | Counter   | `repository`, `branch`, `workflow_name`, `job_status`, `job_conclusion` | Total number of jobs with status          |
+| `promgithub_job_duration`          | Histogram | `repository`, `branch`, `workflow_name`, `job_status`, `job_conclusion` | Duration of jobs runs in seconds          |
+| `promgithub_job_queued`            | Gauge     | `repository`, `branch`, `workflow_name`                       | Number of jobs queued                     |
+| `promgithub_job_in_progress`       | Gauge     | `repository`, `branch`, `workflow_name`                       | Number of jobs in progress                |
+| `promgithub_job_completed`         | Gauge     | `repository`, `branch`, `job_conclusion`, `workflow_name`     | Number of jobs completed                  |
+| `promgithub_commit_pushed`         | Counter   | `repository`                                                  | Total number of commits pushed            |
+| `promgithub_pull_request`          | Counter   | `repository`, `base_branch`, `pull_request_status`            | Total number of pull requests             |
+
 
 ## Using `promgithub` service
 
 For usage information see [Usage documentation](./docs/usage.md)
 
 ## Contributing to `promgithub` service
 
-For contributing guidelines see [Contributing documentation](./docs/contributing.md)
+For contributing guidelines see [Contributing documentation](./docs/contributing.md)

docs/usage.md

@@ -123,7 +123,7 @@ PROMGITHUB_WEBHOOK_SECRET="<your webhook secret>" PROMGITHUB_SERVICE_PORT="<serv
 
 ## Prometheus scraping configuration
 
-Configure prometheus to scrape `promgithub`'s `/metrics` endpoint to extract metrics
+Configure prometheus to scrape `promgithub`'s `/metrics` endpoint to extract metrics.
 
 ### Prometheus configuration
 
@@ -155,4 +155,4 @@ spec:
     - path: /metrics
       interval: 15s
       port: http
-```
+```

src/github.go

@@ -82,6 +82,16 @@ type GithubPullRequest struct {
 	Repository GithubRepo `json:"repository"`
 }
 
+type runMetricDetails struct {
+	repository string
+	branch     string
+	name       string
+	status     string
+	conclusion string
+	startedAt  string
+	endedAt    string
+}
+
 func validateHMAC(body []byte, signature string, secret []byte) bool {
 	h := hmac.New(sha256.New, secret)
 	h.Write(body)
@@ -121,146 +131,118 @@ func githubEventsHandler(w http.ResponseWriter, r *http.Request) {
 	w.WriteHeader(http.StatusOK)
 }
 
-func updateWorkflowMetrics(body []byte) {
-	var payload GithubWorkflow
-
-	if err := json.Unmarshal(body, &payload); err != nil {
-		logger.Error("Failed to unmarshal workflow_run payload", zap.Error(err))
-		return
-	}
-
-	workflowStatusCounter.With(prometheus.Labels{
-		"repository":      payload.Workflow.Repository.FullName,
-		"branch":          payload.Workflow.Branch,
-		"workflow_name":   payload.Workflow.Name,
-		"workflow_status": payload.Workflow.Status,
-		"conclusion":      payload.Workflow.Conclusion,
-	}).Inc()
+func observeRunMetrics(
+	details runMetricDetails,
+	statusCounter *prometheus.CounterVec,
+	queuedGauge *prometheus.GaugeVec,
+	inProgressGauge *prometheus.GaugeVec,
+	completedGauge *prometheus.GaugeVec,
+	durationHistogram *prometheus.HistogramVec,
+) {
+	statusCounter.WithLabelValues(
+		details.repository,
+		details.branch,
+		details.name,
+		details.status,
+		details.conclusion,
+	).Inc()
 
-	// Handle updating the gauges based on workflow status
-	switch strings.ToLower(payload.Workflow.Status) {
+	switch strings.ToLower(details.status) {
 	case "queued":
-		workflowQueuedGauge.With(prometheus.Labels{
-			"repository":    payload.Workflow.Repository.FullName,
-			"branch":        payload.Workflow.Branch,
-			"workflow_name": payload.Workflow.Name,
-		}).Inc()
+		queuedGauge.WithLabelValues(
+			details.repository,
+			details.branch,
+			details.name,
+		).Inc()
 	case "in_progress":
-		workflowInProgressGauge.With(prometheus.Labels{
-			"repository":    payload.Workflow.Repository.FullName,
-			"branch":        payload.Workflow.Branch,
-			"workflow_name": payload.Workflow.Name,
-		}).Inc()
-		workflowQueuedGauge.With(prometheus.Labels{
-			"repository":    payload.Workflow.Repository.FullName,
-			"branch":        payload.Workflow.Branch,
-			"workflow_name": payload.Workflow.Name,
-		}).Dec()
+		inProgressGauge.WithLabelValues(
+			details.repository,
+			details.branch,
+			details.name,
+		).Inc()
+		queuedGauge.WithLabelValues(
+			details.repository,
+			details.branch,
+			details.name,
+		).Dec()
 	case "completed":
-		workflowCompletedGauge.With(prometheus.Labels{
-			"repository":          payload.Workflow.Repository.FullName,
-			"branch":              payload.Workflow.Branch,
-			"workflow_conclusion": payload.Workflow.Conclusion,
-			"workflow_name":       payload.Workflow.Name,
-		}).Inc()
-		workflowInProgressGauge.With(prometheus.Labels{
-			"repository":    payload.Workflow.Repository.FullName,
-			"branch":        payload.Workflow.Branch,
-			"workflow_name": payload.Workflow.Name,
-		}).Dec()
+		completedGauge.WithLabelValues(
+			details.repository,
+			details.branch,
+			details.conclusion,
+			details.name,
+		).Inc()
+		inProgressGauge.WithLabelValues(
+			details.repository,
+			details.branch,
+			details.name,
+		).Dec()
 
-		// Update duration histogram when the workflow is completed
-		createdAt, err1 := time.Parse(time.RFC3339, payload.Workflow.CreatedAt)
-		updatedAt, err2 := time.Parse(time.RFC3339, payload.Workflow.UpdatedAt)
+		startedAt, err1 := time.Parse(time.RFC3339, details.startedAt)
+		endedAt, err2 := time.Parse(time.RFC3339, details.endedAt)
 		if err1 == nil && err2 == nil {
-			duration := updatedAt.Sub(createdAt).Seconds()
-			workflowDurationHistogram.With(prometheus.Labels{
-				"repository":      payload.Workflow.Repository.FullName,
-				"branch":          payload.Workflow.Branch,
-				"workflow_name":   payload.Workflow.Name,
-				"workflow_status": payload.Workflow.Status,
-				"conclusion":      payload.Workflow.Conclusion,
-			}).Observe(duration)
+			duration := endedAt.Sub(startedAt).Seconds()
+			durationHistogram.WithLabelValues(
+				details.repository,
+				details.branch,
+				details.name,
+				details.status,
+				details.conclusion,
+			).Observe(duration)
 		}
 	}
 }
 
-func updateJobMetrics(body []byte) {
-	var payload GithubJob
+func updateWorkflowMetrics(body []byte) {
+	var payload GithubWorkflow
 
 	if err := json.Unmarshal(body, &payload); err != nil {
-		logger.Error("Failed to unmarshal workflow_job payload", zap.Error(err))
+		logger.Error("Failed to unmarshal workflow_run payload", zap.Error(err))
 		return
 	}
 
-	jobStatusCounter.With(prometheus.Labels{
-		"runner":         payload.Job.RunnerName,
-		"repository":     payload.Job.Repository.FullName,
-		"branch":         payload.Job.Branch,
-		"workflow_name":  payload.Job.WorkflowName,
-		"job_name":       payload.Job.Name,
-		"job_status":     payload.Job.Status,
-		"job_conclusion": payload.Job.Conclusion,
-	}).Inc()
+	observeRunMetrics(
+		runMetricDetails{
+			repository: payload.Workflow.Repository.FullName,
+			branch:     payload.Workflow.Branch,
+			name:       payload.Workflow.Name,
+			status:     payload.Workflow.Status,
+			conclusion: payload.Workflow.Conclusion,
+			startedAt:  payload.Workflow.CreatedAt,
+			endedAt:    payload.Workflow.UpdatedAt,
+		},
+		workflowStatusCounter,
+		workflowQueuedGauge,
+		workflowInProgressGauge,
+		workflowCompletedGauge,
+		workflowDurationHistogram,
+	)
+}
 
-	// Handle updating the gauges based on job status
-	switch strings.ToLower(payload.Job.Status) {
-	case "queued":
-		jobQueuedGauge.With(prometheus.Labels{
-			"runner":        payload.Job.RunnerName,
-			"repository":    payload.Job.Repository.FullName,
-			"branch":        payload.Job.Branch,
-			"workflow_name": payload.Job.WorkflowName,
-			"job_name":      payload.Job.Name,
-		}).Inc()
-	case "in_progress":
-		jobInProgressGauge.With(prometheus.Labels{
-			"runner":        payload.Job.RunnerName,
-			"repository":    payload.Job.Repository.FullName,
-			"branch":        payload.Job.Branch,
-			"workflow_name": payload.Job.WorkflowName,
-			"job_name":      payload.Job.Name,
-		}).Inc()
-		jobQueuedGauge.With(prometheus.Labels{
-			"runner":        payload.Job.RunnerName,
-			"repository":    payload.Job.Repository.FullName,
-			"branch":        payload.Job.Branch,
-			"workflow_name": payload.Job.WorkflowName,
-			"job_name":      payload.Job.Name,
-		}).Dec()
-	case "completed":
-		jobCompletedGauge.With(prometheus.Labels{
-			"runner":         payload.Job.RunnerName,
-			"repository":     payload.Job.Repository.FullName,
-			"branch":         payload.Job.Branch,
-			"job_conclusion": payload.Job.Conclusion,
-			"workflow_name":  payload.Job.WorkflowName,
-			"job_name":       payload.Job.Name,
-		}).Inc()
-		jobInProgressGauge.With(prometheus.Labels{
-			"runner":        payload.Job.RunnerName,
-			"repository":    payload.Job.Repository.FullName,
-			"branch":        payload.Job.Branch,
-			"workflow_name": payload.Job.WorkflowName,
-			"job_name":      payload.Job.Name,
-		}).Dec()
+func updateJobMetrics(body []byte) {
+	var payload GithubJob
 
-		// Update duration histogram when the job is completed
-		startedAt, err1 := time.Parse(time.RFC3339, payload.Job.StartedAt)
-		completedAt, err2 := time.Parse(time.RFC3339, payload.Job.CompletedAt)
-		if err1 == nil && err2 == nil {
-			duration := completedAt.Sub(startedAt).Seconds()
-			jobDurationHistogram.With(prometheus.Labels{
-				"runner":         payload.Job.RunnerName,
-				"repository":     payload.Job.Repository.FullName,
-				"branch":         payload.Job.Branch,
-				"workflow_name":  payload.Job.WorkflowName,
-				"job_name":       payload.Job.Name,
-				"job_status":     payload.Job.Status,
-				"job_conclusion": payload.Job.Conclusion,
-			}).Observe(duration)
-		}
+	if err := json.Unmarshal(body, &payload); err != nil {
+		logger.Error("Failed to unmarshal workflow_job payload", zap.Error(err))
+		return
 	}
+
+	observeRunMetrics(
+		runMetricDetails{
+			repository: payload.Job.Repository.FullName,
+			branch:     payload.Job.Branch,
+			name:       payload.Job.WorkflowName,
+			status:     payload.Job.Status,
+			conclusion: payload.Job.Conclusion,
+			startedAt:  payload.Job.StartedAt,
+			endedAt:    payload.Job.CompletedAt,
+		},
+		jobStatusCounter,
+		jobQueuedGauge,
+		jobInProgressGauge,
+		jobCompletedGauge,
+		jobDurationHistogram,
+	)
 }
 
 func updateCommitMetrics(body []byte) {
@@ -271,13 +253,8 @@ func updateCommitMetrics(body []byte) {
 		return
 	}
 
-	for _, commit := range payload.Commits {
-		commitPushedCounter.With(prometheus.Labels{
-			"repository":          payload.Repository.FullName,
-			"branch":              payload.Ref,
-			"commit_author":       commit.Author.Name,
-			"commit_author_email": commit.Author.Email,
-		}).Inc()
+	for range payload.Commits {
+		commitPushedCounter.WithLabelValues(payload.Repository.FullName).Inc()
 	}
 }
 
@@ -289,10 +266,9 @@ func updatePullRequestMetrics(body []byte) {
 		return
 	}
 
-	pullRequestCounter.With(prometheus.Labels{
-		"repository":          payload.Repository.FullName,
-		"base_branch":         payload.PullRequest.Base.Ref,
-		"pull_request_author": payload.PullRequest.User.Login,
-		"pull_request_status": payload.Action,
-	}).Inc()
+	pullRequestCounter.WithLabelValues(
+		payload.Repository.FullName,
+		payload.PullRequest.Base.Ref,
+		payload.Action,
+	).Inc()
 }