From c5ea17918834fc03059a37c8dc1011ed34f42c4e Mon Sep 17 00:00:00 2001
From: Alex Castilio <alexsantos@microsoft.com>
Date: Thu, 23 Jan 2025 17:34:10 +0000
Subject: [PATCH] feat: add metricscardinality to heartbeat (#1235)

# Description

Add `metricscardinality` to heartbeat. This will give visibility on the
number of time series being exposed by retina.

## Related Issue

#1040

## Checklist

- [x] I have read the [contributing
documentation](https://retina.sh/docs/contributing).
- [x] I signed and signed-off the commits (`git commit -S -s ...`). See
[this
documentation](https://docs.github.com/en/authentication/managing-commit-signature-verification/about-commit-signature-verification)
on signing commits.
- [x] I have correctly attributed the author(s) of the code.
- [x] I have tested the changes locally.
- [x] I have followed the project's style guidelines.
- [x] I have updated the documentation, if necessary.
- [x] I have added tests, if applicable.

## Screenshots (if applicable) or Testing Completed

Metrics exported with heartbeat:

![image](https://github.com/user-attachments/assets/6fb4d76b-5780-4751-91c0-46e3c4f0fb85)


## Additional Notes

Metrics of types `histogram` and `summary` expose multiple time series
during a scrape. Code is counting according to number of time series
exposed at /metrics endpoint.


Ref:
https://prometheus.io/docs/concepts/metric_types/#histogram
https://prometheus.io/docs/concepts/metric_types/#summary


---

Please refer to the [CONTRIBUTING.md](../CONTRIBUTING.md) file for more
information on how to contribute to this project.

Signed-off-by: Alex Castilio dos Santos <alexsantos@microsoft.com>
---
 pkg/telemetry/telemetry.go | 48 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 48 insertions(+)

diff --git a/pkg/telemetry/telemetry.go b/pkg/telemetry/telemetry.go
index 4984df42db..b680728784 100644
--- a/pkg/telemetry/telemetry.go
+++ b/pkg/telemetry/telemetry.go
@@ -10,12 +10,15 @@ import (
 	"os"
 	"runtime"
 	"runtime/debug"
+	"strconv"
 	"sync"
 	"time"
 
 	"github.com/microsoft/ApplicationInsights-Go/appinsights"
 	"github.com/microsoft/ApplicationInsights-Go/appinsights/contracts"
+	"github.com/microsoft/retina/pkg/exporter"
 	"github.com/microsoft/retina/pkg/log"
+	io_prometheus_client "github.com/prometheus/client_model/go"
 )
 
 var (
@@ -174,10 +177,55 @@ func (t *TelemetryClient) heartbeat(ctx context.Context) {
 	if err != nil {
 		t.trackWarning(err, "failed to get cpu usage")
 	}
+
+	metricscardinality, err := metricsCardinality()
+	if err != nil {
+		t.trackWarning(err, "failed to get metrics cardinality")
+	}
+
+	props["metricscardinality"] = strconv.Itoa(metricscardinality)
+
 	maps.Copy(props, cpuProps)
 	maps.Copy(props, t.profile.GetMemoryUsage())
 	t.TrackEvent("heartbeat", props)
 }
+func metricsCardinality() (int, error) {
+	metricFamilies, err := exporter.CombinedGatherer.Gather()
+	if err != nil {
+		return 0, fmt.Errorf("failed to gather metrics: %w", err)
+	}
+
+	metricscardinality := 0
+
+	for _, mf := range metricFamilies {
+		switch mf.GetType() { //nolint:exhaustive // 'default' satisfies exhaustiveness
+
+		case io_prometheus_client.MetricType_HISTOGRAM:
+			metrics := mf.GetMetric()
+			for _, m := range metrics {
+				metricscardinality += len(m.GetHistogram().GetBucket()) + 3 // +3 for le="+Inf", _sum and _count
+			}
+
+		case io_prometheus_client.MetricType_GAUGE_HISTOGRAM:
+			metrics := mf.GetMetric()
+			for _, m := range metrics {
+				metricscardinality += len(m.GetHistogram().GetBucket()) + 3 // +3 for le="+Inf", _sum and _count
+			}
+
+		case io_prometheus_client.MetricType_SUMMARY:
+			metrics := mf.GetMetric()
+			for _, m := range metrics {
+				metricscardinality += len(m.GetSummary().GetQuantile()) + 2 // +2 for _sum and _count
+			}
+
+		default:
+			metricscardinality += len(mf.GetMetric())
+
+		}
+	}
+
+	return metricscardinality, nil
+}
 
 func bToMb(b uint64) uint64 {
 	return b >> mbShift