From acf53534fe67eea3892160415ae3513194a989ca Mon Sep 17 00:00:00 2001
From: Edward Welch <edward.welch@grafana.com>
Date: Fri, 11 Feb 2022 16:26:30 -0500
Subject: [PATCH] remove promtail_log_entries_bytes_bucket

Signed-off-by: Edward Welch <edward.welch@grafana.com>
---
 clients/pkg/promtail/targets/file/metrics.go | 16 ++++------------
 clients/pkg/promtail/targets/file/tailer.go  |  2 --
 docs/sources/operations/observability.md     |  1 -
 docs/sources/upgrading/_index.md             |  8 ++++++++
 4 files changed, 12 insertions(+), 15 deletions(-)

diff --git a/clients/pkg/promtail/targets/file/metrics.go b/clients/pkg/promtail/targets/file/metrics.go
index 48e0d6bcd84ac..07a7d0f9381fd 100644
--- a/clients/pkg/promtail/targets/file/metrics.go
+++ b/clients/pkg/promtail/targets/file/metrics.go
@@ -8,11 +8,10 @@ type Metrics struct {
 	reg prometheus.Registerer
 
 	// File-specific metrics
-	readBytes          *prometheus.GaugeVec
-	totalBytes         *prometheus.GaugeVec
-	readLines          *prometheus.CounterVec
-	filesActive        prometheus.Gauge
-	logLengthHistogram *prometheus.HistogramVec
+	readBytes   *prometheus.GaugeVec
+	totalBytes  *prometheus.GaugeVec
+	readLines   *prometheus.CounterVec
+	filesActive prometheus.Gauge
 
 	// Manager metrics
 	failedTargets *prometheus.CounterVec
@@ -45,12 +44,6 @@ func NewMetrics(reg prometheus.Registerer) *Metrics {
 		Name:      "files_active_total",
 		Help:      "Number of active files.",
 	})
-	m.logLengthHistogram = prometheus.NewHistogramVec(prometheus.HistogramOpts{
-		Namespace: "promtail",
-		Name:      "log_entries_bytes",
-		Help:      "the total count of bytes",
-		Buckets:   prometheus.ExponentialBuckets(16, 2, 8),
-	}, []string{"path"})
 
 	m.failedTargets = prometheus.NewCounterVec(prometheus.CounterOpts{
 		Namespace: "promtail",
@@ -69,7 +62,6 @@ func NewMetrics(reg prometheus.Registerer) *Metrics {
 			m.totalBytes,
 			m.readLines,
 			m.filesActive,
-			m.logLengthHistogram,
 			m.failedTargets,
 			m.targetsActive,
 		)
diff --git a/clients/pkg/promtail/targets/file/tailer.go b/clients/pkg/promtail/targets/file/tailer.go
index fe92b37c94b17..6ccad9b6c38b0 100644
--- a/clients/pkg/promtail/targets/file/tailer.go
+++ b/clients/pkg/promtail/targets/file/tailer.go
@@ -150,7 +150,6 @@ func (t *tailer) readLines() {
 		}
 
 		t.metrics.readLines.WithLabelValues(t.path).Inc()
-		t.metrics.logLengthHistogram.WithLabelValues(t.path).Observe(float64(len(line.Text)))
 		entries <- api.Entry{
 			Labels: model.LabelSet{},
 			Entry: logproto.Entry{
@@ -225,5 +224,4 @@ func (t *tailer) cleanupMetrics() {
 	t.metrics.readLines.DeleteLabelValues(t.path)
 	t.metrics.readBytes.DeleteLabelValues(t.path)
 	t.metrics.totalBytes.DeleteLabelValues(t.path)
-	t.metrics.logLengthHistogram.DeleteLabelValues(t.path)
 }
diff --git a/docs/sources/operations/observability.md b/docs/sources/operations/observability.md
index dd851c2553984..1e6c4077cd94a 100644
--- a/docs/sources/operations/observability.md
+++ b/docs/sources/operations/observability.md
@@ -60,7 +60,6 @@ Promtail exposes these metrics:
 | `promtail_encoded_bytes_total`            | Counter     | Number of bytes encoded and ready to send.                                                 |
 | `promtail_file_bytes_total`               | Gauge       | Number of bytes read from files.                                                           |
 | `promtail_files_active_total`             | Gauge       | Number of active files.                                                                    |
-| `promtail_log_entries_bytes`              | Histogram   | The total count of bytes read.                                                             |
 | `promtail_request_duration_seconds_count` | Histogram   | Number of send requests.                                                                   |
 | `promtail_sent_bytes_total`               | Counter     | Number of bytes sent.                                                                      |
 | `promtail_sent_entries_total`             | Counter     | Number of log entries sent to the ingester.                                                |
diff --git a/docs/sources/upgrading/_index.md b/docs/sources/upgrading/_index.md
index 70d4fa798e6e1..a99a568331341 100644
--- a/docs/sources/upgrading/_index.md
+++ b/docs/sources/upgrading/_index.md
@@ -120,6 +120,14 @@ The response body has the following schema:
     e.g. if you previously used `__project_id` then you'll need to update your relabel config to use `__gcp_resource_labels_project_id`.
   - `resource_type` has been moved to `__gcp_resource_type`
 
+#### `promtail_log_entries_bytes_bucket` histogram has been removed.
+
+This histogram reports the distribution of log line sizes by file. It has 8 buckets for every file being tailed.
+
+This creates a lot of series and we don't think this metric has enough value to offset the amount of series genereated so we are removing it.
+
+While this isn't a direct replacement, two metrics we find more useful are size and line counters configured via pipeline stages, an example of how to configure these metrics can be found in the [metrics pipeline stage docs](https://grafana.com/docs/loki/latest/clients/promtail/stages/metrics/#counter)
+
 ## 2.4.0
 
 The following are important changes which should be reviewed and understood prior to upgrading Loki.