feat: opentelemetry-collector-observ-lib

Author: theSuess

opentelemetry-collector-observ-lib/.lint

@@ -0,0 +1,5 @@
+exclusions:
+  target-instance-rule:
+    reason: "The job label is used differently in the otel world (`service.name` maps to `job` and is the same across instances)"
+  template-instance-rule:
+    reason: "The job label is used differently in the otel world (`service.name` maps to `job` and is the same across instances)"

opentelemetry-collector-observ-lib/Makefile

@@ -0,0 +1 @@
+include ../Makefile_mixin

opentelemetry-collector-observ-lib/README.md

@@ -0,0 +1,26 @@
+# OpenTelemetry Collector observability library
+
+This library can be used to generate dashboards, rows, panels, and alerts for [OpenTelemetry Collector internal telemetry](https://opentelemetry.io/docs/collector/internal-telemetry/) .
+
+Supports the following sources:
+
+## Import
+
+```sh
+jb init
+jb install https://github.com/grafana/jsonnet-libs/opentelemetry-collector-observ-lib
+```
+
+## Example: Generate monitoring-mixin
+
+```.jsonnet
+local otelcollib = import 'opentelemetry-collector-observ-lib/main.libsonnet';
+local otelcol =
+  otelcollib.new()
+  + otelcollib.withConfigMixin(
+    {
+        filteringSelector: 'job!=""',
+    }
+  );
+otelcollib.asMonitoringMixin()
+```

opentelemetry-collector-observ-lib/alerts.libsonnet

@@ -0,0 +1,46 @@
+local common = import './signals/common.libsonnet';
+
+{
+  new(this): {
+
+    groups: [
+      {
+        name: this.config.uid + '-rules',
+        rules:
+          [
+            {
+              alert: 'ExporterQueueSaturated',
+              expr: 'sum(otelcol_exporter_queue_size{}) by (data_type,service_instance_id,exporter) / sum(otelcol_exporter_queue_capacity{}) by (data_type,service_instance_id,exporter) > 0.8',
+              'for': '5m',
+              keep_firing_for: '5m',
+              labels: {
+                service_name: 'opentelemetry-collector',
+                severity: 'warning',
+              },
+              annotations: {
+                summary: 'Exporter queue is reaching limit.',
+                description: 'The {{ $labels.data_type }} queue of the {{ $labels.exporter }} exporter is almost full ({{ $value }}). It will start dropping data once full',
+              },
+            },
+          ]
+          + [
+            {
+              alert: '%(capitalized)sFailedToSend' % signal,
+              expr: 'sum(rate(otelcol_exporter_enqueue_failed_%(metric_id)s_total{}[5m])) by (exporter,service_instance_id) > 0' % signal,
+              'for': '5m',
+              keep_firing_for: '5m',
+              labels: {
+                service_name: 'opentelemetry-collector',
+                severity: 'critical',
+              },
+              annotations: {
+                summary: 'Failed to enqueue %(plural)s for export.' % signal,
+                description: 'Collector {{ $labels.service_instance_id }} failed to enqueue %(plural)s for the {{ $labels.exporter }}' % signal,
+              },
+            }
+            for signal in common.signalTypes
+          ],
+      },
+    ],
+  },
+}

opentelemetry-collector-observ-lib/config.libsonnet

@@ -0,0 +1,16 @@
+{
+  local this = self,
+  filteringSelector: 'job!=""',
+  groupLabels: [],
+  instanceLabels: ['job', 'service_instance_id'],
+  uid: 'otelcol',
+  dashboardNamePrefix: '',
+  dashboardTags: ['opentelemetry', 'collector'],
+  metricsSource: 'otelcol',
+  signals+:
+    {
+      receiver: (import './signals/receiver.libsonnet')(this),
+      processor: (import './signals/processor.libsonnet')(this),
+      exporter: (import './signals/exporter.libsonnet')(this),
+    },
+}

opentelemetry-collector-observ-lib/dashboards.libsonnet

@@ -0,0 +1,39 @@
+local g = import './g.libsonnet';
+{
+  local root = self,
+  new(this):
+    {
+      'otelcol-overview.json':
+        g.dashboard.new(this.config.dashboardNamePrefix + 'Collector overview')
+        + g.dashboard.withVariables(this.signals.receiver.getVariablesMultiChoice())
+        + g.dashboard.withVariablesMixin([
+          {
+            datasource: {
+              type: 'prometheus',
+              uid: '${datasource}',
+            },
+            baseFilters: [],
+            filters: [],
+            label: 'Extra filters',
+            name: 'adhoc',
+            type: 'adhoc',
+          },
+        ])
+        + g.dashboard.withTags(this.config.dashboardTags)
+        + g.dashboard.withUid(this.config.uid + '-overview-dashboard')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.overview,
+                this.grafana.rows.process,
+                this.grafana.rows.receivers,
+                this.grafana.rows.processors,
+                this.grafana.rows.exporters,
+              ]
+            )
+          ),
+          setPanelIDs=false
+        ),
+    },
+}