[9.1] (backport #9928) Ingest internal telemetry from the OTel Collector when it is running

changelog/fragments/1759257958-collector-telemetry-monitoring.yaml

@@ -0,0 +1,32 @@
+# Kind can be one of:
+# - breaking-change: a change to previously-documented behavior
+# - deprecation: functionality that is being removed in a later release
+# - bug-fix: fixes a problem in a previous version
+# - enhancement: extends functionality but does not break or fix existing behavior
+# - feature: new functionality
+# - known-issue: problems that we are aware of in a given version
+# - security: impacts on the security of a product or a user’s deployment.
+# - upgrade: important information for someone upgrading from a prior version
+# - other: does not fit into any of the other categories
+kind: enhancement
+
+# Change summary; a 80ish characters long description of the change.
+summary: Include OTel Collector internal telemetry in Agent monitoring
+
+# Long description; in case the summary is not enough to describe the change
+# this field accommodate a description without length limits.
+# NOTE: This field will be rendered only for breaking-change and known-issue kinds at the moment.
+#description:
+
+# Affected component; usually one of "elastic-agent", "fleet-server", "filebeat", "metricbeat", "auditbeat", "all", etc.
+component: elastic-agent
+
+# PR URL; optional; the PR number that added the changeset.
+# If not present is automatically filled by the tooling finding the PR where this changelog fragment has been added.
+# NOTE: the tooling supports backports, so it's able to fill the original PR number instead of the backport PR number.
+# Please provide it if you are adding a fragment for a different PR.
+#pr: https://github.com/owner/repo/1234
+
+# Issue URL; optional; the GitHub issue related to this changeset (either closes or is part of).
+# If not present is automatically filled by the tooling with the issue linked to the PR number.
+#issue: https://github.com/owner/repo/1234

go.mod

@@ -79,11 +79,30 @@ require (
 	go.elastic.co/apm/v2 v2.7.1
 	go.elastic.co/ecszap v1.0.3
 	go.elastic.co/go-licence-detector v0.7.0
+<<<<<<< HEAD
 	go.opentelemetry.io/collector/component/componentstatus v0.132.0
 	go.opentelemetry.io/collector/connector/forwardconnector v0.132.0
 	go.opentelemetry.io/collector/pipeline v1.38.0
 	go.opentelemetry.io/collector/processor/memorylimiterprocessor v0.132.0
 	go.opentelemetry.io/collector/receiver/nopreceiver v0.132.0
+=======
+	go.opentelemetry.io/collector/component/componentstatus v0.135.0
+	go.opentelemetry.io/collector/component/componenttest v0.135.0
+	go.opentelemetry.io/collector/connector/forwardconnector v0.135.0
+	go.opentelemetry.io/collector/exporter/debugexporter v0.135.0
+	go.opentelemetry.io/collector/exporter/nopexporter v0.135.0
+	go.opentelemetry.io/collector/exporter/otlpexporter v0.135.0
+	go.opentelemetry.io/collector/exporter/otlphttpexporter v0.135.0
+	go.opentelemetry.io/collector/extension/extensiontest v0.135.0
+	go.opentelemetry.io/collector/extension/memorylimiterextension v0.135.0
+	go.opentelemetry.io/collector/pipeline v1.41.0
+	go.opentelemetry.io/collector/processor/batchprocessor v0.135.0
+	go.opentelemetry.io/collector/processor/memorylimiterprocessor v0.135.0
+	go.opentelemetry.io/collector/receiver/nopreceiver v0.135.0
+	go.opentelemetry.io/collector/receiver/otlpreceiver v0.135.0
+	go.opentelemetry.io/collector/service v0.135.0
+	go.opentelemetry.io/ebpf-profiler v0.0.202536
+>>>>>>> a441ebee7 (Ingest internal telemetry from the OTel Collector when it is running (#9928))
 	go.uber.org/zap v1.27.0
 	golang.org/x/crypto v0.41.0
 	golang.org/x/exp v0.0.0-20250215185904-eff6e970281f
@@ -686,8 +705,12 @@ require (
 	go.opentelemetry.io/collector/scraper v0.132.0 // indirect
 	go.opentelemetry.io/collector/scraper/scraperhelper v0.132.0 // indirect
 	go.opentelemetry.io/collector/semconv v0.128.1-0.20250610090210-188191247685 // indirect
+<<<<<<< HEAD
 	go.opentelemetry.io/collector/service v0.132.0 // indirect
 	go.opentelemetry.io/collector/service/hostcapabilities v0.132.0 // indirect
+=======
+	go.opentelemetry.io/collector/service/hostcapabilities v0.135.0 // indirect
+>>>>>>> a441ebee7 (Ingest internal telemetry from the OTel Collector when it is running (#9928))
 	go.opentelemetry.io/contrib/bridges/otelzap v0.12.0 // indirect
 	go.opentelemetry.io/contrib/detectors/gcp v1.36.0 // indirect
 	go.opentelemetry.io/contrib/instrumentation/google.golang.org/grpc/otelgrpc v0.62.0 // indirect

internal/pkg/agent/application/application.go

@@ -128,7 +128,11 @@ func New(
 	if err != nil {
 		return nil, nil, nil, fmt.Errorf("failed to create upgrader: %w", err)
 	}
+<<<<<<< HEAD
 	monitor := monitoring.New(isMonitoringSupported, cfg.Settings.DownloadConfig.OS(), cfg.Settings.MonitoringConfig, agentInfo)
+=======
+	monitor := componentmonitoring.New(isMonitoringSupported, cfg.Settings.DownloadConfig.OS(), cfg.Settings.MonitoringConfig, rawConfig.OTel, agentInfo, isOtelExecModeSubprocess)
+>>>>>>> a441ebee7 (Ingest internal telemetry from the OTel Collector when it is running (#9928))
 
 	runtime, err := runtime.NewManager(
 		log,

internal/pkg/agent/application/coordinator/coordinator.go

@@ -88,7 +88,7 @@ type MonitorManager interface {
 	// Enabled when configured to collect metrics/logs.
 	Enabled() bool
 
-	// Reload reloads the configuration for the upgrade manager.
+	// Reload reloads the configuration for the monitoring manager.
 	Reload(rawConfig *config.Config) error
 
 	// MonitoringConfig injects monitoring configuration into resolved ast tree.

internal/pkg/agent/application/monitoring/otel_remap.js

@@ -0,0 +1,139 @@
+// A script for use in the Beats script processor, to remap raw OTel telemetry
+// from its prometheus endpoint to backwards-compatible Beats metrics fields
+// that can be viewed in Agent dashboards.
+
+function process(event) {
+  // This hard-coded exporter name will not work for the general
+  // (non-monitoring) use case.
+  var elastic_exporter = event.Get("prometheus.labels.exporter") == "elasticsearch/_agent-component/monitoring";
+  var elastic_scope = event.Get("prometheus.labels.otel_scope_name") == "github.com/open-telemetry/opentelemetry-collector-contrib/exporter/elasticsearchexporter";
+
+  // We accept general collector fields that are scoped to the elasticsearch
+  // exporter (queue metrics, sent / error stats), or fields specifically
+  // scoped to the elasticsearch exporter (custom elastic metrics).
+  if (!elastic_exporter && !elastic_scope) {
+    event.Cancel();
+    return;
+  }
+
+  // Hack: if the scope is elastic-custom fields, deterministically mangle the
+  // agent.id. Since the label set is different, these are passed through in
+  // different events, and if we don't do this one of the events will be
+  // rejected as a duplicate since they have the same component id, agent id,
+  // and metricset.
+  var id = event.Get("agent.id");
+  if (id != null && id.length > 0) {
+    // Increment / wrap the last hex character of the uuid
+    var prefix = id.substring(0, id.length - 1);
+    var last = id.substring(id.length - 1);
+    var rotated = "0";
+    if (last < "f") {
+      rotated = String.fromCharCode(last.charCodeAt(0) + 1);
+    }
+    id = prefix + rotated;
+    event.Put("agent.id", id);
+  }
+
+  // The event will be discarded unless we find some valid metric to convert.
+	var keep_event = false;
+
+	var queue_size = event.Get("prometheus.metrics.otelcol_exporter_queue_size");
+	var queue_capacity = event.Get("prometheus.metrics.otelcol_exporter_queue_capacity");
+  if (queue_size != null) {
+  	keep_event = true;
+    event.Put("beat.stats.libbeat.pipeline.queue.filled.events", queue_size);
+  }
+  if (queue_capacity != null) {
+  	keep_event = true;
+    event.Put("beat.stats.libbeat.pipeline.queue.max_events", queue_capacity);
+  }
+	if (queue_size != null && queue_capacity != null) {
+		var queue_pct = queue_size / queue_capacity;
+		if (!isNaN(queue_pct)) {
+			event.Put("beat.stats.libbeat.pipeline.queue.filled.pct", queue_pct);
+		}
+	}
+
+  var total_sent = 0;
+  var total_sent_valid = false;
+  // Add send statistics from all source types
+  var sent_logs = event.Get("prometheus.metrics.otelcol_exporter_sent_log_records_total");
+  if (sent_logs != null) {
+    total_sent += sent_logs;
+    total_sent_valid = true;
+  }
+  var sent_spans = event.Get("prometheus.metrics.otelcol_exporter_sent_spans_total");
+  if (sent_spans != null) {
+    total_sent += sent_spans;
+    total_sent_valid = true;
+  }
+  var sent_metrics = event.Get("prometheus.metrics.otelcol_exporter_sent_metric_points_total");
+  if (sent_metrics != null) {
+    total_sent += sent_metrics;
+    total_sent_valid = true;
+  }
+  if (total_sent_valid) {
+    event.Put("beat.stats.libbeat.output.events.acked", total_sent);
+  	keep_event = true;
+  }
+
+  var total_failed = 0;
+  var total_failed_valid = false;
+  // Add failed statistics from all source types
+  var failed_logs = event.Get("prometheus.metrics.otelcol_exporter_send_failed_log_records_total");
+  if (failed_logs != null) {
+    total_failed += failed_logs;
+    total_failed_valid = true;
+  }
+  var failed_spans = event.Get("prometheus.metrics.otelcol_exporter_send_failed_spans_total");
+  if (failed_spans != null) {
+    total_failed += failed_spans;
+    total_failed_valid = true;
+  }
+  var failed_metrics = event.Get("prometheus.metrics.otelcol_exporter_send_failed_metric_points_total");
+  if (failed_metrics != null) {
+    total_failed += failed_metrics;
+    total_failed_valid = true;
+  }
+  if (total_failed_valid) {
+    event.Put("beat.stats.libbeat.output.events.dropped", total_failed);
+  	keep_event = true;
+  }
+
+  var flushed_bytes = event.Get("prometheus.metrics.otelcol_elasticsearch_flushed_bytes_total");
+  if (flushed_bytes != null) {
+    event.Put("beat.stats.libbeat.output.write.bytes", flushed_bytes);
+  	keep_event = true;
+  }
+
+  var retried_docs = event.Get("prometheus.metrics.otelcol_elasticsearch_docs_retried_ratio_total");
+  if (retried_docs != null) {
+    // "failed" in the beats metric means an event failed to ingest but was
+    // not dropped, and will be retried.
+    event.Put("beat.stats.libbeat.output.events.failed", retried_docs);
+  	keep_event = true;
+  }
+
+  var request_count = event.Get("prometheus.metrics.otelcol_elasticsearch_bulk_requests_count_ratio_total");
+  if (request_count != null) {
+    // This is not an exact semantic match for how Beats measures batch count,
+    // but it's close.
+    event.Put("beat.stats.libbeat.output.events.batches", request_count);
+  	keep_event = true;
+  }
+
+  var processed_docs_count = event.Get("prometheus.metrics.otelcol_elasticsearch_docs_processed_ratio_total");
+  if (processed_docs_count != null) {
+    // Approximate semantic match: the otel metric counts all document
+    // ingestion attempts, including success, failure, and retries,
+    // which is a better match for the Beats definition of total events
+    // than otelcol_elasticsearch_docs_received_ratio_total which
+    // includes only unique events seen (regardless of retries etc).
+    event.Put("beat.stats.libbeat.output.events.total", processed_docs_count);
+  	keep_event = true;
+  }
+
+  if (!keep_event) {
+    event.Cancel();
+  }
+}