Ensure monitoring the Otel collector never runs in a beat process (#11204) (#11283)

* Ensure monitoring the Otel collector never runs in a beat process

* Add changelog entry

* Move log lines to constants

(cherry picked from commit a9f4420)

Co-authored-by: Mikołaj Świątek <[email protected]>

Author: mergify[bot]

changelog/fragments/1763382303-prometheus-otel-self-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: bug-fix
+
+# Change summary; a 80ish characters long description of the change.
+summary: Ensure the monitoring input for the Otel collector can only run inside the collector.
+
+# 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

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

@@ -1661,6 +1661,12 @@ func (c *Coordinator) splitModelBetweenManagers(model *component.Model) (runtime
 		case component.OtelRuntimeManager:
 			otelComponents = append(otelComponents, comp)
 		case component.ProcessRuntimeManager:
+			// Hack to fix https://github.com/elastic/elastic-agent/issues/11169
+			// TODO: Remove this after https://github.com/elastic/elastic-agent/issues/10220 is resolved
+			if comp.ID == "prometheus/metrics-monitoring" {
+				c.logger.Warnf("The Otel prometheus metrics monitoring input can't run in a beats process, skipping")
+				continue
+			}
 			runtimeComponents = append(runtimeComponents, comp)
 		default:
 			// this should be impossible if we parse the configuration correctly

testing/integration/ess/beat_receivers_test.go

@@ -819,6 +819,12 @@ agent.monitoring.enabled: false
 	}
 }
 
+// Log lines TestBeatsReceiverProcessRuntimeFallback checks for
+const (
+	otelRuntimeUnsupportedLogLineStart = "otel runtime is not supported"
+	prometheusInputSkippedLogLine      = "The Otel prometheus metrics monitoring input can't run in a beats process, skipping"
+)
+
 // TestBeatsReceiverProcessRuntimeFallback verifies that we fall back to the process runtime if the otel runtime
 // does not support the requested configuration.
 func TestBeatsReceiverProcessRuntimeFallback(t *testing.T) {
@@ -850,7 +856,7 @@ outputs:
     hosts: [http://localhost:9200]
     api_key: placeholder
     indices: [] # not supported by the elasticsearch exporter
-agent.monitoring.enabled: false
+agent.monitoring._runtime_experimental: otel
 `
 
 	// this is the context for the whole test, with a global timeout defined
@@ -874,13 +880,14 @@ agent.monitoring.enabled: false
 		status, statusErr := fixture.ExecStatus(ctx)
 		assert.NoError(collect, statusErr)
 		// we should be running beats processes even though the otel runtime was requested
-		assertBeatsHealthy(collect, &status, component.ProcessRuntimeManager, 1)
+		assertBeatsHealthy(collect, &status, component.ProcessRuntimeManager, 4)
 	}, 1*time.Minute, 1*time.Second)
 	logsBytes, err := fixture.Exec(ctx, []string{"logs", "-n", "1000", "--exclude-events"})
 	require.NoError(t, err)
 
 	// verify we've logged a warning about using the process runtime
-	var unsupportedLogRecord map[string]any
+	var unsupportedLogRecords []map[string]any
+	var prometheusUnsupportedLogRecord map[string]any
 	for _, line := range strings.Split(string(logsBytes), "\n") {
 		line = strings.TrimSpace(line)
 		if line == "" {
@@ -891,9 +898,13 @@ agent.monitoring.enabled: false
 			continue
 		}
 
-		if message, ok := logRecord["message"].(string); ok && strings.HasPrefix(message, "otel runtime is not supported") {
-			unsupportedLogRecord = logRecord
-			break
+		if message, ok := logRecord["message"].(string); ok {
+			if strings.HasPrefix(message, otelRuntimeUnsupportedLogLineStart) {
+				unsupportedLogRecords = append(unsupportedLogRecords, logRecord)
+			}
+			if strings.HasPrefix(message, prometheusInputSkippedLogLine) {
+				prometheusUnsupportedLogRecord = logRecord
+			}
 		}
 	}
 
@@ -904,11 +915,8 @@ agent.monitoring.enabled: false
 		}
 	})
 
-	require.NotNil(t, unsupportedLogRecord, "unsupported log message should be present")
-	message, ok := unsupportedLogRecord["message"].(string)
-	require.True(t, ok, "log message field should be a string")
-	expectedMessage := "otel runtime is not supported for component system/metrics-default, switching to process runtime, reason: unsupported configuration for system/metrics-default: error translating config for output: default, unit: system/metrics-default, error: indices is currently not supported: unsupported operation"
-	assert.Equal(t, expectedMessage, message)
+	assert.Len(t, unsupportedLogRecords, 5, "one log line for each component we try to run")
+	assert.NotEmpty(t, prometheusUnsupportedLogRecord, "should get a log line about Otel prometheus metrics input being skipped")
 }
 
 // TestComponentWorkDir verifies that the component working directory is not deleted when moving the component from