fix(servicegraph): make virtual node tests reliable on Windows

.chloggen/fix_servicegraph-windows-tests.yaml

@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: bug_fix
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: connector/servicegraphconnector
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Make virtual node tests reliable on Windows by improving timing and polling mechanisms.
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [33836]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: [user]

connector/servicegraphconnector/connector_test.go

@@ -647,18 +647,17 @@ func TestExtraDimensionsLabels(t *testing.T) {
 }
 
 func TestVirtualNodeServerLabels(t *testing.T) {
-	if runtime.GOOS == "windows" {
-		t.Skip("skipping test on Windows, see https://github.com/open-telemetry/opentelemetry-collector-contrib/issues/33836")
-	}
-
 	virtualNodeDimensions := []string{"peer.service", "db.system", "messaging.system"}
 	cfg := &Config{
 		Dimensions:                virtualNodeDimensions,
 		LatencyHistogramBuckets:   []time.Duration{time.Duration(0.1 * float64(time.Second)), time.Duration(1 * float64(time.Second)), time.Duration(10 * float64(time.Second))},
 		Store:                     StoreConfig{MaxItems: 10},
 		VirtualNodePeerAttributes: virtualNodeDimensions,
 		VirtualNodeExtraLabel:     true,
-		MetricsFlushInterval:      time.Millisecond,
+		// Reduce flush interval for faster test execution
+		MetricsFlushInterval: 10 * time.Millisecond,
+		// Reduce store expiration interval
+		StoreExpirationLoop: 10 * time.Millisecond,
 	}
 
 	set := componenttest.NewNopTelemetrySettings()
@@ -675,14 +674,19 @@ func TestVirtualNodeServerLabels(t *testing.T) {
 	assert.NoError(t, err)
 	assert.NoError(t, conn.ConsumeTraces(context.Background(), td))
 
-	conn.store.Expire()
-	assert.Eventually(t, func() bool {
-		return conn.store.Len() == 0
-	}, 100*time.Millisecond, 2*time.Millisecond)
-	require.NoError(t, conn.Shutdown(context.Background()))
+	// Wait for metrics to be generated with timeout
+	deadline := time.Now().Add(5 * time.Second)
+	var metrics []pmetric.Metrics
+	for time.Now().Before(deadline) {
+		metrics = conn.metricsConsumer.(*mockMetricsExporter).GetMetrics()
+		if len(metrics) > 0 {
+			break
+		}
+		time.Sleep(10 * time.Millisecond)
+	}
 
-	metrics := conn.metricsConsumer.(*mockMetricsExporter).GetMetrics()
-	require.GreaterOrEqual(t, len(metrics), 1) // Unreliable sleep-based check
+	require.NotEmpty(t, metrics, "no metrics generated within timeout")
+	require.NoError(t, conn.Shutdown(context.Background()))
 
 	expectedMetrics, err := golden.ReadMetrics(expected)
 	assert.NoError(t, err)
@@ -695,18 +699,17 @@ func TestVirtualNodeServerLabels(t *testing.T) {
 }
 
 func TestVirtualNodeClientLabels(t *testing.T) {
-	if runtime.GOOS == "windows" {
-		t.Skip("skipping test on Windows, see https://github.com/open-telemetry/opentelemetry-collector-contrib/issues/33836")
-	}
-
 	virtualNodeDimensions := []string{"peer.service", "db.system", "messaging.system"}
 	cfg := &Config{
 		Dimensions:                virtualNodeDimensions,
 		LatencyHistogramBuckets:   []time.Duration{time.Duration(0.1 * float64(time.Second)), time.Duration(1 * float64(time.Second)), time.Duration(10 * float64(time.Second))},
 		Store:                     StoreConfig{MaxItems: 10},
 		VirtualNodePeerAttributes: virtualNodeDimensions,
 		VirtualNodeExtraLabel:     true,
-		MetricsFlushInterval:      time.Millisecond,
+		// Reduce flush interval for faster test execution
+		MetricsFlushInterval: 10 * time.Millisecond,
+		// Reduce store expiration interval
+		StoreExpirationLoop: 10 * time.Millisecond,
 	}
 
 	set := componenttest.NewNopTelemetrySettings()
@@ -723,14 +726,19 @@ func TestVirtualNodeClientLabels(t *testing.T) {
 	assert.NoError(t, err)
 	assert.NoError(t, conn.ConsumeTraces(context.Background(), td))
 
-	conn.store.Expire()
-	assert.Eventually(t, func() bool {
-		return conn.store.Len() == 0
-	}, 100*time.Millisecond, 2*time.Millisecond)
-	require.NoError(t, conn.Shutdown(context.Background()))
+	// Wait for metrics to be generated with timeout
+	deadline := time.Now().Add(5 * time.Second)
+	var metrics []pmetric.Metrics
+	for time.Now().Before(deadline) {
+		metrics = conn.metricsConsumer.(*mockMetricsExporter).GetMetrics()
+		if len(metrics) > 0 {
+			break
+		}
+		time.Sleep(10 * time.Millisecond)
+	}
 
-	metrics := conn.metricsConsumer.(*mockMetricsExporter).GetMetrics()
-	require.GreaterOrEqual(t, len(metrics), 1) // Unreliable sleep-based check
+	require.NotEmpty(t, metrics, "no metrics generated within timeout")
+	require.NoError(t, conn.Shutdown(context.Background()))
 
 	expectedMetrics, err := golden.ReadMetrics(expected)
 	assert.NoError(t, err)