[exporter/elasticsearch] Add explicit bounds histogram support to metrics

.chloggen/elasticsearchexporter_metrics-histogram-support.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: enhancement
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: elasticsearchexporter
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Add explicit bounds histogram support to metrics
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [34045]
+
+# (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]

exporter/elasticsearchexporter/data_stream_router.go

@@ -8,7 +8,6 @@ import (
 
 	"go.opentelemetry.io/collector/pdata/pcommon"
 	"go.opentelemetry.io/collector/pdata/plog"
-	"go.opentelemetry.io/collector/pdata/pmetric"
 	"go.opentelemetry.io/collector/pdata/ptrace"
 )
 
@@ -60,7 +59,7 @@ func routeLogRecord(
 // routeDataPoint returns the name of the index to send the data point to according to data stream routing attributes.
 // This function may mutate record attributes.
 func routeDataPoint(
-	dataPoint pmetric.NumberDataPoint,
+	dataPoint dataPoint,
 	scope pcommon.InstrumentationScope,
 	resource pcommon.Resource,
 	fIndex string,

exporter/elasticsearchexporter/exporter.go

@@ -179,27 +179,64 @@ func (e *elasticsearchExporter) pushMetricsData(
 			for k := 0; k < scopeMetrics.Metrics().Len(); k++ {
 				metric := scopeMetrics.Metrics().At(k)
 
-				// We only support Sum and Gauge metrics at the moment.
-				var dataPoints pmetric.NumberDataPointSlice
-				switch metric.Type() {
-				case pmetric.MetricTypeSum:
-					dataPoints = metric.Sum().DataPoints()
-				case pmetric.MetricTypeGauge:
-					dataPoints = metric.Gauge().DataPoints()
-				}
-
-				for l := 0; l < dataPoints.Len(); l++ {
-					dataPoint := dataPoints.At(l)
-					fIndex, err := e.getMetricDataPointIndex(resource, scope, dataPoint)
+				upsertDataPoint := func(dp dataPoint, dpValue pcommon.Value) error {
+					fIndex, err := e.getMetricDataPointIndex(resource, scope, dp)
 					if err != nil {
-						errs = append(errs, err)
-						continue
+						return err
 					}
 					if _, ok := resourceDocs[fIndex]; !ok {
 						resourceDocs[fIndex] = make(map[uint32]objmodel.Document)
 					}
-					if err := e.model.upsertMetricDataPoint(resourceDocs[fIndex], resource, scope, metric, dataPoint); err != nil {
-						errs = append(errs, err)
+
+					if err = e.model.upsertMetricDataPointValue(resourceDocs[fIndex], resource, scope, metric, dp, dpValue); err != nil {
+						return err
+					}
+					return nil
+				}
+
+				// TODO: support exponential histogram
+				switch metric.Type() {
+				case pmetric.MetricTypeSum:
+					dps := metric.Sum().DataPoints()
+					for l := 0; l < dps.Len(); l++ {
+						dp := dps.At(l)
+						val, err := numberToValue(dp)
+						if err != nil {
+							errs = append(errs, err)
+							continue
+						}
+						if err := upsertDataPoint(dp, val); err != nil {
+							errs = append(errs, err)
+							continue
+						}
+					}
+				case pmetric.MetricTypeGauge:
+					dps := metric.Gauge().DataPoints()
+					for l := 0; l < dps.Len(); l++ {
+						dp := dps.At(l)
+						val, err := numberToValue(dp)
+						if err != nil {
+							errs = append(errs, err)
+							continue
+						}
+						if err := upsertDataPoint(dp, val); err != nil {
+							errs = append(errs, err)
+							continue
+						}
+					}
+				case pmetric.MetricTypeHistogram:
+					dps := metric.Histogram().DataPoints()
+					for l := 0; l < dps.Len(); l++ {
+						dp := dps.At(l)
+						val, err := histogramToValue(dp)
+						if err != nil {
+							errs = append(errs, err)
+							continue
+						}
+						if err := upsertDataPoint(dp, val); err != nil {
+							errs = append(errs, err)
+							continue
+						}
 					}
 				}
 			}
@@ -238,7 +275,7 @@ func (e *elasticsearchExporter) pushMetricsData(
 func (e *elasticsearchExporter) getMetricDataPointIndex(
 	resource pcommon.Resource,
 	scope pcommon.InstrumentationScope,
-	dataPoint pmetric.NumberDataPoint,
+	dataPoint dataPoint,
 ) (string, error) {
 	fIndex := e.index
 	if e.dynamicIndex {

exporter/elasticsearchexporter/exporter_test.go

@@ -8,6 +8,7 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"math"
 	"net/http"
 	"runtime"
 	"sync"
@@ -495,6 +496,7 @@ func TestExporterMetrics(t *testing.T) {
 			},
 		)
 		metrics.ResourceMetrics().At(0).ScopeMetrics().At(0).Metrics().At(0).SetName("my.metric")
+		metrics.ResourceMetrics().At(0).ScopeMetrics().At(0).Metrics().At(0).SetEmptySum().DataPoints().AppendEmpty().SetIntValue(0)
 		mustSendMetrics(t, exporter, metrics)
 
 		rec.WaitItems(1)
@@ -633,6 +635,103 @@ func TestExporterMetrics(t *testing.T) {
 
 		assertItemsEqual(t, expected, rec.Items(), false)
 	})
+
+	t.Run("publish histogram", func(t *testing.T) {
+		rec := newBulkRecorder()
+		server := newESTestServer(t, func(docs []itemRequest) ([]itemResponse, error) {
+			rec.Record(docs)
+			return itemsAllOK(docs)
+		})
+
+		exporter := newTestMetricsExporter(t, server.URL, func(cfg *Config) {
+			cfg.Mapping.Mode = "ecs"
+		})
+
+		metrics := pmetric.NewMetrics()
+		resourceMetrics := metrics.ResourceMetrics().AppendEmpty()
+		scopeA := resourceMetrics.ScopeMetrics().AppendEmpty()
+		metricSlice := scopeA.Metrics()
+		fooMetric := metricSlice.AppendEmpty()
+		fooMetric.SetName("metric.foo")
+		fooDps := fooMetric.SetEmptyHistogram().DataPoints()
+		fooDp := fooDps.AppendEmpty()
+		fooDp.ExplicitBounds().FromRaw([]float64{1.0, 2.0, 3.0})
+		fooDp.BucketCounts().FromRaw([]uint64{1, 2, 3, 4})
+		fooOtherDp := fooDps.AppendEmpty()
+		fooOtherDp.SetTimestamp(pcommon.NewTimestampFromTime(time.Unix(3600, 0)))
+		fooOtherDp.ExplicitBounds().FromRaw([]float64{4.0, 5.0, 6.0})
+		fooOtherDp.BucketCounts().FromRaw([]uint64{4, 5, 6, 7})
+
+		mustSendMetrics(t, exporter, metrics)
+
+		rec.WaitItems(2)
+
+		expected := []itemRequest{
+			{
+				Action:   []byte(`{"create":{"_index":"metrics-generic-default"}}`),
+				Document: []byte(`{"@timestamp":"1970-01-01T00:00:00.000000000Z","data_stream":{"dataset":"generic","namespace":"default","type":"metrics"},"metric":{"foo":{"counts":[1,2,3,4],"values":[0.5,1.5,2.5,3]}}}`),
+			},
+			{
+				Action:   []byte(`{"create":{"_index":"metrics-generic-default"}}`),
+				Document: []byte(`{"@timestamp":"1970-01-01T01:00:00.000000000Z","data_stream":{"dataset":"generic","namespace":"default","type":"metrics"},"metric":{"foo":{"counts":[4,5,6,7],"values":[2,4.5,5.5,6]}}}`),
+			},
+		}
+
+		assertItemsEqual(t, expected, rec.Items(), false)
+	})
+
+	t.Run("publish only valid data points", func(t *testing.T) {
+		rec := newBulkRecorder()
+		server := newESTestServer(t, func(docs []itemRequest) ([]itemResponse, error) {
+			rec.Record(docs)
+			return itemsAllOK(docs)
+		})
+
+		exporter := newTestMetricsExporter(t, server.URL, func(cfg *Config) {
+			cfg.Mapping.Mode = "ecs"
+		})
+
+		metrics := pmetric.NewMetrics()
+		resourceMetrics := metrics.ResourceMetrics().AppendEmpty()
+		scopeA := resourceMetrics.ScopeMetrics().AppendEmpty()
+		metricSlice := scopeA.Metrics()
+		fooMetric := metricSlice.AppendEmpty()
+		fooMetric.SetName("metric.foo")
+		fooDps := fooMetric.SetEmptyHistogram().DataPoints()
+		fooDp := fooDps.AppendEmpty()
+		fooDp.ExplicitBounds().FromRaw([]float64{1.0, 2.0, 3.0})
+		fooDp.BucketCounts().FromRaw([]uint64{})
+		fooOtherDp := fooDps.AppendEmpty()
+		fooOtherDp.SetTimestamp(pcommon.NewTimestampFromTime(time.Unix(3600, 0)))
+		fooOtherDp.ExplicitBounds().FromRaw([]float64{4.0, 5.0, 6.0})
+		fooOtherDp.BucketCounts().FromRaw([]uint64{4, 5, 6, 7})
+		barMetric := metricSlice.AppendEmpty()
+		barMetric.SetName("metric.bar")
+		barDps := barMetric.SetEmptySum().DataPoints()
+		barDp := barDps.AppendEmpty()
+		barDp.SetDoubleValue(math.Inf(1))
+		barOtherDp := barDps.AppendEmpty()
+		barOtherDp.SetDoubleValue(1.0)
+
+		err := exporter.ConsumeMetrics(context.Background(), metrics)
+		require.ErrorContains(t, err, "invalid histogram data point")
+		require.ErrorContains(t, err, "invalid number data point")
+
+		rec.WaitItems(2)
+
+		expected := []itemRequest{
+			{
+				Action:   []byte(`{"create":{"_index":"metrics-generic-default"}}`),
+				Document: []byte(`{"@timestamp":"1970-01-01T00:00:00.000000000Z","data_stream":{"dataset":"generic","namespace":"default","type":"metrics"},"metric":{"bar":1}}`),
+			},
+			{
+				Action:   []byte(`{"create":{"_index":"metrics-generic-default"}}`),
+				Document: []byte(`{"@timestamp":"1970-01-01T01:00:00.000000000Z","data_stream":{"dataset":"generic","namespace":"default","type":"metrics"},"metric":{"foo":{"counts":[4,5,6,7],"values":[2,4.5,5.5,6]}}}`),
+			},
+		}
+
+		assertItemsEqual(t, expected, rec.Items(), false)
+	})
 }
 
 func TestExporterTraces(t *testing.T) {

exporter/elasticsearchexporter/model.go

@@ -7,6 +7,7 @@ import (
 	"bytes"
 	"encoding/binary"
 	"encoding/json"
+	"errors"
 	"fmt"
 	"hash"
 	"hash/fnv"
@@ -65,7 +66,7 @@ var resourceAttrsToPreserve = map[string]bool{
 type mappingModel interface {
 	encodeLog(pcommon.Resource, plog.LogRecord, pcommon.InstrumentationScope) ([]byte, error)
 	encodeSpan(pcommon.Resource, ptrace.Span, pcommon.InstrumentationScope) ([]byte, error)
-	upsertMetricDataPoint(map[uint32]objmodel.Document, pcommon.Resource, pcommon.InstrumentationScope, pmetric.Metric, pmetric.NumberDataPoint) error
+	upsertMetricDataPointValue(map[uint32]objmodel.Document, pcommon.Resource, pcommon.InstrumentationScope, pmetric.Metric, dataPoint, pcommon.Value) error
 	encodeDocument(objmodel.Document) ([]byte, error)
 }
 
@@ -80,6 +81,11 @@ type encodeModel struct {
 	mode  MappingMode
 }
 
+type dataPoint interface {
+	Timestamp() pcommon.Timestamp
+	Attributes() pcommon.Map
+}
+
 const (
 	traceIDField   = "traceID"
 	spanIDField    = "spanID"
@@ -176,7 +182,7 @@ func (m *encodeModel) encodeDocument(document objmodel.Document) ([]byte, error)
 	return buf.Bytes(), nil
 }
 
-func (m *encodeModel) upsertMetricDataPoint(documents map[uint32]objmodel.Document, resource pcommon.Resource, _ pcommon.InstrumentationScope, metric pmetric.Metric, dp pmetric.NumberDataPoint) error {
+func (m *encodeModel) upsertMetricDataPointValue(documents map[uint32]objmodel.Document, resource pcommon.Resource, _ pcommon.InstrumentationScope, metric pmetric.Metric, dp dataPoint, value pcommon.Value) error {
 	hash := metricHash(dp.Timestamp(), dp.Attributes())
 	var (
 		document objmodel.Document
@@ -188,15 +194,74 @@ func (m *encodeModel) upsertMetricDataPoint(documents map[uint32]objmodel.Docume
 		document.AddAttributes("", dp.Attributes())
 	}
 
+	document.AddAttribute(metric.Name(), value)
+
+	documents[hash] = document
+	return nil
+}
+
+func histogramToValue(dp pmetric.HistogramDataPoint) (pcommon.Value, error) {
+	// Histogram conversion function is from
+	// https://github.com/elastic/apm-data/blob/3b28495c3cbdc0902983134276eb114231730249/input/otlp/metrics.go#L277
+	bucketCounts := dp.BucketCounts()
+	explicitBounds := dp.ExplicitBounds()
+	if bucketCounts.Len() != explicitBounds.Len()+1 || explicitBounds.Len() == 0 {
+		return pcommon.Value{}, errors.New("invalid histogram data point")
+	}
+
+	vm := pcommon.NewValueMap()
+	m := vm.Map()
+	counts := m.PutEmptySlice("counts")
+	values := m.PutEmptySlice("values")
+
+	values.EnsureCapacity(bucketCounts.Len())
+	counts.EnsureCapacity(bucketCounts.Len())
+	for i := 0; i < bucketCounts.Len(); i++ {
+		count := bucketCounts.At(i)
+		if count == 0 {
+			continue
+		}
+
+		var value float64
+		switch i {
+		// (-infinity, explicit_bounds[i]]
+		case 0:
+			value = explicitBounds.At(i)
+			if value > 0 {
+				value /= 2
+			}
+
+		// (explicit_bounds[i], +infinity)
+		case bucketCounts.Len() - 1:
+			value = explicitBounds.At(i - 1)
+
+		// [explicit_bounds[i-1], explicit_bounds[i])
+		default:
+			// Use the midpoint between the boundaries.
+			value = explicitBounds.At(i-1) + (explicitBounds.At(i)-explicitBounds.At(i-1))/2.0
+		}
+
+		counts.AppendEmpty().SetInt(int64(count))
+		values.AppendEmpty().SetDouble(value)
+	}
+
+	return vm, nil
+}
+
+var errInvalidNumberDataPoint = errors.New("invalid number data point")
+
+func numberToValue(dp pmetric.NumberDataPoint) (pcommon.Value, error) {
 	switch dp.ValueType() {
 	case pmetric.NumberDataPointValueTypeDouble:
-		document.AddAttribute(metric.Name(), pcommon.NewValueDouble(dp.DoubleValue()))
+		value := dp.DoubleValue()
+		if math.IsNaN(value) || math.IsInf(value, 0) {
+			return pcommon.Value{}, errInvalidNumberDataPoint
+		}
+		return pcommon.NewValueDouble(value), nil
 	case pmetric.NumberDataPointValueTypeInt:
-		document.AddAttribute(metric.Name(), pcommon.NewValueInt(dp.IntValue()))
+		return pcommon.NewValueInt(dp.IntValue()), nil
 	}
-
-	documents[hash] = document
-	return nil
+	return pcommon.Value{}, errInvalidNumberDataPoint
 }
 
 func (m *encodeModel) encodeSpan(resource pcommon.Resource, span ptrace.Span, scope pcommon.InstrumentationScope) ([]byte, error) {

exporter/elasticsearchexporter/model_test.go

@@ -99,11 +99,14 @@ func TestEncodeMetric(t *testing.T) {
 
 	var docsBytes [][]byte
 	for i := 0; i < metrics.ResourceMetrics().At(0).ScopeMetrics().At(0).Metrics().At(0).Sum().DataPoints().Len(); i++ {
-		err := model.upsertMetricDataPoint(docs,
+		val, err := numberToValue(metrics.ResourceMetrics().At(0).ScopeMetrics().At(0).Metrics().At(0).Sum().DataPoints().At(i))
+		require.NoError(t, err)
+		err = model.upsertMetricDataPointValue(docs,
 			metrics.ResourceMetrics().At(0).Resource(),
 			metrics.ResourceMetrics().At(0).ScopeMetrics().At(0).Scope(),
 			metrics.ResourceMetrics().At(0).ScopeMetrics().At(0).Metrics().At(0),
-			metrics.ResourceMetrics().At(0).ScopeMetrics().At(0).Metrics().At(0).Sum().DataPoints().At(i))
+			metrics.ResourceMetrics().At(0).ScopeMetrics().At(0).Metrics().At(0).Sum().DataPoints().At(i),
+			val)
 		require.NoError(t, err)
 	}