Merge branch 'main' into v2-sanitizer

Signed-off-by: Vamshi Maskuri <[email protected]>

.github/workflows/ci-e2e-spm.yml

@@ -18,6 +18,14 @@ permissions:
 jobs:
   spm:
     runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        mode:
+        - name: v1
+          binary: all-in-one
+        - name: v2
+          binary: jaeger
+
     steps:
     - name: Harden Runner
       uses: step-security/harden-runner@17d0e2bd7d51742c71671bd19fa12bdc9d40a3d6 # v2.8.1
@@ -38,11 +46,6 @@ jobs:
 
     - name: Setup Node.js version
       uses: ./.github/actions/setup-node.js
-
-    - name: Temporary - only run the build
-      run:
-        cd docker-compose/monitor && make build
-
+
     - name: Run SPM Test
-      run: ./scripts/spm-integration-test.sh
-
+      run:  bash scripts/spm-integration-test.sh  -b ${{ matrix.mode.binary }}

CHANGELOG.md

@@ -28,6 +28,11 @@ run `make changelog` to generate content
 
 #### ⛔ Breaking Changes
 
+* The OTEL Collector upgrade brought in a change where OTLP receivers started listening on `localhost` instead of `0.0.0.0` as before. As a result, when running in container environment the endpoints are likely unreachable from other containers (Issue [#5737](https://github.com/jaegertracing/jaeger/issues/5737)). The fix will be available in the next release. Meanwhile, the workaround is to instruct Jaeger to listen on `0.0.0.0`, as in [this fix](https://github.com/jaegertracing/jaeger/pull/5734/files#diff-299f817cc4ab077ddb763f1e6a023d9d042d714e2fd3736cc40af3f218d44f1eR15):
+```
+      - COLLECTOR_OTLP_GRPC_HOST_PORT=0.0.0.0:4317
+      - COLLECTOR_OTLP_HTTP_HOST_PORT=0.0.0.0:4318
+```
 * Update opentelemetry-go to v1.28.0 and refactor references to semantic conventions ([@renovate-bot](https://github.com/renovate-bot) in [#5698](https://github.com/jaegertracing/jaeger/pull/5698))
 
 #### ✨ New Features

README.md

@@ -3,9 +3,9 @@
 <img align="right" width="290" height="290" src="https://www.jaegertracing.io/img/jaeger-vector.svg">
 
 [![Slack chat][slack-img]](#get-in-touch)
-[![Project+Community stats][community-badge]][community-stats]
 [![Unit Tests][ci-img]][ci]
 [![Coverage Status][cov-img]][cov]
+[![Project+Community stats][community-badge]][community-stats]
 [![FOSSA Status][fossa-img]][fossa]
 [![OpenSSF Scorecard][openssf-img]][openssf]
 [![OpenSSF Best Practices][openssf-bp-img]][openssf-bp] 
@@ -16,6 +16,8 @@
 
 # Jaeger - a Distributed Tracing System
 
+💥💥💥 Jaeger v2 is coming! Read the [blog post](https://medium.com/jaegertracing/towards-jaeger-v2-moar-opentelemetry-2f8239bee48e) and [try it out](./cmd/jaeger).
+
 ```mermaid
 graph TD
     SDK["OpenTelemetry SDK"] --> |HTTP or gRPC| COLLECTOR

cmd/all-in-one/all_in_one_test.go

@@ -89,6 +89,7 @@ func healthCheck(t *testing.T) {
 }
 
 func httpGet(t *testing.T, url string) (*http.Response, []byte) {
+	t.Logf("Executing HTTP GET %s", url)
 	req, err := http.NewRequest(http.MethodGet, url, nil)
 	require.NoError(t, err)
 	req.Close = true // avoid persistent connections which leak goroutines

cmd/jaeger/README.md

@@ -3,8 +3,50 @@
 This is experimental Jaeger V2 based on OpenTelemetry collector.
 See https://github.com/jaegertracing/jaeger/issues/4843.
 
+Read the [blog post](https://medium.com/jaegertracing/towards-jaeger-v2-moar-opentelemetry-2f8239bee48e).
+
+Tracking issue: https://github.com/jaegertracing/jaeger/issues/4843.
+
+```mermaid
+flowchart LR
+    Receiver1 --> Processor
+    Receiver2 --> Processor
+    Receiver3 --> Processor
+    Processor --> Exporter
+
+    Exporter --> Database
+    Database --> Query[Query + UI]
+
+    subgraph Pipeline
+        Receiver1[OTLP Receiver]
+        Receiver2[Jaeger Proto Receiver]
+        Receiver3[Zipkin Receiver]
+        Processor[Batch
+            Processor]
+        Exporter[Jaeger
+            Storage
+            Exporter]
+    end
+
+    subgraph JaegerStorageExension[Jaeger Storage Ext]
+        Database[(Storage)]
+    end
+    subgraph JaegerQueryExtension[Jaeger Query Ext]
+        Query
+    end
+```
+
+## Try it out
+
+* Download `docker-compose-v2.yml` from https://github.com/jaegertracing/jaeger/blob/main/examples/hotrod/docker-compose-v2.yml
+* Optional: find the latest Jaeger version (see https://www.jaegertracing.io/download/) and pass it via environment variable `JAEGER_VERSION`. Otherwise `docker compose` will use the `latest` tag, which is fine for the first time you download the images, but once they are in your local registry the `latest` tag is never updated and you may be running stale (and possibly incompatible) verions of Jaeger and the HotROD app.
+* Run Jaeger backend and HotROD demo, e.g.:
+  * `JAEGER_VERSION=1.59 docker compose -f path-to-yml-file-v2 up`
+* Access Jaeger UI at http://localhost:16686 and HotROD app at http://localhost:8080
+* Shutdown / cleanup with `docker compose -f path-to-yml-file down`
+
 ## Compatibility
 
 ### Service Name Sanitizer
 
-In v1, there was a `serviceNameSanitizer` that sanitized the service names in span annotations using a source of truth alias to service cache. This functionality has been removed in v2. If your implementation relies on this sanitizer, you will need to find a different way to integrate this functionality, such as implementing a custom processor.
+In v1, there was a `serviceNameSanitizer` that sanitized the service names in span annotations using a source of truth alias to service cache. This functionality has been removed in v2. If your implementation relies on this sanitizer, you will need to find a different way to integrate this functionality, such as implementing a custom processor.

cmd/jaeger/collector-with-kafka.yaml

@@ -0,0 +1,28 @@
+service:
+  pipelines:
+    traces:
+      receivers: [otlp, jaeger]
+      processors: [batch]
+      exporters: [kafka]
+
+receivers:
+  otlp:
+    protocols:
+      grpc:
+      http:
+  jaeger:
+    protocols:
+      grpc:
+      thrift_binary:
+      thrift_compact:
+      thrift_http:
+
+processors:
+  batch:
+
+exporters:
+  kafka:
+    brokers:
+      - localhost:9092
+    topic: "jaeger-spans"
+    encoding: otlp_proto

cmd/jaeger/ingester-remote-storage.yaml

@@ -0,0 +1,37 @@
+service:
+  extensions: [jaeger_storage, jaeger_query]
+  pipelines:
+    traces:
+      receivers: [kafka]
+      processors: [batch]
+      exporters: [jaeger_storage_exporter]
+  telemetry:
+    metrics:
+      address: 0.0.0.0:8889
+    logs:
+      level: debug
+
+extensions:
+  jaeger_query:
+    trace_storage: some_storage
+
+  jaeger_storage:
+    backends:
+      some_storage:
+        memory:
+          max_traces: 100000
+
+receivers:  
+  kafka:
+    brokers:
+      - localhost:9092
+    topic: "jaeger-spans"
+    encoding: otlp_proto
+    initial_offset: earliest
+
+processors:
+  batch:
+
+exporters:
+  jaeger_storage_exporter:
+    trace_storage: some_storage

cmd/jaeger/internal/all-in-one.yaml

@@ -5,6 +5,13 @@ service:
       receivers: [otlp, jaeger, zipkin]
       processors: [batch]
       exporters: [jaeger_storage_exporter]
+  telemetry:
+    resource:
+      service.name: jaeger
+    metrics:
+      level: detailed
+    # TODO Initialize telemetery tracer once OTEL released new feature.
+    # https://github.com/open-telemetry/opentelemetry-collector/issues/10663 
 
 extensions:
   jaeger_query:

cmd/jaeger/internal/components.go

@@ -6,6 +6,7 @@ package internal
 import (
 	"github.com/open-telemetry/opentelemetry-collector-contrib/connector/spanmetricsconnector"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/exporter/kafkaexporter"
+	"github.com/open-telemetry/opentelemetry-collector-contrib/exporter/prometheusexporter"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/receiver/jaegerreceiver"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/receiver/kafkareceiver"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/receiver/zipkinreceiver"
@@ -90,6 +91,7 @@ func (b builders) build() (otelcol.Factories, error) {
 		// add-ons
 		storageexporter.NewFactory(), // generic exporter to Jaeger v1 spanstore.SpanWriter
 		kafkaexporter.NewFactory(),
+		prometheusexporter.NewFactory(),
 		// elasticsearch.NewFactory(),
 	)
 	if err != nil {

cmd/jaeger/internal/exporters/storageexporter/exporter_test.go

@@ -39,10 +39,13 @@ import (
 )
 
 type mockStorageExt struct {
-	name    string
-	factory *factoryMocks.Factory
+	name           string
+	factory        *factoryMocks.Factory
+	metricsFactory *factoryMocks.MetricsFactory
 }
 
+var _ jaegerstorage.Extension = (*mockStorageExt)(nil)
+
 func (*mockStorageExt) Start(context.Context, component.Host) error {
 	panic("not implemented")
 }
@@ -51,13 +54,20 @@ func (*mockStorageExt) Shutdown(context.Context) error {
 	panic("not implemented")
 }
 
-func (m *mockStorageExt) Factory(name string) (storage.Factory, bool) {
+func (m *mockStorageExt) TraceStorageFactory(name string) (storage.Factory, bool) {
 	if m.name == name {
 		return m.factory, true
 	}
 	return nil, false
 }
 
+func (m *mockStorageExt) MetricStorageFactory(name string) (storage.MetricsFactory, bool) {
+	if m.name == name {
+		return m.metricsFactory, true
+	}
+	return nil, false
+}
+
 func TestExporterConfigError(t *testing.T) {
 	config := createDefaultConfig().(*Config)
 	err := config.Validate()

cmd/jaeger/internal/extension/jaegerquery/config.go

@@ -18,14 +18,12 @@ var _ component.ConfigValidator = (*Config)(nil)
 // Config represents the configuration for jaeger-query,
 type Config struct {
 	queryApp.QueryOptionsBase `mapstructure:",squash"`
-
-	TraceStoragePrimary string `valid:"required" mapstructure:"trace_storage"`
-	TraceStorageArchive string `valid:"optional" mapstructure:"trace_storage_archive"`
-
-	HTTP confighttp.ServerConfig `mapstructure:",squash"`
-	GRPC configgrpc.ServerConfig `mapstructure:",squash"`
-
-	Tenancy tenancy.Options `mapstructure:"multi_tenancy"`
+	TraceStoragePrimary       string                  `valid:"required" mapstructure:"trace_storage"`
+	TraceStorageArchive       string                  `valid:"optional" mapstructure:"trace_storage_archive"`
+	MetricStorage             string                  `valid:"optional" mapstructure:"metric_storage"`
+	HTTP                      confighttp.ServerConfig `mapstructure:",squash"`
+	GRPC                      configgrpc.ServerConfig `mapstructure:",squash"`
+	Tenancy                   tenancy.Options         `mapstructure:"multi_tenancy"`
 }
 
 func (cfg *Config) Validate() error {

cmd/jaeger/internal/extension/jaegerquery/server.go

@@ -14,10 +14,14 @@ import (
 	"github.com/jaegertracing/jaeger/cmd/jaeger/internal/extension/jaegerstorage"
 	queryApp "github.com/jaegertracing/jaeger/cmd/query/app"
 	"github.com/jaegertracing/jaeger/cmd/query/app/querysvc"
+	"github.com/jaegertracing/jaeger/internal/metrics/otelmetrics"
 	"github.com/jaegertracing/jaeger/pkg/jtracer"
+	"github.com/jaegertracing/jaeger/pkg/metrics"
 	"github.com/jaegertracing/jaeger/pkg/telemetery"
 	"github.com/jaegertracing/jaeger/pkg/tenancy"
 	"github.com/jaegertracing/jaeger/plugin/metrics/disabled"
+	"github.com/jaegertracing/jaeger/storage/metricsstore"
+	storageMetrics "github.com/jaegertracing/jaeger/storage/spanstore/metrics"
 )
 
 var (
@@ -45,6 +49,8 @@ func (*server) Dependencies() []component.ID {
 }
 
 func (s *server) Start(_ context.Context, host component.Host) error {
+	mf := otelmetrics.NewFactory(s.telset.MeterProvider)
+	queryMetricsFactory := mf.Namespace(metrics.NSOptions{Name: "query"})
 	f, err := jaegerstorage.GetStorageFactory(s.config.TraceStoragePrimary, host)
 	if err != nil {
 		return fmt.Errorf("cannot find primary storage %s: %w", s.config.TraceStoragePrimary, err)
@@ -54,8 +60,8 @@ func (s *server) Start(_ context.Context, host component.Host) error {
 	if err != nil {
 		return fmt.Errorf("cannot create span reader: %w", err)
 	}
-	// TODO
-	// spanReader = storageMetrics.NewReadMetricsDecorator(spanReader, baseFactory.Namespace(metrics.NSOptions{Name: "query"}))
+
+	spanReader = storageMetrics.NewReadMetricsDecorator(spanReader, queryMetricsFactory)
 
 	depReader, err := f.CreateDependencyReader()
 	if err != nil {
@@ -67,7 +73,12 @@ func (s *server) Start(_ context.Context, host component.Host) error {
 		return err
 	}
 	qs := querysvc.NewQueryService(spanReader, depReader, opts)
-	metricsQueryService, _ := disabled.NewMetricsReader()
+
+	mqs, err := s.createMetricReader(host)
+	if err != nil {
+		return err
+	}
+
 	tm := tenancy.NewManager(&s.config.Tenancy)
 
 	// TODO OTel-collector does not initialize the tracer currently
@@ -81,6 +92,7 @@ func (s *server) Start(_ context.Context, host component.Host) error {
 	telset := telemetery.Setting{
 		Logger:         s.telset.Logger,
 		TracerProvider: tracerProvider.OTEL,
+		Metrics:        queryMetricsFactory,
 		ReportStatus:   s.telset.ReportStatus,
 	}
 
@@ -89,7 +101,7 @@ func (s *server) Start(_ context.Context, host component.Host) error {
 	s.server, err = queryApp.NewServer(
 		// TODO propagate healthcheck updates up to the collector's runtime
 		qs,
-		metricsQueryService,
+		mqs,
 		s.makeQueryOptions(),
 		tm,
 		telset,
@@ -122,6 +134,24 @@ func (s *server) addArchiveStorage(opts *querysvc.QueryServiceOptions, host comp
 	return nil
 }
 
+func (s *server) createMetricReader(host component.Host) (metricsstore.Reader, error) {
+	if s.config.MetricStorage == "" {
+		s.telset.Logger.Info("Metric storage not configured")
+		return disabled.NewMetricsReader()
+	}
+
+	mf, err := jaegerstorage.GetMetricsFactory(s.config.MetricStorage, host)
+	if err != nil {
+		return nil, fmt.Errorf("cannot find metrics storage factory: %w", err)
+	}
+
+	metricsReader, err := mf.CreateMetricsReader()
+	if err != nil {
+		return nil, fmt.Errorf("cannot create metrics reader %w", err)
+	}
+	return metricsReader, err
+}
+
 func (s *server) makeQueryOptions() *queryApp.QueryOptions {
 	return &queryApp.QueryOptions{
 		QueryOptionsBase: s.config.QueryOptionsBase,