Update OpenTelemetry configurations and remove unused modules

Revised otel-collector and docker-compose configurations to streamline tracing and metrics collection, replacing Datadog connectors with OpenTelemetry. Removed unused scraping modules and unnecessary dependencies to improve setup efficiency and resource utilization.

docker/docker-compose-datadog.yml

@@ -1,6 +1,4 @@
-version: '3.7'
-
-services:
+services:
   postgres:
     image: postgres:latest
     command: -c 'max_connections=2000'
@@ -13,17 +11,6 @@ services:
     ports:
       - "5432:5432"
 
-  cockroachdb:
-    image: cockroachdb/cockroach:v22.1.0
-    command: start-single-node --insecure
-    ports:
-      - "26257:26257" # CockroachDB SQL port
-      - "8080:8080"   # CockroachDB UI port
-    volumes:
-      - cockroachdb-data:/cockroach/cockroach-data
-    environment:
-      - COCKROACH_DATABASE=elsa
-
   rabbitmq:
     image: "rabbitmq:3-management"
     ports:
@@ -44,40 +31,22 @@ services:
       - postgres
       - rabbitmq
       - redis
-      - datadog-agent
+      - otel-collector
     environment:
-      DD_AGENT_HOST: datadog-agent
-      DD_ENV: development
-      DD_TRACE_DEBUG: true
-      DD_TRACE_OTEL_ENABLED: true
-      DD_SERVICE: "elsa-server-local"
-      DD_VERSION: "3.2.1-blueberry"
-
-      # Enable priority sampling
-      DD_TRACE_SAMPLING_PRIORITY: "true"
-
-      # Global rate limiting of traces (number of spans per second)
-      DD_TRACE_RATE_LIMIT: 100
-
-      # Global sample rate for all traces (applies to spans that do not match a specific sampling rule)
-      DD_TRACE_SAMPLE_RATE: 1.0  # Keep 100% of the traces globally (adjust as needed)
-
-      # Sampling rules for controlling sampling of specific services and errors
-      DD_TRACE_SAMPLING_RULES: >
-        [
-          {
-            "service": "elsa-server-local",
-            "name": "WorkflowExecution",
-            "sample_rate": 1.0,
-            "condition": {"tags": {"hasIncidents": "true"}}
-          },
-          {
-            "service": "elsa-server-local",
-            "name": "ActivityExecution",
-            "sample_rate": 1.0,
-            "condition": {"tags": {"hasIncidents": "true"}}
-          }
-        ]
+      # OpenTelemetry environment variables
+      OTEL_EXPORTER_OTLP_ENDPOINT: "http://otel-collector:4317"  # Point to OpenTelemetry Collector
+      OTEL_EXPORTER_OTLP_PROTOCOL: "grpc"  # Use gRPC for OTLP
+      OTEL_TRACES_EXPORTER: "otlp"
+      OTEL_METRICS_EXPORTER: "otlp"
+      OTEL_LOGS_EXPORTER: "otpl"
+      OTEL_RESOURCE_ATTRIBUTES: "service.name=elsa-server-local,service.version=3.2.1-blueberry,deployment.environment=development"
+      OTEL_DOTNET_AUTO_TRACES_ADDITIONAL_SOURCES: "Elsa.Workflows"
+      OTEL_DOTNET_AUTO_INSTRUMENTATION_ENABLED: "true"
+      OTEL_LOG_LEVEL: "debug"
+      OTEL_DOTNET_AUTO_RESOURCE_DETECTOR_ENABLED: "true"
+      OTEL_DOTNET_AUTO_LOGS_CONSOLE_EXPORTER_ENABLED: "true"
+      OTEL_DOTNET_AUTO_METRICS_CONSOLE_EXPORTER_ENABLED: "true"
+      OTEL_DOTNET_AUTO_TRACES_CONSOLE_EXPORTER_ENABLED: "true"
 
       ASPNETCORE_ENVIRONMENT: Development
       PYTHONNET_PYDLL: /opt/homebrew/Cellar/python@3.11/3.11.6_1/Frameworks/Python.framework/Versions/3.11/bin/python3.11
@@ -100,28 +69,41 @@ services:
     ports:
       - "14000:8080"
 
+  otel-collector:
+    image: otel/opentelemetry-collector-contrib:latest
+    volumes:
+      - ./otel-collector-config.yaml:/etc/otel-collector-config.yaml
+    command: [ "--config", "/etc/otel-collector-config.yaml", "--feature-gates", "-component.UseLocalHostAsDefaultHost" ]
+    environment:
+      DD_API_KEY: "secret api key"
+      DD_SITE: "datadoghq.eu"
+    ports:
+      - "13133:13133"
+      - "4317:4317"
+      - "4318:4318"
+
   datadog-agent:
-    image: datadog/agent:7.57.1
+    image: datadog/agent:latest
     environment:
-      DD_API_KEY: "YOUR_API_KEY"
+      DD_API_KEY: "secret api key"
       DD_SITE: "datadoghq.eu"
+      DD_HOSTNAME: "otel-collector"
       DD_LOGS_ENABLED: "true"
+      DD_OTLP_CONFIG_LOGS_ENABLED: "true"
       DD_LOGS_CONFIG_CONTAINER_COLLECT_ALL: "true"
       DD_APM_ENABLED: "true"
-      DD_REMOTE_CONFIGURATION_ENABLED: "true"
       DD_APM_NON_LOCAL_TRAFFIC: "true"
-
+      DD_OTLP_CONFIG_RECEIVER_PROTOCOLS_GRPC_ENDPOINT: 0.0.0.0:4317  # The Datadog Agent expects traces from OpenTelemetry Collector
+      DD_OTLP_CONFIG_RECEIVER_PROTOCOLS_HTTP_ENDPOINT: 0.0.0.0:4318
+
       # Service autodiscovery
       DD_AC_INCLUDE: "name:postgres,name:rabbitmq,name:redis,name:elsa-server"
       DD_AC_EXCLUDE: "name:datadog-agent"
-
-    volumes:
-      - /var/run/docker.sock:/var/run/docker.sock
-      - /proc/:/host/proc/:ro
-      - /sys/fs/cgroup/:/host/sys/fs/cgroup:ro
+
     ports:
       - "8126:8126"
+      - "14317:4317"
+      - "14318:4318"
 
 volumes:
-  postgres-data:
-  cockroachdb-data:
+  postgres-data:

docker/otel-collector-config.yaml

@@ -5,47 +5,25 @@ receivers:
         endpoint: 0.0.0.0:4317
       http:
         endpoint: 0.0.0.0:4318
-  hostmetrics:
-    collection_interval: 10s
-    scrapers:
-      paging:
-        metrics:
-          system.paging.utilization:
-            enabled: true
-      cpu:
-        metrics:
-          system.cpu.utilization:
-            enabled: true
-      disk:
-      filesystem:
-        metrics:
-          system.filesystem.utilization:
-            enabled: true
-      load:
-      memory:
-      network:
-      processes:
-  docker_stats:
-    metrics:
-      container.network.io.usage.rx_packets:
-        enabled: true
-      container.network.io.usage.tx_packets:
-        enabled: true
-      container.cpu.usage.system:
-        enabled: true
-      container.memory.rss:
-        enabled: true
-      container.blockio.io_serviced_recursive:
-        enabled: true
 
 processors:
   batch:
     send_batch_max_size: 100
-    send_batch_size: 10
+    send_batch_size: 10  # Increased batch size for efficiency
     timeout: 1s
-
-connectors:
-  datadog/connector:
+  tail_sampling:
+    decision_wait: 10s
+    num_traces: 10000  # Increased from 100 to handle more traces
+    expected_new_traces_per_sec: 100  # Increased from 10
+    decision_cache:
+      sampled_cache_size: 100000
+    policies: [
+      {
+        name: incidents-policy,
+        type: boolean_attribute,
+        boolean_attribute: { key: hasIncidents, value: true }
+      }
+    ]
 
 exporters:
   debug:
@@ -58,22 +36,14 @@ exporters:
 service:
   pipelines:
     metrics:
-      receivers: [ hostmetrics, otlp, datadog/connector ]
+      receivers: [ otlp ]
       processors: [ batch ]
       exporters: [ datadog ]
     traces:
       receivers: [ otlp ]
-      processors: [ batch ]
-      exporters: [ datadog/connector ]
-    traces/sampling:
-      # This pipeline has a Datadog connector, a batch processor and a Datadog exporter.
-      # It receivers all traces from the Datadog connector and sends them to Datadog.
-      # Add any sampling here, so that the generated trace metrics account for all traces.
-      receivers: [ datadog/connector ]
-      # Add any sampling here
-      processors: [ ]
-      exporters: [ datadog ]
+      processors: [ batch, tail_sampling ]  # Added tail_sampling to the main traces pipeline
+      exporters: [ debug, datadog ]  # Directly exporting to debug and datadog
     logs:
       receivers: [ otlp ]
       processors: [ batch ]
-      exporters: [ datadog ]
+      exporters: [ debug, datadog ]

src/bundles/Elsa.Server.Web/Elsa.Server.Web.csproj

@@ -52,8 +52,6 @@
         <PackageReference Include="DistributedLock.Redis"/>
         <PackageReference Include="FluentStorage.Azure.Blobs"/>
         <PackageReference Include="Grpc.Net.Client" />
-        <PackageReference Include="OpenTelemetry.AutoInstrumentation" />
-        <PackageReference Include="OpenTelemetry.Instrumentation.StackExchangeRedis" />
         <PackageReference Include="Proto.Persistence.Sqlite"/>
         <PackageReference Include="Proto.Persistence.SqlServer"/>
     </ItemGroup>