Enable evals with traceloop translate

.gitignore

@@ -59,5 +59,8 @@ _build/
 .mypy_cache/
 target
 
+# Deepeval cache
+.deepeval
+
 # Benchmark result files
 *-benchmark.json

instrumentation-genai/opentelemetry-instrumentation-langchain/examples/multi_agent_travel_planner/.env.traceloop

@@ -0,0 +1,23 @@
+# Copy this file to `.env` and update values before running the sample.
+
+# Required OpenAI API key
+OPENAI_API_KEY=sk-YOUR_API_KEY
+
+# Optional: override default model (defaults to gpt-4.1)
+# OPENAI_MODEL=gpt-4.1
+
+# OTLP exporter configuration (update for your collector)
+# OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4317
+# OTEL_EXPORTER_OTLP_PROTOCOL=grpc
+
+# Traces will use this service.name
+OTEL_SERVICE_NAME=opentelemetry-multi-agent-traceloop-translator-evals
+
+OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318
+OTEL_SEMCONV_STABILITY_OPT_IN=gen_ai_latest_experimental
+OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT=true
+OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT_MODE=SPAN_AND_EVENT
+OTEL_INSTRUMENTATION_GENAI_EMITTERS=span_metric_event,splunk
+OTEL_INSTRUMENTATION_GENAI_EMITTERS_EVALUATION=replace-category:SplunkEvaluationResults
+OTEL_INSTRUMENTATION_GENAI_EVALS_RESULTS_AGGREGATION=true
+OTEL_INSTRUMENTATION_GENAI_DEBUG=true

instrumentation-genai/opentelemetry-instrumentation-langchain/examples/multi_agent_travel_planner/Dockerfile.traceloop

@@ -0,0 +1,68 @@
+# Use Python 3.12 as base image
+FROM python:3.12-slim
+
+# Set working directory
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    git \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy the entire instrumentation-genai and util directories to maintain package paths
+# Build context is at repo root to access both instrumentation-genai/ and util/
+COPY instrumentation-genai/opentelemetry-instrumentation-langchain /app/opentelemetry-instrumentation-langchain
+COPY util/opentelemetry-util-genai /app/opentelemetry-util-genai
+COPY util/opentelemetry-util-genai-traceloop-translator /app/opentelemetry-util-genai-traceloop-translator
+COPY util/opentelemetry-util-genai-evals /app/opentelemetry-util-genai-evals
+COPY util/opentelemetry-util-genai-evals-deepeval /app/opentelemetry-util-genai-evals-deepeval
+COPY util/opentelemetry-util-genai-emitters-splunk /app/opentelemetry-util-genai-emitters-splunk
+
+# Set working directory to the example
+WORKDIR /app/opentelemetry-instrumentation-langchain/examples/multi_agent_travel_planner
+
+# Install Python dependencies from requirements.traceloop.txt (excluding local -e packages)
+# First, create a temporary requirements file without the local editable packages
+RUN grep -v "^-e \.\." requirements.traceloop.txt > /tmp/requirements_external.txt && \
+    pip install --no-cache-dir -r /tmp/requirements_external.txt && \
+    rm /tmp/requirements_external.txt
+
+# Install the local packages in editable mode
+# The Traceloop translator will enable zero-code instrumentation via .pth file
+RUN cd /app/opentelemetry-util-genai && \
+    pip install --no-cache-dir --no-deps -e . && \
+    cd /app/opentelemetry-util-genai-evals && \
+    pip install --no-cache-dir --no-deps -e . && \
+    cd /app/opentelemetry-util-genai-evals-deepeval && \
+    pip install --no-cache-dir --no-deps -e . && \
+    cd /app/opentelemetry-util-genai-emitters-splunk && \
+    pip install --no-cache-dir --no-deps -e . && \
+    cd /app/opentelemetry-instrumentation-langchain && \
+    pip install --no-cache-dir --no-deps -e . && \
+    cd /app/opentelemetry-util-genai-traceloop-translator && \
+    pip install --no-cache-dir --no-deps -e .
+
+# Verify packages are installed correctly
+RUN python3 -c "from opentelemetry.util.genai.handler import get_telemetry_handler; print('✓ GenAI handler available')" && \
+    python3 -c "from opentelemetry.util.genai.evals import create_evaluation_manager; print('✓ Evaluation manager available')" && \
+    python3 -c "import opentelemetry.util.genai.emitters.splunk; print('✓ Splunk emitters available')" && \
+    python3 -c "import opentelemetry.util.evaluator.deepeval; print('✓ Deepeval evaluator module available')" && \
+    python3 -c "import deepeval; print('✓ Deepeval SDK installed')" && \
+    python3 -c "from opentelemetry.util.genai.traceloop import enable_traceloop_translator; print('✓ Traceloop translator available')"
+
+# Make the script executable
+RUN chmod +x main_traceloop.py
+
+# Set default environment variables for OpenTelemetry
+ENV OTEL_PYTHON_LOG_CORRELATION=true \
+    OTEL_PYTHON_LOG_LEVEL=info \
+    OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf \
+    PYTHONUNBUFFERED=1
+
+# Health check (optional)
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+    CMD python3 -c "import sys; sys.exit(0)"
+
+# Run the Traceloop version
+CMD ["python3", "main_traceloop.py"]
+

instrumentation-genai/opentelemetry-instrumentation-langchain/examples/multi_agent_travel_planner/cronjob-traceloop.yaml

@@ -0,0 +1,149 @@
+apiVersion: batch/v1
+kind: CronJob
+metadata:
+  name: travel-planner-tl-translate-evals-splunk
+  namespace: o11y-4-ai-admehra
+  labels:
+    app: travel-planner-tl-translate-evals-splunk
+    component: telemetry
+  annotations:
+    description: "Multi-agent travel planner with Traceloop translator and GenAI evaluation infrastructure using Deepeval"
+    git-commit: "2e5901a"
+spec:
+  # Run every 2 minutes for testing
+  schedule: "*/2 * * * *"
+  suspend: false
+
+  # Keep last 3 successful and 1 failed job for debugging
+  successfulJobsHistoryLimit: 3
+  failedJobsHistoryLimit: 1
+
+  jobTemplate:
+    metadata:
+      labels:
+        app: travel-planner-tl-translate-evals-splunk
+        component: telemetry
+    spec:
+      template:
+        metadata:
+          labels:
+            app: travel-planner-tl-translate-evals-splunk
+            component: telemetry
+        spec:
+          restartPolicy: OnFailure
+
+          containers:
+            - name: travel-planner-traceloop
+              # Multi-platform image (amd64, arm64) with git commit hash tag
+              image: admehra621/travel-planner-tl-translate-evals-splunk:2e5901a
+              imagePullPolicy: Always
+
+              env:
+                # === GenAI Semantic Conventions (REQUIRED) ===
+                - name: OTEL_SEMCONV_STABILITY_OPT_IN
+                  value: "gen_ai_latest_experimental"
+
+                # === OpenTelemetry Resource Attributes ===
+                - name: OTEL_RESOURCE_ATTRIBUTES
+                  value: "deployment.environment=o11y-inframon-ai,git.commit.id=2e5901a"
+
+                # === Service name for telemetry ===
+                - name: OTEL_SERVICE_NAME
+                  value: "travel-planner-tl-translate-evals-splunk"
+
+                # === OpenAI Configuration ===
+                - name: OPENAI_API_KEY
+                  valueFrom:
+                    secretKeyRef:
+                      name: openai-credentials
+                      key: api-key
+
+                - name: OPENAI_MODEL
+                  value: "gpt-4o-mini"
+
+                # === GenAI Content Capture ===
+                - name: OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT
+                  value: "true"
+
+                - name: OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT_MODE
+                  value: "SPAN_AND_EVENT"
+
+                # === GenAI Emitters Configuration ===
+                - name: OTEL_INSTRUMENTATION_GENAI_EMITTERS
+                  value: "span_metric_event,splunk"
+
+                - name: OTEL_INSTRUMENTATION_GENAI_EMITTERS_EVALUATION
+                  value: "replace-category:SplunkEvaluationResults"
+
+                # === Evaluation Settings ===
+                # Note: OTEL_INSTRUMENTATION_GENAI_EVALS_EVALUATORS not set - will use all default evaluations
+                # (bias, toxicity, answer_relevancy, faithfulness, contextual_relevancy)
+
+                - name: OTEL_INSTRUMENTATION_GENAI_EVALS_RESULTS_AGGREGATION
+                  value: "true"
+
+                # === GenAI Debug Flags ===
+                - name: OTEL_GENAI_EVAL_DEBUG_SKIPS
+                  value: "true"
+
+                - name: OTEL_GENAI_EVAL_DEBUG_EACH
+                  value: "true"
+
+                - name: OTEL_INSTRUMENTATION_GENAI_DEBUG
+                  value: "true"
+
+                # === OpenTelemetry Logs Exporter ===
+                - name: OTEL_LOGS_EXPORTER
+                  value: "otlp"
+
+                # === Get the host IP for Splunk OTEL agent ===
+                - name: SPLUNK_OTEL_AGENT
+                  valueFrom:
+                    fieldRef:
+                      fieldPath: status.hostIP
+
+                # === OpenTelemetry OTLP endpoint using Splunk agent ===
+                - name: OTEL_EXPORTER_OTLP_ENDPOINT
+                  value: "http://$(SPLUNK_OTEL_AGENT):4317"
+
+                # === OTLP Protocol (grpc) ===
+                - name: OTEL_EXPORTER_OTLP_PROTOCOL
+                  value: "grpc"
+
+                # === Exclude health check URLs ===
+                - name: OTEL_PYTHON_EXCLUDED_URLS
+                  value: "^(https?://)?[^/]+(/)?$"
+
+                # === Enable Python logging auto instrumentation ===
+                - name: OTEL_PYTHON_LOGGING_AUTO_INSTRUMENTATION_ENABLED
+                  value: "true"
+
+                # === Enable log correlation ===
+                - name: OTEL_PYTHON_LOG_CORRELATION
+                  value: "true"
+
+                # === Enable LangChain content capture ===
+                - name: OTEL_INSTRUMENTATION_LANGCHAIN_CAPTURE_MESSAGE_CONTENT
+                  value: "true"
+
+                # === Enable Splunk profiler ===
+                - name: SPLUNK_PROFILER_ENABLED
+                  value: "true"
+
+                # === Unbuffered Python output ===
+                - name: PYTHONUNBUFFERED
+                  value: "1"
+
+                # === GenAI evaluation sampling rate ===
+                - name: OTEL_GENAI_EVALUATION_SAMPLING_RATE
+                  value: "1"
+
+              # === Resource limits ===
+              resources:
+                requests:
+                  memory: "512Mi"
+                  cpu: "500m"
+                limits:
+                  memory: "1Gi"
+                  cpu: "1000m"
+