SigNoz

SigNoz is an open-source observability platform that helps developers monitor their applications and troubleshoot issues. It provides a rich set of features for monitoring and analyzing traces, metrics, and logs.

​

Setup

1. Install langtrace’s SDK and initialize the SDK in your code.
2. Sign up and start visualizing the traces on SigNoz

​

Environment Variables

Set up the following environment variables to enable OpenTelemetry tracing and send traces to SigNoz:

OTEL_EXPORTER_OTLP_ENDPOINT="ingest.us.signoz.cloud:443"
OTEL_EXPORTER_OTLP_HEADERS="signoz-access-token=<token>"
OTEL_RESOURCE_ATTRIBUTES=service.name="rag chatbot"
OTEL_EXPORTER_OTLP_PROTOCOL="grpc"

Replace <token> with your SigNoz access token.

​

RAG Application Example

Here’s an example of monitoring a Retrieval-Augmented Generation (RAG) chatbot application that uses Elastic Search for document search and Azure OpenAI for generating responses.

​

Initialize Langtrace

from langtrace_python_sdk import langtrace
langtrace.init()

​

Configure Azure OpenAI (if using)

Set up the necessary Azure OpenAI environment variables:

export LLM_TYPE=azure
export OPENAI_VERSION=<version>       # e.g., 2023-05-15
export OPENAI_BASE_URL=<base_url>     # Azure OpenAI endpoint URL
export OPENAI_API_KEY=<api_key>       # Azure OpenAI API key
export OPENAI_ENGINE=<deployment_name> # Azure deployment name

​

Running Your Application

For a Flask application, use the OpenTelemetry instrumentation:

OTEL_RESOURCE_ATTRIBUTES=service.name=Langtrace \
OTEL_EXPORTER_OTLP_ENDPOINT="<SIGNOZE_OTLP_ENDPOINT>" \
OTEL_EXPORTER_OTLP_HEADERS="<SIGNOZ_OTLP_HEADER>" \
OTEL_EXPORTER_OTLP_PROTOCOL=grpc \
opentelemetry-instrument uvicorn main:app

​

Monitoring in SigNoz

After setting up your application, you can monitor your RAG application’s performance in the SigNoz dashboard. The dashboard provides comprehensive visibility into your application’s behavior:

The main dashboard shows key metrics including:

• Request rates and latency
• Error rates and distribution
• Resource utilization patterns
• Service dependencies

Diving deeper into specific traces reveals detailed execution flows:

The trace view helps you:

• Track request propagation
• Identify bottlenecks
• Measure component-wise latency
• Debug error scenarios

For granular analysis, you can examine individual spans:

The span view provides:

• Detailed timing information
• Input/output parameters
• Error messages and stack traces
• Custom attributes and tags

Use these traces for:

• Debugging issues
• Performance monitoring
• System optimization
• Understanding user interactions

​

Troubleshooting

• No traces visible: Verify your SigNoz endpoint and token are correct
• Missing spans: Ensure Langtrace is properly initialized before any LLM operations
• Connection issues: Check if your OTLP endpoint is accessible from your application

That’s it! ✨ 🧙‍♂️ Enjoy the high cardinality traces on SigNoz.