Azurinsight

Azurinsight is a lightweight, local emulator for Azure Application Insights, designed to help developers debug and visualize telemetry data directly within Visual Studio Code without needing a connection to Azure.

Inspired by Azurite, Azurinsight provides a local ingestion endpoint and a built-in Telemetry Viewer to inspect requests, traces, exceptions, and more in real-time.

Features

• Local Telemetry Ingestion: Accepts telemetry via the standard Application Insights SDKs (using a custom endpoint).
• Real-time Visualization: View live telemetry streams directly in VS Code.
• Data Persistence: Telemetry is stored locally in an SQLite database, persisting across sessions.
• Granular Management:
  • Purge All: Clear all data with a single click.
  • Delete Selected: Select and remove specific telemetry items.
  • Delete Range: Remove data within a specific date/time range.
• Search & Filter: Quickly find relevant logs by name or type.
• Detailed Inspection: View full JSON payloads of telemetry items.

Getting Started

Visual Studio Code Extension

Azurinsight is primarily designed as a VS Code extension.

1. Install: Search for "Azurinsight" in the VS Code Marketplace and install it.
2. Start: The emulator starts automatically with VS Code, or you can run the command Azurinsight: Start Server.
3. View Telemetry: Click the "Azurinsight" status bar item or run Azurinsight: Open Telemetry Viewer.

Configuration

You can configure the emulator via VS Code settings:

• azurinsight.port: The port for the ingestion server (default: 5000).
• azurinsight.dbPath: Path to the SQLite database file (default: inside the extension folder).
• azurinsight.autoStart: Whether to start the server automatically on VS Code startup (default: true).

Usage with Application Insights SDK

To send telemetry to Azurinsight, configure your Application Insights SDK to point to the local endpoint.

Python Example:

import logging
from azure.monitor.opentelemetry import configure_azure_monitor

# Configure Azure Monitor to point to the local emulator
# The IngestionEndpoint must point to your local server's base URL.
# The SDK will automatically append /v2.1/track
CONNECTION_STRING = (
    "InstrumentationKey=00000000-0000-0000-0000-000000000000;"
    "IngestionEndpoint=http://localhost:5000/;"
    "LiveEndpoint=http://localhost:5000/"
)

configure_azure_monitor(
    connection_string=CONNECTION_STRING,
)

logger = logging.getLogger(__name__)

def main():
    print("Sending telemetry to http://localhost:5000/ ...")
    
    logger.info("Inside span 'hello-emulator'")
    
    logger.warning("This is a warning message from Python!")
    
    try:
        x = 1 / 0
    except ZeroDivisionError:
        logger.error("Caught a division by zero error", exc_info=True)

print("Telemetry sent. Check the emulator UI.")

if __name__ == "__main__":
    main()

Architecture

Azurinsight consists of three main packages:

• packages/server: A Node.js Express server that handles telemetry ingestion (/v2.1/track) and serves the query API (/api/query, /api/purge, etc.). It uses SQLite for storage and WebSockets for live streaming.
• packages/extension: The VS Code extension that manages the server process and hosts the Webview.
• packages/ui: A React application (Webview) for visualizing the telemetry.

License

This project is licensed under the MIT License.