AWS Application Signals MCP Server

# Debugging with Logs The AppSignals MCP server includes logging to help debug issues. Logs are written to stderr to avoid interfering with the MCP protocol. ## Quick Start Enable debug logging by setting an environment variable: ```bash export MCP_APPSIGNALS_LOG_LEVEL=DEBUG ``` ## Log Levels - **DEBUG**: Detailed information for troubleshooting - **INFO**: General operation status (default) - **ERROR**: Failures and exceptions with stack traces ## What's Logged - Tool execution start/completion with timing - AWS API errors with details - Service/metric counts and query parameters - Performance metrics (execution time) ## Example Output **Successful operation:** ``` 2024-06-11 20:30:15 - INFO - Starting list_application_signals_services request 2024-06-11 20:30:15 - INFO - Retrieved 5 services from Application Signals 2024-06-11 20:30:15 - INFO - list_application_signals_services completed in 0.355s ``` **Error with details:** ``` 2024-06-11 20:31:22 - ERROR - AWS ClientError in get_service_details for 'unknown-service': ResourceNotFoundException - The specified service does not exist ``` ## Troubleshooting Tips 1. **Enable debug mode** when something isn't working: ```bash export MCP_APPSIGNALS_LOG_LEVEL=DEBUG ``` 2. **Check for errors** in the output: - Look for "ERROR" messages - AWS error codes explain what went wrong - Stack traces show where failures occurred 3. **Monitor performance** by looking for "completed in X.XXXs" messages