python-observability

Python observability is implemented by configuring structlog for machine-readable JSON logs and integrating prometheus_client to track the four golden signals: latency, traffic, errors, and saturation. This provides deep production insights for faster debugging.

Configuring structlog for JSON output with consistent fields is done by calling the `configure_logging` function provided by the Skill. This generates machine-readable logs that enable easier querying and alerting across your production applications.

Tracking the four golden signals in a Python application is best achieved by integrating Prometheus to collect key performance indicators. The Skill provides tools to monitor latency, traffic, errors, and saturation to optimize performance and debug systems.

Yes, opentelemetry-sdk and opentelemetry-exporter-otlp are required dependencies for distributed tracing. They work with the OpenTelemetry API to propagate correlation IDs and export trace data, enabling end-to-end request visibility across microservices.

Structured logging and distributed tracing are needed when debugging production systems experiencing high error rates or performance issues. They allow you to analyze JSON logs, trace request flows, and pinpoint the exact service causing failures.