Skip to content

Observability

Monitor and debug your application in production.

Levels

The observability level controls logging and metrics. It is independent of error tracking, which is enabled separately by the "Include Sentry error tracking?" prompt.

Minimal

  • Structured JSON logging
  • Health and readiness endpoints (/health/, /ready/)

Standard

  • Everything in Minimal
  • Production request logging with enriched fields (per-request and Celery task logs)

Full

  • Everything in Standard
  • Prometheus metrics (django-prometheus)
  • OpenTelemetry distributed tracing

With the Kubernetes deployment target, the Full level also generates a Prometheus ServiceMonitor and a Grafana dashboard under deploy/k8s/monitoring/.

Error tracking (Sentry)

Sentry is controlled by its own prompt, not the observability level, so it works at any level. When enabled it is wired in config/settings/prod.py and reads SENTRY_DSN from the environment; at the Full level the Redis integration is added as well.

See the Features Overview for the full option list.