Skip to content

Understanding Usage Logs

Usage logs capture detailed metrics about individual service instance operations. Each log entry records the full context of a request processed by a specific service instance, enabling deep visibility into how your deployments behave under operational conditions.

Each usage log entry records:

  • Request details: The HTTP method, path, headers, query parameters, and how the request was authenticated
  • Response details: The status code, response headers, and response body content
  • Timing: When the operation occurred
  • Service context: Which service and instance processed the request
  • Metadata: Request ID, version, region, and operational scope

This comprehensive capture enables troubleshooting at the request level, understanding operational patterns, and auditing specific interactions with your services.

Trail logs are ingested in near real time but may take up to 5 minutes to appear after an operation completes. Treat brief gaps as ingestion delay rather than missing data.

Usage logs are tied to specific service instances. This means:

  • Per-instance visibility: Each instance records its own operational activity independently
  • Instance-specific behavior: You understand how individual deployments behave
  • Instance troubleshooting: You can investigate instance-specific issues
  • Per-instance monitoring: Monitor individual deployments separately

This instance-scoped design allows you to correlate usage patterns with specific configurations, deployments, or operational conditions within that instance.

Usage logs are designed for queries like:

  • “What requests did my authorizer instance process in the past 24 hours?”
  • “Which requests to my verifier instance returned error responses?”
  • “How did the authentication method distribution change for this instance?”
  • “What was the response time pattern for requests to this instance?”