Observability

What it is

Observability uses logs, metrics, and traces to explain system behavior.

When to use it

Capture enough evidence to diagnose failures without exposing secrets.

Syntax

request ID | latency | error count | queue depth | reconciliation mismatch

Example

// For a local assessment, useful minimum:
// Structured errors with request IDs
// Transfer IDs in relevant logs
// Visible pending/failed state
// A query for unresolved transactions

Common mistakes

Do not log full sensitive payloads; operational metrics and immutable audit records serve different purposes.

Related topics