Error Handling and Information Leakage
Users need a stable safe error; operators need correlated internal diagnostics. Sending the latter to the former reveals implementation and sometimes secrets.
Frame the problem
Security starts with a concrete asset, attacker capability and trust crossing.
Why the system fails
Exceptions, SQL errors, stack traces or downstream bodies are returned verbatim to the caller.
The important question is not “what is Error Handling and Information Leakage?” but “which assumption let untrusted data or an over-scoped identity cross internal failure detail → untrusted response?” Trace the decision at the boundary, then constrain what can happen after the first control fails.
Design the control in layers
Start with the control closest to the interpretation or privilege boundary: Return generic user-facing errors with a correlation ID Then add a control that reduces blast radius and telemetry that proves the decision was enforced.
The resulting design is not labelled secure. Record the identified controls, the known failure paths, the remaining exposure, and the evidence you would need during an incident.
| Prevent | Detect | Recover |
|---|---|---|
| Return generic user-facing errors with a correlation ID · Keep detailed diagnostics in access-controlled logs · Redact downstream and secret material | Responses containing stack/schema/secret patterns | Contain the affected identity or component, scope impact from audit evidence, and preserve a regression test. |
Key points
- Asset: Internal schema, topology, paths, identifiers and diagnostic context.
- Boundary: Internal failure detail → untrusted response
- Primary control: Return generic user-facing errors with a correlation ID
- Detection signal: Responses containing stack/schema/secret patterns
- Always ask what limits damage when the primary control fails.
Boundary control exercise
This lesson uses the shared boundary-control exercise.
Follow the attack
Safe conceptual simulation: capability → missing control → crossed boundary → asset impact.
- 1Attacker starts with: A caller able to trigger error paths repeatedly.
- 2Exceptions, SQL errors, stack traces or downstream bodies are returned verbatim to the caller.
- 3The weak or missing boundary control is crossed: Internal failure detail → untrusted response
- 4Impact: Reconnaissance, sensitive-data leakage and easier exploitation.
- Reconnaissance, sensitive-data leakage and easier exploitation.
Defend, detect, recover
One prevention is a single point of security failure. Layer it and make failure observable.
- • Return generic user-facing errors with a correlation ID
- • Keep detailed diagnostics in access-controlled logs
- • Redact downstream and secret material
- • Responses containing stack/schema/secret patterns
- • Contain the affected identity or component.
- • Scope access from audit evidence.
- • Fix the boundary and add a regression test.
- • Misconfiguration and new access paths can bypass the intended control.
- • A privileged insider or compromised control plane may still reach the asset.