Consider what happens when a system slows down, fails or loses a dependency. Assess whether problems can be detected, diagnosed and recovered from, and whether the product is ready for everyday operation.
All writing
Systems & Security
Reliability & Operations
How systems are maintained.
Pillar 04
Definition
Technical areas
- Monitoring, logging and error handling
- Performance, availability and capacity
- Backups, recovery and operational readiness
What an assessment could examine
These are examples to adapt to the system and its risks, rather than results from an assessment.
- Follow a representative failure through logs, monitoring and alerts to see whether it provides enough context for diagnosis without exposing sensitive data.
- Review response times, resource limits and dependency failures against the product’s availability needs.
- Examine backup coverage, restore procedures and incident responsibilities, including evidence from a recovery exercise.
Possible improvements
Findings might lead to changes such as these. Prioritise the changes supported by evidence and verify their effect.
- Make errors actionable, close monitoring gaps and reduce noisy alerts.
- Address performance bottlenecks and improve handling of unavailable dependencies.
- Clarify recovery procedures and ownership, test restoration and verify that recovery meets the product’s needs.