Skip to content

Latest commit

 

History

History
14 lines (11 loc) · 1.86 KB

File metadata and controls

14 lines (11 loc) · 1.86 KB

Threat Model Analysis for google/sanitizers

  • Description: An attacker, or even unintentional misconfiguration, could lead to sanitizers being enabled in production or performance-critical environments. This would drastically increase resource consumption (CPU, memory) due to runtime checks. An attacker might then exploit this by simply sending normal traffic, which, combined with the sanitizer overhead, overwhelms the application, making it slow, unresponsive, or causing it to crash.
  • Impact: Application becomes unavailable or severely degraded, disrupting service for legitimate users. This can lead to financial losses, reputational damage, and user dissatisfaction.
  • Affected Sanitizer Component: Entire sanitizer instrumentation mechanism (across AddressSanitizer, MemorySanitizer, ThreadSanitizer, UndefinedBehaviorSanitizer). The core runtime instrumentation logic is affected.
  • Risk Severity: High
  • Mitigation Strategies:
    • Strictly disable sanitizers in production builds: Utilize build systems and compiler flags to ensure sanitizers are only enabled for development and testing.
    • Implement robust build and deployment pipelines: Automate the build and deployment process to minimize manual errors and ensure consistent configurations across environments.
    • Conduct performance testing in staging environments: Test application performance in a staging environment that mirrors production, both with and without sanitizers, to identify any accidental sanitizer enablement.
    • Implement production monitoring and alerting: Monitor key performance indicators (KPIs) in production to detect unexpected performance degradation that could indicate accidentally enabled sanitizers.