a practice that allows systems to provide basic functionality even when core processes are not available; this is in contrast to being completely unavailable when one function fails.
— DevOps for the Modern Enterprise, Mirco Hering
graceful degradation
1 of our books defines this term.
See also blameless postmortem · concept of error budgets · incident · mean time to discovery (MTTD) · mean time to recovery (MTTR) · site reliability engineering
Filed under Reliability & Incidents