Failure modes rarely arrive alone. The most destructive incidents involve three or more failure modes interacting.
Pattern 1: The silent cascade FM1 (SPOF fails) → FM2 (cascade to dependants) → FM11 (nobody can see why) → hours of downtime.
Pattern 2: The thundering herd amplifier FM7 (thundering herd on cache miss) → FM3 (database connections exhausted) → FM2 (services time out) → FM11 (alerts fire too late to guide response).
Pattern 3: The slow corruption FM8 (schema change silently misinterpreted) → FM9 (incorrect data propagates) → FM4 (systems disagree on state) → discovered weeks later during reconciliation.
The compounding table below shows the most common dangerous pairs:
| Pair | Risk pattern |
|---|---|
| FM1 + FM2 | SPOF triggers cascade — most common full outage pattern |
| FM7 + FM2 | Thundering herd prevents recovery — outage self-perpetuates |
| FM8 + FM9 | Contract violation causes silent corruption — discovered late |
| FM4 + FM12 | Consistency failure caused by split-brain — hardest to recover |
| Any + FM11 | Any failure combined with blindness — extends recovery time dramatically |