Alert fatigue happens when teams receive so many alerts they stop paying attention. Real problems hide in the noise.
Signs of alert fatigue:
- Alerts acknowledged without investigation
- Email filters that auto-archive alerts
- "The team knew about that" after outages
Solutions:
Review every alert. If nobody acts on it, remove it.
Aggregate related alerts. One router failure should not generate interface alerts.
Suppress during maintenance. Known events should pause alerting.
Track alert volume. If on-call receives more than a few alerts per shift, something is wrong.