Common bad-alert patterns to avoid.
mediumAnswer
- (1) Alert on threshold that fires daily (noise).
- (2) Alert without runbook (no action).
- (3) Alert on symptoms, not causes ('page fires slower' rather than 'db p99 up').
- (4) Alerting on averages (hides tails).
- (5) Static thresholds ignoring seasonality.
- (6) No de-duplication (spam).
- (7) Waking humans for auto-recoverable.
- (8) Missing correlation (10 alerts for one root cause).
- (9) Alert on things with no fix.
- Rule: every alert = 'someone must do something now'.
Check yourself — multiple choice
- More alerts
- Daily fires (noise) / no runbook / symptom vs cause / averages / static thresholds / no dedup / no auto-recover / no correlation / no fix; every alert = must act now
- Random
- Not real
Bad alerts: noise / no runbook / symptom / average / static / no dedup / duplicates.
#monitoring#observability
Practise MLOps & Data Quality
215 interview questions in this topic.