Avoid Alert Fatigue with Smart Actions Cooldown

The Alert Fatigue Problem

A rule fires once — good. Then it fires again 5 minutes later because the window rolled. Then again. By the time your team checks, they've already trained themselves to ignore it.

Smart Actions solves this with a cooldown — a minimum time between firings of the same rule.

How Cooldown Works

After a rule fires:

1. LogIt sets a cooldown lock for the duration you configured
2. Even if the threshold is crossed again during that window, the rule stays silent
3. Once the cooldown expires, the rule is live again

This is implemented as an atomic Redis key with a TTL — no extra queries, no database polling.

Choosing the Right Cooldown

Scenario	Recommended cooldown
Transient error spike (recovers in minutes)	15 minutes
Payment processor outage	1–6 hours
Churn cluster (needs investigation)	6 hours
Deployment health check	30 minutes
Viral launch signup surge	1 hour

Example: Tiered Rules for the Same Event

You can create multiple rules on the same channel with different thresholds and cooldowns for different severity levels:

Rule 1 — Warning (fires first, short cooldown)

Field	Value
Threshold	3 errors
Window	5 minutes
Cooldown	15 minutes
Action	Push notification — warning

Rule 2 — Critical (fires at higher count, longer cooldown)

Field	Value
Threshold	10 errors
Window	5 minutes
Cooldown	1 hour
Action	Slack webhook — critical

Tips

• Match cooldown to your team's realistic response time. If it takes 30 minutes to triage and fix, a 15-minute cooldown just means two identical alerts.
• "Never" cooldown (set to 0) is almost always the wrong choice — you'll regret it the first time you have an incident.
• After a rule fires, check Smart Actions → Trigger History to see the exact count that triggered it and when the cooldown expires.