Skip to Content
AlertingOverview

Alerting

exit1.dev provides multi-channel alerting to keep you informed when your checks change status.

Alert Channels

ChannelFreeNanoProAgency
Email10/hr, 10/mo50/hr, 1,000/mo500/hr, 10,000/mo1,000/hr, 50,000/mo
SMS25/hr, 50/mo50/hr, 100/mo
Webhooks152550
SlackYesYes
DiscordYesYes
Microsoft TeamsYesYes

How Alerting Works

  1. A check runs and detects a status change (e.g., up to down)
  2. Smart verification re-checks after 30 seconds to confirm
  3. The consecutive failure threshold is evaluated
  4. If the threshold is met, alerts are dispatched to all configured channels
  5. When the check recovers, recovery alerts are sent

Flap Suppression

To prevent alert fatigue from flapping services (rapidly alternating between up and down), exit1.dev uses the minConsecutiveEvents setting. This requires a minimum number of consecutive events before triggering an alert, reducing noise from unstable endpoints.

Alert Configuration

Alerts can be configured at two levels:

  • Account level - Default alert settings applied to all checks
  • Per-check level - Override settings for individual checks

Check Severity (P1–P5)

Each check has a severity from P1 (Critical) to P5 (Minimal), set with a slider under Alert behavior in its settings. Severity lets channels that support notification priority treat important sites differently from low-stakes ones — page me for a P1 outage, stay quiet for a P5 dev box — without configuring separate integrations. Today severity drives Pushover priority.

The slider sits behind a Use default priority switch that is on by default, so a new check inherits whatever priority its integrations are configured with. Turn the switch off to set an explicit severity, which then becomes a hard ceiling on every alert that check sends — including outages. That makes P3 a real choice (“tell me, but never interrupt me”), not a neutral one; leave the switch on if you want an outage to keep its usual High floor.

Whatever you pick, a recovery notification is always capped below the level that bypasses do-not-disturb. Coming back up is never urgent.

Getting Started

Last updated on