---
title: PagerDuty Integration for Uptime Monitoring
description: Open PagerDuty incidents from confirmed WatchFor incidents over Events API v2 — recovery resolves the same incident, US and EU regions supported.
canonical: https://watchfor.io/integrations/pagerduty
---

[← All alert channels](/integrations)

# PagerDuty incidents that close themselves when the site comes back

If PagerDuty already owns your rota, monitoring should feed it rather than compete with it. A confirmed incident here opens one in PagerDuty over the Events API v2, keyed so that the recovery resolves the same incident instead of leaving a stale page behind — which is the failure mode that makes people stop trusting an integration.

[Start free](/auth/signup)[Setup guide](/docs/notifications/pagerduty)

## Setting it up

Events API v2 integration key .

- 1 In PagerDuty, open the service that should own these alerts and add an Events API V2 integration.
- 2 Copy the integration key into Alerting → Channels & contacts → Add Channel → PagerDuty, and pick the US or EU region.
- 3 Attach contact groups and send a test.

The full procedure, including every optional field, is in the [setup guide](/docs/notifications/pagerduty).

## What the alert looks like in PagerDuty

A PagerDuty incident with the diagnosis as its summary, severity mapped to PagerDuty's, and the target, expected and detected values and confirming locations as custom details. The recovery resolves that same incident.

## Specific to PagerDuty

- Deduplicated by incident key: one WatchFor incident is one PagerDuty incident, not a page per check.
- US and EU endpoints, for organisations that must keep data in the EU.
- Recovery auto-resolves rather than leaving the page open.

## True of every channel

- Alerts fire on confirmed incidents, not on a single failed check, so one blip from one probe does not wake anyone.
- During a storm, several incidents inside the grouping window arrive as one summary rather than as twenty separate messages.
- Recovery closes the same thread it opened, with the outage duration attached.
- Every channel is on every plan, including the free one. There is no per-channel upsell.

## Questions people ask

Will it page me for every failed check? No. A page is opened for a confirmed incident, which requires agreement between probes rather than one failed request. That distinction is the difference between an integration you keep and one you mute.

Does the PagerDuty incident resolve on its own? Yes. The recovery sends a resolve event with the same key, so the incident closes without anyone touching it.

Do I still need your on-call scheduling? No — if PagerDuty runs your rota, use it. Our on-call schedules, rotations and escalation chains are there for teams who do not want a second product, and they are included rather than billed per responder.

## Wire it up in a few minutes

15 monitors free, every alert channel included, no card required.

[Start free](/auth/signup)[See all channels](/integrations)

---

Canonical page: https://watchfor.io/integrations/pagerduty · Site guide: https://watchfor.io/llms.txt