Send alerts to PagerDuty

A PagerDuty channel sends each Pingstack incident to one PagerDuty service. PagerDuty’s schedules, escalation policies and urgency rules then decide who is paged. PagerDuty channels are available on plans that include incident-management integrations; other plans show Available on Team with an Upgrade link. An Editor or Admin can create and change the channel.

Prerequisites

• A PagerDuty service that should receive Pingstack alerts.
• An Events API V2 integration on that service. Copy its 32-character Integration Key from the service’s Integrations tab.

Create the channel

1. Open Notification channels and choose Add a channel, or choose Add destination on the PagerDuty card.
2. Select PagerDuty, paste the key into Integration key and choose the notification group.
3. Under Severity, choose the PagerDuty severity for When a check is down, When a check is late and When a check is degraded.
4. Save, open the channel and choose Send test alert.
5. Assign the notification group to the checks that should page through PagerDuty.

Expected result: Pingstack shows “Test alert sent to PagerDuty, then resolved.” The service’s incident list shows a resolved Test alert from Pingstack with severity info, and the channel’s recent deliveries list two successful sends.

After saving, Pingstack shows the key only as its last four characters, for example PagerDuty key ••••1234. Leave the field blank when editing to keep the saved key, or enter a new key to replace it.

How incidents map to PagerDuty

Pingstack event	PagerDuty result
An incident opens	A new PagerDuty incident at the channel’s severity for the check’s status
The check’s status changes while the incident is open	The same PagerDuty incident is updated with the new severity and summary
The check recovers	The PagerDuty incident resolves

Every event for one Pingstack incident uses the same deduplication key, so PagerDuty keeps one incident per outage. Pingstack checks whether the incident has already recovered before each send, so a delayed retry resolves the alert instead of reopening it.

Each alert includes the check name, a summary of what failed, when it started, the check’s status and cause, and a View check in Pingstack link.

Severity

Check status	Default severity
Down	Critical
Late	Error
Degraded	Warning

You can choose Critical, Error, Warning or Info for each status. Severity is not PagerDuty’s P1–P5 priority, which PagerDuty assigns through its own rules. Whether an alert pages someone urgently depends on the service’s urgency setting, for example high urgency for every incident or urgency based on severity.

Delivery failures

PagerDuty rejecting the key or the request is not retried; check the key and send a test alert. Rate limits and PagerDuty outages are retried automatically with increasing delays. Every attempt and its error appear in the channel’s recent deliveries and in notification history.

Plan changes

If the account moves to a plan without incident-management integrations, existing PagerDuty channels keep their keys but show Paused and send nothing. Moving back to an eligible plan resumes delivery.