PagerDuty integration

Overview

PagerDuty is the dominant incident-management platform – on-call schedules, alert routing, escalation policies, incident lifecycle. The Automize PagerDuty connector covers incidents (CRUD + actions), services, schedules, oncalls, escalation policies, teams, users, and the Events API v2 for alert ingestion.

• Regions: PagerDuty is global (api.pagerduty.com) with EU + US data residency options on Enterprise.
• Plans: Free, Professional, Business, Digital Operations – all expose the REST API. Some features (analytics, AIOps) need higher tiers.

Setting up the connection

Auth type: Bearer Token.

1. Sign in to PagerDuty → Integrations → API Access Keys → Create New API Key.
2. Pick scope: General Access (read+write) or read-only.
3. Copy the key.
4. In Automize, open Settings → Connectors → PagerDuty → Add connection. Paste key. Connector sends as 'Token token=<key>' header.
5. For Events API (sending alerts), use a Service's Integration Key (routing_key) – separate from API key.

Find your credentials at https://yourcompany.pagerduty.com/api_keys.

Rate limits

PagerDuty enforces 960 req/min per account on REST API. Events API is higher. Connector retries 429 with backoff.

Data model

PagerDuty primitives:

    Service ─── (an alerting target) ── Escalation Policy ─── Schedules ─── Users
    Incident ─── (alert that fired) ─── Alerts (individual events bundled)
                                    ├─ Status (triggered, acknowledged, resolved)
                                    ├─ Assignments
                                    └─ Notes
    Teams (org grouping)

Available operations

16 operations available. Click any row to jump to its detail.

Operations

Acknowledge incident

Acknowledge a PagerDuty incident.

Tips

• Shortcut for Update with status=acknowledged. Stops escalation.

Create Incident

Create a new incident

Tips

• incident: {type:'incident', title, service:{id, type:'service_reference'}, urgency:'high'|'low', body}.
• Manually trigger an incident (vs Events API which auto-triggers from a monitoring alert).

Create maintenance window

Create a PagerDuty maintenance window.

Tips

• maintenance_window: {start_time, end_time, services:[{id, type:'service_reference'}], description}.
• Mutes incidents during planned maintenance – alerts still fire but don't notify.

Get Incident

Get incident details

Tips

• incident_id (P-style ID or numeric). Returns incident + assignments + escalation chain.

List alerts

List alerts for a PagerDuty incident.

Tips

• Alerts grouped under an incident. An incident bundles multiple alerts from the same alerting source.

List Escalation Policies

List escalation policies

No input parameters.

Tips

• Policies route incidents through layers (e.g. primary oncall → backup → manager).

List Incidents

List incidents

No input parameters.

Tips

• Filter by date_range (1d, 1w), statuses[] (triggered, acknowledged, resolved), service_ids[].
• Pages 25 per call (max 100).

List On-Calls

List current on-call entries

No input parameters.

Tips

• Who's on call right now. Filter by schedule_ids[], escalation_policy_ids[], time_zone.
• Use for 'page the current oncall' automation.

List Schedules

List on-call schedules

No input parameters.

Tips

• On-call schedules. Use Get Schedule (separate, not exposed here) for shift details.

List Services

List all services

No input parameters.

Tips

• Returns services (alerting endpoints). Filter by team_ids[], include=['integrations', 'escalation_policies'].

List Teams

List all teams

No input parameters.

Tips

• Org grouping. Services + escalation policies can be team-scoped.

List Users

List all users

No input parameters.

Tips

• All users in the org. include=['teams', 'contact_methods'] for richer payload.

Merge incidents

Merge PagerDuty incidents.

Tips

• Merge source_incidents into the target (incident_id). Useful for de-duplicating alert storms.

Resolve incident

Resolve a PagerDuty incident.

Tips

• Resolves the incident. Final state – re-opens require a new incident.

Send Event

Send a trigger/resolve event via Events API v2

Tips

• Events API v2 – separate from REST. routing_key is a Service's Integration Key (NOT API key).
• event_action: 'trigger', 'acknowledge', 'resolve'.
• payload: {summary, source, severity, custom_details}.
• Most common path for system → PagerDuty alerts.

Update Incident

Update/resolve an incident

Tips

• PUT-style. Set status: acknowledged/resolved. assignments[] to reassign.

FAQ

Events API vs Incidents API?

Events API (Send Event) is what monitoring systems push to – fires alerts on a Service's routing_key. Incidents API is read/manage incidents once created. Use Events API to TRIGGER; Incidents API to MANAGE.

Token format – REST vs Events?

REST uses an API key (your user-scoped, Authorization: Token token=&lt;key&gt;). Events uses a Service's Integration Key (routing_key) – different per Service. Don't mix.

Why do my alerts not notify?

Check (a) Service is connected to an Escalation Policy, (b) Policy has a Schedule with people in it, (c) people have valid contact methods, (d) no Maintenance Window covering now.

Related connectors

• Datadog
• Sentry
• Grafana
• Aws-cloudwatch