Prometheus
Create Runframe incidents from Prometheus alert rules via Alertmanager webhooks.
Prometheus
Create incidents from Prometheus alert rules via Alertmanager.
How it works
Prometheus Alert Rule → Alertmanager → Webhook POST → Runframe → Incident CreatedPrometheus evaluates alert rules and sends firing alerts to Alertmanager. Alertmanager forwards them to Runframe's webhook endpoint, which creates or resolves incidents automatically.
Setup from Runframe
- Navigate to Integrations Hub
- Click Set up webhook next to Prometheus
- Copy the unique webhook URL (format:
https://api.runframe.io/webhooks/{routingKey}) - Link services to route Prometheus alerts to specific Runframe services
Alertmanager configuration
Add Runframe as a webhook receiver in your Alertmanager config:
receivers:
- name: 'runframe'
webhook_configs:
- url: 'https://api.runframe.io/webhooks/YOUR_ROUTING_KEY'
send_resolved: trueAssociate the receiver with your alert routes:
route:
group_by: ['alertname', 'cluster', 'service']
group_wait: 10s
group_interval: 10s
repeat_interval: 12h
receiver: 'runframe'Keep your routing key secure
The routing key in your webhook URL is a secret token. Treat it like a password and use secrets management to avoid hardcoding it in config files.
Field mapping
| Prometheus field | Runframe field |
|---|---|
| Alert name | Incident title |
| Alert annotations | Description |
service_id label | Routes to specific service |
| Alert severity label | Severity (see mapping below) |
| Firing time | Detection time |
Severity mapping
| Prometheus severity | Default Runframe severity |
|---|---|
| Critical | Critical (SEV0) |
| Warning | Medium (SEV2) |
| Info | Minimal (SEV4) |
Auto-resolve
When send_resolved: true is set in Alertmanager, Runframe automatically resolves the incident when the alert clears. This keeps your incident list clean without manual intervention.
Testing
- Create a test alert rule in Prometheus that fires immediately
- Verify Alertmanager forwards it to Runframe
- Confirm the incident is created with correct severity and service routing
- Resolve the alert and verify the incident auto-resolves