Sentry integration
Turn Sentry issues and alert rules into HowlOps incidents, so an error spike pages on-call the same way an outage does.
What this brings you
An error spike in Sentry is often the first sign something is actually down, not just noisy. Connect Sentry as an inbound source and its issue and alert events land in HowlOps as incidents, on the same on-call rotation, status page, and incident history as your uptime and heartbeat monitors. You stop treating "Sentry is yelling" and "the service is down" as two different workflows.
Setup
- In HowlOps, open Integrations in the sidebar (the Integration Hub).
- Under Bring alerts into HowlOps, click the Sentry source card, then Generate endpoint to get a token-scoped webhook URL.
- In Sentry, go to Settings > Developer Settings and create an Internal Integration. Set its Webhook URL to the endpoint from step 2.
- Subscribe the integration to the Issue resource to receive lifecycle events (created, resolved, ignored, assigned, unresolved), and/or enable it as an action on an alert rule to receive Issue Alert notifications when that rule matches.
- Trigger a test issue or alert and confirm it appears as an incident in HowlOps.
Only an explicit resolved status in Sentry closes the matching HowlOps incident. ignored does not, since the underlying problem may still exist. Sentry's level maps to HowlOps severity: fatal and error become critical, warning stays warning, and info/debug become info.
See also
- Integrations reference for the endpoint format and severity mapping details.
- Incidents & severities for how an inbound alert becomes an incident.
Was this page helpful?