> ## Documentation Index
> Fetch the complete documentation index at: https://docs.scaling.cloud/llms.txt
> Use this file to discover all available pages before exploring further.

# Send Sentry alerts to on-call

> Install the scaling.cloud Sentry integration and add it as an issue-alert action so Sentry errors open incidents and page your on-call engineer.

This guide turns [Sentry](/integrations/sentry) issue alerts into paged scaling.cloud [incidents](/concepts/incidents). When a Sentry alert rule fires, scaling.cloud opens an incident and routes it to whoever is on-call through your [escalation policy](/concepts/escalation-policies).

## Prerequisites

* The **Admin** role in your scaling.cloud organization and in your Sentry organization.
* At least one [on-call schedule](/concepts/oncall-schedules) and an [escalation policy](/concepts/escalation-policies) configured in scaling.cloud — an alert with nowhere to page is rejected.

<Steps>
  <Step title="Install the scaling.cloud integration">
    Install scaling.cloud from the Sentry integration directory (or from **Settings → Integrations** in scaling.cloud, **Connect Sentry**) and approve the authorization for your Sentry organization. This single install covers both sending alerts and [reading for investigations](/integrations/sentry#enrich-investigations-read).
  </Step>

  <Step title="Confirm on-call is set up">
    In scaling.cloud, make sure an [on-call schedule](/guides/set-up-oncall-schedule) and an [escalation policy](/guides/configure-escalation-policy) exist for the team that should answer Sentry alerts. This is where paged incidents will route.
  </Step>

  <Step title="Add the alert action in Sentry">
    In Sentry, go to **Alerts → Create Alert → Issues** (or edit an existing issue-alert rule). Under **Then perform these actions**, add **Send a notification via scaling.cloud**. Set your trigger conditions and **Save** the rule.
  </Step>

  <Step title="Verify a page">
    Trigger a matching error (or use Sentry's test action). Within a few seconds an incident appears in scaling.cloud and the on-call engineer is paged through your escalation policy. **Acknowledge** the page, then **resolve** the Sentry issue — the incident resolves automatically.
  </Step>
</Steps>

## How alerts map to incidents

| Sentry                                                        | scaling.cloud                                                                                     |
| ------------------------------------------------------------- | ------------------------------------------------------------------------------------------------- |
| Issue-alert rule fires                                        | Incident opened and routed through your [escalation policy](/concepts/escalation-policies).       |
| Event level `fatal` / `error` / `warning` / `info` \| `debug` | Severity `critical` / `high` / `medium` / `low`.                                                  |
| Same Sentry issue fires again                                 | Deduplicated — no duplicate incident.                                                             |
| Sentry issue resolved                                         | Matching incident resolved automatically.                                                         |
| Sentry **project slug**                                       | Mapped to the [component](/concepts/components) of the same name; component-less if none matches. |

## Troubleshooting

| Symptom                                                          | Fix                                                                                                                                        |
| ---------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------ |
| "Send a notification via scaling.cloud" isn't in the action list | Confirm the scaling.cloud integration is **installed** in this Sentry organization.                                                        |
| Alert fires in Sentry but no incident appears                    | The receiving org has no escalation policy. Add one under [escalation policies](/guides/configure-escalation-policy).                      |
| Incident opens but nobody is paged                               | Check the [on-call schedule](/concepts/oncall-schedules) has someone on call and a [contact method](/concepts/contact-methods) configured. |
| Incident has no component                                        | No component matches the Sentry project slug. Create a component named after the project, or leave it component-less.                      |
| Alerts stopped after working before                              | The integration may have been disconnected. Reconnect under **Settings → Integrations → Sentry**.                                          |

## Related

* [Sentry integration overview](/integrations/sentry) — both directions in one place.
* [Connect Sentry for investigations](/guides/connect-sentry) — the read direction.
* [Escalation policies](/concepts/escalation-policies) and [routing policies](/concepts/routing-policies) — how a paged incident reaches the right person.