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

# PagerDuty integration

> Connect PagerDuty so RubixKube agents can trigger incidents for critical insights, sync acknowledgement and resolution, and keep the on-call picture in step with RubixKube.

Connect PagerDuty so RubixKube agents can page the right on-call when a high-severity insight lands, and so acknowledgement and resolution stay in step between the two systems.

## What it unlocks

<CardGroup cols={2}>
  <Card title="Trigger incidents" icon="bell">
    Agents can create PagerDuty incidents for insights that cross the severity bar you set.
  </Card>

  <Card title="Acknowledge and resolve" icon="arrows-rotate">
    State flows both ways. Acking in PagerDuty reflects in RubixKube, and vice versa, so the same incident is not worked twice.
  </Card>

  <Card title="RCA attached" icon="file-magnifying-glass">
    Every incident triggered by RubixKube carries a link to the RCA report (or the insight, if an RCA has not landed yet).
  </Card>

  <Card title="Tenant-wide by default" icon="globe">
    Enable once. Every agent across every environment can use PagerDuty.
  </Card>
</CardGroup>

## Connect it

<Steps titleSize="h3">
  <Step title="Open the Integrations page">
    In the RubixKube console, go to **Integrations** and select **PagerDuty**.
  </Step>

  <Step title="Authenticate">
    Authorise with PagerDuty. An admin completes the flow and picks the services RubixKube can reach.
  </Step>

  <Step title="Configure access">
    Choose the PagerDuty services and the default severity threshold. Anything at or above the threshold can trigger a page; anything below stays inside RubixKube and your chat tools.
  </Step>
</Steps>

Once connected, PagerDuty is available tenant-wide.

## How the agents use it

* When an insight crosses the severity bar, agents can trigger the right PagerDuty service and attach the RCA (or the insight, if the RCA is still in flight).
* Ack and resolve state stays in sync, so the on-call person does not see a ghost incident in RubixKube after they have worked it in PagerDuty.

## Build custom workflows with skills

The connection exposes trigger and update capabilities. Specific workflows (per-team service routing, tighter thresholds for regulated environments, quieter thresholds for lab, auto-follow-up if the page has not been acked within N minutes) are built as [Skills](/concepts/skills). Skills are where you shape paging behaviour to your on-call model.

## Troubleshooting

<AccordionGroup>
  <Accordion title="Pages do not fire">
    Confirm the service is still connected in **Integrations → PagerDuty** and the severity threshold is set to a level the insight actually crosses.
  </Accordion>

  <Accordion title="Ack does not sync back to RubixKube">
    Reauthorise the integration to refresh webhooks. Webhook delivery from PagerDuty can also be delayed briefly under load.
  </Accordion>
</AccordionGroup>

## Related guides

<CardGroup cols={2}>
  <Card title="Slack" icon="slack" href="/integrations/slack">
    Pair PagerDuty with Slack for day-to-day comms.
  </Card>

  <Card title="Skills" icon="sparkles" href="/concepts/skills">
    How to compose paging capabilities into custom workflows.
  </Card>
</CardGroup>