On Sentry event

View as Markdown

Sentry event triggers run a pipeline when a supported event occurs for a Sentry issue. The pipeline runs when the event and issue match the configured trigger and filters. For example, you can roll back a release when a fatal issue appears in production or notify the team when an issue is resolved.

The trigger supports the following Sentry issue events:

  • created - a new issue was created
  • resolved - an issue was resolved
Info
The trigger requires a Sentry integration. Events are delivered through the integration selected in the trigger.

Configure trigger

  1. Open the pipeline and go to the Workflow tab.
  2. Click + in the Triggers section and select the Sentry trigger.
  3. Select the Integration and the events that should trigger the pipeline.
  4. Add filters to restrict which issues can trigger the pipeline, if needed.
  5. Click Apply changes.

Filters

  • Projects - Sentry projects the issue belongs to.
  • Levels - severity levels of the issue, e.g. error, fatal.
  • Statuses and Substatuses - status and substatus of the issue, for example unresolved with the new or escalating substatus.
  • Count - the minimum number of times the issue has occurred.
  • User count - the minimum number of affected users.
  • Snooze - prevents the same issue from triggering the pipeline more than once within the configured time window. The value is set in minutes, up to 20160 (14 days).
Warning
Count and User count are compared with the values included in the event. Sentry sends the created event once per issue, usually with a count of 1, and does not send another created event when the count later reaches the configured threshold. For resolved events, these values contain the totals at the time of resolution.

YAML

The event type is SENTRY. The integration field requires the integration identifier, not its display name:

yaml
- pipeline: "Roll back on fatal error"
  events:
    - type: "SENTRY"
      integration: "my-sentry"
      projects:
        - "backend"
      levels:
        - "fatal"
      events:
        - "created"
      snooze_minutes: 60
  actions:
    - action: "Roll back release"
      type: "SSH_COMMAND"
      targets:
        - "prod-server"
      commands: |-
        ./rollback.sh
    - action: "Notify team"
      type: "SLACK"
      integration: "slack"
      channel: "C0123456789"
      content: "Rolled back after $BUDDY_RUN_SENTRY_TITLE in $BUDDY_RUN_SENTRY_PROJECT"

In this example the pipeline starts when a new fatal issue appears in the backend project. The same issue will not start the pipeline again for 60 minutes.

Variables

A run started by a Sentry event receives these variables:

Name Description
BUDDY_RUN_SENTRY_EVENT Sentry event that triggered the run, for example created
BUDDY_RUN_SENTRY_ISSUE_ID Identifier of the issue in Sentry
BUDDY_RUN_SENTRY_TITLE Title of the issue
BUDDY_RUN_SENTRY_URL API URL of the issue (https://sentry.io/api/0/...). The link to the issue in Sentry is data.issue.web_url in the payload
BUDDY_RUN_SENTRY_PROJECT Sentry project the issue belongs to
BUDDY_RUN_SENTRY_LEVEL Severity level of the issue
BUDDY_RUN_SENTRY_STATUS Status of the issue
BUDDY_RUN_SENTRY_SUBSTATUS Substatus of the issue. Not exported for resolved
BUDDY_RUN_SENTRY_COUNT Number of times the issue occurred
BUDDY_RUN_SENTRY_USER_COUNT Number of users affected by the issue
BUDDY_RUN_SENTRY_ASSIGNED_TO Person the issue is assigned to. Not exported when the issue is unassigned
BUDDY_RUN_SENTRY_PAYLOAD Body of the Sentry webhook that started the run

See also

Last modified on Nov 13, 2026