​

Sentry Exporter

​

Supported Telemetry

​

Overview

​

How routing works

• Telemetry is grouped per project and sent to that project’s OTLP ingestion endpoint.
• The project slug comes from a resource attribute (default: service.name). If the attribute is missing or empty, the spans/logs in that resource are dropped with a warning.
• You can remap attribute values to project slugs with a lookup table.
• When auto_create_projects is enabled and a project is missing, the exporter enqueues it for asynchronous creation and drops data for that project until creation succeeds. If disabled, export fails permanently when a project is unknown.
• Project endpoints and keys are cached; a 403 for an unknown project ID triggers cache eviction and retry.
• Rate limits from Sentry response headers (or HTTP 429) are honored per DSN and category; limited data is dropped with a throttle error and retried by the queue.

​

Configuration

• url: Base URL for the Sentry organization (required).
• org_slug: Target organization slug (required).
• auth_token: Authentication token with access to Sentry APIs (required).
  • basic functionality requires both project:read and org:read or higher
  • auto_create_projects requires project:write or higher. If your organization has disabled member project creation, the org:write or team:admin scope is required.
• auto_create_projects (default: false): Create missing projects using the first team found in the org.
• routing: Controls how telemetry is mapped to projects.
  • project_from_attribute: Resource attribute to use (default: service.name).
  • attribute_to_project_mapping: Optional map from attribute value to project slug.
• http: Standard confighttp client settings (timeout default 30s, TLS, headers, etc.).
• timeout: Exporter timeout (defaults to 30s).
• sending_queue: Optional queue settings (enabled by default with collector defaults).

exporters:
  sentry:
    url: https://sentry.io
    org_slug: my-org
    auth_token: ${SENTRY_AUTH_TOKEN}
    auto_create_projects: true
    routing:
      project_from_attribute: service.name
      attribute_to_project_mapping:
        api-service: backend-api
        web-service: frontend-web
    http:
      timeout: 30s
      tls:
        insecure_skip_verify: false
    timeout: 30s
    sending_queue:
      enabled: true

​

Known Limitations

• Manually deleting a project in Sentry while the exporter cache still holds its endpoints can drop data until a 403 is observed and the cache entry is evicted.
• When auto_create_projects is enabled, the first data for a missing project can be dropped while the async creation worker provisions the project and fetches its endpoints; pre-create projects to avoid startup loss.
• Retries are not possible, since a single collector batch can route to multiple projects. If one project export fails but others succeed, a retry would resend already accepted data, leading to potential duplication for the successful projects.

​

Associating with Sentry Errors

​

Configuration

​

Example Configuration

# Minimal dynamic configuration
sentry:
  url: https://sentry.io
  org_slug: my-org
  auth_token: test-auth-token-12345

# Dynamic configuration with routing options
sentry/with_routing:
  url: https://sentry.io
  org_slug: my-org
  auth_token: test-auth-token-12345
  auto_create_projects: true
  routing:
    project_from_attribute: service.name
    attribute_to_project_mapping:
      api-service: backend-api
      web-service: frontend-web

# Dynamic configuration with all options
sentry/full:
  url: https://sentry.example.com
  org_slug: example-org
  auth_token: full-test-token
  timeout: 45s
  auto_create_projects: true
  http:
    timeout: 20s
    tls:
      insecure: true
  routing:
    project_from_attribute: deployment.environment.name
    attribute_to_project_mapping:
      production: prod-project
      staging: stage-project