name: Pantheon Rate Limits
description: >-
  Pantheon enforces rate limits at several layers: the Terminus REST API
  (terminus.pantheon.io), the Global CDN edge, and per-site traffic thresholds
  tied to Site Plan capacity. Specific numeric limits on the Terminus API are
  not publicly documented; limits below reflect known platform constraints and
  best-practice guidance from Pantheon documentation.
url: https://docs.pantheon.io/terminus

api_rate_limits:
  - scope: Terminus REST API (terminus.pantheon.io)
    description: >-
      The Terminus CLI communicates with terminus.pantheon.io over HTTPS using
      machine tokens for authentication. Pantheon does not publish explicit
      per-minute or per-hour request quotas for the management API. Scripted
      automations using Terminus in CI/CD pipelines should implement retry
      logic with back-off to handle transient 429 or 503 responses.
    limits:
      - type: Concurrent requests
        value: not publicly specified
        notes: >-
          Burst scripting against the API (e.g., looping over hundreds of sites)
          may result in temporary throttling. Pantheon recommends adding delays
          between bulk operations.
      - type: Paged request entry limit
        value: 100 entries per page
        header: null
        notes: >-
          The Terminus client enforces PAGED_REQUEST_ENTRY_LIMIT = 100 when
          paginating collection endpoints.

cdn_and_traffic_limits:
  - scope: Global CDN — per-site traffic
    description: >-
      Traffic capacity is governed by the Site Plan purchased. Exceeding plan
      limits triggers overage notifications and may require a plan upgrade.
      Pantheon's Global CDN handles caching at the edge; cached responses do
      not count against container capacity.
    limits:
      - plan: Basic
        monthly_visits: '25,000–35,000'
      - plan: Performance Small
        monthly_visits: '25,000–35,000'
      - plan: Performance Medium
        monthly_visits: '50,000–70,000'
      - plan: Performance Large
        monthly_visits: '150,000–210,000'
      - plan: Performance XL
        monthly_visits: '300,000–420,000'
      - plan: Elite
        monthly_visits: unlimited

authentication:
  - type: Machine Token
    description: >-
      Long-lived tokens generated in Pantheon Personal Settings. Equivalent to
      username/password credentials. Used in CI/CD and automation; revoke when
      no longer needed.
    expiry: never
    url: https://docs.pantheon.io/machine-tokens

headers_and_errors:
  - code: 429
    meaning: Too Many Requests — slow down API scripting; add back-off delay
  - code: 503
    meaning: Service Unavailable — transient; retry with exponential back-off

notes:
  - Pantheon does not publish a public rate-limit policy document for the management API.
  - For bulk site operations via Terminus, Pantheon recommends pacing requests.
  - Status of all API services is tracked at https://status.pantheon.io