instructure/canvas-lms · error
endpoint undefined
Error message
endpoint undefined
What it means
When publishing final grades, the grade_export plugin settings must include a publish_endpoint URL; if settings[:publish_endpoint] is blank, this error stops publishing because there is no SIS endpoint to post to.
Solutions
- Set publish_endpoint in the grade_export plugin settings to the SIS grades URL
- Check settings[:publish_endpoint].present? before invoking publishing
- Verify plugin settings persist correctly in the target environment (DB vs YAML)
Example fix
// before plugin.settings # publish_endpoint: nil // after plugin_settings['publish_endpoint'] = 'https://sis.example.com/grades' plugin.save_settings
Defensive patterns
Strategy: validation
Validate before calling
plugin = Canvas::Plugin.find!('grade_export')
raise 'publish_endpoint not configured' if plugin.enabled? && plugin.settings[:publish_endpoint].blank? Type guard
null
Try / catch
begin course.publish_final_grades(user) rescue RuntimeError => e raise unless e.message == 'endpoint undefined' # guide admin to plugin settings to set the SIS endpoint end
Prevention
- Configure publish_endpoint during plugin setup and verify it
- Diff plugin settings when migrating environments
- Add a settings health-check before scheduling publishing
When it happens
Trigger: grade_export plugin enabled but its publish_endpoint setting is empty/blank when publish_final_grades runs (via its worker).
Common situations: Plugin enabled during setup but endpoint URL never configured; endpoint cleared during maintenance; copying plugin settings between environments without the endpoint.
Understand the failure class
Background: "is required", "must be set", "missing required field": configuration validation errors across open-source libraries — this error's family across 36 libraries.
Related errors
- unknown format type: #
- final grade publishing disabled
- grade publishing requires a grading standard
- A new_id, '# ', referenced an existing # and the # with #…
- A new_integration_id, '#
AI-assisted analysis of instructure/canvas-lms@1c9f0bb801 (2026-09-15).
Data as JSON: /api/errors/446225e8886526b1.
Report an issue: GitHub.
Appendix: source
Thrown at app/models/course.rb:2639
def send_final_grades_to_endpoint(publishing_user, user_ids_to_publish = nil)
# actual grade publishing logic is here, but you probably want
# 'publish_final_grades'
recompute_student_scores_without_send_later(user_ids_to_publish)
enrollments = student_enrollments.not_fake.eager_load(:user).preload(:course_section).order_by_sortable_name
enrollments = enrollments.where(user_id: user_ids_to_publish) if user_ids_to_publish
errors = []
posts_to_make = []
posted_enrollment_ids = []
all_enrollment_ids = enrollments.map(&:id)
begin
raise "final grade publishing disabled" unless Canvas::Plugin.find!("grade_export").enabled?
settings = Canvas::Plugin.find!("grade_export").settings
raise "endpoint undefined" if settings[:publish_endpoint].blank?
format_settings = Course.valid_grade_export_types[settings[:format_type]]
raise "unknown format type: #{settings[:format_type]}" unless format_settings
raise "grade publishing requires a grading standard" if !grading_standard_enabled? && format_settings[:requires_grading_standard]
publishing_pseudonym = SisPseudonym.for(publishing_user, self)
raise "publishing disallowed for this publishing user" if publishing_pseudonym.nil? && format_settings[:requires_publishing_pseudonym]
callback = Course.valid_grade_export_types[settings[:format_type]][:callback]
posts_to_make = callback.call(self, enrollments, publishing_user, publishing_pseudonym)
rescue => e
Enrollment.where(id: all_enrollment_ids).update_all(grade_publishing_status: "error", grade_publishing_message: e.to_s)
raise e
end
default_timeout = Setting.get("send_final_grades_to_endpoint_timelimit", 15.seconds.to_s).to_f
timeout_options = { raise_on_timeout: true, fallback_timeout_length: default_timeout }View on GitHub (pinned to 1c9f0bb801)