we-promise/sure · critical · ImportSession::EnqueueError

import_enqueue_failed

import_enqueue_failed

Error message

Import session could not be queued.

What it means

ImportSession#publish_later (app/models/import_session.rb:248-259) wraps ImportSessionJob.perform_later in a rescue: if the ActiveJob adapter cannot enqueue (Redis/SolidQueue database down, adapter misconfiguration, serialization failure), it rolls the session back to its previous status, stores error_details {code: 'import_enqueue_failed'}, logs 'ImportSession enqueue failed import_session_id=... exception=...', and raises EnqueueError. The controller maps it to HTTP 503 {error: 'import_enqueue_failed', message: 'Import session could not be queued.'}.

Source

Thrown at app/models/import_session.rb:258

      previous_status = status
      update!(status: :importing, error_details: {})
      should_enqueue = true
    end

    return unless should_enqueue

    begin
      ImportSessionJob.perform_later(self)
    rescue => error
      with_lock do
        reload
        if importing?
          update!(status: previous_status, error_details: enqueue_error_details)
        end
      end
      Rails.logger.error("ImportSession enqueue failed import_session_id=#{id} exception=#{error.class}")
      raise EnqueueError, "Import session could not be queued."
    end
  end

  def publish
    return unless prepare_for_publish!

    Rails.logger.info("ImportSession publish started import_session_id=#{id}")

    imports.ordered_by_sequence.each do |import|
      process_chunk!(import)
    end

    update!(status: :complete, summary: aggregate_chunk_summaries, error_details: {})
    enqueue_family_sync
    Rails.logger.info("ImportSession publish completed import_session_id=#{id}")
  rescue => error
    update!(
      status: :failed,

View on GitHub (pinned to e69894adb9)

Solutions

1. Restore the queue backend (Redis reachable, SolidQueue migration ran, worker running), then call publish again — the status rollback makes the retry safe
2. Grep logs for 'ImportSession enqueue failed import_session_id=… exception=…' to identify the underlying adapter error
3. Verify config.active_job.queue_adapter and queue names match the running workers
4. Confirm the session's error_details show import_enqueue_failed and its status returned to pending/failed before re-publishing

Example fix

# Ruby caller, before
session.publish_later # raises ImportSession::EnqueueError during a Redis outage

# after
begin
  session.publish_later
rescue ImportSession::EnqueueError
  retry_after_backend_restored { session.reload.publish_later } # session status was rolled back

# No request-side validation can guarantee a backend enqueue — check health first when possible
api.get('/up') # or your Redis/SolidQueue healthcheck, before triggering publish

retries = 0
begin
  session.publish_later
rescue ImportSession::EnqueueError
  retries += 1
  raise if retries > 3
  sleep(2**retries) # queue backend transient — session status was rolled back, retry is safe
  session.reload
  retry
end

When it happens

Trigger: POST /api/v1/import_sessions/:id/publish while the queue backend is unavailable or rejecting jobs: Redis connection refused, SolidQueue tables not migrated, Sidekiq/worker processes not running, or a queue-adapter mismatch.

Common situations: Redis outage or maintenance window; docker-compose restarted without the queue service; production using a different queue adapter than local; job arguments failing serialization after a model change.

Related errors

• import_session_conflict
• record_not_found
• validation_failed
• record_not_found
• record_not_found

AI-assisted analysis of we-promise/sure@e69894adb9 (2026-08-21). Data as JSON: /api/errors/ea1e7c831117f48c. Report an issue: GitHub.