we-promise/sure · warning · Provider::Redbark::RateLimitError

rate_limited

rate_limited

Error message

Rate limit exceeded

What it means

Raised by Provider::Redbark's handle_response on HTTP 429 as RateLimitError. Unlike most clients, Redbark's with_retries explicitly includes RateLimitError in its retry rescue list, so a 429 is retried up to 3 times with exponential backoff+jitter before this error surfaces to the caller.

Source

Thrown at app/models/provider/redbark.rb:248

    # Redbark error envelope: { error: { message, code, details } }
    # Error messages carry the parsed provider message only, never the raw
    # response body - callers log and re-log these strings.
    def handle_response(response)
      case response.code
      when 200, 201
        JSON.parse(response.body, symbolize_names: true)
      when 400
        raise Error.new("Bad request: #{error_message_from(response)}", :bad_request)
      when 401
        raise AuthenticationError.new("Invalid API key", :unauthorized)
      when 403
        raise AuthenticationError.new("Access forbidden - your Redbark plan may not include API access", :access_forbidden)
      when 404
        raise Error.new("Resource not found", :not_found)
      when 410
        raise Error.new("Endpoint requires an accountId: #{error_message_from(response)}", :bad_request)
      when 429
        raise RateLimitError.new("Rate limit exceeded", :rate_limited)
      when 500..599
        raise ServerError.new("Redbark server error (#{response.code})", :server_error)
      else
        raise Error.new("Unexpected response #{response.code}: #{error_message_from(response)}", :unknown)
      end
    end

    def error_message_from(response)
      parsed = JSON.parse(response.body)
      parsed.dig("error", "message") || "no error message provided"
    rescue JSON::ParserError
      "unparseable error response"
    end
end

View on GitHub (pinned to e69894adb9)

Solutions

1. Reduce sync concurrency (fewer parallel jobs per API key)
2. Stagger job schedules so bulk account syncs don't align
3. Back off at the caller level when this surfaces — the library already retried 3 times with backoff
4. Check the Redbark dashboard for the key's rate tier if limits are consistently tight

def redbark_rate_limited?(error)
  error.is_a?(Provider::Redbark::RateLimitError)
end

begin
  redbark.get_transactions(connection_id: cid, account_id: aid, start_date: from, end_date: to)
rescue Provider::Redbark::RateLimitError
  RedbarkSyncJob.perform_in(jittered_delay(10.minutes), account_id: aid) # library already retried 3x
end

When it happens

Trigger: Burst-syncing many accounts in parallel, multiple Sidekiq jobs sharing one API key, or hammering /transactions across a large import — enough sustained 429s to exhaust the internal retries.

Common situations: Background sync jobs overlapping at schedule boundaries; one API key shared across staging and production; retry storms amplifying request volume.

Related errors

• rate_limited
• rate_limited
• rate_limited
• rate_limited
• rate_limited

AI-assisted analysis of we-promise/sure@e69894adb9 (2026-08-21). Data as JSON: /api/errors/d90b941faf69cb3e. Report an issue: GitHub.