Skip to content

provider_unavailable

HTTP status: 502 Bad Gateway

  1. Retry with exponential backoff. Start at 5 seconds, double on each retry, cap at 5 minutes. Provider outages typically clear within 10-15 minutes.
  2. Check the provider's status page (Bright Data / Oxylabs / Nimble Way / Zyte / Decodo) — see Providers for current vendor links.
  3. Configure workflow fallback so jobs cascade to a backup provider when the primary fails. Add a service node with a fallback edge to a secondary provider's service node in your workflow.
  4. For sustained outages, pause the affected workflow via Workflows → Toggle rather than letting jobs pile up in the queue.

We dispatched the request to a third-party scraping provider (Bright Data, Oxylabs, Nimble Way, Zyte, Decodo) and the provider returned a 5xx, timed out, or refused the connection. This is upstream — our infrastructure is fine. This is distinct from internal_error (our bug) and service_unavailable (our infrastructure outage).

Contact support with your request_id (looks like req_01HF3JKB...). We'll surface which provider failed and what the upstream returned.