Inside No Timeout On Outgoing Fetch() To RERUM

TinyPEN’s outgoing fetch calls to RERUM currently run without a timeout - leaving workers stuck waiting when RERUM stalls. This silent block isn’t just inconvenient; it’s a real risk. In real tests, blocked TinyPEN workers waited indefinitely when RERUM responded slowly or hung. The danger? A stuck RERUM connection means no response - no signal, no timeout, just a frozen wait. No timeout is a gamble with user trust. Adding AbortController with a 30-second timeout turns fate into control. Wrap every fetch in a controlled signal: abort after 30 seconds if no reply. That tiny fix prevents entire systems from freezing. When outgoing calls lack boundaries, even a fast server can become a black hole. How often do we assume speed equals reliability - without safeguards? Remember: in digital systems, patience isn’t virtue. It’s a vulnerability. The bottom line: protect your flow - add timeouts before downtime hits.