What this tests

timeouts

Point your client at this specimen and find out — before prod does — whether it has a timeout, and whether the timeout has a timeout.

Anatomy

Nothing is wrong. Everything is right. It's just five seconds late — which in distributed-systems time is roughly one geological epoch and change. The response, when it arrives, is perfect: correct status, correct body, headers immaculate. A Michelin meal served by a waiter who took the scenic route through a hedge maze. This is the fault that actually takes production down, because nobody alerts on slow. A dependency drags its feet, every caller's timeout — or embarrassing lack of one — gets exercised simultaneously, thread pools fill like a car park on cup final day, and the cascade writes itself in whatever language your services happen to speak. If your client has a deadline, this specimen is a light workout. If it doesn't — and many don't, shipped by people who believed "the API is usually fast" constituted a resilience strategy — enjoy watching it discover mortality.

Replay it

curl -i https://faultydata.com/r/FD-2026-016

// response appears here

Signature

{Status:200 Headers:map[] Body: Corrupt: Delay:0x69e720af2210 Flaky:<nil> Sequence:<nil> Drip:<nil> StallAfter:0 Hang:false Reset:false Raw: Phases:[]}

Keep digging

FD-2026-006 — Total reachability loss (Meta BGP withdrawal, Oct 2021) FD-2026-017 — Connection accepted, never replies FD-2026-018 — Partial body then stall FD-2026-025 — Edge DNS unreachable (Akamai, Jul 2021) FD-2026-043 — Datacenter literally on fire (OVH, Mar 2021) FD-2026-057 — Drip-fed response body