What this tests

parsing

Point your client at this specimen and find out — before prod does — whether it actually validates what arrives, or just hopes.

Anatomy

The id field is 9223372036854775807 — int64 max, a perfectly valid JSON number that fits nowhere comfortably. JavaScript's Number is a float64, which rounds this to a different integer entirely — the id you stored is not the id you received, it's a nearby id, like a delivery to the wrong flat that insists it's the right one. This is how Twitter IDs famously broke JavaScript clients in 2011: tweet IDs exceeded 2^53, JS rounded them, and suddenly everyone's replies pointed at different tweets. The fix — serialize IDs as strings — exists precisely because this happened at scale, in public, to a company with a large API. Clients that parse into int64 handle it. Clients that parse into float64 get a number that's off by one or two, which is more than enough to join the wrong records, bill the wrong account, or reply to the wrong tweet. The difference between int64 max and its float64 neighbour is the difference between a database and a disaster.

Replay it

curl -i https://faultydata.com/r/FD-2026-056

// response appears here

Signature

{Status:200 Headers:map[] Body:{"id":9223372036854775807,"name":"ada","balance":0.1} Corrupt: Delay:<nil> Flaky:<nil> Sequence:<nil> Drip:<nil> StallAfter:0 Hang:false Reset:false Raw: Phases:[]}

Keep digging

FD-2026-011 — Truncated JSON body FD-2026-012 — Malformed JSON (trailing comma) FD-2026-013 — Wrong types in a valid envelope FD-2026-014 — Encoding mismatch (claims UTF-8, sends Latin-1) FD-2026-015 — UTF-8 BOM prefix FD-2026-028 — Malformed channel file update (CrowdStrike, Jul 2024)