What this tests
retry-logicmonitoring
Point your client at this specimen and find out — before prod does — whether it backs off, retries, and knows when to stop.
Anatomy
Niantic planned for success and got fifty times success instead — the most traffic a Google Cloud customer had ever generated, arriving in about a week, powered by the entire planet simultaneously deciding to hunt cartoon rats in car parks. The failure mode was pure capacity arithmetic: logins 503ing and players retrying so hard that the retry traffic itself became a meaningful fraction of the load. Every rejected trainer became a retrying trainer. It is still the canonical answer to "what does a launch that succeeds too hard look like?"
Source: Google Cloud blog · 2016-07-01
Replay it
curl -i https://faultydata.com/r/FD-2026-089
// response appears hereSignature
{Status:0 Headers:map[] Body: Corrupt: Delay:<nil> Flaky:<nil> Sequence:<nil> Drip:<nil> StallAfter:0 Hang:false Reset:false Raw: Phases:[{Weight:65 Signature:{Status:503 Headers:map[Retry-After:10] Body:{"error":"trainer capacity reached — try again"} Corrupt: Delay:<nil> Flaky:<nil> Sequence:<nil> Drip:<nil> StallAfter:0 Hang:false Reset:false Raw: Phases:[]}} {Weight:35 Signature:{Status:200 Headers:map[] Body:{"status":"ok","trainers_online":"all of them"} Corrupt: Delay:<nil> Flaky:<nil> Sequence:<nil> Drip:<nil> StallAfter:0 Hang:false Reset:false Raw: Phases:[]}}]}
Keep digging
FD-2026-002 — CDN edge degradation (Fastly, June 2021) FD-2026-021 — S3 us-east-1 subsystem removal (AWS, Feb 2017) FD-2026-024 — Network config rollback storm (Cloudflare, Jun 2022) FD-2026-044 — Three-day outage from a consensus bug (Roblox, Oct 2021) FD-2026-059 — East-west network partition (GitHub, Oct 2018) FD-2026-003 — Elevated latency + partial 5xx (AWS us-east-1, Nov 2020)