On Sunday 11 December 2022 at 04:28 UTC, a HTTP GET flood targeted a B2B SaaS customer in the UK. The attack peaked at 93.5 million requests per second and lasted 92 minutes. Traffic originated from 3057 autonomous systems in 74 countries, predominantly a rented booter service.
| Vector | HTTP GET flood |
| Peak | 93.5 million requests per second |
| Duration | 92 min |
| Time to mitigation | 0.165 s |
| Attack traffic reaching origin | 0.058% |
| Legitimate traffic challenged | 0.50% |
Timeline
The attack was preceded by a competitor’s product launch. Request rates on the targeted routes exceeded their hourly baseline by a factor of 468 within 26 seconds. The risk score of participating clients crossed the challenge threshold automatically and proof-of-work difficulty rose with origin load.
What the customer saw
A brief increase in p99 latency of 193 ms during the first minute, then normal service.
Recommendations
- Lower challenge thresholds on authentication endpoints during high-risk events.
- Add a dedicated rate limit for the targeted route.
- Enable authenticated origin pulls.
Would love a follow-up on how you handle HTTP/3 fingerprinting.
Verified good bots and allow-listed partners bypass challenges entirely.
Would love a follow-up on how you handle HTTP/3 fingerprinting.
Verified good bots and allow-listed partners bypass challenges entirely.
Our auditors asked for exactly this kind of incident evidence under DORA.
This matches what we see in iGaming around big matches.
Do you publish the edge IP ranges in a machine-readable format?