On Wednesday 31 May 2023 at 11:12 UTC, a search endpoint flood targeted a tax authority portal customer in Latin America. The attack peaked at 534.9 million requests per second and lasted 6 minutes. Traffic originated from 2316 autonomous systems in 20 countries, predominantly a Mirai-derived IoT botnet.
| Vector | search endpoint flood |
| Peak | 534.9 million requests per second |
| Duration | 6 min |
| Time to mitigation | 0.816 s |
| Attack traffic reaching origin | 0.032% |
| Legitimate traffic challenged | 0.24% |
Timeline
The attack was preceded by a hacktivist channel announcing the target. Request rates on the targeted routes exceeded their hourly baseline by a factor of 285 within 8 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 37 ms during the first minute, then normal service.
Recommendations
- Keep origin IPs out of public DNS history.
- Enable authenticated origin pulls.
- Lower challenge thresholds on authentication endpoints during high-risk events.
The billing model is what got our finance team on board, honestly.
Would love a follow-up on how you handle HTTP/3 fingerprinting.
Would love a follow-up on how you handle HTTP/3 fingerprinting.
Great to hear, thanks for sharing your experience.
Carpet bombing is nasty. Good to see a clear explanation of it.
Thanks! Yes — the risk score and its components are included in every log record.
Could you share the dataset behind the percentages?