On Tuesday 22 February 2022 at 14:50 UTC, a HTTP/2 rapid reset targeted a news publisher customer in the Nordics. The attack peaked at 654.7 million requests per second and lasted 156 minutes. Traffic originated from 1893 autonomous systems in 16 countries, predominantly a rented booter service.
| Vector | HTTP/2 rapid reset |
| Peak | 654.7 million requests per second |
| Duration | 156 min |
| Time to mitigation | 0.221 s |
| Attack traffic reaching origin | 0.074% |
| Legitimate traffic challenged | 0.57% |
Timeline
The attack was preceded by a publicly announced sales event. Request rates on the targeted routes exceeded their hourly baseline by a factor of 338 within 29 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
The customer’s status page stayed green throughout; they learned about the attack from our notification.
Recommendations
- Review allow-listed partner ranges quarterly.
- Enable authenticated origin pulls.
- Keep origin IPs out of public DNS history.
Do you publish the edge IP ranges in a machine-readable format?
Any plans to support per-tenant limits keyed on a JWT claim?
Carpet bombing is nasty. Good to see a clear explanation of it.
Interesting that most attacks are under ten minutes. Our experience is similar.
Do you publish the edge IP ranges in a machine-readable format?
Would love a follow-up on how you handle HTTP/3 fingerprinting.
We moved from a scrubbing provider to always-on last year; time to mitigation went from minutes to basically nothing.
Solid runbook advice. The DNS-at-2am point hit home.