Cloudflare Mitigated Record-Setting 17.2 Million RPS DDoS Attack
Security and web performance services provider Cloudflare says it identified and mitigated what appears to be the largest volumetric distributed denial-of-service (DDoS) attack to date, which peaked at 17.2 million requests-per-second (RPS).
Almost three times larger than any previous volumetric DDoS attack, the assault reached 68% of the average 25 million HTTP requests per second of legitimate traffic that Cloudflare served daily during the second quarter of the year. Within seconds, it reached a total of more than 330 million attack requests.
Targeting an organization in the financial sector, the assault was launched by a Mirai-like botnet, with more than 20,000 bots in 125 countries sending attack requests. Roughly 15% of the traffic originated from Indonesia, while another 17% came from India and Brazil combined.
According to Cloudflare, the botnet appears to have lost roughly 2,000 of its 30,000 bots, but it remained capable of launching impressive traffic bursts over short periods of time (seconds).
Last week, the same botnet was used to launch an HTTP DDoS attack that peaked at roughly 8 million RPS. Two weeks ago, another Mirai-variant botnet was responsible for “over a dozen UDP and TCP based DDoS attacks that peaked multiple times above 1 Tbps, with a max peak of approximately 1.2 Tbps,” Cloudflare says.
The security services provider says that the number of Mirari-based DDoS attacks has increased significantly over the past weeks. In July, L3/4 attacks went up by 88%, while L7 attacks increased by 9%. By the end of August, the company expects a 71% increase in L3/4 Mirai attacks and a 185% spike in L7 DDoS attacks.
Mirai was first observed in 2016 targeting Internet of Things (IoT) devices such as CCTV cameras. Numerous variants of the botnet have emerged ever since, expanding the list of targeted devices to include Linux routers and servers, Android devices, and more.
While the first variants of the malware would propagate through open Telnet ports 23 and 2323, brute-forcing known credentials such as default usernames and passwords, newer versions also target publicly disclosed vulnerabilities in IoT devices.