After pushing malware, ad networks also used for DDoS

The DDoS mitigation team at CloudFlare recently observed a large-scale attack which they believe was the result of malicious ads being loaded inside apps and browsers on mobile devices.

The attack, which targeted one of the company's customers, peaked at 275,000 HTTP requests per second and was launched from over 650,000 unique IP (Internet Protocol) addresses, most of them from China.

What was interesting about this attack was that the requests appeared to be generated by real browsers, not scripts or malware, as are typically used in HTTP-based DDoS attacks. Furthermore, an analysis of the request headers indicated that almost 80 percent of the devices generating the traffic were smartphones and tablets.

The investigation revealed that the referrer for the requests was an attack page with multiple ads on it and embedded JavaScript code that loaded an additional script from other domains. That secondary script instructed the browsers to make Ajax (XHR) cross-origin calls in a loop to the victim site.

"There is no way to know for sure why so many mobile devices visited the attack page, but the most plausible distribution vector seems to be an ad network," said Marek Majkowski, a member of CloudFlare's DDoS mitigation team, in a blog post. "It seems probable that users were served advertisements containing the malicious JavaScript. These ads were likely showed in iframes in mobile apps, or mobile browsers to people casually browsing the internet."

Security researchers have warned about the possibility of mobile-based DDoS attacks for years, especially as mobile data connections become faster and more stable.

In a presentation at the Black Hat security conference in 2013, security researchers Jeremiah Grossman and Matt Johansen warned about the possibility of malicious advertisements being used to create a browser-based botnet that could be used to launch DDoS attacks, among other things. It appears that attackers might have finally caught up with that idea.

Malvertising attacks, which involve tricking ad networks into distributing advertisements that contain malicious code, have been used to infect computers with malware for years. Despite efforts from the networks to stop such incidents, attackers still seem to find a way in, most likely due to the complexity of the online advertising ecosystem.

A 2014 investigation by the U.S. Senate into malvertising concluded that online advertisements typically go through five or six intermediaries before they are delivered to users' browsers, and ads can be replaced with malicious ones at any point in that chain.

"The online advertising industry has grown in complexity to such an extent that each party can conceivably claim it is not responsible when malware is delivered to a user's computer through an advertisement," the Senate's Committee on Homeland Security and Governmental Affairs said in its report.