Botnet likely caused spike in number of Tor clients
The spike in the number of clients using the Tor anonymity network was likely caused by a botnet, according to Tor and third-party security researchers.
Around Aug. 20, the number of Tor clients jumped. There are now millions of new Tor clients and the number continues to rise, said Tor project leader Roger Dingledine, writing as “arma” in a blog post on Thursday. The spike is likely being caused by a botnet, wrote Dingledine, who often blogs under the “arma” handle and is one of the original Tor developers.
Tor obscures a user’s IP address by routing traffic through a series of encrypted volunteer relays that are selected at random. People have been using it to protect their privacy online but the same features make it attractive for those with more malicious intentions.
“Some people have speculated that the growth in users comes from activists in Syria, Russia, the United States, or some other country that has good reason to have activists and journalists adopting Tor en masse lately. Others have speculated that it’s due to massive adoption of the Pirate Browser (a Tor Browser Bundle version that discards most of Tor’s security and privacy features),” Dingledine wrote.
“The fact is, with a growth curve like this one, there’s basically no way that there’s a new human behind each of these new Tor clients,” Dingledine wrote.
The clients were installed onto millions of computers pretty much overnight. Since no large software or OS vendors have come forward to say they just bundled Tor with all their products, that leaves one conclusion: somebody infected millions of computers and as part of their plan they installed Tor clients on them, Dingledine wrote.
The suspicion that the uptake in Tor usage is caused by a botnet is shared by Dutch security firm Fox-IT.
“We found that it is very likely that it is a botnet,” said Ronald Prins, director and co-founder of the company.
“It seems to be a general-purpose botnet,” Prins said, adding that a general purpose botnet is often used to harvest data such as log-in credentials that can be used later, or sold to another party. But what the botnet is trying to achieve is unknown at this point, he said.
Using Tor to control a botnet can be convenient because it makes it hard to detect, Prins said. The botnet’s command and control (C&C) server is hidden by Tor, he noted. “This hinders the take-down very much,” he said.
While Tor can be helpful, it also has a significant drawback, Prins said: “Traffic is very slow.”
Theories
Fox-IT researchers said the name of the botnet could be “Mevade.A.” But they also found old references that suggest the name is “Sefnit,” which dates back to at least 2009 and also included Tor connectivity, they said in a blog post.
“We have found various references that the malware is internally known as SBC to its operators,” they wrote, adding that they assume that it originates from a Russian-speaking area, and is likely to be financial-crime related. The researchers did not specify where they found the references.
Tor also thinks it is plausible that the botnet is running its C&C point as a hidden service, according to Dingledine.
While the Tor network is still working for now, the botnet could cause trouble, according to Tor.
The biggest problems are not caused by the amount of traffic added to the network, but rather by new circuits that are being made, Dingledine wrote.
“Tor clients build circuits preemptively, and millions of Tor clients means millions of circuits. Each circuit requires the relays to do expensive public key operations, and many of our relays are now maxed out on CPU load,” Dingledine wrote.
This sets up a possible dangerous cycle. “When a client tries to build a circuit but it fails, it tries again. So if relays are so overwhelmed that they each drop half the requests they get, then more than half the attempted circuits will fail (since all the relays on the circuit have to succeed), generating even more circuit requests,” Dingledine wrote.
To deal with these issues, Tor took several temporary measures to mitigate the problem. But for the future, other options need to be explored, Dingledine said. Tor could for example limit the circuit-create requests or learn to recognize the circuit building signature of a bot client.
“It would be great if botnet researchers would identify the particular characteristics of the botnet and start looking at ways to shut it down (or at least get it off of Tor),” Dingledine said.
“And finally, I still maintain that if you have a multi-million node botnet, it’s silly to try to hide it behind the 4,000-relay Tor network. These people should be using their botnet as a peer-to-peer anonymity system for itself,” Dingledine wrote.