Honeypot Stings Attackers With Counterattacks
Researchers test the controversial concept of hacking back and gathering intelligence on attackers
March 26, 2013
A Russian researcher who built an aggressive honeypot to test the ability to hack back at attackers mostly ensnared fellow white hat researchers, script kiddies, and some of his friends in his experiment—but he discovered that he had also counterattacked the desktop of an intelligence agency in a nation formerly part of the Soviet Union.
Alexey Sintsov, a security researcher and co-founder of DefCon Russia, ran an experimental homegrown "aggressive" honeypot from the second quarter of 2011 through the third quarter of 2012 on the DefCon Russia website he manages in anticipation of the site being targeted by attackers. Turns out that it's not only easy to build a honeypot that attacks back, Sintsov says, but it was also relatively simple to gather the attackers' network adapter settings, trace routes, and login names.
"When you start a site about IT security, you should be ready, that [a] big part of people will try to hack you. Not because they are evil guys or black hats, just because they can," Sintsov says. "My thought was like, 'Okay, it will be fun to play with them in a new game they do not expect'" and to test the theory of "reverse-penetration," he says.
"Actually, it was awareness testing for [the] community. It was interesting to find that attackers do not expect reverse attacks on themselves," he says.
The honeypot—which is no longer active--basically installed a backdoor agent via Java applet and exploited JSONP hijacking vulnerabilities in two email services.
[Georgia's CERT tricks alleged Russian hacker with phony file, records him via his computer, and ID's him. See Say 'Cheese': Georgian Nation Makes Offense Its Defense.]
The trap was specifically set for SQL injection attacks. Sintsov used two basic lures for potential attackers on the site: a PHP-based honeypot server that included a social engineering element and an automated attack that grabbed the attackers' email addresses if he or she used two Russian email services, mail.ru and yandex.ru, exploiting now-patched vulnerabilities in those services.
The PHP portion included a field for "members" to enter their "secret code" to enter the "private zone," he explains. "So it's a good idea to try a SQL injection attack" there, he says of the lure.
"My script had [a] few checks for some patterns, and when a SQL injection attempt was detected, the script [threw the] Java applet, 'GUI for member zona. Welcome,'" he says. The Java applet then installed a backdoor on the attacker's Windows machine, he says.
While it is possible to grab the attackers' internal IP addresses and resources, scan for his files, BSSIDs, and make audio and video recordings from his laptop, among other things with the attacking honeypot, Sintsov didn't go that far because that would be considered over the top in the information security industry, he says.
"It was fun when you got your friends' logins, e-mails and PC names. Or it is always a pleasure to see how [an] antivirus vendor runs your software on their VMware. For script kiddies, it was predictable," he says. "But some attacks and results [were] really unexpected, like reverse-penetrated [a] desktop that belong [to an] intelligence service of one of ex-USSR republic" nations, he says.
Sintsov says he later learned that the intel agent's workstation had likely been compromised and used as a cover for another attacker. He and his team found the hacker's "nickname" as well.
They also spotted SQL injection attempts from an external IP address of Russia's Ministry of Defense, but luckily, those "attackers" didn't run his backdoor for some reason, possibly because they were on a non-Windows machine, he says.
But aggressive and offensive honeypots are a controversial concept and the legal ramifications are tricky. Sintsov, who presented his honeypot experiment findings at Black Hat Europe this month in Amsterdam, says the legal issues are up for interpretation. The Java applet and email grabber were housed in a "private zone" on the website, for example, so why would the honeypot operator be at fault if someone hacked into that area and ran something—a SQL injection attack, he says.
And what about when an attacker breaks into the area of the site with the binary/backdoor and downloads that backdoor? "This page is defended with a password and I do not ask visitors to enter this page. Even more, I have a message that it is [a] forbidden zone, please enter the password or go away. But someone hacks us and downloaded this backdoor," he says. "I am not a law guy, but think that it can be [an] interesting discussion."
Still, Sintsov admits running an aggressive honeypot that attacks back at attackers would be difficult for a typical enterprise to justify. "But if we can exclude law questions, I think it is a good thing to try for the oil sector or smart grid," he says.
In his research paper (PDF) on the experiment, Sintsov explains it this way: "Obviously, reverse penetration has a number of moral, ethical and legal issues. However, it can be applied with a sufficiently high efficiency for detection and identification of attackers, at least against those who are not careful enough to expect a counterattack. I think that these methods can be used for protection at the state level and, probably, already exist these days."
Have a comment on this story? Please click "Add Your Comment" below. If you'd like to contact Dark Reading's editors directly, send us a message.
About the Author
You May Also Like
The State of Attack Surface Management (ASM), Featuring Forrester
Nov 15, 2024Applying the Principle of Least Privilege to the Cloud
Nov 18, 2024The Right Way to Use Artificial Intelligence and Machine Learning in Incident Response
Nov 20, 2024Safeguarding GitHub Data to Fuel Web Innovation
Nov 21, 2024