Police arrest Mariposa botnet masters, 12M+ hosts compromised
Summary: Spanish Ministry of Interior arrests 3 botnet masters operating a 12M+ infected hosts botnet that managed to steal sensitive data from 800,000 users across 190 countries, some of which include Fortune 1000 companies and 40 major banks.
According to a statement published by the Spanish Ministry of Interior, the botnet masters behind a 12M+ infected hosts botnet dubbed Mariposa, were arrested in a cooperative effort between law enforcement, security vendors and the academic community.
Following the arrest of one of the botnet masters, law enforcement officers seized sensitive data belonging to 800,000 users across 190 countries, and found evidence of infected hosts located within the networks of 500 of the US Fortune 1,000 companies and more than 40 major banks.
Just how sophisticated were the botnet masters behind Mariposa? You'll be surprised to find out.
- On December 23 2009, in a joint international operation, the Mariposa Working Group was able to take control of Mariposa. The gang’s leader, alias Netkairo, seemingly rattled, tried at all costs to regain control of the botnet. As I mentioned before, to connect to the Mariposa C&C servers the criminals used anonymous VPN services to cover their tracks, but on one occasion, when trying to gain control of the botnet, Netkairo made a fatal error: he connected directly from his home computer instead of using the VPN. Netkairo finally regained control of Mariposa and launched a denial of service attack against Defence Intelligence using all the bots in his control. This attack seriously impacted an ISP, leaving numerous clients without an Internet connection for several hours, including several Canadian universities and government institutions.
The initial reports describe the group as not so technically sophisticated "normal people" making a lot of money through cybercrime. A logical question emerges - how is it possible that a group of "normal people" can build such a massive botnet? By outsourcing.
- Go through related posts: Malware Infected Hosts as Stepping Stones; The Cost of Anonymizing a Cybercriminal's Internet Activities; The Cost of Anonymizing a Cybercriminal's Internet Activities - Part Two; Zeus Crimeware as a Service Going Mainstream; Managed Polymorphic Script Obfuscation Services
The name Mariposa actually means butterfly, which is the original name of a commercially distributed DIY malware kit, sold online for 800/1000 EUR, unless of course the arrested botnet masters weren't using a pirated version, which is ironically, a common practice within the cybercrime ecosystem these days. What's particularly interesting about the malware was the fact that it was using its own UDP-based protocol for communication, which according to the original author was developed with stealthiness in mind since UDP connections are rarely logged.
Moreover, the bot has typical for modern malware releases anti-debugging features, as well as built-in DDoS functionality relying on TCP and UDP flood tactics. The three main propagation vectors include MSN, removable media, and through P2P, targeting the following networks - Ares, Bearshare, Imesh, Shareaza, Kazaa, Dcplusplus, Emule, Emuleplus, Limewire.
Just like the majority of commercial DIY malware releases, this one also includes a disclaimer attempting to position it as a "tool for educational purposes only", with the DDoS option itself described as a tool for "stress testing" your own infrastructure. Also, despite the initial claims that the "mastermind of 'botnet' scam remains a mystery", commercial releases by the original coder of the bot have been circulating in the underground marketplace since 2007. With the recent bust of his customers, he'll be definitely keeping a low profile for a while.
The Mariposa botnet is the tip of iceberg in respect to DIY botnets (Research: Small DIY botnets prevalent in enterprise networks; Inside the botnets that never make the news - A Gallery) aggregated using commercially, or freely available malware kits.
What this incident proves is that not only is cybercrime becoming easier to outsource in 2010, but also, that even inexperienced people can quickly gain access to capabilities once reserved for sophisticated attackers.
Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Talkback
This story is already old, but...
How to identify the targets?
computers?
This site is relative bot spam free.
Now if we could only get rid of the Redmond bots. ;)
lol... :D
Here's at least one example. (nt)
Except for that produced by the ABMbots
What is more baffling, however...
And yes, I know that if they keep doing it, it must be because someone is stupid enough to do business with a Spammer.
They've destroyed usenet, made email a pain in the ass, and are now working on message boards and social networking sites.
They're all scum, and if they all died tonight I wouldn't shed a tear.
Scum, scum, scum, scum...
Funny story.. I met a programmer a couple of days ago at a coffee shop. When i asked what kind of apps he writes, he told me he writes email harvesting apps. I instantly had this visual of shoving my fist down his throat and pulling his heart out through his esophagus. Instead, I "accidentally" spilled my coffee on him!
Thats What I'm Saying
I guess these are the same people who believe they really had a South African relative leave them several godszillion dollars. A South AFrican relative they never knew about, at that. And I guess the fact it's sent to "undisclosed recipients" isn't enough a clue that being the "last known relative" is bull. lol.
I assume it's 1 in a million
The sad thing is the number of banks and real businesses that have infected computers. Their executives should be held accountable to their shareholders. If they have a virus infection it should be reported with their quarterly results. It's not always IT that's to blame, since it's the executives that make the policies and hold back IT sometimes.
Barnum's Principle
This way to the egress !!
Interesting.
Let me guess..
And for the record.. I use Linux on my home server and would use on all my machines if it weren't such PITA to set up..
One GUI installer for all Linux platforms and Linux <i>could</i> gain significant market share.
Asking an open ended question is zealotry? I'm sorry then. Forget I asked.
The Old Moronic Marketshare argument again
Look at Apache, dominant market share, relativly few hacks compared to IIS, which has less market share.
Show us the data.
O Rly?
Why is it that Linux sites are compromised more than twice as often then, according to the site statistics?
Thanks
Where on zone-h.con does it say..
All I see are countless pages of "site X was visibly defaced", "site Y was visibly defaced", etc.
It doesn't seem to say anywhere on that site how many were noticeably defaced per webserver. It also doesn't seem to say how the noticeable defacement occurred (was it a vulnerability in the OS? In the webserver? In a scripting language? In some unrelated program the owner was running? A weak password being guessed? A weak admin giving out the password???), and the only thing it lists anywhere on the whole website are noticeable defacements. Meaning when some script kiddie replaces the front page with "HAHAHA PWNT", not when someone hacks into the server and steals all the data on it or anything like that.
Completely useless.