Monday, February 13, 2006

Recent Malware developments

In some of my February's streams :) "The War against botnets and DDoS attacks" and "CME - 24 aka Nyxem, and who's infected?" I covered some of the recent events related to malware trends in the first months of 2006. This is perhaps the perfect time to say a big thanks to everyone who's been expressing ideas, remarks and thoughts on my malware research. While conducting the reseach itself I realized that I simply cannot include everything I want it, as I didn't wanted to release a book to have its content outdated in less than an year, but a "stick to the big picture" representation of the things to come. The best part is that while keeping daily track of the trends and trying to compile a summary to be released at the end of the year, many more concepts that I didn't include come to my mind, so I feel I'll have enough material for a quality summary and justification of my statements. So what are some of the recent developments to keep in mind?

A lot of buzz on the CME-24 front, and I feel quite a lot of time was spent on speculating on the infected population out of a web counter whose results weren't that very accurate as originally though. And as vendors closely cooperated to build awareness on the destructive payload, I think that's the first victory for 2006, no windows of opportunity The best is that CAIDA patiently waited until the buzz is over to actually come up with reliable statistics on Nyxem.

It's rather quiet on the AV radars' from the way I see it, and quickly going through F-Secure's, Kaspersky's (seem to be busy analyzing code, great real-time stats!), Symantec's I came across the similarities you can feel for yourself in "the wild" :) Symantec's ThreatCon is normal, what's interesting to note is VirusTotal's flood of detected WMF's, which is perhaps a consequence of the *known* second vulnerability.

James Ancheta's case was perhaps the first known and so nicely documented on botnet power on demand. Recently, a botnet, or the participation in such shut down a hospital's network, more over I think StormPay didn't comply with a DDoS extortion attempt during the weekend?

Joanna Rutkowska provided more insights on stealth malware in her research (slides, demo) about "about new generation of stealth malware, so called Stealth by Design (SbD) malware, which doesn't use any of the classic rootkit technology tricks, but still offers full stealth. The presentation also focuses on limitations of the current anti-rootkit technology and why it’s not useful in fighting SbD malware. Consequently, alternative method for compromise detection is advocated in this presentation, Explicit Compromise Detection (ECD), as well as the challenges which Independent Software Vendors encounter when trying to implement ECD for Windows systems – I call it Memory Reading Problem (MRP). "

How sound is the possibility of malware heading towards the BIOS anyway? An "Intelligent P2P worm's activity" that I just across to also deserves to be mentioned, the concept is great, still the authors have to figure out how to come up with legitimate file sizes for multimedia files if they really want to fake its existence, what do you think on this?

Some recent research and articles worth mentioning are, Kaspersky's Malware - Evolution : October - December 2005 outlines the possibilities for cryptoviral extortion attacks, 0days vulnerabilities, and how the WMF bug got purchased/sold for $4000. There's also been quite a lot of new trojans analyzed by third-party researchers, and among the many recent articles that made me an impression are "Malicious Malware: attacking the attackers, part 1" and part 2, from the article :

"This article explores measures to attack those malicious attackers who seek to harm our legitimate systems. The proactive use of exploits and bot networks that fight other bot networks, along with social engineering and attacker techniques are all discussed in an ethical manner."

Internet worms and IPv6 has nice points, still I wish there were only network based worms to bother about. Besides all I've missed important concepts in various commentaries, did you? Malware is still vulnerabilities/social engineering attacks split at least for the last several months, still the increased corporate and home IM usage will inevitable lead to many more security threats to worry about. Web platform worms such as MySpace and Google's AdSense Trojan, are slowly gaining grounds as a Web 2.0 concept, so virus or IDS signatures are to look for, try both!

During January, David Aitel reopened the subject of beneficial worms out of Vesselin Bontchev's research on "good worms". While I have my reservations on such a concept that would have to do with patching mostly the way I see it, could exploiting a vulnerability in a piece of malware by considered useful some day, or could a network mapping worm launched in the wild act as an early response system on mapped targets that could end up in a malware's "hitlist"? And I also think the alternative to such an approach going beyond the network level is Johnny Long's (recent chat with him) Google Dorks Hacking Database, you won't need to try to map the unlimited IPv6 address space looking for preys. Someone will either do the job for you, or with the time, transparancy in IPv6, one necessary for segmented and targeted attacks will be achieved as well.

Several days ago, Kaspersky released their summary for 2005, nothing ground breaking in here compared to previous research on how the WMF vulnerability was purchased/sold for $4000 :) but still, it's a very comprehensive and in-depth summary of 2005 in respect to the variables of a malware they keep track of. I recommend you to go through it. What made me an impression? 
- on average, 6368 malicious programs detected by month

- +272% Trojan-Downloaders 2005 vs 2004

- +212% Trojan-Dropper 2005 vs 2004

- +413% Rootkit 2005 vs 2004

- During 2005, on average 28 new rootkits a month

- IM worms 32 modifications per month

- IRC worms are on -31%

- P2P worms are on -43%, the best thing is that Kaspersky labs also shares my opinion on the reason for the decline, P2P busts and general prosecutions for file-sharing. What's also interesting is to mention is the recent ruling in a district court in Paris on the "legality of P2P" in France and the charge of 5 EUR per month for access to P2P, but for how long? :) P2P filesharing isn't illegal and if you cannot come up with a way to release your multimedia content online, don't bother doing at all. In previous chats I had with Eric Goldman, he also makes some very good points on the topic.

- +68% Exploit, that is software vulnerabilities and the use of exploits both known or 0day's with the idea to easily exploit targeted PC, though I'm expecting the actual percentage to be much higher

- Internet banking malware reached a record 402% growth rate by the end of 2005 The Trojan.Passwd is a very good example, it clearly indicates that it is written for financial gains. E-banking can indeed prove dangerous sometimes, and while I'm not being a paranoid in here, I'd would recommend you go through Candid's well written "Threats to Consider when doing E-banking" paper

- A modest growth from 22 programs per month in 2004 to 31 in 2005 on the Linux malware front

I feel today's malware scene is so vibrant that it's getting more and more complex to keep track of possible propagation vectors, ecosystem here and there, and mostly communicating what's going on to the general public(actually this one isn't). 
What's to come and what drives the current growth of malware?
- money!
- the commercialization of the market for software vulnerabilities, where we have the first underground purchase of the WMF exploit, so have software vulnerabilities always been the currency of trade in the security world or they've started getting the necessary attention recently?
- is stealth malware more than an issue compared to utilizing 0day vulnerabilities, and is retaining current zombie PCs a bigger priority than to infecting new ones?
- business competitors, enemies, unethical individuals are actively seeking for undetected pieces of malware coded especially for their needs, these definitely go beneath the sensors
- Ancheta's case is a clear indication of a working Ecosystem from my point of view, that goes as high as to provide after-sale services such as DDoS strength consultations and 0day malware on demand

To sum up, malware tends to look so sneaky when spreading and zoomed out :) I originally came across the VisualComplexity project in one of my previous posts on visualization. Feel I've missed something that's worth mentioning during the last two months? Than consider expanding the discussion!
You can also consider going through the following resources related to malware :