DDoS Attack Graphs from Russia vs Georgia's Cyberattacks

Part of Georgia's information warfare campaign aiming to minimize the bandwidth impact on its de-facto media platforms such as the web site of  their Ministry of Foreign Affairs, I've just received a report part of Georgia's "Russian Invasion of Georgia" series entitled "Russian Cyberwar on Georgia", which is quoting me on page 4 in regard to the "too good to be courtesy of Russia's cyber militia" creative that appeared on the defaced Georgian President's web site. The report also includes DDoS attack graphs and related details worth going through :

"The last large cyberattack took place on 27 August. After that, there have been no serious attacks on Georgian cyberspace. By that is meant that minor attacks are still continuing but these are indistinguishable from regular traffic and can certainly be attributed to regular civilians. On 27 August, at approximately 16:18 (GMT +3) a DDoS attack against the Georgian websites was launched. The main target was the Georgian Ministry of Foreign Affairs. The attacks peaked at approx 0,5 million network packets per second, and up to 200–250 Mbits per second in bandwidth (see attached graphs). The graphs represent a 5-minute average: actual peaks were higher.

The attacks mainly consisted of HTTP queries to the http://mfa.gov.ge website. These were requests for the main page script with randomly generated parameters. These requests were generated to overload the web server in a way where every single request would need significant CPU time. The initial wave of the attack disrupted services for some Georgian websites. The services became slow and unresponsive. This was due to the load on the servers by these requests. As you see from the graphs above the attacks started to wind down after most of the attackers were successfully blocked. The latest attack may have been initiated as a response to the media coverage on the Russian cyber attacks."

In case you're interested in more factual evidence about what was happening at the particular moment in time, go through the following assessment - "Coordinated Russia vs Georgia cyber attack in progress", as well as through the following posts - "The Russia vs Georgia Cyber Attack"; "Who's Behind the Georgia Cyber Attacks?"; "Georgia President’s web site under DDoS attack from Russian hackers".

The Cost of Anonymizing a Cybercriminal's Internet Activities

What would the perfect traffic anonymity service provider targeting cybercriminals consist of? A service operating in Russia that is on purposely not logging any of its user's activities, next to allowing direct spamming from the socks servers, automatic rotation of the VPN servers which they operate in a RBN style hosting provider, or a service using actual malware infected hosts as VPN tunnels not only securing the cybercrime traffic, but also, forwarding the responsibility for the malicious activities to the end user?

Long gone are the days of socks chaining, the practice of automatically connecting to multiple malware infected hosts in order to use them as stepping stones, in between the rest of the malicious activities going on their behalf.

The possibilities for building point-to-point or server-to-multiclient encrypted tunnels between malware infected hosts by using already available Socks5 functions has always been there. As of August, the coders behind a relatively popular web based malware originally started as a DDoS kit, but later on started introducing new features on a "module basis", they have started offering a BETA module for building a VPN network of malware infected hosts, including an admin panel for reselling access to these hosts in order to better monetize their botnet.

This VPN-owning of malware infected hosts is not only resulting in improved anonymity for botnet masters and anyone else having access to the network, but is also contributing to the growth of VPN services designed specifically to be accessed by cybercriminals created on the foundatiosn of such admin panels offering easier reselling of access to the network.

So, what's the cost of anonymizing a cybercriminal's Internet activities? Starting from $40 and going to $300 for a quarter of access, with the price increasing based on the level of anonymity added.

Quality Assurance in Malware Attacks - Part Two

Surprisingly, while opportunistic cybercriminals have long embraced the malware as a service model, and are offering managed lower detection rate services for a customer's malware, or DIY ones where the customer can take advantage of popular tools ported to the Web, others are still trying to innovate at a faddish market niche - multiple offline AV scanners tools aiming to ensure that their malware doesn't end up in the hands of vendors/researchers.

Multiple offline AV scanning tools like this very latest release, naturally using pirated copies of popular antivirus software, are faddish, due to the fact that during the last two years, the underground has been busy working on several paid web based services, that not only make sure vendors and researchers never get the chance to obtain the samples, but also, are already offering scheduled scanning of malware and automatic ICQ/Jabber notifications for QA of the campaign, next to the rest of unique features disintermediating legitimate multiple AV scanning services.

Certain features within such services clearly speak for the intentions of the people behind the service. For instance, among one of these features is the ability to fetch a binary from a set of given dropper URLs like malwaredomain.com/binary.exe, the result of the scan can then alert the malware campaigner about the current state of detection.

What's on these proprietary multiple AV scanning service's to-do list? Let's say anything that a legitimate multiple AV scanning service would never offer, like the following according to one of the services in question :

- DIY heuristic scanning level settings for each of the software in place
- upcoming sets of anti spyware and personal firewalls with detailed statistics of the sandboxing
- behavior-based detection results

The possibilities for integrating such proprietary multi AV scanning services within the QA process of a malware campaign are countless, and both, the customers and the sellers seem to have realized the potential of this ecosystem.

Cybercriminals Abusing Lycos Spain To Serve Malware

Spanish cybercriminals have recently started taking advantage of the bogus accounts at Lycos Spain, which they seem to be registering on their own, by releasing a do-it-yourself malicious link generator redirecting to fake YouTube and Adobe Flash video pages. Whereas the concept of abusing legitimate web services for infection and propagation isn't new, what's new is the fact that the FTP access is efficiently abused. 

Here's a description of the link generator :

"Download the program and run it asks for an ID (identifier), then copy it and paste it there, then press' Create Installer 'and the program will create the Installer! (this program to run a simulation that is installing the Adobe Flash and indicates to our page that "has been installed Adobe Flash," in order to show the video when YouVideo refresh the page, this you must file tie it in with your server! and what flames or Installer Setup (simulating being an installer)!  Now you need to upload that file you've joined an FTP, click Next and put the path of that file in the next step!"

Whereas the tool is exclusively relying on Lycos Spain to host the binaries and the campaign itself, the recent blackhat SEO campaign relying on pre-registered Windows Live Spaces and AOL Journals syndicating hot Google Trends keywords, further indicates the malicious attacker's capabilities of efficiently abusing legitimate services. And with the process of bogus accounts registration performed automatically, or outsourced entirely, malicious services aiming to automate the abuse process are only going to get more efficient.

Commoditization of Anti Debugging Features in RATs - Part Two

Yet another piece of malware promoted as a RAT (remote access tool) includes what's turning into the defacto set of anti-debugging features within RATs.

As the authors point out, the Anti Virtual PC, VMware, Virtualbox, Sandboxie, ThreatExpert, Anubis, CWSandbox, Joebox, Norman Sandbox features inevitably increase the server size. Next to the product, there's always the managed service of ensuring a lower detection rate for binaries submitted to the authors.

Summarizing Zero Day's Posts for September

As usual, here's September's summary of all of my posts at Zero Day. You may also want to catch up and go through August's and July's summaries, next to adding my personal RSS feed or Zero Day's main feed to your RSS reader.

Notable article for September - Spamming vendor launches managed spamming service.

01. DoS vulnerability hits Google's Chrome, crashes with all tabs
02. Malware and spam attacks exploiting Picasa and ImageShack
03. Spamming vendor launches managed spamming service
04. Facebook introducing new security warning feature
05. Google downplays Chrome's carpet-bombing flaw
06. Targeted malware attack against U.S schools intercepted
07. The most "dangerous" celebrities to search for in 2008
08. Norwegian BitTorrent tracker under DDoS attack
09. Attacker: Hacking Sarah Palin's email was easy
10. Bill O'Reilly's web site hacked, attackers release personal details of users
11. India's government: At last, we've cracked Blackberry's encryption
12. Memory exhaustion DoS vulnerability hits Google's Chrome
13. 44% of second hand mobile devices still contain sensitive data
14. Spammers attacking Microsoft's CAPTCHA -- again