General Discussion

[Botnet computing power]

FYI...

Botnet computing power...
- http://blog.trendmicro.com/the-internet-infestation-how-bad-is-it-really/
Sep. 16, 2009 - "Industry experts have previously estimated that, on average, a compromised machine remains infected for 6 weeks. However, our latest research indicates that this estimate is far from accurate. During the analysis of approximately 100 million compromised IP addresses, we identified that half of all IP addresses were infected for at least 300 days. That percentage rises to eighty percent if the minimum time is reduced to a month... The news only gets worse from that point. While three-fourths of the IP addresses in our study were identified with consumer users, the remaining quarter belong to enterprise users. Because a single IP address for these users is typically identified with a single gateway which may, in turn, be connected to multiple machines in an internal network, the actual percentage of enterprise machines affected by malware may be higher than the IP address data suggests. Once a machine becomes compromised, it is not unusual to find it has become part of a wider botnet. Botnets frequently cause damage in the form of malware attacks, fraud, information theft and other crimes. In 2009, virtually all malware tracked by Trend Micro experts are used by cybercriminals to steal information... Overall, botnets control more compromised machines than had been previously believed. Only a handful of criminals (likely a few hundred) have more than 100 million computers under their control. This means that cybercriminals have more computing power at their disposal than the entire world’s supercomputers combined. It’s no wonder then that more than 90 percent of all e-mail worldwide is now spam..."

(More detail and charts available at the URL above.)

 

[Conficker still defeats experts]

FYI...

Conficker still defeats experts
- http://www.theinquirer.net/inquirer/news/1534307/conflicker-defeats-experts
22 September 2009 - "... The worm has infected more than five million computers in a botnet that could take out the Internet in some countries... Rodney Joffe, a director of the Conficker Working Group formed to defeat the worm said, "The general agreement in the security world is that Conficker is the largest threat facing us from a cyber crime point of view." The worm, which spreads rapidly among personal computers by exploiting a flaw in Microsoft Windows, first surfaced last November. According to Joffe, "it has proven to be extremely resilient. It's almost impossible to remove." Infected PCs are dragooned into a "botnet" controlled by the Conficker worm's unknown authors, which security researchers fear could be used to launch cyber attacks over the Internet..."
- http://www.confickerworkinggroup.org/wiki/pmwiki.php/ANY/InfectionTracking
2009-09-21

Conficker C P2P Protocol and Implementation
- http://mtc.sri.com/Conficker/P2P/
Last Update: 21 September 2009
> http://mtc.sri.com/Conficker/P2P/#overview

- http://isc.sans.org/tag.html?tag=conficker

Conficker Eye Chart
- http://www.confickerworkinggroup.org/infection_test/cfeyechart.html

 

[Botnet hides its commands]

FYI...

Botnet hides its commands...
- http://www.secureworks.com/research/blog/index.php/2009/09/29/monkifdlkhora-botnet-hiding-its-commands-as-jpeg-images/
September 29, 2009 - "SecureWorks... has been carefully monitoring the activity of the Monkif/DlKhora botnet. This bot is an example of a Downloader trojan, in that its primary purpose is to receive instructions to download and execute other malware. The trojan also attempts to disable anti-virus and personal firewall software to maintain its foothold on the system. One interesting technique the Monkif botnet utilizes to hide its intent on the network is to encode the instructions to appear as if the command and control server is returning a JPEG file. The server sets the HTTP Content-Type header to “image/jpeg” and prefaces the bot commands with a fake 32-byte JPEG header. The bot checks if the header matches and decodes the rest of the response to retrieve its commands. The commands are encoded using a single byte XOR with 0×4. The malware that CTU has observed being installed by Monkif is a BHO (Browser Helper Object) trojan commonly referred to as ExeDot, which performs Ad Hijacking and Ad Clicking. The botnet makes no attempt to pad the commands to make the data size representative of a true JPEG. In addition, the data will not parse to a legitimate JPEG. These attributes may provide opportunities for generic countermeasures to detect the traffic by identifying malformed image data..."

 

[2,000 legit sites serve malware]

FYI...

2,000 legit sites serve malware
- http://www.theregister.co.uk/2009/10/16/gumblar_mass_web_compromise/
16 October 2009 - "Cybercriminals have laced about 2,000 legitimate websites with a potent malware cocktail that surreptitiously attacks people who browse to them, a security researcher warned Friday. Unlike past outbreaks of the mass web attack known as Gumblar, this round actually plants exploit code on the website servers themselves. Curiously, the directory and file name of the malicious payload is in most cases unique and identical to a legitimate file that existed on the website. The trick makes it extremely difficult for webmasters and anti-malware programs to detect the threats. "This is an ugly can of worms," said Mary Landesman, the ScanSafe security researcher who warned of the mass attack*. "Any time you see a new technique evolve like this the concern is we'll be seeing much more of this in the future, and certainly it complicates the remediation of the compromised website." Previously, Gumblar planted links in thousands of compromised websites that silently redirected users to a handful of servers that hosted the exploits. That method allowed white hats to foil the attack by shutting down one or two domains. With the malware embedded directly in the compromised websites, the take-down process is significantly more time consuming. Also making matters hard for Landesman to get the sites cleaned up: Most of the websites belong to small businesses that cater to non-English speakers..."
Gumblar Website Botnet Awakes
* http://blog.scansafe.com/journal/2009/10/15/gumblar-website-botnet-awakes.html

** http://www.virustotal.com/analisis/7c9d0a66a44beb3f6713ae731708fcd041fe1d46b84eaae44caf7f0866f905d1-1255452285
File HiwA7.dat received on 2009.10.13 16:44:45 (UTC)
Result: 7/41 (17.07%)

Zeus Bot Joins Gumblar Attacks
- http://blog.scansafe.com/journal/2009/10/20/zeus-bot-joins-gumblar-attacks.html
October 20, 2009 - "... unlike traditional compromises which simply inject pointers to malware hosted on an attacker-owned domain, in these attacks the compromised domain is also acting as host for the malware itself. This method of attack complicates remediation via technologies that rely on blacklisting because the number of compromised websites (now acting as malware hosts) is in the thousands. It also makes the Gumblar compromised websites a triple threat - potentially exposing visitors to the malware contained on the compromised site, and the malware loaded from ncenterpanel.cn*, and the malware loaded from other compromised sites."
* http://google.com/safebrowsing/diagnostic?site=ncenterpanel.cn/
"...the last time suspicious content was found on this site was on 2009-10-28. Malicious software includes 1209 trojan(s), 876 scripting exploit(s)... this site has hosted malicious software over the past 90 days. It infected 487 domain(s)..."

 

[Zbot botnet - new phishing attacks]

FYI...

Zbot botnet - new phishing attacks
- http://www.darkreading.com/shared/printableArticle.jhtml?articleID=220700200
Oct. 19, 2009 - "The massive Zbot botnet that spreads the treacherous Zeus banking Trojan has been launching a wave of relatively convincing phishing attacks during the past few days - the most recent of which is a phony warning of a mass Conficker infection from Microsoft that comes with a free "cleanup tool." The wave of attacks began early last week targeting corporations in the form of email messages that alerted victims of a "system upgrade." Email is accompanied by poisoned attachments and links; in some cases it poses as a message from victims' IT departments, including their actual email domains, and alerts them about a "security upgrade" to their email accounts. The message then refers victims to a link to reset their mailbox accounts, and the link takes them to a site that looks a lot like an Outlook Web Access (OWA) page (PDF), but instead infects them with the Zeus Trojan. Today, researchers at F-Secure spotted the botnet spamming out malware-laden email that tries to trick recipients with a convincing lure messages that says, "On October 22, 2009 server upgrade will take place"... The Shadowserver Foundation has seen multiple versions of Zeus-related attacks lately, including the Conficker "cleanup utility" that poses as an email from Microsoft, according to Andre DiMino, director of Shadowserver. And the targeted Outlook attacks use real domains: "What is also interesting about the recent campaign is that the email comes from the targeted user's own domain with an 'administrator' prefix. The link is disguised to look like it's from an update server on the local domain, but instead points to the malicious location"..."

- http://atlas.arbor.net/
"... We are also seeing email spam attacks to spread malware from the Bredolab botnet, from the ZBot botnet, and a Rogue AV downloader purporting to be an anti-conficker system update. "

 

[Bredolab trojan - botnet targets Facebook users]

FYI...

Bredolab trojan - botnet targets Facebook users
- http://www.computerworld.com/s/article/9140058/Massive_bot_attack_spoofs_Facebook_password_messages?taxonomyId=16
October 28, 2009 - "A massive bot-based attack has been hitting Facebook users, with nearly three-quarters of a million users receiving fake password reset messages (email SPAM), according to security researchers. The attack, which began Monday afternoon, according to e-mail security vendor Cloudmark*, targets Facebook users with a spoofed message that claims recipients' Facebook passwords have been reset as a security measure. The messages, which come bearing subject lines such as "Facebook Password Reset Confirmation," include a file attachment that supposedly contains the new password. In fact, the attached .zip file includes a Trojan downloader, dubbed "Bredlab" by some antivirus companies, "Bredolab" by others... At least 8% of the users who have received one of the fake messages have tagged it as legitimate, going to the trouble of pulling the message from their junk folder - where Cloudmark has placed it - because they think it's real... Cloudmark has no data on how many users were actually duped into opening the .zip file and running the enclosed .exe that installs Bredolab..."
* http://news.cnet.com/8301-27080_3-10385498-245.html

> http://boards.cexx.org/index.php?msg=79740;topic=17533.180

 

[Conficker infects 7M computers]

FYI...

Conficker infects 7M computers
- http://www.computerworld.com/s/article/9140171/After_one_year_Conficker_infects_7M_computers?
October 30, 2009 - "The Conficker worm has passed a dubious milestone. It has now infected more than 7 million computers, security experts estimate. On Thursday, researchers at the volunteer-run Shadowserver Foundation logged computers from more than 7 million unique IP addresses*, all infected by the known variants of Conficker. They have been able to keep track of Conficker infections by cracking the algorithm the worm uses to look for instructions on the Internet and placing their own "sinkhole" servers on the Internet domains it is programmed to visit. Conficker has several ways of receiving instructions, so the bad guys have still been able to control PCs, but the sinkhole servers give researchers a good idea how many machines are infected..."
* http://www.confickerworkinggroup.org/wiki/pmwiki.php/ANY/InfectionTracking

Conficker Eye Chart
- http://www.confickerworkinggroup.org/infection_test/cfeyechart.html

 

[Gumblar attacks spread to thousands of new sites]

FYI...

Gumblar attacks spread to thousands of new sites
- http://threatpost.com/en_us/blogs/gumblar-attacks-spread-thousands-new-sites-103009
October 30, 2009 - "Gumblar, the nasty bit of malware that was part of a mass SQL injection on legitimate Web sites this spring, is continuing to spread and its creators have been busy lately, compromising hundreds of new sites, leading to a massive new wave of infections of end-user PCs... In Gumblar's case, the iFrame redirection is the tactic of choice and it has been quite effective. In its original form Gumblar was redirecting victims to one of two remote sites, Gumblar .cn or Martuz .cn. The latest incarnation is pointing victims to thousands of servers in more than 200 countries that are now spreading Gumblar, according to research by Michael Molsner of Kaspersky Lab*. More than 7,200 servers spreading Gumblar are in the U.S., and many of the sites compromised around the globe are in the .gov and .edu domains. "Our accumulated data for one week showed 443748 access hits in total - and that is only a part of the whole incident. For several days after we noticed this new threat and added detection of the malicious files targeting Adobe Reader and Flash Player, there was surprisingly little talk about it in IT security circles. The 'new gumblar' took some time to get noticed more widely and _still_ seems unnoticed by many. However, it is very active indeed and as a side effect several PC vendors support lines have been flooded with queries about sudden reboots etc. There are also reports that machines infected with a buggy version of gumblar fail to boot completely, leaving the screen black and only the mouse pointer visible." Experts say that many of the machines that have been infected with Gumblar and other similar pieces of malware often are re-infected once they've been cleaned as users don't realize that their browsers are vulnerable and that the seemingly safe sites they're visiting are in fact serving malware."
* http://www.viruslist.com/en/weblog?weblogid=208187886
October 30, 2009

- http://google.com/safebrowsing/diagnostic?site=gumblar.cn/
"... last time Google visited this site was on 2009-11-01, and the last time suspicious content was found on this site was on 2009-11-01... It infected 6073 domain(s)..."
- http://google.com/safebrowsing/diagnostic?site=martuz.cn/
"... last time Google visited this site was on 2009-11-01, and the last time suspicious content was found on this site was on 2009-11-01... It infected 8328 domain(s)..."

 

[Fast Flux by the Numbers - Q3 2009]

FYI...

Fast Flux by the Numbers - Q3 2009
- http://asert.arbornetworks.com/2009/11/q3-2009-fast-flux-by-the-numbers/
November 2, 2009 - "... This year’s seen a huge uptick in Avalanche domains**, and the release of notes from ICANN on the Fast Flux Working Group* as well as a specific note around Avalanche. Arbor, like a few others, has been actively working with registries to address fast flux... Comparing to Q2... the biggest gainers are .tk and .eu, with .uk coming in as a new top 10 player. We’ve been trying to work with .eu as they are being targeted, along with .uk, by the Avalanche guys. However, our efforts in .eu are largely fruitless while Nominet in the UK has defended .uk quite handily. The .tk stuff we’re looking at, as it could be a false positive due to the way that .tk hosts stuff... Across all domain names, in Q3 we saw more TLDs hit, some 34 (against Q2’s 26 distinct TLDs). The attackers are striking at more TLDs in hopes of finding the soft spots, ones that just don’t respond. The average lifetime of a fast flux domain name: 418063 seconds, or about 9.7 days. CN domains are taken down within 7.8 days, EU domain names within 1.6 days, COM domains within 7.23 days, and TK domains within 1.44 days... Average lifetime of all domains in Q2: 21 days. Three weeks! That’s success now that we’re down to under 10 days. A cursory examination of this data suggests that while numbers are up, response times are getting better. This may be something worth cheering. Also, it appears that fast flux is still being used for the same old stuff: phishing, malware, malvertising, child pron, and the like."

(Charts available at the URL above.)

- http://atlas.arbor.net/summary/fastflux

Fast Flux Working Group
* https://st.icann.org/pdp-wg-ff/index.cgi?fast_flux_pdp_wg

** http://threatpost.com/en_us/blogs/avalanche-crimeware-kit-fuels-phishing-attacks-102309

 

[Gumblar authors crash WordPress sites]

FYI...

Gumblar authors crash WordPress sites
- http://www.networkworld.com/news/2009/110409-botnet-authors-crash-wordpress-sites.html?hpg1=bn
11/04/2009 - "Webmasters who find an annoying error message on their sites may have caught a big break, thanks to a slip-up by the authors of the Gumblar botnet. Tens of thousands of Web sites, many of them small sites running the WordPress blogging software, have been broken, returning a "fatal error" message in recent weeks. According to security experts those messages are actually generated by some buggy malicious code sneaked onto them by Gumblar's authors... Gumblar's authors apparently made some changes to their Web code... and as a result "the current version of Gumbar effectively breaks WordPress blogs"*... WordPress sites that have crashed because of the buggy code display the following error message: Fatal error: Cannot redeclare xfm() (previously declared in /path/to/site/index.php(1) : eval()'d code:1)
in /path/to/site/wp-config.php(1) : eval()'d code on line 1
Other sites running software such as Joomla get different fatal-error messages... In effect, the messages warn Gumblar's victims that they've been compromised..."    
* http://blog.unmaskparasites.com/2009/11/04/gumblar-breaks-wordpress-blogs-and-other-complex-php-sites/
04 Nov 09

WordPress Exploit Scanner
- http://wordpress.org/extend/plugins/exploit-scanner/
• Version: 0.6
• Last Updated: 2009-11-4
• Requires WordPress Version: 2.7.1 or higher
• Compatible up to: 2.8.5

 

[iframe injection]

FYI...

- http://blog.scansafe.com/journal/2009/11/5/gumblarcn-its-baaaack.html
November 5, 2009 - "... some of the compromises were following a different pattern than we'd been seeing over the past couple of weeks. Further investigation revealed the newest iframe injection was pointing once again to gumblar .cn - the malware domain that originally earned Gumblar its name. The domain's reactivation occurred less than 24 hours ago, but it has ramifications that could stretch back for months. Any sites compromised in the May Gumblar attacks that were not yet cleaned up (unfortunately an all-to-common occurrence) could now start becoming vectors of Gumblar infection once again. This is in addition to new compromises pointing to the newly activated gumblar .cn and the already very active Gumblar compromises which are using compromised websites as malware hosts*...
Edited to add: This is not the first example of registrars releasing malware domain names back into use..."
* http://blog.scansafe.com/journal/2009/10/15/gumblar-website-botnet-awakes.html
October 15, 2009

- http://www.iss.net/threats/gumblar.html

- http://google.com/safebrowsing/diagnostic?site=gumblar.cn/
"... last time Google visited this site was on 2009-11-06, and the last time suspicious content was found on this site was on 2009-11-06... It infected 5918 domain(s)..."
- http://google.com/safebrowsing/diagnostic?site=martuz.cn/
"... last time Google visited this site was on 2009-11-06, and the last time suspicious content was found on this site was on 2009-11-06... It infected 8558 domain(s)..."

- http://www.sophos.com/blogs/sophoslabs/v/post/7342
November 8, 2009

 

[The Gumblar system]

FYI...

The Gumblar system
- http://www.viruslist.com/en/weblog?weblogid=208187897
November 11, 2009 - "... Analysis of some infected websites showed that the only way to inject the infection of Gumblar was by using FTP access, because those websites have no server-side scripting. Later this was proved by an analysis of FTP log files... it's a fully automated system. It's a new generation of self-building botnets. This system is actively attacking visitors of a website and once these visitors have been infected with the Windows executable, it grabs FTP credentials from the victim machines. The FTP accounts are then used to infect every webpage on new webservers. This way the system extends the number of infected pages, thus attacking more and more computers. The entire process is automated and the owner of the system just needs to adjust the system and update the Trojan executable which steals passwords and the exploits used to attack the browser. The system works in a constant loop of attacking new computers, getting new FTP accounts and infecting new servers..."

(Screenshots available at the URL above.)

 

[Gumblar infection count]

FYI...

Gumblar infection count
- http://www.viruslist.com/en/weblog?weblogid=208187923
December 04, 2009 - "We've now analyzed more than 600 MB of collected data related to the recent resurrection of the Gumblar threat. Overall, we've identified 2000+ Infectors (computers hosting the malicious *.php files and payload) and 76100+ 'Redirectors' (computers with links leading back to the malicious sites). Most Infectors are also part of the group of Redirectors, they serve one *.php file and additionally contain the link to another Infector in their own entry page..."

 

[Max Power - many malware domains]

FYI...

Max Power - many malware domains
- http://isc.sans.org/diary.html?storyid=7693
Last Updated: 2009-12-04 19:46:31 UTC - "Who Max Power is? Well, we don't know either. It's a pseudonym of a gang or guy who has a decent-sized spyware racket going. Max has been sitting on the same IP address for the past three months, 210.51.166.119, in AS9929. ChinaNet. Even Google knows that 10% of the sites in this AS are malicious. Looking at the IP address in Reverse DNS or MalwareURL.com, we can see the many malware domains "Max Power" has been using in the recent past. Some of the names are associated with the Koobface and Zeus malware families. The address lay dormant for the last week of November, but just woke up again yesterday morning, and is currently serving the malware domain "tempa3-dot-cn". This domain is at the moment linked to from various questionable "pharmaceuticals" web sites, and it currently pushes a bunch of exploits which, if successful, download and run a backdoor of the "TDSS"/"Tidserv" family. Detection was dismal at first*, but has improved a bit over the last 24 hours**."
* http://www.virustotal.com/en/analisis/f7152a760e6182ad4881c5f77d50df1aa1221b5fa0482f5678ddb3101d2c643f-1259872180
File load.exe received on 2009.12.03 20:29:40 (UTC)
Result: 6/40 (15.00%)
** http://www.virustotal.com/en/analisis/f7152a760e6182ad4881c5f77d50df1aa1221b5fa0482f5678ddb3101d2c643f-1259949728
File load.exe received on 2009.12.04 18:02:08 (UTC)
Result: 18/41 (43.90%)

 

[Zeus bot using Amazon as C&C server]

FYI...

Zeus bot using Amazon as C&C server
- http://www.theregister.co.uk/2009/12/09/amazon_ec2_bot_control_channel/
9 December 2009 - "... a new variant of the Zeus banking trojan has been spotted using the popular Amazon service as a command and control channel for infected machines. After marks get tricked into installing the password-logging malware, their machines began reporting to EC2 for new instructions and updates, according to researchers from CA's internet security business unit*... Over the past few months, accounts on Twitter, Google's app engine, and Facebook have also been transformed into master control channels for machines under the spell of surreptitious malware... According to analysis** from Zero Day blogger Dancho Danchev, the cybercriminals behind Zeus appear to have plugged into Amazon's Relational Database Service as a backend alternative in case they lose access to their original domain..."
* http://community.ca.com/blogs/securityadvisor/archive/2009/12/09/zeus-in-the-cloud.aspx

** http://blogs.zdnet.com/security/?p=5110

- http://sunbeltblog.blogspot.com/2009/12/is-botnet-c-and-c-headed-for-cloud.html
December 10, 2009