Welcome Guest. | Log In| Register | Membership Benefits
  • Email this page E-mail this page
  • |  Print Print this page
  • |   Bookmark and Share

New 'Drive-By' Attack Is Remote

Symantec and Indiana University researchers build proof-of-concept exploit preying on businesses and homes with broadband routers

Feb 15, 2007 | 12:50 PM

By Kelly Jackson Higgins
DarkReading

They're calling it "drive-by pharming." But unlike war driving, an attacker doesn't have to be anywhere in the vicinity to stage this newly discovered type of attack, which can give him/her access to personal data such as your bank account.

The invasion is a simple one. The attacker uses a broadband router vendor's default passwords -- some of which can be found with a Net search -- to take control of such routers, which are found all over small businesses, homes, and WiFi hotspots. But first, the victim must visit a Website infected with malicious JavaScript code, which uses the dreaded Cross Site Request Forgery (CSRF) technique in the browser to log into your local broadband router and change its settings to control where you go on the Web.

Millions of users worldwide could be vulnerable to this attack, says Zulfikar Ramzan, senior principal researcher for Symantec Security Response, who recently co-developed the proof-of-concept for this attack.

"Instead of a real bank address, it sends them to a fake bank Web page that looks real, or to [fake versions of] major brands like PayPal and eBay and it can see anything you're typing," Ramzan explains. "The attacker can control whatever server you go to... And you'd see the attacker's Web page, but never know the difference."

The POC attack simulates login access to routers with Web management interfaces, and it changes their DNS settings to a rogue DNS server that controls where the user goes, sending them to its infected sites instead of the intended ones. Then the attacker can grab any sensitive data the victim provides on those sites, such as credit card or bank account data. But the attack can only be executed via broadband routers and wireless access points with Web management interfaces, Ramzan says.

The good news is that it's easy to protect yourself from falling victim to this attack -- all it takes is changing the default password on your broadband router or wireless access point, something many users don't bother doing. Some research suggests that half of these routers use default passwords.

Ramzan, who worked with researchers at Indiana University on the POC, says the concept builds on previous research by WhiteHat Security CTO Jeremiah Grossman and researcher T.C. Niedzialkowskion using JavaScript for profiling and attacking an internal network via the Web. "It occurred to me that with the techniques he [Grossman] talked about... you could mount a war driving attack without the physical proximity," Ramzan says. "It makes the attack a lot more potent."

And once one machine on the WLAN or home LAN is exposed, they all are. "All the machines [linked to] that router can be owned," Ramzan says. "So if kids searching the Web are exposed, their parents are potentially susceptible" from their computers as well.

What makes the attack even more disturbing is it changes Domain Name Service settings. "Once you can mess with DNS, you're in trouble," Ramzan says. "It starts giving you the wrong information and it's hard to figure out what's going on at that point."

Ramzan, who plans to blog about the POC today, says he hasn't seen this type of attack in the wild, although there's no way to be sure because it's tough to detect. "One can obfuscate JavaScript," he says. "And it's so simple that it's really a matter of time before people start" executing this type of attack.

— Kelly Jackson Higgins, Senior Editor, Dark Reading


Subscribe to RSS










Bugs
ENTERPRISE VULNERABILITIES
Vulnerability:cxf
Published:2010-08-19
Severity:High
Description:Apache CXF 2.0.x before 2.0.13, 2.1.x before 2.1.10, and 2.2.x before 2.2.9, as used in Apache ServiceMix, Apache Camel, Apache Chemistry, Apache jUDDI, Apache Geronimo, and other products, does not properly reject DTDs in SOAP messages, which allows remote attackers to read arbitrary files, send HTTP requests to intranet servers, or cause a denial of service (CPU and memory consumption) via a crafted DTD, as demonstrated by an entity declaration in a request to samples/wsdl_first_pure_xml, a similar issue to CVE-2010-1632.
Vulnerability:libvirt
Published:2010-08-19
Severity:Medium
Description:Red Hat libvirt, possibly 0.6.1 through 0.8.2, looks up disk backing stores without referring to the user-defined main disk format, which might allow guest OS users to read arbitrary files on the host OS, and possibly have unspecified other impact, via unknown vectors.
Vulnerability:libvirt
Published:2010-08-19
Severity:Medium
Description:Red Hat libvirt, possibly 0.7.2 through 0.8.2, recurses into disk-image backing stores without extracting the defined disk backing-store format, which might allow guest OS users to read arbitrary files on the host OS, and possibly have unspecified other impact, via unknown vectors.
Vulnerability:libvirt
Published:2010-08-19
Severity:Medium
Description:Red Hat libvirt, possibly 0.6.0 through 0.8.2, creates new images without setting the user-defined backing-store format, which allows guest OS users to read arbitrary files on the host OS via unspecified vectors.
Vulnerability:libvirt
Published:2010-08-19
Severity:Low
Description:Red Hat libvirt 0.2.0 through 0.8.2 creates iptables rules with improper mappings of privileged source ports, which allows guest OS users to bypass intended access restrictions by leveraging IP address and source-port values, as demonstrated by copying and deleting an NFS directory tree.


Briefing Centers
POWERFUL INFORMATION
AT YOUR FINGERTIPS
(SPONSORED LINKS)