Risk
7/9/2008
06:59 PM
Connect Directly
LinkedIn
Twitter
Google+
RSS
E-Mail
50%
50%

Security Community Cooperates To Fix 'DNS Poisoning'

Armed with knowledge of DNS transaction IDs, an attacker could reroute requests for certain Web sites to Web sites of his or her choosing or hijack e-mail.

US-CERT, the government's cyber security arm, on Tuesday warned of a serious weakness in the Domain Name System (DNS) protocol that could be used to send Internet users to malicious sites.

In an unusual move that reflects the seriousness of the security flaw, news of the vulnerability was delayed for months to allow software vendors like Microsoft, Cisco, and Sun -- more than 80 are affected -- to release coordinated fixes.

As part of its monthly patch schedule, Microsoft on Tuesday fixed the DNS flaw in Windows with patch MS08-037, "Vulnerabilities in DNS Could Allow Spoofing."

Amol Sarwate, manager of vulnerability labs at Qualys, considers the vulnerability to be significant. "I wouldn't characterize it as an end of the world scenario but it is a very important vulnerability," he said.

While details about the vulnerability and how to exploit it have been deliberately withheld, Sarwate said that the issue appears to be that the transaction ID generated in a DNS request -- querying a DNS server to link an IP address with an Internet domain name -- is insufficiently random to avoid being guessed by a knowledgeable attacker.

Armed with knowledge of DNS transaction IDs, an attacker could reroute requests for certain Web sites to Web sites of his or her choosing or hijack e-mail. The technique is called DNS poisoning.

US-CERT says that "per-query source port randomization" can mitigate the risk posed by the vulnerability.

"The effect of the vulnerability, if it was ever exploited, is enormous," explained Jeff Kalwerisky, chief security evangelist of Alpha Software. "Someone who exploited the vulnerability would be able to reroute every single transaction in your computer."

"The good news is that for the average user on the desktop, if you've got automated patching on Windows, you've got the patch already," said Kalwerisky. "The hard part is a large corporation with multiple DNS servers and routers and switchers."

Kalwerisky advises patching as soon as possible.

Dan Kaminsky of IO Active, the researcher who discovered the flaw, has posted an online vulnerability checker at doxpara.com. And if checking one's site can be done so easily, expect cyber criminals to scout the Net for vulnerable machines just as soon as they figure out how the flaw can be exploited.

Kalwerisky said that the way the vulnerability has been handled shows the security community can work together effectively. "Dan has proved the utter value of independent secure researchers," he said "Security should not be locked up in Microsoft or other software houses."

Kaminsky is scheduled to disclose details about the vulnerability at the Black Hat Conference in Las Vegas on August 6th.

Comment  | 
Print  | 
More Insights
Register for Dark Reading Newsletters
White Papers
Cartoon
Current Issue
Flash Poll
Video
Slideshows
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2015-0714
Published: 2015-05-02
Multiple cross-site scripting (XSS) vulnerabilities in Cisco Finesse Server 10.0(1), 10.5(1), 10.6(1), and 11.0(1) allow remote attackers to inject arbitrary web script or HTML via unspecified parameters, aka Bug ID CSCut53595.

CVE-2014-3598
Published: 2015-05-01
The Jpeg2KImagePlugin plugin in Pillow before 2.5.3 allows remote attackers to cause a denial of service via a crafted image.

CVE-2014-8361
Published: 2015-05-01
The miniigd SOAP service in Realtek SDK allows remote attackers to execute arbitrary code via a crafted NewInternalClient request.

CVE-2015-0237
Published: 2015-05-01
Red Hat Enterprise Virtualization (RHEV) Manager before 3.5.1 ignores the permission to deny snapshot creation during live storage migration between domains, which allows remote authenticated users to cause a denial of service (prevent host start) by creating a long snapshot chain.

CVE-2015-0257
Published: 2015-05-01
Red Hat Enterprise Virtualization (RHEV) Manager before 3.5.1 uses weak permissions on the directories shared by the ovirt-engine-dwhd service and a plugin during service startup, which allows local users to obtain sensitive information by reading files in the directory.

Dark Reading Radio
Archived Dark Reading Radio
Join security and risk expert John Pironti and Dark Reading Editor-in-Chief Tim Wilson for a live online discussion of the sea-changing shift in security strategy and the many ways it is affecting IT and business.