How Twitter Dodged Attack That Took Down New York TimesJordan Robertson
Chalk one up for Twitter Inc.
While the New York Times and Google Inc. had visitors to their sites redirected this week by hackers, the microblogging service was better able to deflect attacks because of a simple tool called a registry lock. Like alerts sent to credit-card users when something bad happens, the feature notifies website managers of attempts by intruders to tamper with critical information, such as Web-address data.
The cost? As little as $50 a year.
Large banks, e-commerce companies, gambling sites and pornographers have used registry locks from VeriSign Inc. and NeuStar Inc. to prevent unauthorized changes. Attacks by the Syrian Electronic Army routed New York Times readers to a site that displayed the group’s initials and altered some registration data. They underscore how vulnerable many companies are to relatively unsophisticated attacks, which can take down sites and harm their businesses.
“This is certainly an ah-ha moment,” said Rodney Joffe, a senior technologist at NeuStar. The Sterling, Virginia-based company began offering registry locks in 2010 and requires that website domain information be accompanied by two layers of verification, such as additional codes from security tokens.
“It is a niche business but there’s no reason for it to be,” he said. “It’s the kind of thing you have to do today.”
While Twitter’s site operated normally, twitter.co.uk was inaccessible for some users. The Syrian Electronic Army, which backs the country’s president, Bashar al-Assad, claimed responsibility for the New York Times and Twitter intrusions, as well as the Washington Post this month and the Financial Times in early May. Unknown hackers altered Google’s website in the Palestinian territories, displaying a map without Israel.
The attacks exploited weaknesses in a registration network called the Domain Name System, exposing risks that site operators face because they’re relying on third parties to handle their online addresses. Weaknesses in DNS, which was created in the 1980s to help computers find websites using names instead of numbers, haven’t been seen as a significant threat outside of the financial-services and retail sectors up to now, according to John Pescatore, director of emerging-security trends at the SANS Institute in Stamford, Connecticut.
“There are still a lot of sloppy practices,” Pescatore said. “There’s a lot of room to raise the bar.”
Because Twitter, based in San Francisco, monitors its DNS information in real time and had implemented a registry lock, it was better prepared than the New York Times, according to HD Moore, chief research officer at Rapid7, a Boston-based security firm. Since the attacks, many other companies have moved to institute similar safeguards, he said.
Twitter has had its DNS records hacked before. The company acknowledged in 2009 that its DNS records were compromised by hackers who defaced the site with a message about Iran. Jim Prosser, a spokesman for Twitter, declined to comment on the company’s security measures.
A vast system that underpins how computers locate each other, DNS is often called the phone book of the Internet. In 2008, Dan Kaminsky, a security researcher, uncovered a flaw in the system that would let hackers easily impersonate legitimate sites. He worked with technology companies to fix it. The finding prompted several companies that process financial transactions online to adopt additional security measures to ensure their domain information is secure, while others stayed on the sidelines, according to SANS’s Pescatore.
NeuStar and VeriSign, another provider of registry lock services, declined to identify the companies using its registry lock services. Danny McPherson, chief security officer of VeriSign, said in a statement that the technology gives customers more control over who can change information.
Eileen Murphy, a spokeswoman for the New York Times Co., said the newspaper is looking at additional measures.
“In light of this attack and the apparent vulnerability even at what had been highly secure registrars, we are tightening all of our security,” she said.
Jay Nancarrow a spokesman for Google, declined to comment on the company’s security. The company’s Palestine site itself wasn’t hacked and Google is talking with the domain manager to resolve the issue, he said.
One complication of hosting sites with addresses of specific countries or regions is that many of the registration providers don’t use registry locks and other protective steps, said Paco Hope, a principal consultant with Cigital Inc.
“When you’re a company like the New York Times or Twitter or Google, your stock in trade is the Internet, it’s the service you offer, and that’s why it makes sense to put in a lot more security,” Hope said.
The rise in sophisticated hacking attacks is helping fuel a market for computer-security technology that is expected to exceed $65.7 billion this year, according to Gartner Inc.
Many companies that didn’t prioritize a threat involving their DNS records are now rethinking that approach, SANS’s Pescatore said.
“It’s one of several Achilles’ heels of using the Internet,” he said.