Dispatches from the front line of the internet

RSS

Twitter.com, NYTimes.com and more taken down in Syrian hack

The SEA strikes through DNS servers.

What happens if you visit NYTimes.com
What happens if you visit NYTimes.com

Hackers took down the New York Times, Twitter and Huffington Post websites overnight through a method known as DNS hijacking. Although the NYT's website is still down this morning, the rest appear to be back up, albeit with continued problems on some subsystems. The Syrian Electronic Army (SEA) hacking collective is obliquely claiming responsibility on Twitter.

The SEA is famous for finding novel entry-points into a company's online presence, and this is no different. Rather than hacking into the companies' servers directly, DNS hijacking allows an attacker to redirect the web address which normally points to the servers on which the site is stored.

Every server on the internet has a unique IP address, a 12-digit code which refers to its virtual location. But in order to avoid having to remember all these numbers, there's a second system which sits on top of IP addresses, which lets us type in the alphanumeric domain names we all know and love. When someone enters nytimes.com into their address bar, the browser looks up the domain name using a Domain Name System (DNS) server; that server then tells your browser what IP address the URL points to, the two computers connect, and everything works happily.

What happened overnight is that the SEA managed to break into the website of Melbourne IT, the company which the New York Times and others used to register those domain names. They then changed the records so that instead of pointing to the New York Times' website, the address pointed to theirs.

On the one hand, that's a lot less bad than it would be if the servers themselves were broken into. The New York Times continued to publish normally to their IP address, 170.149.168.130, and don't appear to have lost any data or sensitive information. On the other hand, the sites were still down, and the redirect still exposed users to potential security risks. For instance, it would be possible to build a passable version of a log-in page and steal a lot of passwords. When it comes to Twitter, one of the affected companies, the problems are even greater: the site has a lot of code embedded throughout the internet, in the form of tweet buttons and single-sign-in services. If the SEA had wanted, that could have been the beginning of a much more serious collection of hacks.

As it is, the group appears to have limited themselves to their normal operations, the digital equivalent of graffiti. Albeit graffiti in a very prominent place. But that it was so easy to take down the sites of such huge media organisations should give us all the shivers. The internet is a long way from secure, and some of the biggest problems left are fundamental to how the whole thing works.