The fallout from the Twitter hack is still, well, falling out. Twitter admitted on Friday that 250,000 of its user accounts may have been hacked. Who's to blame? Java? Chinese hackers?
On the heels of The New York Times and The Wall Street Journal Chinese hacking revelations, Twitter said it also detected unusual access patterns that led it to identify unauthorized access attempts to Twitter user data. Twitter even shut down a live attack in progress.
"This attack was not the work of amateurs, and we do not believe it was an isolated incident," said Bob Lord, Twitter's director of Information Security, writing in a blog post. "The attackers were extremely sophisticated, and we believe other companies and organizations have also been recently similarly attacked."
Is Java to Blame?
Paul Henry, a and forensic analyst at Lumension, said the Twitter hack was an odd one.
"There's not much information about it yet, so we don't know exactly what information the attackers got, but we do know that a whole bunch of people had to change their passwords," Henry said. "While there's been speculation that this was an organized attack, we don't know for certain, since very little information about the attack has been disclosed."
Henry noted that there has also been speculation that Java might have played a role, but he doesn't believe that Java was a factor. Since Java is used to attack and compromise single users, he explained, it's unlikely that it would have been the vector for an attack compromising 250,000 users. Still, he said, the warnings about Java that have been circulating for the last month or so should be heeded.
"Java is a flawed component and is one of the biggest attack vectors out there right now. Unfortunately, there's not much you can do to get around it, as many sites require it to function properly. Oracle has yet to fix many of its underlying flaws and we probably won't see a truly secure version of Java for another year or two. By then, I hope that developers have moved away from Java. Otherwise, the problem with Java is only going to get worse."
Henry also echoed Twitter's advice to practice "good password hygiene." In addition to not using the same passwords across multiple sites and using a combination of letters -- capitalized and lower case -- numbers and symbols, he recommended that changing passwords regularly. A monthly change is ideal for best practices, he said, but changing your passwords quarterly at a minimum is a good idea."
Follow the Money
Chris Petersen, chief technology officer at LogRhythm, told us it should come as no surprise that network intrusion attempts are on the rise given hackers' continued success in monetizing their efforts that often times yield high returns.
"Social media platforms are not immune to the IT Security challenges facing small. medium and large companies each and every day. In essence, they face greater challenges attempting to protect not only the proprietary information on their users, but also mission-critical corporate info as well," Petersen said.
"Regardless of the level of hackers' sophistication, the best policy enterprises and blue-chip organizations can have in place includes continuous network monitoring. Detecting anomalous activity in its early stages is critical and often times the difference between fast containment and mitigation or longer-term, more severe consequences."
Posted: 2013-02-20 @ 1:37pm PT
"A monthly change is ideal for best practices, he said, but changing your passwords quarterly at a minimum is a good idea."
There are some contrary ideas circulating. One basic argument is that requiring overly frequent password changes, prompts users to write them down and paste them on their monitor or in some other obvious location. If you set up passwords to expire monthly and then give say two weeks notice that passwords are going to expire, you give people at most half a month without having to think about a password change.
A good read to a least start rethinking this can be found at: http://www.cerias.purdue.edu/site/blog/post/password-change-myths/