LinkedIn passwords posted online following massive breach

LinkedIn has confirmed that “some” of its 6.5 million users’ passwords have been compromised following a massive data breach yesterday.

In a carefully worded blog post, LinkedIn director Vicente Silveira said the company has confirmed that an unspecified number of hashed passwords posted publicly on a Russian hacker forum earlier this week, “correspond to LinkedIn accounts.”

Silveira made no mention of how the passwords may have ended up on the forums but noted that LinkedIn is continuing to investigate.

“Members that have accounts associated with the compromised passwords will notice that their LinkedIn account password is no longer valid,” Silveria said.

Users of the social networking site for professionals will also receive an email from LinkedIn with instructions on how to reset their passwords. The email will not contain any links that users will need to click on to reset their password, he noted. Affected customers will also receive a note from LinkedIn with more information on what happened and why they are being asked to reset their passwords, Silveira said.

Earlier Silveira had posted a separate note urging LinkedIn members to change their passwords and providing them with tips on how to create strong passwords.

Silveira was responding to numerous reports earlier Wednesday that hackers accessed close to 6.5 million hashed passwords from a LinkedIn database and posted it publicly on a Russian hacker forum. According to security researchers who had seen the compromised data, more than 300,000 of the hashed passwords have already been decrypted and posted online in clear text.

LinkedIn had earlier said it was looking into those reports but had not confirmed the breach.

Tal Be’ery, security research leader at Imperva, claims to have seen the stolen data and said much more than 6.5 million passwords might have been compromised.

According to Be’ery, the passwords that have been posted online appear to be only those passwords that the hackers needed help in cracking. What the breached password list is missing are the usual easy-to-guess passwords that people commonly use to control access to online accounts, he said.

The LinkedIn password file does not contain any of the common passwords that Imperva’s researchers have typically run across when analysing similar password breaches, he said.

“Most likely, the hacker has figured out the easy passwords and needs help with less common ones.” So it’s likely that only the more complicated passwords have been revealed so far, he theorised.

The breached list shows that LinkedIn did not use best practices in protecting the passwords, he said. The hashes that were used to mask the real passwords were so-called unsalted SHA-1 hashes. SHA-1 is a hashing algorithm that is used to protect passwords. Because SHA-1 isn’t foolproof, security experts have for some time recommended that organisations use a technique called “salting” to make passwords harder to crack.

With salting, an application applies a random string of characters to a password before it is hashed. The process ensures that even if two passwords are identical, their hashes will be unique.

In an apparent response to the focus on the unsalted hashing issue, Silveira noted that LinkedIn recently added enhanced security measures for salting and hashing its password databases. Silveira’s post does not indicate when LinkedIn began the practice.

The compromise is a big deal for LinkedIn users, said John Pescatore, an analyst with Gartner. “LinkedIn definitely had to have some kind of serious security incident for this to happen. And they probably had lax security policies or controls for a simple unsalted hash file like this to exist,” he said.

One worrisome aspect of the breach is that it could enable more targeted phishing attacks, he said. “LinkedIn is a great research site for hackers creating targeted phishing attacks to go after system administrators, CFOs, etc.” he said. “If they had access to the non-public parts of people’s LinkedIn profiles we will see even better targeted phishing attacks.”