LinkedIn investigating reports that 6.46 million hashed passwords have leaked online (update)

Macbook keyboard macro

UPDATE: LinkedIn confirms hacking. Read more here.

A user in a Russian forum is claiming to have hacked LinkedIn to the tune of almost 6.5 million account details. The user uploaded 6,458,020 hashed passwords, but no usernames. It's not clear if they managed to download the usernames, but it's likely that both have been downloaded.There is a possibility that this could be a hoax, but several people have said on Twitter that they found their real LinkedIn passwords as hashes on the list. Many of the hashes include "linkedin," which seems to add credence to the claims.

We spoke with Mikko Hypponen, Chief Research Officer at F-Secure, who thinks this is "a real collection." He told us he is "guessing it's some sort of exploit on their web interface, but there's no way to know. I am sure sure LinkedIn will fill us in sooner or later."

It's worth noting that the passwords are stored as unsalted SHA-1 hashes. SHA-1 is a secure algorithm, but is not foolproof. LinkedIn could have made the passwords more secure by 'salting' the hashes, which involves merging the hashed password with another combination and then hashing for a second time. Even so, unless your password is a dictionary word, or very simple, it will take some time to crack. We've reached out to LinkedIn to determine the accuracy of the claims, but in the meantime, we recommend changing your password just in case.

Update: LinkedIn has just tweeted that it is looking into the matter.

Update 2: LinkedIn has tweeted again, and has been unable to confirm any security breach yet. Given the growing number of users that have found their password in the hashes, that's worrying news.

Update 3: Security researcher Steve Gibson has highlighted a website which will check if your password can be found on the list of stolen hashes. Bear in mind if you have a common password a positive result may not mean that your account has been compromised.

Thanks, Sponplat!

Recommended by Outbrain

Comments

That’s annoying. I never change my passwords because I always forget them, but I might have to make an exception.

If you cannot remember your password use this technique:

Which of the following two passwords is stronger,
more secure, and more difficult to crack?

D0g…………………

PrXyc.N(n4k77#L!eVdAfp9

You probably know this is a trick question, but the answer is: Despite the fact that the first password is HUGELY easier to use and more memorable, it is also the stronger of the two! In fact, since it is one character longer and contains uppercase, lowercase, a number and special characters, that first password would take an attacker approximately 95 times longer to find by searching than the second impossible-to-remember-or-type password!

Source: https://www.grc.com/haystack.htm

I love it when an XKCD is actually relevant.

they got their math wrong but indeed that is the idea :)

If you’re saying that Randall Schwartz got his math wrong in that XKCD, you’re calculating the probability wrong. Schwartz is assuming that there is a 2048-word vocabulary, and that the adversary knows your vocabulary and that your password is a random combination of words from that vocabulary (in other words, he’s applying Kerckhoffs’ principle correctly – the attacker knows the algorithm and the namespace of the key, but not the value of the key). That gives you 2048^4 or (2^11)^4 or 2^44. That “D0g…………………” password? The Kolmogorov complexity of that is equivalent to a 5-byte password; once the attacker knows to try dictionary words with vowels replaced by numerics and with repeated periods at the end, the “strength” of that password becomes much, much lower.

Randall Munroe, FYI

In related news: “correct horse battery staple” is now the 3rd most common password online. :)

The SHA-1 hash for “correcthorsebatterystaple”, bfd3617727eab0e800e62a776c76381defbc4145 does actually appear in the LinkedIn password leak.

It actually says it wasn’t leaked.

Thats not what the website used to say, but in any event, “CorrectHorseBatteryStaple” does appear as well.

Good examples. We should all be aware that passwords should not have personal relevance but do not need to be random. More often than not, passwords are cracked through brute-force and dictionary attacks. Thus, we need to deter the computer, not a human. If anything, your second example just confuses both. Like I said, good examples.

You need to read further down on the page.
“The #1 most commonly used password is "123456", and the 4th most common is "Password." So any password attacker and cracker would try those two passwords immediately. Yet the Search Space Calculator above shows the time to search for those two passwords online (assuming a very fast online rate of 1,000 guesses per second) as 18.52 minutes and 17.33 centuries respectively! If "123456" is the first password that’s guessed, that wouldn’t take 18.52 minutes. And no password cracker would wait 17.33 centuries before checking to see whether "Password" is the magic phrase.”

Not correct. Any brute force password cracker software I’ve seen tries plaintexts in a Markov order rather than alphabetically, so frequently occurring parts like a bunch of dots in a row would fail pretty quickly.

It would still go through all passwords at min char length, then min+1, min+2, etc so the extra security comes from the length. It might fail quicker than the second example but it’s still secure enough that it would take some order of years to brute force.

But then wouldn’t adding a ‘4’ to the end of the second password make it much stronger than the first?

From what I understand, but the question should really be what the cutoff is to determine whether or not a password is “secure enough.”

Never, ever use GRC as a source for security. They’re usually wrong, or very wrong, as you can see by the other replies to your post.

I’d love to see some evidence that GRC is “usually wrong”. Furthermore, I don’t see any replies that prove anything.

There were several, including even the amusing XKCD with statistical math applied.

XKCD and GRC were in concurrence.

Great tip. I’ve changed my password to the first example, so everything should be secure now.

It is NOT a "Password Strength Meter."

I wonder if there is a hybrid password strength meter that takes this into account

You need to use a password manager – 1password, lastpass, something like this. You’d still need to remember a password but it’ll only be one. And the rest you can make as secure as you want, e.g. 24 characters, upper and lower case, digits and symbols – nobody’s cracking that anytime soon.

As for the one password, try this – a nonsensical but memorable phrase made out of 3-4 words, each word capitalized and the spaces between replaced with digits. It’s easier to remember and while may not be optimal, it’s still a hell of a lot harder to crack than random 8 characters with a couple of digits.

View All Comments
Back to top ↑