The passwords look like this:

    pbkdf2_sha256$12000$zhMKabMgayvK$iniviUCcX9y2PYJcm0AoB3MhybRA1z2Cec1DZnLWxWc=

I do not know how much time it would take to bruteforce these. Can any experienced HNers weigh in?

Looks to me like the $ signs are acting as delimiters, and that that hash is made up of:

cipher/hash: pbkdf2_sha256

cost factor: 12000

salt: zhMKabMgayvK

hash: iniviUCcX9y2PYJcm0AoB3MhybRA1z2Cec1DZnLWxWc=

This exact technique (pretty much) is described here: http://exyr.org/2011/hashing-passwords/. It's a decent, secure way to hash passwords.

Cost factor of 12000 seems solid to me (depends on the hardware they're running on but I'd say brute forcing your way through that would be pretty impossible)

Let's do the math:

A PBKDF2 cost factor/iteration count of 12000 and 32-byte output means each candidate passphrase costs 12002 SHA256 blocks.

I can buy a crappy bitcoin miner which will do 2GH/s for about USD19.

Let's say we're going to use the Gawker leak as our dictionary. That's ~200,000 candidate passwords.

For a given user, I can therefore find their password (if it exists in the Gawker set) in 12002 * 200000 = 2.4GH SHA256 applications. That will take 1.2 seconds.

So for all 125 million eBay users, that's about 4 years. This work is trivially parallelisable, so buying more or faster hardware is brutally effective.

Note: there is obviously, and hopefully, a non-negligible probability that a user's password isn't in that set. Brute force of (say) the whole 8 printable-ASCII character password space would take longer but would be guaranteed to find to find about 50% (from Adobe leak) of user's passwords.

Bitcoin miners can not be re-purposed. They take a header input and output only nonces, not hash results.

I think the point is that if there's a bitcoin miner doing 2GH/sec for $19, the operation is cheap.

No, the cost is due to scale. Producing ASIC chips involves millions of dollars of setup.

    <algorithm>$<cost>$<salt>/<encrypted_password>

It's more than I expected from eBay, I think because of the likes of LinkedIn and their user credential leak, I expected bad habits from larger companies.

Rainbow tables can't be used against the passwords, so each one will need to be computed individually to either find the result or a collision. That's likely why the seller is only asking for $1000.

This is actually about as good as it gets for password hashes, so kudos to eBay.

Since these are salted and require 12000 iterations, cracking individual passwords will be quite time consuming. The preferred method in this case, though, is to go after low hanging fruit.

The way one would do this is to try something like the 500 most common passwords against all entries in the table. This won't take very long (compared to trying to brute force a bunch of individual passwords), and will probably yield a ton of passwords.

I am not an expert by any means, but I believe pbkdf2 is a recommended key stretching function for a hashing method (which looks to be sha256). http://en.wikipedia.org/wiki/PBKDF2

I think cracking difficulty depends on how many "iterations" they use though.

I assume 12000 iterations by the looks of that $ delimited string, I assume they're storing the parameters as,

hash_func$iterations$salt$derived_key

The iterations are listed as 12000. This value is supposed to double every two years and is around 128k I believe currently.

This is probably to try to offset Moore's law, by keeping the hash cracking difficulty in line with technology progrss. But it's funny how this works. If you think about Moore's law, it's basically describing the number of transistors on an IC, those doubling every two years. But it doesn't address expansion in the ways we use our technology. If new machines come out which allow us to stack even more GPUs into a single machine, performance capacity per cracking host will rise even farther than double per year.

One person estimated an 8-GPU cracking machine two years ago at about 539 billion hashes per minute. At 128k hashes for one password, you could make about 70,182 attempts per second.

But here[1] is a five-machine cluster from a year and a half ago with 25 GPUs. Its speed? 63 billion per second against SHA1. This results in 492,187 attempts per second. Assuming SHA256 is about 50% slower, this would be around 246,093 per second.

Some password dictionaries contain millions of words. But if your password is '0Password', it'll probably be cracked in a couple of seconds on modern hardware.

[1] http://arstechnica.com/security/2012/12/25-gpu-cluster-crack...

it's going to be computationally difficult but not impossible to break an individual password, but breaking all of them is unlikely in our (solar system's) lifetime.

However, if you were targeting a specific user and they didn't use a particularly strong password, it's possible that you could brute force it.

There aren't any plain text passwords, but there are email addresses, physical addresses, phone numbers, and DOBs.

It's the DoB which is the giveaway (that this is fake). eBay doesn't collect date of birth, but does collect lots of other authentication information.

EDIT: I stand corrected, see the reply by sp332 below.

eBay says the data includes DoB. https://blog.ebay.com/ebay-inc-ask-ebay-users-change-passwor... [edit: better link] Maybe they only collect it for certain customers?

In that case ebay users should probably change their date of birth too ;-).

And physical address and phone number.

Fake your own death and run away to Argentina.

My guess would be this is less about passwords (which - wild guess here - are probably fairly well-secured) and more about ancillary user meta data.

I guess I'm going to have to change my birthdate.

From the linked sample data, passwords are pbkdf2-sha256 while name/address/phone/birth date are clear text.

properly salted hashs can still harm many peple. You could still run a brute force on the top 10k most common passwords and with ~150m averge computer users, you're going to get a few. And you have their email address. There's a good chance they're using the same password for email. Now you have their email.

With user-individual salts and assuming 100ms for a password trial, 100ms * 150e6 * 10e3 is approximately 4750 years, and while this can be done in parallel, it's not exactly a small amount of resources you'd have to devote to get these relatively simple passwords.

Yeah, that's true. Due respect, but I'm not really worried about those other users, only me. Because I'm a human and I'm selfish.

Reports indicated that the passwords were encrypted, but I do not believe they've mentioned what the encryption scheme is.

Plaintext uniquely identifying information like Date of Birth was included, however.

Judging by the prefix to one of the link pasted hashes, looks like PBKDF2 with sha256?

I've had an eBay account for 10+ years. I thought a "hey, we screwed up, please change your password" type email would be nice.

They did force a password reset on everyone trying to log in yesterday; even if they already reset their password when they first heard the news, effectively locking them out for the evening because their password reset system was understandably overwhelmed with requests. Thanks eBay!

eBay didn't say what kind of encryption they had, only that they were encrypted. Personally I'm more concerned about the personal information in the leaked data that wasn't protected at all! So sure, change your password, but it won't change the fact that someone got your name, email address, physical address, phone number, and date of birth.

On the plus side, think of how nice it will be to get a happy birthday phone call from a hacker.

sp332 to be honest here, I am pretty sure if you search your real life name on the internet you will definitely find somewhere that you can buy the information you just mentioned. I am not trying to be mean here, I am just trying to let you know in case you didn't know.

Plus you already have your email publicly displayed here ... and i found some weird stuff about amateur ... xxx movies when i googled you T_T

He took down his reply, but he shares a name with a fairly famous actor. Gonna take a stab that the "amateur ... xxx" was related to that guy.

Also, you're an idiot.

It's not just the actor, it's a pretty common name. People mistype my email address into all kinds of things, judging by the random emails I get for pets in Texas and real estate in the UK etc.

As with any other hacked company, assume the worst: everything was leaked as plaintext.