Researchers Find Same RSA Encryption Key Used 28,000 Times

itwbennett writes In the course of trying to find out how many servers and devices are still vulnerable to the Web security flaw known as FREAK, researchers at Royal Holloway of the University of London found something else of interest: Many hosts (either servers or other Internet-connected devices) share the same 512-bit public key. In one egregious example, 28,394 routers running a SSL VPN module all use the same 512-bit public RSA key.

Comments

I imagine ....

[PPH]

... some vendor built a router or server up to the point of generating the public/private key pair, tested it, saved the image and started copying it to production units.

Similar mistake have been made before.

Poor first sentence

[in10se]

First line of the article:
"What if the key to your house was shared with 28,000 other homes?"

The fact is, you very well might share the key to your house with more than 28000 other homes. Common lock brands you can buy at Home Depot, Lowe's, etc. create a surprisingly low number of different key/tumbler combinations.

Re:Poor first sentence

[93 Escort Wagon]

Cars, too.

Funny anecdote: Quite some years ago, my wife and I were over at another couple's house for our semi-regular game of Pinocle. After we called it an evening, I went out the door and accidentally got into their Ford Escort (at the time, they owned one that was a very similar color to ours - plus it was night). They stood there and laughed at me... and then I started their car with my key.

Re:Poor first sentence

[Obfuscant]

turns out my key worked on *every* upper back unit in each building.

That's just lazy on the part of your landlord. It's easier for him if all his units share the same lock so he has only one key to carry around.

But common house keys? Yes, relatively few "combinations". I'm looking at mine, bought from a big-box home outlet store. Five lands -- that's the flat areas where the pins rest when the key is inserted. I didn't count them when I rekeyed my locks, but it's about five pin lengths. Let's see, 5^5 is 3125 different keys. Six pin lengths would be only about 15,000 different sets.

My work keys have 6 or 7 lands, but the security of those is reduced because each pin has at least two valid lengths. There is actually a published method for taking a bunch of key blanks and a valid key and figuring out the master.

If you want to know how locks work, go buy a new lock for a house and the rekey kit for it. It's fun. While each kit is "different" (or is supposed to be), with a bit of looking you can find two kits with the same pin lengths just in a different order so you can rekey two locks the same. (The kits I bought had colors for the pins.)

For cars, I heard a long time ago that Toyotas were prime theft targets not because of the value but because there were a limited number of dealer master keys and the crooks had copies.

So easy to find

[kooky45]

Just scanned the /16 next to my home broadband and found a number of repeated certificate hashes and all belonging to systems identifying themselves as

*.myfoscam.org/organizationName=ShenZhen Foscam Intelligent Technology Co,Ltd

Seems to be a network enabled camera.

Re:Know what's worse? Cleartext.

[Anonymous Coward]

Like this: https://xkcd.com/538/

Re:Know what's worse? Cleartext.

[msauve]

I suspect his problem with it is that he confuses it with WEP.

Re:Know what's worse? Cleartext.

[chrysosphinx]

Weak, bad or fake encryption is infinitely much worse than none, because it makes people believe they are safe while they are not.

Re:Know what's worse? Cleartext.

[TechyImmigrant]

You are talking about breaking passwords, not the encryption scheme, which comes later.

Password -> PMK -> 4 way handshake (session key establishment) -> Authenticated encryption (link cipher).

A 12 character, alphanumeric + special character password, uniformly generated is about 70 bits of entropy. The pbkdf2 invocation to generate the PMK has 4096 iterations, causing the brute force attack to need to perform on average ~ 2^81 hashes before finding a password. This would not happen over lunch.

Did your friend's tool actually break WEP instead of WPA-2? Or did you have a weak password? Or were you using a weak EAP method? Or what other form of BS are you talking?