[Cryptography] encrypting bcrypt hashes

[Robin Wood]

On Wed, 8 Mar 2017 at 21:26 Ron Garret <ron at flownet.com> wrote:

> On Mar 8, 2017, at 8:22 AM, Robin Wood <robin at digininja.org> wrote:
>
> We talked about storage of the encryption key and they basically said that
> is out of scope, don't worry about it.
>
>
> Hoo boy.  Then it doesn’t much matter what you tell them because, as Ray
> Dillinger already pointed out, this is the limiting factor.
>
> Talking to a friend who knows crypto, he says that the encryption should
> not have a negative interaction with the bcrypt hash and so adding it will
> slow down the process of recovering a PIN.
>
>
> Not if your adversary steals the encryption key along with the hashed
> PINs.
>
> [2] 10 sec per hash works out at just over a day to brute force a 4 digit
> PIN if I've done my maths right.
>
>
> That’s about right, but earlier you wrote:
>
> The encryption is fast enough to not affect login times
>
>
> That made it sound like latency matters.  10 seconds is a pretty long time
> to make a user wait.
>

The dev team have been given a time, under 1 second from what I heard, to
get the processing done and a response ready, either log in or fail. I did
the maths on 10 seconds just to show that even with a high work factor, an
attacker could get through a single hash fairly quickly.

Robin


>
> rg
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20170308/c03f68ad/attachment.html>