| 1 |
On 14/10/2013 21:17, Michael Orlitzky wrote: |
| 2 |
> On 10/14/2013 02:49 PM, Martin Vaeth wrote: |
| 3 |
>> |
| 4 |
>>> Hiding the salt would just be security through obscurity. |
| 5 |
>> |
| 6 |
>> And yet it is stupid if you do not do it and give away a |
| 7 |
>> huge constant factor for no advantage. |
| 8 |
>> |
| 9 |
> |
| 10 |
> (I'll just agree to disagree about the rest.) |
| 11 |
> |
| 12 |
> Keeping the salt secret makes your application more complex. Rather than |
| 13 |
> "SELECT hash, salt FROM users WHERE...", you now have to "SELECT hash |
| 14 |
> FROM users WHERE..." and then pull the salt from somewhere else. (Where? |
| 15 |
> The filesystem? Do you encrypt that? How?) |
| 16 |
> |
| 17 |
> What's stupid is going to all that effort for a 2x improvement when you |
| 18 |
> could twiddle a bit and get a 340282366920938463463374607431768211456x |
| 19 |
> improvement. |
| 20 |
> |
| 21 |
> |
| 22 |
|
| 23 |
|
| 24 |
Keep in mind the actual original purpose of a salted hash. |
| 25 |
|
| 26 |
If two users happen to use the same password[1], the hashes are the same |
| 27 |
and this is revealed to anyone who can read /etc/passwd[2] i.e everyone. |
| 28 |
Salt obscures this 1-to-1 mapping and does it in a way that it is not |
| 29 |
computationally worth while to try get around it for the general |
| 30 |
case[3]. It's not quite the same thing as security by obscurity - that |
| 31 |
is hiding something in a place you think no-one will think of looking |
| 32 |
but usually turns out to be viable to try and guess. Salt works because |
| 33 |
brute force now doesn't need just one expensive calculation, it needs |
| 34 |
many thousands of expensive calculations. |
| 35 |
|
| 36 |
If the actual problem is that salt is inadequate, the solution is not to |
| 37 |
try and hide it, but to use a more complex hashing algorithm with larger |
| 38 |
salt. It's a race between white and black hats - they build bigger and |
| 39 |
better rainbow tables, we implement bigger and better hashes. The |
| 40 |
constraint is how much cpu grunt is available for purchase at a |
| 41 |
realistic cost. |
| 42 |
|
| 43 |
|
| 44 |
[1] This is not uncommon. The domain size of all possible passwords for |
| 45 |
a implementation is very very large. Human psychology says that the |
| 46 |
actual domain size of passwords people will pick is a tiny fraction of |
| 47 |
the whole. Hence salt. |
| 48 |
|
| 49 |
[2] Nowadays we use shadow, but the development of salt pre-dates shadow |
| 50 |
|
| 51 |
-- |
| 52 |
Alan McKinnon |
| 53 |
alan.mckinnon@×××××.com |
Archives