That actually makes me wonder what you would need to do to recover a
token secret.  If you don't have any access to traffic (i.e., if
you're off-path), then you basically have to try all possible secrets
against the Provider -- which the Provider can prevent by limiting
query rates to anything reasonable.

If you're on-path, then it depends on the signature option.  If the
signers are using PLAINTEXT, then they're hosed no matter what (no
time needed).  Otherwise, an attacker has to crack either HMAC-SHA1 or
RSA-SHA1, neither of which are currently believed feasible in O(many,
many years).

Am I misunderstanding something, or is the two week thing kind of
pointless?  Just limit the rate of authentication (to, say, 1000/sec
if you've got 25 bits or so of entropy; you'll be good for at least a
year) and don't use PLAINTEXT signatures.

--Richard


On Thu, Jul 23, 2009 at 3:42 PM, Mr-Yellow<[email protected]> wrote:
>
> The spec doesn't define a max length... Instead....
>
> "For example, if Token Secrets are valid for two weeks, Service
> Providers should ensure that it is not possible to mount a brute force
> attack that recovers the Token Secret in less than two weeks. Of
> course, Service Providers are urged to err on the side of caution, and
> use the longest secrets reasonable. "
>
> However with most providers making their tokens infinite expiry this
> is kind of defeated as an approach.
>
> It's also impossible to determine what can be done in X weeks. For a
> big botnet owner they might have enough CPU to crack the longest keys
> in less then any token expiry period measured in weeks.
>
> It'd be nice from a developers standpoint to have some indication of
> how large to make our database fields for recording tokens. Maybe the
> brute-force aspects can be better mitigated elsewhere, as it seems no-
> one is following the spec in regards to expiry recommendations.
>
> Just spent hours looking for what turned out to be a truncated token
> because I didn't make a database field long enough. I don't often jump
> straight to 255 characters for a varchar field.
>
> -Ben
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"OAuth" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/oauth?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to