[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Taler] repurchase detection
From: |
Christian Grothoff |
Subject: |
Re: [Taler] repurchase detection |
Date: |
Tue, 23 Feb 2016 08:56:40 +0100 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Icedove/38.5.0 |
On 02/22/2016 11:59 AM, Marcello Stanisci wrote:
>> If it's just the correlation ID, what if it clashes for two merchants?
>>
>> Should we use the hostname of the fulfillment URL? The hostname of the
>> site that offered the contract (with taler-confirm-contract) in the
>> first place? What if the merchant's hostname changes?
>
> There is also a field in the contract that encompasses the triple
> (address, name, jurisdiction) associated with a merchant. This value
> is less likely to change than keys and hostnames, and should not make
> any clash happen. So just use its hash?
>
I think that might be dangerous, as it is most easily faked. I can put
a fake address, name and jurisdiction trivially, a fake DNS name is
harder; in contrast, if we tie this to the EdDSA public key (ideally
X.509-certified) it becomes nearly impossible to fake (assuming it is
combined with an EdDSA signature in the right spot...). So
security-wise, not supporting key rotation is a really minor issue
compared to using insecure information.
signature.asc
Description: OpenPGP digital signature