Re: encryption question
- From: "Rael" <rael@xxxxxxxxxx>
- Date: Tue, 29 Jan 2008 18:58:39 +0200
"Henrick Hellström" <henrick@xxxxxxxxxxxx> wrote in message
That might be an extremely bad idea if not done correctly:
Setup:
1. Select password P
2. Store H = Hash(P)
3. Store C = E(Hash(P),T)
Decryption
1. User inputs P'
2. If H <> Hash(P') return Error
3. Return T = E(Hash(P),C)
The flaw in this scheme ought to be fairly obvious. An attacker who finds
a H,C pair might simply calculate T = E(H,C) and be done with it, without
any information whatsoever about P.
Thanks for the input. I imply from your comment that using the hash function
method will work fine for symmetrical encryption.
Rael
.
- Follow-Ups:
- Re: encryption question
- From: Henrick Hellström
- Re: encryption question
- References:
- encryption question
- From: Rael
- Re: encryption question
- From: SiegfriedN
- Re: encryption question
- From: Henrick Hellström
- encryption question
- Prev by Date: Re: encryption question
- Next by Date: Re: ANN: AnyDAC 2.0 is now available for pre-order
- Previous by thread: Re: encryption question
- Next by thread: Re: encryption question
- Index(es):
Relevant Pages
|