185 reputation
7
bio website Toobusy
location Abersychan, United Kingdom
age 69
visits member for 1 year, 8 months
seen 2 days ago

Old-ish IT Geezer, young at heart, memoir fanboy


Sep
4
comment How to renew keys for a statically-encrypted database
@SquareRootOfTwentyThree : I've just seen that the latest release of my security module allows for GCM, and will also provide a crypto-good random nonce.
Sep
1
comment How to renew keys for a statically-encrypted database
Would you also care to comment on this idea: Galois/Counter Mode (GCM) with a strong-crypto-random nonce?
Sep
1
comment How to renew keys for a statically-encrypted database
Thank you for some enticing new (to me, of course!) ideas.
Sep
1
comment Will varying plaintext compensate for a fixed initialisation vector?
Would the person who downvoted please care to say why, so I can learn to do better?
Sep
1
comment How to renew keys for a statically-encrypted database
Thanks. Re-encrypting everything in one go is not practical for many reasons, which I won't bore you with here. I do, however, accept that solving the problem needs professional crypto oversight.
Sep
1
comment How to renew keys for a statically-encrypted database
Thank you. I fully accept the need to involve an expert.
Nov
21
comment Will varying plaintext compensate for a fixed initialisation vector?
Thanks for your interest. As I'm in the process of doing a 'responsible disclosure' to the manufacturer, I don't want to go into more detail yet, but will definitely come back once the process is concluded.
Nov
20
comment Will varying plaintext compensate for a fixed initialisation vector?
Hmm, I can't quite get my head round that. Obviously, assuming we can only use whole byte offsets, then 8-bits is the 'least bad'. But I still can't quite see how we get beyond the first character that doesn't match the single known plaintext. ?
Nov
20
comment Will varying plaintext compensate for a fixed initialisation vector?
On further reflection, although I'm not defending the option, how do you get more than the first character of the PAN, if the offset is 8 bits?
Nov
20
comment Will varying plaintext compensate for a fixed initialisation vector?
A small detail: if one uses a CFB offset of 8 bits, am I right in thinking it's a little bit more complex that just C=EK(0)⊕P?
Nov
20
comment Will varying plaintext compensate for a fixed initialisation vector?
Thank you. In the light of your excellent answer I've (a) Moved my accept tick (again), and (b) added further detail to the question.
Nov
19
comment Relative merits of AES ECB and CBC modes for securing data at rest
Thank you. The lesson 'Crypto is hard' is well taught, and, I hope, well learned.
Nov
19
comment Relative merits of AES ECB and CBC modes for securing data at rest
@StephenTouset : Thank you. Would two copies of an eight-byte timer, appended together, provide an adequate IV? The timer changes, as a minimum, every microsecond, and only repeats after 130 years.
Nov
18
comment Relative merits of AES ECB and CBC modes for securing data at rest
Thank you for an really good answer. "Don't roll your own" is, of course, extremely good advice, worth a forbidden extra point.
Nov
18
comment Relative merits of AES ECB and CBC modes for securing data at rest
Thank you. Yes, I understand the problem of a body of different messages with some matching plaintext blocks, but (just) in this case, the messages are all only one block long, and the only matches will be between completely identical messages. Does this affect your comment? Please add to your answer, if you believe this significant. Of course, a general rule of "don't use ECB mode" is sensible.