0
$\begingroup$

Questions are as following:

  1. Is 3DES still feasible (in terms of security and performance) to use?
  2. Is 3DES safe to use in ECB mode?
  3. Every time on giving the same input (plain text) to encrypt via 3DES results same output (cipher text). What would the reason be for that? For example: Every time I gave input plain text 'wealth' the output generated was 'VxDfWEQA'. Or is that even possible?
  4. If one would want to discourage a company to not to use 3DES with cipher mode ECB and padding PKCS7 what would that reason would be and why?
  5. If a company/entity still insists to use 3DES. What configuration would be the most suitable to be set for it e.g. cipher mode and padding.

Would like to have the answer of third question considering all possible scenario's if applicable e.g. considering all 3 keys for 3DES were same.

$\endgroup$

closed as too broad by e-sushi Sep 28 '18 at 22:58

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ Point 3 does not sound as truthful: there's one chance against >340000 that a DES ciphertext is 8 letters. $\endgroup$ – fgrieu Sep 26 '18 at 13:44
  • $\begingroup$ @fgrieu, I have updated the question. $\endgroup$ – Ali Qureshi Sep 26 '18 at 13:55
  • $\begingroup$ 1. slow 2. no; ECB is ECB, 3. because ECB. 4. see wikipedia on modes of opreations, 5. depend on the what to enrypt. $\endgroup$ – kelalaka Sep 26 '18 at 14:16
1
$\begingroup$

Is 3DES still feasible (in terms of security and performance) to use?

3DES is very slow except if you have specialised hardware, so it at most wins in speed on legacy hardware that has dedicated DES circuitry but no AES circuitry. If no specialised hardware is available, AES or most other modern lightweight ciphers will be significantly faster.

3DES also has a very small block size, meaning it can start leaking data after just a few GB of encrypted data, see also Sweet32. Additionally the 112-bit key is considered to be quite short these days even though technically still sufficient, but it could be too short for e.g. reasonable security against multi-target attacks.

Is 3DES safe to use in ECB mode?

No block cipher is safe to be used in ECB mode, as you will always see which plaintext blocks appear multiple times. Additionally deterministic encryption (such as when ECB is used) cannot achieve security against chosen-plaintext attacks (a rather weak security notion).

Every time on giving the same input (plain text) to encrypt via 3DES results same output (cipher text). What would the reason be for that? For example: Every time I gave input plain text 'wealth' the output generated was 'VxDfWEQA'. Or is that even possible?

(3)DES is a deterministic function, so if you give it the same input (that is the same message-key pair) twice you will get the same output twice. Randomness usually is only brought into the picture when using the block cipher. Also note that the probability of getting an ASCII-text ciphertext is $(52/256)^8\approx 0.000002898$.

If one would want to discourage a company to not to use 3DES with cipher mode ECB and padding PKCS7 what would that reason would be and why?

See (1) and (3).

If a company/entity still insists to use 3DES. What configuration would be the most suitable to be set for it e.g. cipher mode and padding.

EAX or CCM mode or if neither are an option CTR mode with a counting (context-dependent) IV and then a secure message authentication code (MAC) applied on the ciphertext (which is what CCM and EAX do). These modes require no padding.

$\endgroup$
  • $\begingroup$ Why CCM? It's quite inefficient, which is worse since 3DES is so slow. $\endgroup$ – forest Sep 26 '18 at 22:41
  • $\begingroup$ @forest I didn't want to assume availability of something like SHA256 and GCM isn't applicable (at least I think the standard doesn't specify a standard polynomial for 64-bit?) which leaves us with CCM and EAX. $\endgroup$ – SEJPM Sep 26 '18 at 22:46
  • $\begingroup$ Or OCB or Poly1305. $\endgroup$ – forest Sep 26 '18 at 22:47

Not the answer you're looking for? Browse other questions tagged or ask your own question.