![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
Hi, We are looking for install access 2007. For that I would like to know : - Can we encrypt only some of tables in a database? - Can we give our own key? - How use encryption function via API (C# in visual studio) ? What is the complexity? |
#3
| |||
| |||
|
|
Hi, We are looking for install access 2007. For that I would like to know : - Can we encrypt only some of tables in a database? |
|
- Can we give our own key? |
#4
| |||
| |||
|
|
We are looking for install access 2007. For that I would like to know : - Can we encrypt only some of tables in a database? - Can we give our own key? - How use encryption function via API (C# in visual studio) ? What is the complexity? |
#5
| |||
| |||
|
#6
| |||
| |||
|
|
That crypto really provides a false sense of security. You might as well use ROT13 as the basis. I am sure you can achieve that with a formula in SQL directly and be done with the function altogether. All you have here is obfuscation, not crypto. |
#7
| |||
| |||
|
#8
| |||
| |||
|
|
Indeed I am referencing Arvin's post and not yours. The windows crypto API, although not 'perfect' is a damn sight better than what is offered above. Given the OP's original question I think you are right on the money. The only other method I can think of is a roll-your-own, however I am not confident that the OP has the background knowledge to achieve this. |
|
Still, I wonder if there is a ROT13 for SQL so it can be run inline??? I must have a google around and see what I can find. |
#9
| |||
| |||
|
#10
| |||||
| |||||
|
|
The Windows crypto API is not without its detractors. |
|
There are concerns about the implementation of the algorithms, in particular about in-built exploits known to the manufacturer but not to the public. There are also issues surrounding the origin of some of the base elements used in the construction number handling aspects - in particular that somone knowing the 'master' (although claimed by MS that this is unknown) can actually subvert the entire crypto system provided. From a security point of view, in particular with regards to risk assessment, the lack of peer review and clarity of the implementation leads to doubt about the quality of the API and its security. Depending on needs this can be a serious concern. |
|
A roll your own can be a superior way of dealing with risk, but only if you know what you are actually doing with crypto. |
|
If you muck it up then you can end up with a false sense of security, and even if the algorithms are correct the implementation you choose to follow can render them useless if not handled properly. Havign said all this, I am not aware of a single dedicated security company that would rely solely on the windows crypto API. The majority of 'serious' players in that field design their own implementations, and some use modified implementations of open source methods and code. A lot of them will actually run their crypto on a separate hardware chip, sometimes embedded into a token. |
|
In short the windows crypto API is not the be-all and end-all of security, but for many it is a good start. The usability depends on teh risk assessment, not on anything else. As for the ROT13 - since it is just obfuscation the simpler and faster it is, the less complex to deploy, the more reliable and flexible it will be. If the OP simply wants to 'hide' data from a casual read then ROT13 might well suit the task. Again it comes down to the risk assessment. FWIW I have never seen a risk assessment that has recommended the implementation of the windows crypto API, and that is after many years working in the security industry. Risk assessment is the key here, and I suppose only OP can really make that decision. |
![]() |
| Thread Tools | |
| Display Modes | |
| |