Is AES256 sufficient for password hashing?
by moot - July 30, 2021 at 03:30 AM
#13
(August 04, 2021 at 06:54 PM)pompompurin Wrote:
(July 30, 2021 at 03:33 AM)STARTEXMISLEAD Wrote: AES is fine obviously if you have a long enough char pass.
Honestly tho id suggest looking into the others i guarantee agencies are trying their hardest on AES rn because its the most mainstream known "safe" one.

Your reply makes no sense, please elaborate

If you have a pass under 20 characters long i consider that poor practice, There's going to be some idiot out there that sets their password as something stupid like a basic numeric sequence, People will still attempt bruteforce attacks unless you have a means of full overwrite of everything on that drive including unallocated space after a certain amount of attempts, even then could still be a threat if its bitch easy to guess. My second point was essentially speaking about how AES is high on an agencies "to crack" list, Loads of people have been given advice to use AES because of how it works and how safe its supposedly assumed to be if bad human error has not been made. This results in loads of criminals using these well known hard drive / SSD encryption methods (Since I believe we are talking about local machine encryption here i think like storage space encryption). So you can bet money that they are attempting to work on the more commonly used ones.
#14
(August 04, 2021 at 08:37 PM)STARTEXMISLEAD Wrote:
(August 04, 2021 at 06:54 PM)pompompurin Wrote:
(July 30, 2021 at 03:33 AM)STARTEXMISLEAD Wrote: AES is fine obviously if you have a long enough char pass.



Honestly tho id suggest looking into the others i guarantee agencies are trying their hardest on AES rn because its the most mainstream known "safe" one.







Your reply makes no sense, please elaborate







If you have a pass under 20 characters long i consider that poor practice, There's going to be some idiot out there that sets their password as something stupid like a basic numeric sequence, People will still attempt bruteforce attacks unless you have a means of full overwrite of everything on that drive including unallocated space after a certain amount of attempts, even then could still be a threat if its bitch easy to guess. My second point was essentially speaking about how AES is high on an agencies "to crack" list, Loads of people have been given advice to use AES because of how it works and how safe its supposedly assumed to be if bad human error has not been made. This results in loads of criminals using these well known hard drive / SSD encryption methods (Since I believe we are talking about local machine encryption here i think like storage space encryption). So you can bet money that they are attempting to work on the more commonly used ones.







in all fairness there is no need to even attempt to break the encrypted drive if they are able to pull off a cold boot attack since that can be used to get the encryption keys from the memory; they are rare and weird.. like, the feds will legit freeze your ram modules with liquid nitrogen in order for them to retain data for longer periods of time since the RAM will only retain it for x amount of minutes when the PC is turned off. There are flaws to that type of attack but it is successful a lot of the time because there aren't that many known workarounds. You could say that a LKM such as Tresor would work or even Ramcrypt? since TRESOR will store the keys in a much more volatile area of memory in the CPU Registers, and RAMcrypt which does a similar thing to TRESOR also encrypting data in the memory. Veracrypt came out with their own method of encryption of keys and passwords too. Either way, Drive encryption has been beaten before at some points such as when you could bypass full disk encryption done by Bitlocker last year.



I'll post some papers here on the subjects too since they explain these things better than I can.



Papers:



https://www.usenix.org/legacy/event/sec1...Muller.pdf



https://www.semanticscholar.org/paper/Ra...7da575759f



https://jhalderm.com/pub/papers/coldboot-sec08.pdf



The bitlocker vulnerability which I was referring to:



https://portswigger.net/daily-swig/bitlo...encryption
#15
(August 07, 2021 at 11:39 PM)Sosolapute Wrote:
(August 07, 2021 at 11:37 PM)de4d Wrote:
(August 04, 2021 at 08:37 PM)STARTEXMISLEAD Wrote:
(August 04, 2021 at 06:54 PM)pompompurin Wrote:
(July 30, 2021 at 03:33 AM)STARTEXMISLEAD Wrote: AES is fine obviously if you have a long enough char pass.
Honestly tho id suggest looking into the others i guarantee agencies are trying their hardest on AES rn because its the most mainstream known "safe" one.

Your reply makes no sense, please elaborate

If you have a pass under 20 characters long i consider that poor practice, There's going to be some idiot out there that sets their password as something stupid like a basic numeric sequence, People will still attempt bruteforce attacks unless you have a means of full overwrite of everything on that drive including unallocated space after a certain amount of attempts, even then could still be a threat if its bitch easy to guess. My second point was essentially speaking about how AES is high on an agencies "to crack" list, Loads of people have been given advice to use AES because of how it works and how safe its supposedly assumed to be if bad human error has not been made. This results in loads of criminals using these well known hard drive / SSD encryption methods (Since I believe we are talking about local machine encryption here i think like storage space encryption). So you can bet money that they are attempting to work on the more commonly used ones.

in all fairness there is no need to even attempt to break the encrypted drive if they are able to pull off a cold boot attack since that can be used to get the encryption keys from the memory; they are rare and weird.. like, the feds will legit freeze your ram modules with liquid nitrogen in order for them to retain data for longer periods of time since the RAM will only retain it for x amount of minutes when the PC is turned off. There are flaws to that type of attack but it is successful a lot of the time because there aren't that many known workarounds. You could say that a LKM such as Tresor would work or even Ramcrypt? since TRESOR will store the keys in a much more volatile area of memory in the CPU Registers, and RAMcrypt which does a similar thing to TRESOR also encrypting data in the memory. Veracrypt came out with their own method of encryption of keys and passwords too. Either way, Drive encryption has been beaten before at some points such as when you could bypass full disk encryption done by Bitlocker last year.

I'll post some papers here on the subjects too since they explain these things better than I can.

https://www.usenix.org/legacy/event/sec1...Muller.pdf
https://www.semanticscholar.org/paper/Ra...7da575759f
https://jhalderm.com/pub/papers/coldboot-sec08.pdf
i'm proud of that reply
thanks my love
#16
I reccomend ROT13 for the best encrytpioj standard
#17
Yes I think it will be sufficient for you
#18
(August 07, 2021 at 11:37 PM)Opalstones Wrote:
(August 04, 2021 at 08:37 PM)STARTEXMISLEAD Wrote:
(August 04, 2021 at 06:54 PM)pompompurin Wrote:
(July 30, 2021 at 03:33 AM)STARTEXMISLEAD Wrote: AES is fine obviously if you have a long enough char pass.







Honestly tho id suggest looking into the others i guarantee agencies are trying their hardest on AES rn because its the most mainstream known "safe" one.















Your reply makes no sense, please elaborate















If you have a pass under 20 characters long i consider that poor practice, There's going to be some idiot out there that sets their password as something stupid like a basic numeric sequence, People will still attempt bruteforce attacks unless you have a means of full overwrite of everything on that drive including unallocated space after a certain amount of attempts, even then could still be a threat if its bitch easy to guess. My second point was essentially speaking about how AES is high on an agencies "to crack" list, Loads of people have been given advice to use AES because of how it works and how safe its supposedly assumed to be if bad human error has not been made. This results in loads of criminals using these well known hard drive / SSD encryption methods (Since I believe we are talking about local machine encryption here i think like storage space encryption). So you can bet money that they are attempting to work on the more commonly used ones.















in all fairness there is no need to even attempt to break the encrypted drive if they are able to pull off a cold boot attack since that can be used to get the encryption keys from the memory; they are rare and weird.. like, the feds will legit freeze your ram modules with liquid nitrogen in order for them to retain data for longer periods of time since the RAM will only retain it for x amount of minutes when the PC is turned off. There are flaws to that type of attack but it is successful a lot of the time because there aren't that many known workarounds. You could say that a LKM such as Tresor would work or even Ramcrypt? since TRESOR will store the keys in a much more volatile area of memory in the CPU Registers, and RAMcrypt which does a similar thing to TRESOR also encrypting data in the memory. Veracrypt came out with their own method of encryption of keys and passwords too. Either way, Drive encryption has been beaten before at some points such as when you could bypass full disk encryption done by Bitlocker last year.







I'll post some papers here on the subjects too since they explain these things better than I can.







Papers:







https://www.usenix.org/legacy/event/sec1...Muller.pdf







https://www.semanticscholar.org/paper/Ra...7da575759f







https://jhalderm.com/pub/papers/coldboot-sec08.pdf







The bitlocker vulnerability which I was referring to:







https://portswigger.net/daily-swig/bitlo...encryption


Cold boot attacks only works on old DDRM. New memory is much more volatile on power loss. I wouldn't doubt a similar attack strategy exists (but why even go that far when you have evil maid, phishing, bad configurations, etc).

https://xkcd.com/538/
#19
(August 21, 2021 at 08:54 PM)Intravert Wrote:
(August 07, 2021 at 11:37 PM)Opalstones Wrote:
(August 04, 2021 at 08:37 PM)STARTEXMISLEAD Wrote:
(August 04, 2021 at 06:54 PM)pompompurin Wrote:
(July 30, 2021 at 03:33 AM)STARTEXMISLEAD Wrote: AES is fine obviously if you have a long enough char pass.







Honestly tho id suggest looking into the others i guarantee agencies are trying their hardest on AES rn because its the most mainstream known "safe" one.















Your reply makes no sense, please elaborate















If you have a pass under 20 characters long i consider that poor practice, There's going to be some idiot out there that sets their password as something stupid like a basic numeric sequence, People will still attempt bruteforce attacks unless you have a means of full overwrite of everything on that drive including unallocated space after a certain amount of attempts, even then could still be a threat if its bitch easy to guess. My second point was essentially speaking about how AES is high on an agencies "to crack" list, Loads of people have been given advice to use AES because of how it works and how safe its supposedly assumed to be if bad human error has not been made. This results in loads of criminals using these well known hard drive / SSD encryption methods (Since I believe we are talking about local machine encryption here i think like storage space encryption). So you can bet money that they are attempting to work on the more commonly used ones.















in all fairness there is no need to even attempt to break the encrypted drive if they are able to pull off a cold boot attack since that can be used to get the encryption keys from the memory; they are rare and weird.. like, the feds will legit freeze your ram modules with liquid nitrogen in order for them to retain data for longer periods of time since the RAM will only retain it for x amount of minutes when the PC is turned off. There are flaws to that type of attack but it is successful a lot of the time because there aren't that many known workarounds. You could say that a LKM such as Tresor would work or even Ramcrypt? since TRESOR will store the keys in a much more volatile area of memory in the CPU Registers, and RAMcrypt which does a similar thing to TRESOR also encrypting data in the memory. Veracrypt came out with their own method of encryption of keys and passwords too. Either way, Drive encryption has been beaten before at some points such as when you could bypass full disk encryption done by Bitlocker last year.







I'll post some papers here on the subjects too since they explain these things better than I can.







Papers:







https://www.usenix.org/legacy/event/sec1...Muller.pdf







https://www.semanticscholar.org/paper/Ra...7da575759f







https://jhalderm.com/pub/papers/coldboot-sec08.pdf







The bitlocker vulnerability which I was referring to:







https://portswigger.net/daily-swig/bitlo...encryption


Cold boot attacks only works on old DDRM. New memory is much more volatile on power loss. I wouldn't doubt a similar attack strategy exists (but why even go that far when you have evil maid, phishing, bad configurations, etc).

https://xkcd.com/538/
yes, some exist. Also, that comic is pretty relatable lmao - i know people who were bribed with drugs by federal entities in turn for the names of their associates and their personal information.
#20
AES isn't for hashing, it's for encrypting. Ideally, you don't want passwords to be encrypted, because anyone with the decryption key would be able to decrypt it. One-way hash functions such as bcrypt or what have you are good for this as they only go one way. So you'd hash the user's password when they sign up, then from then on, you compare the hash of whatever password they try to sign in with with the hash you already have saved. In the case of bcrypt, a salt is also used by default and included with the hash to make sure no two hashes are the same even if the plaintext is. If you use another hashing algorithm, make sure you use a salt in that case as well.
#21
That's called breaking the key-derivation function (the method used to convert a small password to a long key). The password is usually short to be easily remembered, and the goal of a modern KDF is to take a very long time to make brute forcing harder (SHA is a bad example, because it's designed to be fast).
#22
(September 23, 2021 at 08:33 PM)septaneno Wrote: That's called breaking the key-derivation function (the method used to convert a small password to a long key). The password is usually short to be easily remembered, and the goal of a modern KDF is to take a very long time to make brute forcing Nox Vidmate VLC harder (SHA is a bad example, because it's designed to be fast).

I don't know if this is the correct place to post this, but have any breakthroughs been done on AES-256 or lower and is it likely that this cipher can really protect data for thousands of years(assuming you have good entropy)

Possibly Related Threads…
Thread Author Replies Views Last Post
Hashing bcrypt xpzkcs 4 468 June 30, 2021 at 02:45 PM
Last Post: ejdicu73848
RTX vs GTX hashing syler 0 693 February 14, 2019 at 05:18 PM
Last Post: syler
Hashing and Salts Silhouette 4 949 September 24, 2018 at 01:31 PM
Last Post: bitsandbytes

 Users browsing this thread: 1 Guest(s)