How does SSH (Secure Shell) work?

[MadRat]

I understand that there is a private and a public key. The public key is given freely out to the rest of the world. The private key is needed to decode the information from the public key. But if someone knows the public key then why is their message safe in any way, and why can't one decode the information back to the original message from the public key? It doesn't make sense to me.

 

[CTho9305]

Take some math classes, or look up the RSA algorithm online.

edit: the basic idea is that it is very easy to multiply two large primes (duh, it's just multiplication), but very difficult to factor the products of large primes into their component primes. I don't know exactly how it works, because you can't rapidly generate a list of large primes, so I don't know where the primes come from. If it's just a list stored somewhere, searching the list would be trivial.

 

[n0cmonkey]

I'm not positive about all of this, but reading up on it shouldn't be much of a problem. Check openssh.com, see if they have any information on it. But I think, the public key encrypts the data so only the private key can decrypt it.

 

[n0cmonkey]

Originally posted by: CTho9305
Take some math classes, or look up the RSA algorithm online.

edit: the basic idea is that it is very easy to multiply two large primes (duh, it's just multiplication), but very difficult to factor the products of large primes into their component primes.

That shouldn't be necessary for a basic understanding of the process, but it would definitely help for in depth knowledge 😉

 

[MadRat]

Originally posted by: CTho9305
the basic idea is that it is very easy to multiply two large primes (duh, it's just multiplication), but very difficult to factor the products of large primes into their component primes. I don't know exactly how it works, because you can't rapidly generate a list of large primes, so I don't know where the primes come from. If it's just a list stored somewhere, searching the list would be trivial.

Awww, so its not that its perfectly secure, its that its secure for some while. That makes alot of sense then.

 

[Barnaby W. Füi]

http://www-106.ibm.com/developerworks/linux/library/l-keyc.html

 

[AmigaMan]

Originally posted by: BingBongWongFooey
http://www-106.ibm.com/developerworks/linux/library/l-keyc.html

Great info, thanks. There's a link on the bottom for the second article of this series that talks about key caching. I've been using ssh for a while but never knew about the ssh-agent.

 

[Nothinman]

Awww, so its not that its perfectly secure, its that its secure for some while. That makes alot of sense then.

Nothing is perfectly secure, given enough time you can bruteforce any encryption key no matter how big it is.

 

[drag]

Originally posted by: Nothinman

Awww, so its not that its perfectly secure, its that its secure for some while. That makes alot of sense then.

Nothing is perfectly secure, given enough time you can bruteforce any encryption key no matter how big it is.

Yep.

For example that's why when you go to check out the top-500 most powerfull computers, they are actually only top-500 that are PUBLICY KNOWN. 😉

Then again, the human half of security is usually a lot easier to crack then the encrypted half.

 

[CTho9305]

Originally posted by: Nothinman

Awww, so its not that its perfectly secure, its that its secure for some while. That makes alot of sense then.

Nothing is perfectly secure, given enough time you can bruteforce any encryption key no matter how big it is.

Not a one-time-pad with keylength >= message length, using "truly random" numbers in the key.

 

[Fiveohhh]

Originally posted by: drag

Originally posted by: Nothinman

Awww, so its not that its perfectly secure, its that its secure for some while. That makes alot of sense then.

Nothing is perfectly secure, given enough time you can bruteforce any encryption key no matter how big it is.

Yep.

For example that's why when you go to check out the top-500 most powerfull computers, they are actually only top-500 that are PUBLICY KNOWN. 😉

Then again, the human half of security is usually a lot easier to crack then the encrypted half.

Ever read Dan Brown's "Digital Fortress"?😀