To view parent comment, click here.
To read all comments associated with this story, please click here.
People, I think that's what he meant. You took it too literally.
Remote vulnerabilities are rare in most OSes including current Windows versions, and in Linux you don't often download and run nakit15yo.jpg.exe files, so the best approach is to brute force dumb passwords for services such as SSH that are left open by default by many admins and some distributions.
If you leave OpenSSH open, you must make sure that either password authentication or remote login to weak accounts is disabled. Non-default ports do work, but it isn't really safe.
Once a non-admin user is compromised the hacker has a lot of locally exploitable bugs to choose from.
He doesn't even need to do that as he can run a distributed spam botnet from non-root users all the same.
Edited 2009-01-23 22:56 UTC






Member since:
2006-01-04
"The typical cracked Linux machine is a Server, and it is usually manually cracked. "
. You'd be surprised about how often Linux machines are brute-forced by automated daemons in order to distribute spam and Windows malware. Linux is almost never the end-goal, but rather a way of getting to the Windows machines.
Really? Then there's a hell of a lot of people devoting their life to ssh dictionary attacks on my subnet
I hate Windows and love Linux, but still, I accept there is automated malware for Linux too.. it just tends to not be categorised and hunted in the same way as Windows malware. After all, most Linux boxes have a very good intrusion detection system: a vigilant administrator.
I do think, however, that the OS X/Linux security model itself is more secure than Windows, and helps prevent against serious attacks.