Messages in this thread | | | Date | Mon, 27 Dec 1999 18:29:15 -0800 (PST) | From | Robert Dinse <> | Subject | Unexecutable Stack / Buffer Overflow Exploits... |
| |
Boy, I'd kill for some generic buffer overflow exploit prevention that was 100% foolproof and didn't depend on applications being written properly, while still maintaining reasonable functionality of the computing environment. I.E. turning the power off and locking the machine in a physically secured environment would be secure but not very functional.
To the person that said most modern clients are smart enough to change UID before accepting user input, I have to ask, how many times have you had to upgrade sendmail, bind, ftpd, popd, imapd, rstatd, rlockd, portmap, .... because of buffer overflow exploits that keep popping up? And even for example, if you run bind non-root, if they can clobber your name server files and screw up name service, even that will suffice to raise major havoc. Besides if they can get a shell as ANY UID that gives them an inside base from which to advance their attack.
It's pretty clear to me that applications programmers will never adequately secure their code; and it's beyond the scope of what I can do to re-write every application I use or to be absolutely sure there are no buffer overflow or other exploits.
If you can stop the script kiddies; that's about 99.99% of the problem solved.
I've looked at Stack Guard, but unfortunately, it can be gotten around fairly easily and I'm sure at best it would just result in a new generation of exploit scripts AND it's only applicable to x86 presently (most of my equipment is Sparc based). I'd use it if it worked on Sparc anyway, because it's another layer to get around.
I don't see unexecutable stack as a 100% fix, but it's another layer that the script kiddies would have to get around. The more the better. It's always going to be more difficult to stop a determined real hacker that actually knows what they're doing, there are just too many bugs in any complex OS that someone can creatively find ways to exploit, but for every real hacker there are 10,000 script kiddies out there.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |