> ... Which shouldn't be oversold as a barrier. There > are tricks (some > fairly trivial) to bypass that protection.
Sure. With enough assembler, almost anything can be defeated. Lucky us that all we have today are a bunch of braindead "X-mas tree experts", nobody even remembers how to write assembler code anymore. With that written, the NX bit is just one hurdle, and we haven't even touched upon others. So even if somebody did defeat the "NX hurdle", there are pile of others waiting to be "braved". > The best way I've seen it put is: "If someone > can run code on your > computer without your consent, it isn't your computer > anymore". I used to know a guy that also grew up with Solaris. Now, there probably wasn't a UNIX system out there he couldn't bust or break into; this guy was reeeaaally good. One such episode involved breaking the unbreakable Digital OSF/1 national telco server, and to the point where the sysadmin got so desperate that he was begging him to leave the system alone. When I asked why the sysadmin simply didn't kick him out and locked his account, the guy just shrugged his shoulders: "hw would if I existed on the system, and if he could see my processes running. How's he gonna kick me out if I'm invisible to him?" So I asked him if that was it, game over. He shrugged his shoulders again: "I can do whatever, but the person on the other hand ALWAYS has the upper hand: there's NOTHING I can do when they pull that network cable." This message posted from opensolaris.org _______________________________________________ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
