|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
Re: Security through obscurity, etc.
Leo Bicknell (bicknell
ussenterprise.async.vt.edu)Tue, 13 Dec 1994 15:27:15 -0500 (EST)
- Messages sorted by: [ date ][ thread ][ subject ][ author ]
- Next message: Mitch Wright: "Re: this is interesting..."
- Previous message: Timothy Newsham: "Re: Stallman and enable-local-variables in bugtraq-digest V1 #64"
- In reply to: Oliver Friedrichs: "Re: Security through obscurity, etc."
- Next in thread: Oliver Friedrichs: "Re: Security through obscurity, etc."
> The difference is too large to even argue about. A CERT advisory doesn't > give root to someone on any unprotected system on the Internet. Perhaps > 1 in 10 people will figure out the problem, would you rather have 10 out > of 10 people be guaranteed to? It doesn't matter if 1 in 10, or 10 in 10 can get into your site as root. One person with root access, can, in one command obliterate everything on your system. Frankly, I look at it this way. If the advisory doesn't tell you specifically what the problem is, someone will have to go look for it. If they look and find it, this tells me they have some intelegence/experience -- ie might be able to cover up their tracks, at least for a little while. With exploit scripts the odds are some bozo who doesn't know what it is will run it wrong and you'll notice right way because it's such a botched attempt. If one person knows how to get root on my site, I want to know too. And if that means that 10 other people learn in the process that's ok, because knowing is the only way I'll be able to stop that first person from doing something I don't want them to do. Keeping people in the dark only keeps those who don't already know from finding out. Those who do already know are still just as dangerous (if not more so because no one is looking for them). Also, vendors are (in a relative sense) slow to fix problems. As bad as it may sound, things will get fixed a lot faster if someone breaks into 50 of vendor x's systems and makes the news. I've seen vendors not release a patch for months because "no one knew about it". Perhaps a newspaper headline like "50 sites running x wiped out last night" would make them work a little faster. Of course, I wouldn't want it to be my site, but that's a risk you run being on the Internet, at any moment you might be destroyed. -- Leo Bicknell - bicknellvt.edu | Make a little birdhouse bicknell
csugrad.cs.vt.edu | in your soul...... bicknell
ussenterprise.async.vt.edu | They Might http://ussenterprise.async.vt.edu/~bicknell/ | Be Giants
- Next message: Mitch Wright: "Re: this is interesting..."
- Previous message: Timothy Newsham: "Re: Stallman and enable-local-variables in bugtraq-digest V1 #64"
- In reply to: Oliver Friedrichs: "Re: Security through obscurity, etc."
- Next in thread: Oliver Friedrichs: "Re: Security through obscurity, etc."