Home page logo

fulldisclosure logo Full Disclosure mailing list archives

Re: Not even the NSA can get it right
From: Steve Kudlak <stevex11 () sbcglobal net>
Date: Thu, 26 May 2005 20:59:51 -0700

Way back when I worked for government agencies for a living all the easy to get to sites had nothing sensitive on them. Everything that had sensitive stuff was not on the ARPANET or was behind multiple gsteways. Right now even normal citizens like you and me can build pretty secure systems that will stop a lot of stuff. I assume the NSA does the same too but can do better. I come from the "Rainbow Books" era and those have been replaced by other things at this point. But there were a few bugs in Sun's C-2 Security and that's low level.

Now it could be they hired some standard webdesign firm to do it and that the website is only its sort of public face. There are Intanets with much better security and there are secure Networks that run on nice BSD variants that are very good. BSD is good because a lot of it is people who every morning or evening;) they get up for the past 20+ years they have thought about security issues and watched what happened and all that stuff. I have been giggling at the teenagers who have been attacking my website as of late. I learned a lot by reading the logs. But but we have secure passwords that are not in any dictionary and all that good stuff. It is also completely seperate from public accounts like this one I use for day to chattering about on the Internet..

Have Fun,
Sends Steve

Have Fun,
Sends Steve

Valdis.Kletnieks () vt edu wrote:

On Wed, 25 May 2005 12:58:37 EDT, Dan Margolis said:

Right, but why is XSS interesting? Why would they *want* a "suspected
script kiddie" list? Honeypots are good for learning about what sorts of
attacks are in the wild, *not* for learning who the attackers are.

So watching the console logs on a tempting target like www.nsa.gov for
a month isn't going to give a *really* good idea of what's out there?

Consider - of those who went and tried the XSS that got posted, what percent
probably tried some *other* tricks to see what *else* they could get it to do?

Yes, the NSA crew almost certainly know the attacks themselves - but by keeping
an eye on what tricks have made it out to the script kiddies, they can measure
how fast the tricks propagate. Any attack they see on *that* server they can
safely conclude that it's part of the script kiddie canon (as it's very unlikely
that a black hat would blow a 0-day attacking that server when everybody *knows*
there's probably nothing worthwhile on there...)

Remember - we're talking about the organization that provided guidance on the
design of DES's S-boxes, which made *no* sense at the time.  Many years later,
we find out that the NSA knew about differential cryptanalysis, the IBM crew
independently discovered it, but kept quiet at the NSA's urging, and then when
differential cryptanalysis came out in the open literature, the S-boxes made
sense.  This gave the NSA a *very* good measure of how far ahead they were
at the time.

Or the public website is just maintained by low-pay civil servants (after
all, there's no need for a security clearance for any of those pages ;)

Granted, we don't know everything the NSA does, but I see little to gain
from a public XSS hole, however insignificant. Occam's razor, folks; why
should I buy into such a twisted conspiracy theory?

I never said you should.  I merely implied that immediately concluding that
it was a stupid mistake might in itself be stupid.  Remember - we *know* that
many black hats try to stay under the radar by leaving tracks that look like
common script kiddies (so all the recon probes disappear in the noise).  Why
shouldn't the world leader in spreading and recognizing disinformation do the
same once in a while? ;)

Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/

Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/

  By Date           By Thread  

Current thread:
[ Nmap | Sec Tools | Mailing Lists | Site News | About/Contact | Advertising | Privacy ]