RE: Why I should have the right to kill a malicious process on your machine

[InfoSec News <isn () c4i org>]

Forwarded from: "Deus, Attonbitus" <Thor () HammerofGod com>
Cc: jasonc () science org


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

At 08:35 PM 1/15/2003, Jason Coombs wrote:
> Aloha, Tim.

Hey Jason-


> Rights in and liability for abandoned property is a complex subject of law.
> Nobody would argue that you don't have the right to perform remote system
> administration on abandoned property that happens to still be connected to a
> power source and the Internet, but a server that has been owned by a worm or
> an anonymous third-party attacker is not clearly abandoned. As physical
> property it still belongs to its legal owner. If we allow anonymous remote
> system administration that is allegedly benign or even beneficial to
> information security why shouldn't we also encourage the coding of
> self-replicating concept worms and viruses that exploit security
> vulnerabilities for the sole purpose of demonstrating that such
> vulnerabilities exist?

A self-replicating worm, even if its purpose is to inoculate some other 
parent worm, still carries with it the issues of bandwidth and process 
utilization.  It also requires "scan-based" propagation where it must first 
connect to a host to determine if it is infected with the target 
worm(s).  The same initial group of systems would end up hosting the new 
process; the end result being an event that is no less of a worm than the 
worm it set out to destroy in the first place.

The neutralizing technology we are talking about acts in singularity to a 
specific attacking host as a defensive measure.  The intent is to only 
affect the attacking process, not to perform "remote administration."   I 
know that there are many hypothetical scenarios one may present that 
illustrates where this type of action will fail, but for the most part, I 
think we have done a satisfactory job of addressing most of the 
"real-world" issues.

I'd really like it if you would check out the whitepaper (and all others on 
the list that are interested) and give me some feedback:
http://www.hammerofgod.com/strikeback.txt


> Code Red was a concept worm. It did no real harm. Its spread could have
> educated IIS administrators as to the threat of their unpatched boxes, but
> it didn't. Code Red II DID result in widespread awareness of the security
> risk of unpatched IIS boxes because it did cause widespread harm. It's not
> difficult to see the slippery slope that begins with your good intentions
> and ends with the logical conclusion that in order to cause real security
> for the good of your nation and the world, you have to write malicious code
> that self-replicates and causes global electronic paralysis. Otherwise
> nobody will listen, nobody will acknowledge the threat even though you see
> it clearly, and nobody will act to prevent more severe penetrations before
> they occur.

Again, we are not suggesting self-replicating code, so I'm not clear on how 
one could arrive as that being the resultant logical conclusion.  This is 
not a long term goal- it is a short term solution; a means an end... 
Ultimately, I would like to see a level of responsibility and 
accountability assigned to system owners for the state their assets when 
they choose to participate in a public network.   Ideally, people would 
practice due diligence in system security- the current state of the 
Internet shows that this is simply not being done.  To me, if you have 
something that needs to be done, but those who should be responsible can't 
or won't do it, then the only logical conclusion I can come to is that 
someone else has to do it for them.  That doesn't mean that it gets to be 
me or you, but it does mean that we need to take a look at other options to 
ensure that the integrity of the systems and services we have come to 
depend on is maintained.  This is one of those options that I think should 
be explored.

 From a legal standpoint ( as I understand it from many conversations with 
counsel) one can not prove a "duty to perform" due diligence in system 
security unless one can prove that such duty would be expected by a general 
portion of the populous.  From a security standpoint, the 
competence:installations ratio has simply not permeated the computing 
community to the degree that would be necessary to reasonably expect such 
duty.   A general understanding of "if you don't secure your system you may 
be responsible" just might raise that level.

> When electronic trespassing is permitted in violation of other people's
> reasonable legal rights under the condition that the trespasser must be
> attempting to do something beneficial to the security of the property in
> which she trespasses the entire notion of illegal electronic trespassing
> disappears, to be replaced with forensic arguments made by expert witnesses
> in front of juries.

I obviously don't consider it trespassing.  This is an ACK to a SYN- a 
defensive ACK to an attacking SYN.  There are certainly grounds for 
disagreement here, and it is something that must be discussed- particularly 
since issues pertaining to the degree cyber-events are related to the 
physical continuum are just going to become more complex.
I must say thank you, however, for not bringing up the "If a burglar breaks 
into your neighbor's house" analogy!

>  You do not want a jury of your peers to decide whether
> or not the prosecution's interpretation of the computer evidence is accurate
> or whether your defense expert witness is correct in her forensic
> counter-analysis that proves your innocence. This is a losing situation for
> you, the accused, because law enforcement will always appear to be the more
> credible witness.
>
> "Ladies and gentlemen of the jury, who are you going to believe?
> THOR () HAMMEROFGOD COM or the FBI?"

That is another question altogether... But that is the main reason I am 
looking for, at the least, some sort of legal backing in case one should 
find himself in a position of having to educate a jury as to the difference 
between physical trespass and digital self-defense.


> Advocating white hack hacker penetrations of other people's property for the
> purpose of remote system administration also fails the common sense test.
> Common sense tells us that we can defend ourselves adequately from all of
> our badly-configured and compromised peers simply by unplugging our
> computers from the network that connects us to them.

I don't think that defensive measure has any common sense to it at 
all.  Though I am loath to bring another physical analogy into the 
discussion, that is like saying that if I don't want to receive 
telemarketing calls that common sense would dictate that we just unplug our 
phones.

> If the only viable
> solution to the world's information security problems includes automated
> legalized trespassing, then the world needs brand new computer products
> designed from the ground up with infosec in mind. The fact that we will soon
> see the first generation of these systems enter the marketplace may be proof
> of the fundamental insecurity of existing programmable computers; though the
> jury is still out deliberating this point.

Well, I would say that the world needs new computer products with security 
built into the manufacturing and development models anyway- but we are a 
ways off from that.  We have worm problems now, and need to find a solution 
now.  Will this work?  Maybe, maybe not.  But I would rather spend some 
time and energy looking for a solution, even if not ideal, than to just sit 
around and take it.

I very much appreciate the comments, Jason- please give the whitepaper a 
once-over and let me know what you think.

All comments appreciated.

ttyl

Tim

-----BEGIN PGP SIGNATURE-----
Version: PGP 7.1

iQA/AwUBPiciYIhsmyD15h5gEQIgTACg2EkFmnM1XDQq/fgtCSkAMbJH5EkAoP8Z
5aP+xjvBbfDUorU6q5GZjb6S
=jaBJ
-----END PGP SIGNATURE-----



-
ISN is currently hosted by Attrition.org

To unsubscribe email majordomo () attrition org with 'unsubscribe isn'
in the BODY of the mail.