Message-ID: <4D347607.7050905@isecom.org>
Date: Mon, 17 Jan 2011 18:01:59 +0100
From: Pete Herzog <lists@...com.org>
To: "Thor (Hammer of God)" <thor@...merofgod.com>
Cc: "full-disclosure@...ts.grok.org.uk" <full-disclosure@...ts.grok.org.uk>,
	"Valdis.Kletnieks@...edu" <Valdis.Kletnieks@...edu>
Subject: Re: Getting Off the Patch


> No, I do not run a patch management company, but despite that,

I don't feel I scrutinized patch management in any way other than to 
say doing patch management costs something and not doing it does not 
cost that something. I think that's a fair assessment regardless of my 
patch management experience.

> Coming up with some way of creating a dependency on new, additional

I see examples out there of those less successful than you at 
implementing controls properly and in the right places. One of the 
things about the model of patching I don't like is how it requires 
constant administration and one that I'm hoping to avoid by either 
combining it with existing change control or, where there is none, to 
bring a bit of order to a stochastic environment. You're apparently 
not my target audience then.

> The fact that patching changes code is a point so obvious that it

When we create models we do it on the prospect of improving something. 
We don't expect much to shift right away but we will see the shift in 
5 to 10 years time. This no-patching we tried on a small scale (few 
servers and a few desktops) and there's ever more people implementing 
it that I hear about on ever growing scales. I have heard of a 
university looking to implement this for their computer labs which 
suffer many infections during the school year. They also won't upgrade 
their systems and are worried about when support ends and the patches 
stop. But that's just one example and one reason why and really I 
haven't seen this yet on the scale you're looking for. ISECOM 
certainly doesn't have the funding to afford a server farm to try it out.

I know this isn't something you find particularly useful. You made 
that clear. It's not for you, and then again, why would you change if 
you're happy with the way things are going for you? New models exist 
for people who have a problem that they haven't been able to solve 
under the existing means. Apparently you have. So this is research 
into new models for those who the old model doesn't work for.

>
> When you go to management with a paradigm shift that will require

Organizations who are looking for better security have come to us and 
begun implementing this piece by piece in their problem areas. I don't 
think anyone anywhere would completely change on the spot. That makes 
no sense. It's a gradual thing. People use new models, like this, in 
their problem areas first. As it works for them and they adapt to it, 
then they move forward applying it in other places. Many times, they 
have an emotional attachment to a process or are so deeply integrated 
into another model that anything else sounds crazy. I understand that 
and I'm not looking for those people to just jump on board.

Just to be clear, one doesn't need a server farm to prove something. 
There's many other ways besides a server farm. Yes, a server farm is a 
good test environment but not one we can afford. In this case we did 
get it to work consistently on various servers and desktops, in the 
real world, over the Internet, for over 5 years. We began to share 
this with others who slowly adopted it in places where they needed it 
or where it wouldn't hurt to try it. Some it took years to get over 
the feeling that they should be patching or running anti-virus just 
because. The money that was saved was not just from patching alone but 
from licenses and new software, specifically those who had to buy the 
newer OS versions to keep getting support patches, new updated app 
licenses, sometimes new hardware, and all the auxiliary costs from 
having newer, untested stuff in house still administered at the same 
level as before.

Now, my goal is not to get you to turn over your business to the model 
but instead, to get more people to try it and learn about op controls 
and OpSec. Clearly it makes you uncomfortable and even find it 
"wacky". So don't do it.

>
> How exactly is this going to be presented to management? "Hey,

Just change as quickly as you are comfortable with. From what I know 
is that many businesses don't like to change things that work. Even 
me. However most people are more than happy to attack problems that 
never seem to go away. That's how you try it. You first approach the 
problem areas that defied other solutions or are absorbing too much of 
your time.

>
> How is anyone supposed to actually consider this when you have

People will consider this if they have a problem where the old model 
of patching as security and other black-list approaches is not helping 
them. People will consider this who need perfectly balanced security 
with their operations. Then they will try it somewhere small first and 
grow it as they need it.

>
> I know this is all a harsh response, but your continued dialog

I expected nothing less from you.

Sincerely,
-pete.

-- 
Pete Herzog - Managing Director - pete@...com.org
ISECOM - Institute for Security and Open Methodologies
www.isecom.org - www.osstmm.org
www.hackerhighschool.org - www.badpeopleproject.org

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

Powered by blists - more mailing lists