lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <200401162044.i0GKi74a012586@ms-smtp-01-eri0.texas.rr.com>
From: mailinglists at wjnconsulting.com (Wes Noonan)
Subject: Re: January 15 is Personal Firewall Day, help the cause

> Am Fre, den 16.01.2004 schrieb Wes Noonan um 18:32:
> > Did you really just propose that a viable solution is to remove network
> > access?
> 
> For some systems: plain and simple yes. If the supplier of a software
> fails to deliver it in a "secure by default" state and even cuts the
> supply of patches (Windows NT4/95/98) these systems should go offline
> immediately. There is no compromise.

In a world that only security mattered, maybe. In the real world however
there is always compromise. Too many folks in the computer security business
seem to over look this point, or confuse compromise with concession. They
hear compromise and think concession, which isn't the same.
 
> This "Personal Firewall Day", aimed at the end consumer, may actually
> plant the idea in people's head that their unpatched and non-supported
> Windows 98 might be safe for the future as soon as they install a
> personal firewall. Well, this is just plain BS.

I haven't seen anyone saying that personal firewalls are the end solution.
As Ron (and others) pointed out and to paraphrase that bastion of good
security Shrek, security is like onions, there are lots of layers. Personal
firewalls are just another component in addition to those other layers, and
personal firewalls do mitigate potential exploits that haven't been patched.
In fact, that last point is a major reason for running any firewall.

> This is how people with exactly those "popular" systems perceive the
> message that they should switch to a more _secure_ system.

Yes, because in many cases the message is not what the user is wanting from
their software. Don't blame the users in that case; blame the people who
continue to fail to grasp the needs of the users. Users don't want to switch
to an operating system that is less functional (by perception or by fact)
solely for security. As you said, security is a trade off.

> This is not the same. Novell has been a propriety system and Windows NT
> has been a propriety system. They both suffered from the closed
> development and security assessment process. This is how Linux and other
> open OS differ.

Perhaps the biggest disservice being done to security and to Linux in
general is the mistaken belief that somehow being an open development system
insulates the product from the same kinds of mistakes that happens to every
other piece of software out there. Take DNS for example. How many exploits
exist and continue to be discovered for it? Belief that merely by being open
source there is some kind of protection from exploits is a myth. Sure, you
might find more bugs before you hit the street, but it is hardly a panacea. 

> Linux isn't safer than Windows because it is less popular. It is safer
> because it doesn't have all doors open by default and vendors can define
> the level of security they want for their distribution. 

Actually, these two points go hand in hand. Linux can afford to restrict or
not open doors because it doesn't have the broad customer bases that Windows
does that have to be catered to. Does this mean that Microsoft doesn't need
to do anything? No, of course not, but it merely illustrates the difficulty.
If Linux had to cater to the same needs it would, and frankly it is, find
that it isn't always a simple undertaking.

> Linux is far from being perfect. Being near perfect I'd raise my vote
> for OpenBSD yet something even slips past them. But MS Windows is just
> the plain opposite of OpenBSD yet Microsoft has the potential to do
> better!

I agree that Microsoft has the potential to do better. Again though, near
perfect is in the eye of the beholder. If OpenBSD was so perfect, it would
be more than just a specialized OS, but it isn't. Again, we come back to the
security compromise bit. Being totally secure, but non-functional in the
user's eyes is not perfection.
 
> The sin is that Microsoft's solution to this problem isn't closing
> unnecessary services BY DEFAULT but promoting additional third party
> software to put in between Windows and the Internet which the end user
> has to pay, deploy and operate. This is pathetic.

Again, you fail to grasp the difficulties in maintaining a broad customer
base. MS RPC isn't unnecessary in many, many cases. You want proof? Go block
port 135 on all of your internal network routers and let me know the result.
Furthermore, just because someone is promoting a 3rd party solution is
hardly pathetic. If you don't want to pay for something you can always turn
the services off or implement the port filtering/firewalling functions that
have existed in every NT flavor since NT 4.0, and maybe even 3.51 though I
am too lazy to verify that one.
 
> > This security through obscurity mantra is laughable.
> 
> Changing topics... not so fast! What happened to the old one ;-)

Actually, I see them both as the same. The constant pushing of the false
belief that by merely changing the operating system to something other than
Microsoft (obscurity) instantly buys you protection (security) is security
through obscurity by every sense of the phrase. "Choose something that most
people aren't running and you won?t be affected by the same things that most
people are". The problem is that more threats will continue to be released,
and as more folks move to the platform of choice, the threats will follow. 
 
> Linux is following (or should be) a strict open source philosophy. How
> is that to be "security by obscurity"?

I explained it above. In this case it's the concept of "by running something
few people do, you aren't as susceptible to threats". That only works as
long as the choice you made remains obscure. As more people move, more
threats will occur. This is true in all market segments. Hell, look at
firewalls. First people stopped using Check Point because it was constantly
being attacked. Then they stopped using PIX because it started constantly
being attacked. Who is the golden boy of the week now? SonicWALL? Netscreen?
No matter what choice you make, there will always be threats. 
 
> > The top dogs always get the most exploits.
> 
> No. The most lousy systems get the most exploits. Face it.
> 
> Take the market for webservers.
> 
> Apache virtually owns the market with more than 60%. How come that
> Microsoft IIS gets the most exploits? When I look into my Snort logs I
> don't get any Code Reds from Apache installations trying to sneak into
> my net. Funny, isn't it? Why isn't there a Code Red with the level of
> spreading for Apache as there is for IIS yet Apache is deployed on more
> than 60% of webservers?

Because they are Microsoft. No one distinguishes one Microsoft brand from
the other, that's why. This all in addition to the well known fact that
Microsoft shipped web services on virtually every system before Windows 2000
on by default, something which isn't counted in your 60% market share
number.

Besides, Apache has had more than its share of bugs which further
illustrates my point - you are not going to protect yourself by simply
running something different. Sure, you may lessen the quantity or types of
threats, but there will always be new ones waiting for you regardless of
vendor.
 
> > Accept the reality. When everyone
> > else starts using Firebird, Thunderbird or whatever other obscure
> program
> > you want to mention as your own personal bestest solution, then it will
> get
> > hacked and exploited beyond belief. History proves this.
> 
> In fact, "history" or better reality has proven you wrong. Or is the
> Apache case just an exception?! I don't think so. It only differs from
> the Linux-Windows comparison as Apache _already has_ an advantage in
> market share.

Actually, history hasn't as I explained above. This all in addition to the
fact that Apache (or should we call it a patchy) has had more than it's
share of exploits. 
 
> Why is delivering a system with all doors shut an unrealistic
> expectation? Why is delivering Windows XP Home with a closed RPC port an
> unrealistic expectation?

Why? Because it is already delivered. Besides, when you have many of the
same people bitching and moaning about how Microsoft is dropping support for
antiquated products and trying to move people to more secure operating
systems at the same time that they bemoan why security is an unrealistic
expectation, it kind of sends a mixed message. Which one do you want?
 
> > And people wonder why users don't understand, but certainly fear, a good
> > chunk of computer security...
> 
> Because they are told they have bought a secure operating system and
> some time later they are told to buy a virus scanner, a personal
> firewall, keep track of updating the OS, the virus scanner, the personal
> firewall, ...

They were never told that. They were told that it is more secure, and it is.
As for patches and updates, this is true for every product. Finally, as for
the need to buy third party applications, I would be willing to bet that you
are one of the folks who would complain that if Microsoft started offering
everything they were being an anti-competitive monopoly...
 
> > Wes Noonan
> > mailinglists@...consulting.com
> > http://www.wjnconsulting.com
> 
> Now, of course this is from someone who is listing Microsoft operating
> systems and applications in second place for vendors...
> 

Indeed, if the best that you can offer is a critique that I know Microsoft
operating systems, I'd say you have run out of valid points to make.

All this typing and the bottom line remains the same.

If you think that by merely switching products (pick the scenario, it isn't
just operating systems) that you are somehow protecting yourself, you are
foolishly na?ve. There is far more to it than that.


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ