mailing list archives
Re: Web sites compromised by IIS attack
From: Paul Schmehl <pauls () utdallas edu>
Date: Wed, 30 Jun 2004 21:08:27 -0500
--On Wednesday, June 30, 2004 6:27 PM -0500 Frank Knobbe <frank () knobbe us>
I'm right there with you, Frank, on one condition. You hold *every*
software vendor to the same standard. IOW, "Apache should be required to
fix their own, broken products"..."RedHat Linux should be
required"......"Oracle should be
etc., ad infinitum, ad nauseum.
Instead of requiring the consumer to install patches, Microsoft should
be required to fix their own, broken products. That means that they
should send their army of engineers (a lot of which are now carrying the
CISSP certification) to the consumers and have their engineers correct
the flaws in their products. They sold flawed products, they should fix
Be careful what you wish for. You may actually get it.
I just upgraded my workstation from RedHat 9.0 to Fedora Core 1. I then
ran up2date and found that there were 142 software packages that needed to
be updated. Just before I did that, I run portupgrade on one of my FreeBSD
boxes. It had 17 programs that had to be updated.
If we're going to require that software vendors produce flawless products,
we're not going to have many software products. Even Postfix, which *to my
knowledge* has never had a security issue, has had numerous bug fixes.
(And I think so highly of Postfix that the first thing I do when I install
a new OS is replace sendmail with Postfix.)
I attended a presentation yesterday for a security product in the
application firewall field. During the presentation, the CISSP stated that
"in every 1000 lines of code there will be 15 errors". I don't know if I'd
agree with that - I suspect most coders are a bit better than that - but I
had to chuckle, because, of course, I immediately thought, "So you admit
that your code is riddled with holes!"
We need better methodologies for finding bugs in software. We need better
training of programmers. We need established standards for coding that
would define things like bounds checking. We need a *lot* of improvements
in software development, and those improvements need to be *industry-wide*,
not just Microsoft.
Every time I read about a security vendor with a remote hole in their
products, I think, "How in the world can they identify attacks, if they
can't even see them in their own code?"
Clearly the problem is a *lot* bigger than Microsoft alone.
Paul Schmehl (pauls () utdallas edu)
Adjunct Information Security Officer
The University of Texas at Dallas
AVIEN Founding Member
Full-Disclosure - We believe in it.