Educause Security Discussion mailing list archives

Re: BlueCoat with McAfee or Kaspersky Malware Engines


From: David Harley <dharley () SMALLBLUE-GREENWORLD CO UK>
Date: Sat, 17 Jan 2009 10:22:07 -0000

(Declaration of interest: I currently work for ESET, who make NOD32.
Nonetheless, this is a vendor-neutral rant...)

I think you're right. This is a problem across the board.

* Sheer glut has seriously reduced the value of the
one-signature-to-each-malicious -program model: it doesn't make sense to
produce a signature for every short-life binary that flares up for a few
minutes (often literally) and is never seen again, even if we had the lab
resources to do it that way. The most effective detection nowadays is either
generic (detection of whole families and sub-families), proactive
(heuristics, sandboxing, emulation etc), or hybrid. That doesn't mean there
isn't a use for known-malware detection, but it's infinitely less useful now
than it was when malware was more specialized, much rarer, and spread much
less slowly.
* Unfortunately, while the sheer numbers of detections we add daily would be
mind-boggling if we could list them for each individual malicious binary,
they represent a much smaller percentage of the totality of current malware.
In the 90s, a good heuristic scanner  could claim to detect something like
70-80% of new malware: clearly, that's no longer the case. :(
* Serverside polymorphism, where the malicious binary is replaced on the
server in a new form at regular intervals, isn't susceptible to the same
algorithmic detection approaches as the polymorphic engines of the '90s.
Binaries do change very quickly.
* The more effective a scanner's detection is, the more R&D time (really!)
badware gangs will put into tweaking their binaries till they evade the
current with the latest updates. I've noticed this trend particularly in the
past year with adware trojans like Virtumonde and fake
anti-virus/anti-spyware apps.

Which is why, while marketing departments sometimes stretch the truth,
hard-core anti-malware researchers will almost always advocate
defense-in-depth rather than kid you that "our product is all the protection
you need."

--
David Harley BA CISSP FBCS CITP
Small Blue-Green World



-----Original Message-----
From: The EDUCAUSE Security Constituent Group Listserv
[mailto:SECURITY () LISTSERV EDUCAUSE EDU] On Behalf Of Russell Fulton
Sent: 17 January 2009 07:53
To: SECURITY () LISTSERV EDUCAUSE EDU
Subject: Re: [SECURITY] BlueCoat with McAfee or Kaspersky
Malware Engines


On 17/01/2009, at 4:07 AM, Stanclift, Michael wrote:

We use McAfee Enterprise on our desktops and have been consistently
disappointed with its levels of detection. However,
Kaspersky has an
engine superior to nearly everyone else on the market and
both in my
use and in testing it's rated to have one of the best
detection rates
in the industry.


I think you will find that all AV products are having trouble
keeping up now. We use Nod32 and have noticed an increasing
failure rate over the past few months.  Malware is now
changing so frequently that AV vendors are struggling to keep
up and once the malware get installed it then protects itself
so that the AV never finds it (root kits) or hooks itself
into Windows so that AV can not remove the files because they
are held open by the OS.

I would suggest that changing your AV vendor will not solve
this problem.

Russell




Current thread: