|























| |
The Battle for Your
Browser
By Larry
Seltzer
2008-04-30
Attackers
are doing a
drive-by on
your
browser, but
the defenses
against such
attacks are
good and
getting
better.
Windows
users have
to look at
the Internet
as a source
of unending
attacks. You
can defend
yourself
with some
software and
some common
sense, and
the defenses
are set to
get even
better.
There are
two basic
popular
types of
malware
infection
these days:
the Trojan
horse
program
marketed
through
links in an
e-mail
and drive-by
browser
hijackings.
I have a
hard time
getting my
hands around
how
effective
one or the
other is.
The
drive-by
method uses
a bag of
JavaScript
that throws
a stream of
attacks at
the browser,
one after
another,
hoping one
will
compromise
it. At
almost all
times these
attacks are
patched
vulnerabilities,
meaning that
you're
basically
safe from
them if you
keep your
browser and
other
software up
to date.
Some of that
software,
like Flash,
Acrobat and
RealPlayer,
are more
likely to
linger in
old,
unpatched
versions, so
you need to
be
assiduous.
A big
part of the
consideration
for
vulnerability
exploits is,
if they
happen to
execute, how
much damage
can they do?
Microsoft
has done a
lot of work
in this area
over the
last few
years,
aiming to
restrict the
ability of
exploits to
do much
damage if
they get
through
initial
defenses.
One of my
favorite
Microsoft
bloggers,
Robert
Hensing, who
works in the
Security
Vulnerability
Research and
Defense
group,
argues that
these
second-level
defenses are
good and
getting
better.
The
active
defenses on
by default
in Vista are
pretty good:
IE Protected
Mode, ASLR
(Address
Space Layout
Randomization)
and DEP (Data
Execution
Protection)
are all
examples of
this. But
applications
have to be
set up to
use ASLR and
DEP, and
both
Microsoft
and ISVs
have been
slow to do
so. These
techniques
are so good
at rooting
out bugs as
well as
exploits
that they
would be too
disruptive
to turn on
in a blanket
fashion.
But
they've been
turning on
more and
more. Even
Apple has
turned on
ASLR and DEP
in some
parts of
QuickTime,
an option it
doesn't have
in OS X. And
IE8 will
have DEP
turned on by
default,
which puts
all the
ActiveX
control
authors on
notice.
ASLR may
prevent it
from finding
an execution
point. If it
can find the
exploit
code, DEP
will likely
prevent it
from
running.
Protected
mode
prevents it
from
modifying
anything in
the system
or becoming
persistent.
As Hensing
points out,
faced with
an
environment
like this,
exploit code
can't get a
lot of work
done.
The
reflexive
advice you
get from a
lot of
sources,
including
the U.S.
government,
is to shut
off
JavaScript
and ActiveX.
I just can't
agree that
this is good
advice;
despite what
the average
Noscript
advocate
says, the
Internet is
unusable
with
JavaScript
turned off.
I try it
every now
and then and
it's just
not worth
surfing the
Web. With
ActiveX
turned off,
you can get
some work
done, but
you'll
notice that
lots of
pages work
badly,
mostly for
excessive
reliance on
Flash.
I often
wonder how
many
completely
updated PCs
get
exploited. I
bet the
numbers are
small, and I
bet almost
all of the
ones that do
get
exploited
are through
social
engineering
where the
user gets
tricked into
lowering
defenses and
running a
Trojan
horse. It's
hard to get
through
otherwise.
|
|
|
Finnish security vendor
F-Secure has collected twice as many
malicious software samples this year than it has
over the last 20 years, a trend that highlights the
growing danger of malicious software on the
Internet.
Through the end of 2006 and 20 years
prior, F-Secure counted a total of 250,000 samples,
said
Mikko Hypponen, F-Secure's chief research
officer. This year alone, 250,000 samples have been
counted, he said.
Statistics on malware from antivirus companies
can vary since the data is often derived from what
their customers experience while using their
software, and it depends on how widely that software
is used.
But other security vendors have also noted the
flood of new malware on the Internet over the last
few years.
Symantec said earlier this year that it detected
212,101 new malicious code threats between January
and June, an increased of 185 percent over the same
period a year prior.
The astounding increase shows that hackers "are
generating large number of different [malware]
variants on purpose to make the lives of antivirus
vendors more difficult," Hypponen said.
A variant is a piece of malware that has a unique
look but belongs to a known family of malware,
sharing common code and functions. Hackers use
techniques such as obfuscation, which jumbles up
code and makes it hard to determine what the program
is, and encryption, to trick security programs.
"Genuine innovation appears to be on the decline
and is currently being replaced with volume and
mass-produced kit malware," according to
F-Secure's report, which covers the second half
of 2007.
Higher numbers of malware samples put more
pressure on vendors to ensure they have fine-tuned
products. To handle the surge, F-Secure has hired
more security analysts as well as continued to
develop automated tools to evaluate malicious
software, Hypponen said.
Any new malware must first undergo an analysis.
Then most security software vendors companies create
a signature, or an indicator, that allows its
software to detect the malware.
Automation makes the task of analyzing malware
somewhat easier, but "in the end, a human makes the
decision where we add detection [signatures],"
Hypponen said.
Malware is Getting Sneakier
StopBadware.org warns that Web's 'dark corners' are everywhere, even on legitimate sites.
Robert McMillan, IDG News Service Wednesday, October 03, 2007 09:00 AM PDT
It's getting harder and harder to know who to trust on the World Wide Web, according to online safety advocates StopBadware.org. On Tuesday, the group released its 2007 Trends in Badware report, saying the bad guys are finding new ways to place their malicious software on our computers -- often by compromising Web sites that we trust.
With the help of one of its sponsor companies, Google Inc., StopBadware maintains a list of 200,000 Web sites that are known to be associated with malicious downloads. According to Max Weinstein, a project manager with StopBadware, more than half of these sites have been hacked and don't even realize it.
In fact, this move to delivering malicious software on legitimate sites has been a disturbing trend over the past year, he said.
"It used to be that the advice to the end-user was 'keep your software up to date and then don't go to bad Web sites,'" he said. "You still don't want to go to those sites, but what we seen now is that you can be on a very legitimate site and have a problem."
Web surfers know that visiting gambling or pornographic sites could harm their computers, but lately attack code can be downloaded from almost anywhere.
In January, for example, the Web sites of Dolphin Stadium and the Miami Dolphins, hosts to the 2007 Super Bowl U.S. football championship, were found to have been hacked and were serving up malicious software, just days before the Super Bowl.
And the bad guys are even sneakier than you might imagine. In June and July, Web sites that had been linked on the popular Boing Boing blog were compromised, a tactic called 'linkjacking.'
Weinstein says criminals don't necessarily have to hack a site to have it serve up malicious software. Part of the problem is in the Web 2.0 world, where sites are built up of many different components pulled from different parts of the Web, it's becoming easier to sneak badware onto a legitimate site.
StopBadware has seen this happening with Web advertising networks, which can easily be subverted by attackers to serve up maliciously encoded scripts and images, he said. "What we're seeing is a lot of cases where a legitimate Web site has an ad network, and that ad network itself, or sometimes even a subcontractor of that ad network, contains an ad that is providing badware."
"It's certainly something we are seeing in increasing numbers, probably in the past several months," Weinstein said.
eBay Inc. is looking into ways of curbing a similar problem. The online auction giant allows users to put their own images and HTML code on its site, but sometimes this leads to "bad code," said eBay Chief Information and Security Officer Dave Cullinane, speaking at an online security symposium held Tuesday at Santa Clara University. The company is looking at including security ratings for users as part of its reputation system to help prevent novice users from accidentally putting malicious or unwanted code on the site. "One of the things we are looking at bundling in is your level of security. As a user goes up, we'll allow you to do more things."
Under the proposed system, eBay power sellers with good security ratings would be given more free rein on the types of features they could add to their stores, Cullinane said.
Another growing source of concern is social networking.
Users should also be wary of fake accounts set up on legitimate social networking sites, which are often designed with one thing in mind: to lure unsuspecting users to malicious Web sites, Weinstein said.
So with all this badware, is the Internet a more dangerous place to be?
It's a tough question, Weinstein says, but he believes things are getting better, largely because people are getting smarter about what they do online. "I think the bad guys are always trying to stay a step ahead of the average users," he said. However, "people are learning, and I think that is having an effect."
"I'd like to think that our effort, and other efforts like ours, are actually making a substantial difference."
Top 10 tricks causing spyware
epidemic
Spyware tricks have
become increasingly devious,
making spyware and adware stick
to machines longer, more
difficult to remove and
sometimes impossible to see with
ordinary methods. In the
spyware tricks series I
wrote about seeing installations
with multiple resuscitators,
increasing numbers of randomly
named files, even randomly named
folders. Internet Explorer
security settings are being
changed by spyware and hosts
files are being hijacked. We've
recently seen installations of
keyloggers and spam bots along
with your garden variety of
adware. Now add rootkits to that
list. Let's look back at the
top 10 tricks of 2005…
10. Spyware spread
through Windows Media files
as described by
Ben Edelman,
Eric Howes and
Ed Bott in January. The
Windows Media Player flaw that
allowed the exploit involved DRM
and has since been patched by
Microsoft.
9. Adware
companies hide their dirty work
using rootkit technology,
examples
Enternet Media's Elitetoolbar
and
ContextPlus' Apropos and
PeopleonPage.
8. Internet Explorer
infected through Firefox
as
documented by Paperghost,
aka Chris Boyd. This story
stirred up quite a bit of
controversy. The real
culprit was a
Java-based malware installer,
which did, in fact, infect the
machine while browsing with
Firefox.
7. Direct Revenue
unleashed Aurora, see Got
Aurora? Nail.exe? for
details and more
here about the massive
impact of the Aurora software,
including a file named nail.exe,
which kept spyware help forums
and HijackThis experts
busy for months and
generated an unprecedented
number of
comments including threats
of violence against Direct
Revenue on my Spyware Warrior
blog.
6. Spam bots,
keyloggers, kiddie porn connect
with major adware companies
– 180solutions, Direct Revenue,
SurfSidekick, BullsEye Network
and ShopAtHomeSelect installed
in conjunction with a spam
zombie and rogue anti-spyware
program, all of which started
from a child porn site and were
installed through an
exploit as illustrated at SunbeltBLOG and
Spyware Warrior.
5. Spazbox domain
installs massive spyware/adware
– using IRC
as documented by
Paperghost and
Spyware Warrior (complete
with video), dissected by Wayne
Porter here
and again
here.
4. Anti-spyware
spread by spyware and trojans,
details
here about super rogues PSGuard,
Razespyware, SpySheriff, Spy
Trooper, WorldAntiSpy and more
recently SpyAxe
here.
3. Direct Revenue
adware distributed through
BitTorrent, (or more
aurora and nail.exe)
exposed by Paperghost and
told by
eWeek.
2. AIM worm carries
backdoor, rootkit and adware, found
to be powered by world wide bot
net with ties to the Middle
East. See
write up from CNET,
Paperghost's analysis and
FaceTime's
press release.
And now, drum roll please,
the top spyware trick of
2005…
1. Sony BMG infects
users with DRM rootkit
originally
reported by Mark Russinovich
at SysInternals. The fallout of
this debacle continues with
artists revolting and plenty
of
legal action against Sony
BMG in the works.
Top 10 rogue anti-spyware (Programs
to Stay Away from)
What is rogue anti-spyware?
Rogue anti-spyware programs are
defined by spyware and
anti-spyware expert Eric Howes
on the
Rogue/Suspect Anti-Spyware
Products and Sites page.
"Rogue/Suspect" means that
these products are of unknown,
questionable, or dubious value
as anti-spyware protection.
Some of the products listed
on this page simply do not
provide proven, reliable
anti-spyware protection or may
be prone to ridiculous false
positives. Others may use
unfair, deceptive, high pressure
sales tactics to scare up sales
from gullible, confused users. A
very few of
these products are either
associated with known
distributors of spyware/adware
or have been known to install
spyware/adware themselves.
A bit of history about the
Rogue Anti-Spyware page, if you
will. I had been loosely
tracking complaints on the web
about anti-spyware apps for some
time when the first "super
rogue" was unleashed just over 2
years ago. In late November
2003, complaints about a program
called
Spy Wiper started popping up
by the dozens in
forums and
blogs all
over the
net. I had an entire
blog category devoted to Spy
Wiper and its successor
Spy Deleter. Eventually the
Center for Democracy &
Technology (CDT) filed a
complaint about the two, and
later the
FTC took action and that
operation was shut down.
Due to the Spy Wiper/Spy
Deleter attacks I was really
fired up about rogue
anti-spyware and started
blogging about rogue apps.
A while later I learned that
Eric Howes had also been
tracking anti-spyware complaints
and testing the applications. We
began collaborating and the
Rogue/Suspect Anti-Spyware
page was officially launched on
June 26, 2004 with about 50 apps
listed.
Less than a year later, on
June 9, 2005, the rogue list
reached 200 apps. If you are
wondering why there are so many
rogue anti-spyware apps, click
here and scroll down a bit.
The list currently stands at 241
programs including 19 that have
been de-listed but remain on the
page with
notes about why they were
listed and later de-listed.
This year we have seen a
proliferation of what I call
super rogues, blogged
here and
here. These super rogues are
usually seen on pages designed
to look like a Windows security
center, seen
here and
here. The super rogues are
also known for hijacking
desktops and being installed via
security exploits, along with a
myriad of spyware and adware
apps, and are usually part of an
infestation called
smitfraud.
Let me say that choosing the
top few was very difficult
because they are all nearly
identical in behavior and
installation methods. I've
ranked them in part by their
pervasiveness and the number of
complaints found about them on
the web. They are apps that
debuted this year, except for
one honorable mention, an app
that's been around for about 2
1/2 years but continues to
appear regularly in spyware
infestations. The names of the
programs are linked to a
complaint or example of the app,
not the website of the vendor or
program.
Without further ado, I
present to you the top 10 rogue
anti-spyware applications of
2005.
Honorable mention goes to
VirtualBouncer/AdDestoyer for
its 2 1/2 year history of being
stealth installed in exploits
without notice or consent.
10.
Spyware Bomber brought to us
by the same folks behind
Enternet Media, the spyware
company
shut down recently by the
FTC.
9.
SlimShield tied with
Winhound Spyware Remover for
hijacking and stealth
installation.
8.
WinAntiVirus and its companion
WinAntiSpyware 2005 for
hijacking, aggressive
advertising and inappropriate
collection of personally
identifying information.
7.
SpywareNo and its clone SpyDemolisher
for stealth installation and
deceptive aggressive
advertising.
6.
Razespyware for stealth
installs, desktop hijacks and
aggressive advertising.
5.
Spy Trooper for stealth
installs, desktop hijacks and
aggressive advertising.
4.
WorldAntiSpy for stealth
installs, desktop hijacks and
aggressive advertising.
3.
PSGuard for stealth
installs, desktop hijacks and
aggressive advertising.
2.
SpySheriff for stealth
installs, desktop hijacks and
aggressive advertising.
1.
SpyAxe for desktop hijacks,
stealth installs and deceptive,
aggressive advertising.
Note: For anyone landing on
this this page while searching
for help with removing these
rogues, I'd suggest going to one
of the reputable spyware help
forums and posting for help.
SpyWareBeware, the
home of ASAP, the
Alliance
of Security Analysis
Professionals lists
member sites where users can get
expert help with spyware removal
from trained volunteers.
The List ... of known fake
Anti-Spyware... (once going
to this link ... it will not
return)
|
|
|