Security Watch
Skype Security Hype?
Findings show that the popular VoIP service is full of security problems -- some of which are intentional.
Hacking
IT Warned Against Skype Hype: A recently published report claims to debunk
the hype about Skype. However, it's hard to see how the report debunks anything,
considering it both recommends for and against Skype's deployment, depending
on whether an adopter can accept the risks that Skype poses.
Cybertrust has written our own report on Skype that clearly outlines
the numerous security issues of Skype. These include its tenacity at defying
perimeter restrictions by modifying the protocol it uses to traverse firewalls;
its proprietary encryption mechanisms, which make monitoring the content virtually
impossible; and its peer-to-peer trust model that would permit a compromised
Skype installation to compromise other Skype installations. In fact, Skype has
so many issues regarding security in an enterprise environment that eBay, its
recent purchaser, has acknowledged it doesn't intend on marketing an enterprise-ready
version. There are many other solutions that offer better and more effective
enterprise features, making Skype something that should definitely be avoided.
Coverity, Stanford University and Symantec recently released
the results of a study they conducted looking for software bugs in open source
code. The study used bug-finding techniques defined by Stanford, and searched
the code of 40 popular software packages including Linux and Apache. The average
result was 0.43 flaws per thousand lines of code, and none of the 40 packages
had zero bugs.
Yeah, OK, so what? It took $1.2 million of taxpayer money to figure this out?
Besides, what does anyone learn from such a study, other than the fact that
all software has bugs and the more lines of code you have the more of them there
will be? All this study does is allow me, as the resident Windows bigot, say,
"See, millions of potential code reviewers and open source software still
has bugs!"
Symantec recently decided that its product, L0phtcrack, "no
longer fits into [the company's] future product strategy."
Interesting! Does Symantec think password auditing is past being needed?
Human Factors
A group of MIT engineers created the company SiteAdvisor in 2005
intent on protecting consumers from malware, spam and adware. The result of
their work is now available as a free plug-in for your browser. The SiteAdvisor
toolbar overlays the results of their scans of sites on search results from
popular search engines like Google or MSN. A green checkmark means they've
checked the site fully and found nothing awry, a red "X" warns of
potential problems. More details about scans can be found by simply clicking
on the icon next to the search result. The company says it has personally scanned
95 percent of all Web traffic.
Anything we can get to give us a heads up is a good thing, no doubt about it.
Certainly one could argue the pros and cons over the methods and results SiteAdvisor
displays, and we'll certainly see some company complain about its ranking
in the future in the way some adware vendors have complained about AV program
ratings. Regardless, it's hard to deny the power of a page full of red
Xs.
IBM and Novell have announced their support for the Higgins
Trust Framework, an open source project under Eclipse that states
as its design goals a set of APIs that will allow the exchange of identity,
profile and relationship information across disparate environments.
Some stories suggest that Higgins will interoperate with Microsoft's
InfoCard effort, an identity management system evolving out of Microsoft
Live, which is a replacement for Microsoft Passport and currently in use in
the Xbox environment.
The stated goals of Higgins are lofty indeed, but the project still lacks significant
support, software development and even well-defined use criteria. IBM has stated
it hopes to change that and "kickstart" the project as well as bring
more vendors into the fold.
IBM hopes to leverage Higgins for its Tivoli product, and Novell obviously
wants to extend its Directory services into a realm that can compete
with Microsoft's InfoCard proposals.
With the backing of supporters like IBM and Novell, Higgins may well turn into
something, but it still looks to be a long way off.
A Virginia U.S. Department of Education Auditor has admitted to installing
surveillance software on his boss' computer without permission. Prosecutors
state they believe the man did so for his own amusement, not profit. The man
faces five years in prison and fines up to $250,000.
Of most interest to me in this case was the fact that prosecutors believed
the defendant was acting "for simple amusement" and not for profit.
The reasons that someone hacks someone else shouldn't, in my opinion,
be a part of their prosecution. It has seemed, however, that it has often played
a mitigating role in sentencing in the past. As this case falls under federal
sentencing guidelines, there's a strong chance he'll still get a stiff sentence,
serving as a strong warning to others who think their "fooling around"
isn't going to yield tough sentences.
Physical Security
A Finnish military security researcher has announced that he can steal
fingerprints from the Microsoft Fingerprint Reader. The new Microsoft
device is actually licensed from Digital Persona, and Microsoft states
that the device should not be considered a security device nor should it be
used to protect sensitive information.
Want
More Security? |
This
column was originally published in our weekly Security Watch
newsletter. To subscribe, click here. |
|
|
This is another one of those situations where a researcher has performed research
that would appear more intended to garner media attention than expand anyone's
knowledge. For example, the device in question, when sold by its original manufacturer
Digital Persona, includes an option to encrypt all images scanned. Microsoft's
implementation doesn't include this option. In my opinion, Microsoft was
either not permitted to OEM the encrypted version because Digital Persona hopes
to up-sell users of the Microsoft version, or, Microsoft purposely wanted to
avoid complaints over the device's ability to secure what it scanned.
Either way, such a device and its operation are never likely to be 100 percent
securable given that the entire transaction (e.g., scan to authentication) is
likely to happen within the PC itself, meaning replay is likely impossible to
prevent. Encryption certainly can add security to the equation, but there would
have to be verification of the fact that a warm finger was on the device when
a scan was taken and a certificate chain ensuring the image scanned of that
warm finger wasn't altered end-to-end.
I think Microsoft's position on the device's security role is accurate,
namely that it shouldn't be thought of as having one.
About the Author
Russ Cooper is a senior information security analyst with Verizon Business, Inc.
He's also founder and editor of NTBugtraq, www.ntbugtraq.com,
one of the industry's most influential mailing lists dedicated to Microsoft security.
One of the world's most-recognized security experts, he's often quoted by major
media outlets on security issues.