Regcast training : Hyper-V 3.0, VM high availability and disaster recovery

The first independent tests of anti-malware products on 64-bit Windows Vista revealed a rash of false positives.

Of the 20 products submitted for testing to independent security certification body Virus Bulletin, six generated false positives when scanning a set of known clean files.

Trend Micro submitted three of its anti-virus products, all of which falsely identified a Microsoft development tool as spyware. Other products to generate false positives were Fortinet's FortiClient, Ikarus Utilities, and VirusBuster, from the Hungarian firm of the same name.

John Hawes, technical consultant at Virus Bulletin, said: "The items added to our set of known clean files this month mostly consisted of common items taken from the 'most-popular' lists of free download sites, so it is a concern that the additions have caused such an upsurge in false detections.

"A false positive can cause as much disruption as a virus infection. False warnings often lead end-users to delete valid files in the belief that they are some form of attack and the resultant damage can be significant," he added.

Hawes blamed increased reliance on heuristic (generic) detection techniques for the rash of false alarms. "Anti-malware vendors must work hard to minimise false detections."

Other factors might also be in play.

Some anti-virus vendors complained of lack of access to the PatchGuard kernel protection system, as well as other security measures, included in the 64-bit version of Vista prior to its release. Opinions over the issue - even within the anti-virus industry - were sharply split.

Virus Bulletin reckons poor results of tests from anti-malware products on 64-bit versions of Vista suggest Microsoft's efforts at locking down the operating system have made life tougher for security firms.

Microsoft's enterprise anti-malware technology Forefront put in a strong performance and was awarded VB100 status for Vista x64 in the latest tests. By contrast, OneCare, its consumer anti-virus product, received a good shoeing after flunking Virus Bulletin's tests on 32-bit Vista products earlier this year.

Testing times

Virus Bulletin's VB100 test pit submitted anti-virus products against a test set of viruses from the WildList, a publicly available list of viruses known to be in circulation. To earn VB100 certification, products must be able to detect 100 per cent of the malware samples contained in the WildList test set while avoiding any false alarms in scanning a set of clean files.

Unlike other certification schemes, Virus Bulletin tests all products free of charge and does not allow re-testing. Virus Bulletin's comparative reviews also cover other performance aspects including detection rates against a selection of viruses never seen outside the lab of anti-virus vendors as well as looking at scanning speeds and performance overheads for anti-malware products.

The results of the VB100 certification of products for Windows Vista x64 Business Edition can be found here (free registration required). Detailed results tables are available to Virus Bulletin subscribers. ®

Agentless Backup is Not a Myth