Posted by Martijn Grooten on Aug 24, 2016
Among the security community a lot of research effort is dedicated to analysing exploit kits and their constantly evolving methods of frustrating researchers while infecting ordinary users with malware. A lot of this research is then used to design and build products that aim to prevent exploit kits from infecting users. But how well do such products work?
In our VBWeb tests, we look at web security products and their ability to block malicious traffic, in particular exploit kits. This week, we published reports on Trustwave Secure Web Gateway and Fortinet's FortiGate, two products that both blocked almost all of the exploit kits they were served, and which duly earned VBWeb certification.
In our lab, we are testing several more products. Late last month, we looked at five web security products and tried to answer the question: how likely are exploit kits to bypass such network-based defences?
Of the five products, three were running in our lab with two others being cloud-based. The exploit kits that were served were live* at the time the requests were made.
During the period 28 July to 2 August, we tested 54 instances of four prominent exploit kits: 31 instances of RIG, 12 of Neutrino, 7 of Magnitude and 4 of Sundown.
The five products blocked between 47 (87%) and 54 (100%) of these exploit kits, with Neutrino being the hardest to block and Magnitude being blocked by all products.
This is certainly good news. Of course, good security hygiene, such as keeping devices, software and plug-ins up to date, is the first and most important step in preventing exploit kits from infecting your systems and devices. But for those who can't trust themselves or their employees always to practise this (and, let's be honest, who does?) it is good to know that security products can provide an important extra layer of defence.
For more information about submitting your product to our VBWeb tests, please contact Martijn Grooten (firstname.lastname@example.org).
* For the locally hosted products, we were able to confirm that the responses they were served would indeed have infected the computer accessing the exploit kit. This is not the case for cloud-based solutions, where we can't control the traffic sent to the product. In a majority of cases, we have reasons to believe the exploit kit wasn't fully sent, possibly due to the IP address being "blacklisted". Of course, for the end-user, this doesn't make a difference.