All checks against policies are done before a user is allowed to log on to the network, said Felman, and done without installing any software on the client. Instead, ActiveX controls are loaded onto the system at the beginning of each Web or VPN session, then unloaded at the end. The control can be optionally cached by the administrator to the client for a performance boost the next time the system logs on.
Systems that don't meet the required policies -- which administrators set at a central software console -- can be completely blocked from logging on, or other administrator-defined options can come into play, such as allowing a log on but with notification to both the end-user and the IT staff that the system isn't up to par. Users can also be automatically redirected to other enterprise assets, such as a patch-providing server or the company's anti-virus definition files, to bring their systems into compliance before log on is permitted, said Felman.
On the firewall side, Clientless Security will check that a firewall is present on the outside system, and if one is absent, the user can be asked to install one, or the administrator can set policies so that one is automatically installed prior to log on.
Other verifications new to version 2.0 said Felman, include status checks of the client's overall security, and warnings if the system has not been patched against currently-known vulnerabilities.
Clientless Security works on guest PCs that are running Windows 98 and later using Internet Explorer 5.0 or later to access an enterprise Web site or application, or the network through a VPN tunnel.