printer friendly version

Well-defined security policies essential

It is not enough merely to secure one's IT infrastructure - telephone systems, physical security and social engineering must also all be considered as part of a comprehensive security policy. "The best IT security in the world is useless if a help desk agent gives out a password to a hacker who calls in pretending to be the boss's secretary, or if sensitive areas are unlocked, allowing access to everyone," says Gary Middleton, National Security Business Development Manager at Dimension Data.

He advises that oganisations treat their voicemail systems in the same way in which they would their IT systems. The telephone system can be the weakest link within the organisation, enabling an attacker to gain valuable information about the business or even insight to assist in preparation of a computer system attack. "Companies should assess what kind of information could be obtained via the telephone system, and put security procedures in place accordingly," he says. "In some ways, voicemail systems are much more vulnerable than computer systems, which most organisations now protect to some degree, for example, by using firewalls. The voicemail system, on the other hand, is often unprotected."

Robert Cherry, writing for the SANS (System Administration, Networking, and Security) Institute, says companies must treat their voicemail and phone systems as valid targets of attack. "We must include these systems in our threat assessments and perform penetration testing against these systems." An equally important component of any security policy should be the management of human weaknesses.

"Social engineering is the term that has been coined for the phenomenon of tricking someone into helping an unauthorised person gain access to IT systems." According to a recent article by the SANS Institute, while most companies focus on technology such as upgrades, security kits and high-end encryption, "a popular means of gaining access bypasses the technical systems completely. It is based on the long-time con or confidence game but has a new name and new face - social engineering."

The article notes that it preys on the weakest link in a security system - the human being. "Social engineers are con artists who exploit human vulnerabilities such as ignorance, naiveté and an individual's natural desire to be liked and helpful," she writes. Some of the most common social engineering scenarios involve the help desk, which is particularly vulnerable because it is the first line of support for problems on the network.

It is easy for con artists to pretend to be senior managers in the organisation, or their personal assistants. If the information they seek is not imparted, they threaten to report the help desk agent to his supervisor, or to the senior managers they claim to represent.

"To protect themselves against threats ranging from hacking to social engineering to voicemail invasion, companies must have a well-defined security policy document," says Middleton. "Merely having policies in place is not enough, however, and can lead to a false sense of security. The security policy must be a living document that is constantly re-visited, and companies must come up with innovative ways of ensuring that they educate their employees about the policy on a regular basis."

The SANS (System Administration, Networking and Security) Institute is a cooperative research and education organisation through which more than 96 000 system administrators, security professionals, and network administrators share the lessons they are learning and find solutions to the challenges they face. SANS was founded in 1989.

For further details contact Gary Middleton, Dimension Data on tel: (011) 709 1000.

Share this article:

Further reading: