Trends in Firewall Configuration Errors: Measuring the Holes in Swiss Cheese

Abstract

The first quantitative evaluation of the quality of corporate firewall configurations appeared in 2004, based on Check Point Firewall-1 rule sets. In general, that survey indicated that corporate firewalls often enforced poorly written rule sets. This article revisits the first survey. In addition to being larger, the current study includes configurations from two major vendors. It also introduces a firewall complexity. The study's findings validate the 2004 study's main observations: firewalls are (still) poorly configured, and a rule -set's complexity is (still) positively correlated with the number of detected configuration errors. However, unlike the 2004 study, the current study doesn't suggest that later software versions have fewer errors.