Port scanning, the act of sweeping computer ports to discover what port may be open, has long been assumed to be a principal first step of any hacker.
But that may not necessarily be the case, according to researchers at the University of Maryland (UM) who found that port scans are not reliable predictors of an attack.
In UM’s test environment, port scans preceded attacks in only 5 percent of cases. The report also found that more than
half of all attacks are not actually predicated by any type of scan. “Hackers don’t necessarily look before they leap,” a new UM report asserts.
“I was surprised that the percentage was that low,” Michel Cukier, assistant professor and one of the report’s authors, said.
Cukier noted that the results may have been the attributable to some the decisions made in UM’s testbed.
“It would be interesting to repeat the experiment on other locations with other choices,” Cukier told internetnews.com.
Though port scanning was not a predictor for attacks, the study found that vulnerability scans do in fact lead to attacks in a significant percentage of cases. The UM report defines vulnerability scanning as a scan “used to fingerprint the presence or absence of an exploitable vulnerability.”
On their own, 21 percent of vulnerability scans led to an attack. When combined with ports scans (that is a port and vulnerability scan), vulnerability scans led to an attack in 71 percent of cases.
The UM research team used the formerly open
source Nessus vulnerability scanner as their network vulnerability scanner though they came to the conclusion that there was a need for a host vulnerability scanner. Cukier and his team have created an open source
project called Ferret to address that need.
“It focuses on Windows vulnerabilities. It is an open source tool,” Cukier said. “Our goal is to build a community similar to the Nessus one but for Windows vulnerabilities.”