ntopng 1.2.0 XSS injection using monitored network traffic

ntopng is the next generation version of the original ntop, a network 
traffic probe and monitor that shows the network usage, similar to what 
the popular top Unix command does.

The web-based frontend of the software is vulnerable to injection of 
script code via forged HTTP Host: request header lines in monitored 
network traffic.

HTTP Host request header lines are extracted using nDPI traffic 
classification library and used without sanitization in several places 
in the frontend, e.g. the Host overview and specific subpages for each 
monitored host.

The injected code might be used to execute javascript and to perform 
management actions with the user-rights of the current ntopng user, 
which can be used to disable the monitoring function or deletion of 
accounts making the monitoring system unusable.

To give a coarse idea of the vulnerability the following python script 
can be used on the monitored network, afterwards the victim needs to 
browse to the Host overview / Host details in the ntopng frontend.

import httplib

conn = httplib.HTTPConnection("example.com")
headers = {"Host": "<SCRIPT>alert(\"xss\")</SCRIPT>", "Accept": 
"text/plain"}
conn.request("GET", "/", None, headers)
r1 = conn.getresponse()
print(r1.status, r1.reason)
data1 = r1.read()

Other users of the nDPI code might be affected as well.

Steffen Bauch
Twitter: @steffenbauch
http://steffenbauch.de