doc/README.md
See README.compilation for more information.
Please make sure that you have a Redis server installed and active on the same host
where ntopng will be running. If you plan to use a remote Redis server, please consider
using the --redis option to specify a remote Redis server IP address and port
or a local socket. We suggest you run Redis as a service so that you do not have
to start it every time you want to use ntopng.
In order to use ntopng as a flow collector with nprobe you need to start the apps as follows:
collector
ntopng -i tcp://127.0.0.1:5556probe (nProbe)
nprobe --zmq "tcp://*:5556" -i ethX -n none -b 2You can instruct ntopng to merge onto the same interface multiple endpoints by separating them with a comma. Example:
ntopng -i tcp://127.0.0.1:5556,tcp://192.168.0.1:5556
You can create a hierarchy of ntopngs (e.g. on a star topology, where you have many ntopng processes on the edge of a network and a central collector) as follows:
Remote ntopng's
Host 1.2.3.4 ntopng -i ethX -I "tcp://*:3456"Host 1.2.3.5 ntopng -i ethX -I "tcp://*:3457"Host 1.2.3.6 ntopng -i ethX -I "tcp://*:3458"Central ntopng
ntopng -i "tcp://1.2.3.4:3456" -i "tcp://1.2.3.5:3457" -i "tcp://1.2.3.6:3458"Note that on the central ntopng you can add -i ethX if you want the central ntopng
monitor a local interface as well.
You need to specify the user and password as specified below (please note the space in the cookie). Note that you can optionally also specify the interface name.
curl --cookie "user=admin; password=admin" "http://127.0.0.1:3000/lua/rest/get/interface/data.lua?ifid=0"
The ntopng default user is 'admin' (without ') and the default
password is also 'admin' (without ')
Check out https://www.ntop.org/guides/ntopng/faq.html#cannot-login-into-the-gui
In order to run multiple ntopng instances independently (i.e. they do not interfere each other), each instance must:
-d-r-wExample:
ntopng -d /path1 -r 127.0.0.1:6379@1 -w 3001ntopng -d /path2 -r 127.0.0.1:6379@2 -w 3002
...In case you save timseries in InfluxDB, make sure you use a different DB per instance in order not to mix data across instances.
Suppose you want to start ntopng as follows -i eth0 -i eth1. ntopng will show you traffic
of these two interfaces without any merge so you can see exactly what happens on each interface.
If you also need an aggregated view of both interfaces you can start ntopng
as ntopng -i eth0 -i eth1 -i view:eth0,eth1 so ntopng will create a virtual interface
that merges information from the two physical interfaces.
If you apply a sampling rate to capture traffic on an interface, say x100, the traffic volume you see on ntopng will be 100 times smaller. In order to simulate more traffic to match real traffic volume, you can apply a scaling factor to the size of each received packet. The scaling factor can be specified through the GUI, into the interface settings.
handle SIGPIPE nostop noprint pass