3. sFlow/NetFlow Collection¶
ntopng, in combination with nProbe, can be used to collect sFlow/NetFlow. The use of ntopng in combination with nProbe is described in Section Using ntopng with nProbe.
3.1. sFlow Collection¶
An introduction to sFlow and its collection in ntopng is provided in this blog post: https://www.ntop.org/nprobe/sflow-collection-and-analysis-with-nprobe-and-ntopng
Supposing that there are sFlow devices - or agents, in the sFlow parlance - exporting sFlow on port 6343 of the host where ntopng is running, in order to collect such data it’s necessary to configure nProbe to collect flows on port 6343 and ntopng to talk with nProbe:
nprobe -i none -n none --collector-port 6343 --zmq tcp://127.0.0.1:5556 ntopng -i tcp://127.0.0.1:5556
The ntopng web dashboard will shortly populate with collected data, including top senders and top destinations, as well as the top layer-7 application protocols detected from the traffic. Similarly, all the other ntopng pages will populate with data.
3.2. sFlow Devices¶
When capturing sFlow traffic, ntopng displays aggregated information regarding the sFlow devices and their ports. This information is accessible from the Collectors menu, sFlow Devices entry:
Each row shows a single sFlow device. If the device is also configured as an SNMP device, ntopng will show its name, description and location information. A chart of the device traffic is shown provided that the “Flow Devices” timeseries are enabled from the ntopng preferences. By clicking on the sFlow device IP address it’s possible to inspect the per-interface statistics of the SNMP device.
For each interface, ntopng shows the following information:
Index: the interface numeric index
Name: the interface name
Chart: the interface traffic chart, if the “Flow Devices” timeseries are enabled
Speed: the interface speed
Duplex: interface duplex information, half or full duplex
Status: the current interface status
Promisc: if the interface is in the promiscuous mode
In Bytes: ingress bytes counter of the interface
Out Bytes: egress bytes counter of the interface
In Errors: ingress errors packets counter of the interface
Out Errors: ingress errors packets counter of the interface
Counters/SNMP Ratio: the ratio between the sFlow counters by the total traffic read via SNMP
3.3. NetFlow Devices¶
3.3.1. How the Flow/SNMP Ratio is calculated?¶
Flow/SNMP Ratio column is calculated by dividing the total interface traffic obtained using flows,
by the total traffic of the same interface read via SNMP. As in flow-based analysis non-IP and layer 2 headers are not accounted,
typical ratio values are between 0.8 and 0.9 (i.e. 0.9 means that 90%% of the received traffic as observed via SNMP has been reported in flows).
Ratio is computed using the traffic over the latest 5 minutes, as periodic SNMP walks are performed at 5-minute intervals.
Ratio is only computed when the following conditions are met:
- Flow Device and SNMP timeseries are enabled
- The device selected must support SNMP and must be configured in the SNMP devices page