NG_NETFLOW(4) MidnightBSD Kernel Interfaces Manual NG_NETFLOW(4)


ng_netflow — Cisco’s NetFlow implementation


#include <sys/types.h>
#include <netinet/in.h>
#include <netgraph/netflow/ng_netflow.h>


The ng_netflow node implements Cisco’s NetFlow export protocol on a router running FreeBSD. The ng_netflow node listens for incoming traffic and identifies unique flows in it. Flows are distinguished by endpoint IP addresses, TCP/UDP port numbers, ToS and input interface. Expired flows are exported out of the node in NetFlow version 5 UDP datagrams. Expiration reason can be one of the following:

RST or FIN TCP segment.

Active timeout. Flows cannot live more than the specified period of time. The default is 1800 seconds (30 minutes).

Inactive timeout. A flow was inactive for the specified period of time. The default is 15 seconds.

Export information is stored in NetFlow version 5 datagrams.


This node type supports up to NG_NETFLOW_MAXIFACES hooks named iface0, iface1, etc., and the same number of hooks named out0, out1, etc., plus a single hook named export. The node does NetFlow accounting of data received on iface* hooks. If corresponding out hook is connected, unmodified data is bypassed to it, otherwise data is freed. If data is received on out hook, it is bypassed to corresponding iface hook without any processing. When full export datagram is built it is sent to the export hook. In normal operation, the export hook is connected to the inet/dgram/udp hook of the ng_ksocket(4) node.


This node type supports the generic control messages, plus the following:


Returns some node statistics and the current timeout values in a struct ng_netflow_info.


Returns information about the ifaceN hook. The hook number is passed as an argument.


Sets data link type on the ifaceN hook. Currently, supported types are raw IP datagrams and Ethernet. This messsage type uses struct ng_netflow_setdlt as an argument:

struct ng_netflow_setdlt {

uint16_t iface;

/* which iface to operate on */

uint8_t dlt;

/* DLT_XXX from bpf.h */


The requested ifaceN hook must already be connected, otherwise message send operation will return an error.


In some cases, ng_netflow may be unable to determine the input interface index of a packet. This can happen if traffic enters the ng_netflow node before it comes to the system interface’s input queue. An example of such a setup is capturing a traffic between synchronous data line and ng_iface(4). In this case, the input index should be associated with a given hook. The interface’s index can be determined via if_nametoindex(3) from userland. This message requires struct ng_netflow_setifindex as an argument:

struct ng_netflow_setifindex {

u_int16_t iface;

/* which iface to operate on */

u_int16_t index;

/* new index */


The requested ifaceN hook must already be connected, otherwise the message send operation will return an error.


Sets values in seconds for NetFlow active/inactive timeouts. This message requires struct ng_netflow_settimeouts as an argument:

struct ng_netflow_settimeouts {







This control message asks a node to dump the entire contents of the flow cache. It is called from flowctl(8), not directly from ngctl(8). See also BUGS section.


Most binary control messages have an ASCII equivalent. The supported ASCII commands are:




"ifinfo %u"


"setdlt { iface = %u dlt = %u }"


"setifindex { iface = %u index = %u }"


"settimeouts { inactive = %u active = %u }"


This node shuts down upon receipt of a NGM_SHUTDOWN control message, or when all hooks have been disconnected.


The simplest possible configuration is one Ethernet interface, where flow collecting is enabled.

/usr/sbin/ngctl -f- <<-SEQ

mkpeer fxp0: netflow lower iface0

name fxp0:lower netflow

connect fxp0: netflow: upper out0

mkpeer netflow: ksocket export inet/dgram/udp

msg netflow:export connect inet/


This is a more complicated example of a router with 2 NetFlow-enabled interfaces fxp0 and ng0. Note that the ng0: node in this example is connected to ng_tee(4). The latter sends us a copy of IP packets, which we analyze and free. On fxp0: we do not use tee, but send packets back to ether node.

/usr/sbin/ngctl -f- <<-SEQ

# connect ng0’s tee to iface0 hook

mkpeer ng0:inet netflow right2left iface0

name ng0:inet.right2left netflow

# set DLT to raw mode

msg netflow: setdlt { iface=0 dlt=12 }

# set interface index (5 in this example)

msg netflow: setifindex { iface=0 index=5 }

# Connect fxp0: to iface1 and out1 hook

connect fxp0: netflow: lower iface1

connect fxp0: netflow: upper out1

# Create ksocket node on export hook, and configure it

# to send exports to proper destination

mkpeer netflow: ksocket export inet/dgram/udp

msg netflow:export connect inet/



netgraph(4), ng_ether(4), ng_iface(4), ng_ksocket(4), ng_tee(4), flowctl(8), ngctl(8)


The ng_netflow node type was written by Gleb Smirnoff 〈〉, based on ng_ipacct written by Roman V. Palagin 〈〉.


Cache snapshot obtained via NGM_NETFLOW_SHOW command may lack some percentage of entries under severe load.

The ng_netflow node type does not fill in AS numbers. This is due to the lack of necessary information in the kernel routing table. However, this information can be injected into the kernel from a routing daemon such as GNU Zebra. This functionality may become available in future releases.

MidnightBSD 0.3 March 2, 2006 MidnightBSD 0.3