1.\" Copyright (c) 2004-2005 Gleb Smirnoff <glebius@FreeBSD.org> 2.\" All rights reserved. 3.\" 4.\" Redistribution and use in source and binary forms, with or without 5.\" modification, are permitted provided that the following conditions 6.\" are met: 7.\" 1. Redistributions of source code must retain the above copyright 8.\" notice, this list of conditions and the following disclaimer. 9.\" 2. Redistributions in binary form must reproduce the above copyright 10.\" notice, this list of conditions and the following disclaimer in the 11.\" documentation and/or other materials provided with the distribution. 12.\" 13.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 14.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 15.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 16.\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 17.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 18.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 19.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 20.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 21.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 22.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 23.\" SUCH DAMAGE. 24.\" 25.\" $FreeBSD$ 26.\" 27.Dd March 2, 2011 28.Dt NG_NETFLOW 4 29.Os 30.Sh NAME 31.Nm ng_netflow 32.Nd Cisco's NetFlow implementation 33.Sh SYNOPSIS 34.In sys/types.h 35.In netinet/in.h 36.In netgraph/netflow/ng_netflow.h 37.Sh DESCRIPTION 38The 39.Nm 40node implements Cisco's NetFlow export protocol on a router running 41.Fx . 42The 43.Nm 44node listens for incoming traffic and identifies unique flows in it. 45Flows are distinguished by endpoint IP addresses, TCP/UDP port numbers, 46ToS and input interface. 47Expired flows are exported out of the node in NetFlow version 5/9 UDP datagrams. 48Expiration reason can be one of the following: 49.Bl -dash 50.It 51RST or FIN TCP segment. 52.It 53Active timeout. 54Flows cannot live more than the specified period of time. 55The default is 1800 seconds (30 minutes). 56.It 57Inactive timeout. 58A flow was inactive for the specified period of time. 59The default is 15 seconds. 60.El 61.Pp 62Node supports IPv6 accounting (NetFlow v9 only) and is aware of multiple fibs. 63Different fibs are mapped to different domain_id in NetFlow V9 and different engine_id in NetFlow V5. 64.Pp 65.Sh HOOKS 66This node type supports up to 67.Dv NG_NETFLOW_MAXIFACES 68(default 65536) hooks named 69.Va iface0 , iface1 , 70etc., 71and the same number of hooks named 72.Va out0 , out1 , 73etc., 74plus two export hooks: 75.Va export 76(for NetFlow version 5) and 77.Va export9 78(for NetFlow version 9). Export can be done simultaneously for all supported 79export hooks. By default (ingress NetFlow enabled) node does NetFlow accounting of data 80received on 81.Va iface* 82hooks. 83If corresponding 84.Va out 85hook is connected, unmodified data is bypassed to it, otherwise data is freed. 86If data is received on 87.Va out 88hook, it is bypassed to corresponding 89.Va iface 90hook without any processing (egress NetFlow disabled by default). 91When full export datagram for an export protocol is built it is sent to the 92.Va export 93or 94.Va export9 95hook. 96In normal operation, one (or more) export hook is connected to the 97.Va inet/dgram/udp 98hook of the 99.Xr ng_ksocket 4 100node. 101.Sh CONTROL MESSAGES 102This node type supports the generic control messages, plus the following: 103.Bl -tag -width indent 104.It Dv NGM_NETFLOW_INFO 105Returns some node statistics and the current timeout values in a 106.Vt "struct ng_netflow_info" . 107.It Dv NGM_NETFLOW_IFINFO 108Returns information about the 109.Va iface Ns Ar N 110hook. 111The hook number is passed as an argument. 112.It Dv NGM_NETFLOW_SETDLT 113Sets data link type on the 114.Va iface Ns Ar N 115hook. 116Currently, supported types are raw IP datagrams and Ethernet. 117This message type uses 118.Vt "struct ng_netflow_setdlt" 119as an argument: 120.Bd -literal -offset 4n 121struct ng_netflow_setdlt { 122 uint16_t iface; /* which iface to operate on */ 123 uint8_t dlt; /* DLT_XXX from bpf.h */ 124}; 125.Ed 126.Pp 127The requested 128.Va iface Ns Ar N 129hook must already be connected, otherwise message send operation will 130return an error. 131.It Dv NGM_NETFLOW_SETIFINDEX 132In some cases, 133.Nm 134may be unable to determine the input interface index of a packet. 135This can happen if traffic enters the 136.Nm 137node before it comes to the system interface's input queue. 138An example of such a setup is capturing a traffic 139.Em between 140synchronous data line and 141.Xr ng_iface 4 . 142In this case, the input index should be associated with a given hook. 143The interface's index can be determined via 144.Xr if_nametoindex 3 145from userland. 146This message requires 147.Vt "struct ng_netflow_setifindex" 148as an argument: 149.Bd -literal -offset 4n 150struct ng_netflow_setifindex { 151 u_int16_t iface; /* which iface to operate on */ 152 u_int16_t index; /* new index */ 153}; 154.Ed 155.Pp 156The requested 157.Va iface Ns Ar N 158hook must already be connected, otherwise the message 159send operation will return an error. 160.It Dv NGM_NETFLOW_SETTIMEOUTS 161Sets values in seconds for NetFlow active/inactive timeouts. 162This message requires 163.Vt "struct ng_netflow_settimeouts" 164as an argument: 165.Bd -literal -offset 4n 166struct ng_netflow_settimeouts { 167 uint32_t inactive_timeout; 168 uint32_t active_timeout; 169}; 170.Ed 171.It Dv NGM_NETFLOW_SETCONFIG 172Sets configuration for the specified interface. 173This message requires 174.Vt "struct ng_netflow_setconfig" 175as an argument: 176.Bd -literal -offset 4n 177struct ng_netflow_setconfig { 178 u_int16_t iface; 179 u_int32_t conf; 180#define NG_NETFLOW_CONF_INGRESS 1 181#define NG_NETFLOW_CONF_EGRESS 2 182#define NG_NETFLOW_CONF_ONCE 4 183#define NG_NETFLOW_CONF_THISONCE 8 184}; 185.Ed 186.Pp 187Configuration is a bitmask of several options. Option NG_NETFLOW_CONF_INGRESS 188enabled by default enables ingress NetFlow generation (for data coming from 189ifaceX hook). Option NG_NETFLOW_CONF_EGRESS enables egress NetFlow (for data 190coming from outX hook). Option NG_NETFLOW_CONF_ONCE defines that packet should 191be accounted only once if it several times passes via netflow node. Option 192NG_NETFLOW_CONF_THISONCE defines that packet should be accounted only once 193if it several times passes via exactly this netflow node. Last two options are 194important to avoid duplicate accounting when both ingress and egress NetFlow 195are enabled. 196.It Dv NGM_NETFLOW_SETTEMPLATE 197Sets various timeouts to announce data flow templates 198(NetFlow v9-specific). This message requires 199.Vt "struct ng_netflow_settemplate" 200as an argument: 201.Bd -literal -offset 4n 202struct ng_netflow_settemplate { 203 uint16_t time; 204 uint16_t packets; 205}; 206.Ed 207.Pp 208Value of time field represents time in seconds to re-announce data templates. 209Value of packets field represents maximum packets count between 210re-announcing data templates. 211.It Dv NGM_NETFLOW_SETMTU 212Sets export interface MTU to build packets of specified size (NetFlow v9-specific). 213This message requires 214.Vt "struct ng_netflow_setmtu" 215as an argument: 216.Bd -literal -offset 4n 217struct ng_netflow_settemtu { 218 uint16_t mtu; 219}; 220.Ed 221.Pp 222Default is 1500 bytes 223.It Dv NGM_NETFLOW_SHOW 224This control message asks a node to dump the entire contents of the flow cache. 225It is called from 226.Xr flowctl 8 , 227not directly from 228.Xr ngctl 8 . 229See also 230.Sx BUGS 231section. 232.El 233.Sh ASCII CONTROL MESSAGES 234Most binary control messages have an 235.Tn ASCII 236equivalent. 237The supported 238.Tn ASCII 239commands are: 240.Pp 241.Bl -tag -width ".Dv NGM_NETFLOW_SETTIMEOUTS" -compact 242.It Dv NGM_NETFLOW_INFO 243.Qq Li info 244.It Dv NGM_NETFLOW_IFINFO 245.Qq Li "ifinfo %u" 246.It Dv NGM_NETFLOW_SETDLT 247.Qq Li "setdlt { iface = %u dlt = %u }" 248.It Dv NGM_NETFLOW_SETIFINDEX 249.Qq Li "setifindex { iface = %u index = %u }" 250.It Dv NGM_NETFLOW_SETTIMEOUTS 251.Qq Li "settimeouts { inactive = %u active = %u }" 252.It Dv NGM_NETFLOW_SETCONFIG 253.Qq Li "setconfig { iface = %u conf = %u }" 254.It Dv NGM_NETFLOW_SETTEMPLATE 255.Qq Li "settemplate { time = %u packets = %u }" 256.It Dv NGM_NETFLOW_SETMTU 257.Qq Li "setmtu { mtu = %u }" 258.El 259.Sh SHUTDOWN 260This node shuts down upon receipt of a 261.Dv NGM_SHUTDOWN 262control message, or when all hooks have been disconnected. 263.Sh EXAMPLES 264The simplest possible configuration is one Ethernet interface, where 265flow collecting is enabled. 266.Bd -literal -offset indent 267/usr/sbin/ngctl -f- <<-SEQ 268 mkpeer fxp0: netflow lower iface0 269 name fxp0:lower netflow 270 connect fxp0: netflow: upper out0 271 mkpeer netflow: ksocket export inet/dgram/udp 272 msg netflow:export connect inet/10.0.0.1:4444 273SEQ 274.Ed 275.Pp 276This is a more complicated example of a router with 2 NetFlow-enabled 277interfaces 278.Li fxp0 279and 280.Li ng0 . 281Note that the 282.Va ng0: 283node in this example is connected to 284.Xr ng_tee 4 . 285The latter sends us a copy of IP packets, which we analyze and free. 286On 287.Va fxp0: 288we do not use tee, but send packets back to ether node. 289.Bd -literal -offset indent 290/usr/sbin/ngctl -f- <<-SEQ 291 # connect ng0's tee to iface0 hook 292 mkpeer ng0:inet netflow right2left iface0 293 name ng0:inet.right2left netflow 294 # set DLT to raw mode 295 msg netflow: setdlt { iface=0 dlt=12 } 296 # set interface index (5 in this example) 297 msg netflow: setifindex { iface=0 index=5 } 298 299 # Connect fxp0: to iface1 and out1 hook 300 connect fxp0: netflow: lower iface1 301 connect fxp0: netflow: upper out1 302 303 # Create ksocket node on export hook, and configure it 304 # to send exports to proper destination 305 mkpeer netflow: ksocket export inet/dgram/udp 306 msg netflow:export connect inet/10.0.0.1:4444 307SEQ 308.Ed 309.Sh SEE ALSO 310.Xr netgraph 4 , 311.Xr setfib 2 , 312.Xr ng_ether 4 , 313.Xr ng_iface 4 , 314.Xr ng_ksocket 4 , 315.Xr ng_tee 4 , 316.Xr flowctl 8 , 317.Xr ngctl 8 318.Rs 319.%A B. Claise, Ed 320.%T "Cisco Systems NetFlow Services Export Version 9" 321.%O RFC 3954 322.Re 323.Pp 324.Pa http://www.cisco.com/en/US/docs/ios/solutions_docs/netflow/nfwhite.html 325.Sh AUTHORS 326.An -nosplit 327The 328.Nm 329node type was written by 330.An Gleb Smirnoff Aq glebius@FreeBSD.org , 331.An Alexander Motin Aq mav@FreeBSD.org , 332.An Alexander Chernikov Aq melifaro@ipfw.ru . 333The initial code was based on 334.Nm ng_ipacct 335written by 336.An Roman V. Palagin Aq romanp@unshadow.net . 337.Sh BUGS 338Cache snapshot obtained via 339.Dv NGM_NETFLOW_SHOW 340command may lack some percentage of entries under severe load. 341IPv6 flows are not shown. 342.Pp 343The 344.Nm 345node type does not fill in AS numbers. 346This is due to the lack of necessary information in the kernel routing table. 347However, this information can be injected into the kernel from a routing daemon 348such as GNU Zebra. 349This functionality may become available in future releases. 350