softflowd.8

.\" Copyright (c) 2002 Damien Miller.  All rights reserved.
.\" Portions Copyright (c) 2001 Kevin Steves.  All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
.\" IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
.\" DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
.\"
.Dd October 14, 2002
.Dt SOFTFLOWD 8
.Os
.Sh NAME
.Nm softflowd
.Nd Traffic flow monitoring
.Sh SYNOPSIS
.Nm softflowd
.Op Fl 6dDh
.Op Fl L Ar hoplimit
.Op Fl T Ar track_level
.Op Fl c Ar ctl_sock
.Bk -words
.Oo Fl i\ \&
.Sm off
.Oo Ar if_ndx : Oc
.Ar interface
.Sm on
.Oc
.Ek
.Op Fl m Ar max_flows
.Op Fl n Ar host:port
.Op Fl p Ar pidfile
.Op Fl r Ar pcap_file
.Op Fl t Ar timeout_name=seconds
.Op Fl v Ar netflow_version
.Op Fl s Ar sampling_rate
.Op bpf_expression
.Sh DESCRIPTION
.Nm
is a software implementation of a flow-based network traffic monitor.
.Nm
reads network traffic and gathers information about active traffic flows.
A "traffic flow" is communication between two IP addresses or (if the
overlying protocol is TCP or UDP) address/port tuples.
.Pp
The intended use of
.Nm
is as a software implementation of Cisco's NetFlow(tm) traffic account
system.
.Nm
supports data export using versions 1, 5 or 9 of the NetFlow protocol.
.Nm
can also run in statistics-only mode, where it just collects summary
information.
However, too few statistics are collected to make this
mode really useful for anything other than debugging.
.Pp
Network traffic may be obtained by listening on a promiscuous network
interface or by reading stored
.Xr pcap 3
files, such as those written by
.Xr tcpdump 8 .
Traffic may be filtered with an optional
.Xr bpf 4
program, specified on the command-line as
.Ar bpf_expression .
.Nm
is IPv6 capable and will track IPv6 flows if the NetFlow export protocol
supports it (currently only NetFlow v.9 possesses an IPv6 export capability).
.Pp
.Nm
tries to track only active traffic flows.
When the
flow has been quiescent for a period of time it is expired automatically.
Flows may also be expired early if they approach their traffic counts
exceed 2 Gib or if the number of flows being tracked exceeds
.Ar max_flows
(default: 8192).
In this last case, flows are expired oldest-first.
.Pp
Upon expiry, the flow information is accumulated into statistics which may
be viewed using
.Xr softflowctl 8 .
If the
.Fl n
option has been specified the flow information is formatted in a UDP datagram
which is compatible with versions 1, 5 or 9 of Cisco's NetFlow(tm) accounting
export format.
These records are sent to the specified
.Ar host
and
.Ar port .
The host may represent a unicast host or a multicast group.
.Pp
The command-line options are as follows:
.Bl -tag -width Ds
.It Fl n Ar host:port
Specify the
.Ar host
and
.Ar port
that the accounting datagrams are to be sent to.
The host may be specified using a hostname or using a numeric IPv4 or
IPv6 address.
Numeric IPv6 addresses should be encosed in square brackets to avoid ambiguity
between the address and the port.
The destination port may be a portname listed in
.Xr services 5
or a numeric port.
.It Fl i Xo
.Sm off
.Oo Ar if_ndx : Oc
.Ar interface
.Sm on
.Xc
Specify a network interface on which to listen for traffic.
Either the
.Fl i
or the
.Fl r
options must be specified.
.It Fl r Ar pcap_file
Specify that
.Nm
should read from a
.Xr pcap 3
packet capture file (such as one created with the
.Fl w
option of
.Xr tcpdump 8 )
file rather than a network interface.
.Nm
processes the whole capture file and only expires flows when
.Ar max_flows
is exceeded.
In this mode,
.Nm
will not fork and will automatically print summary statistics before
exiting.
.It Fl p Ar pidfile
Specify an alternate location to store the process ID when in daemon mode.
Default is
.Pa /var/run/softflowd.pid
.It Fl c Ar ctlsock
Specify an alternate location for the remote control socket in daemon mode.
Default is
.Pa /var/run/softflowd.ctl
.It Fl m Ar max_flows
Specify the maximum number of flows to concurrently track.
If this limit is exceeded, the flows which have least recently seen traffic
are forcibly expired.
In practice, the actual maximum may briefly exceed this limit by a
small amount as  expiry processing happens less frequently than traffic
collection.
The default is 8192 flows, which corresponds to slightly less
than 800k of working data.
.It Fl t Ar timeout_name=time
Set the timeout names
.Ar timeout_name
to
.Ar time .
Refer to the
.Sx Timeouts
section for the valid timeout names and their meanings.
The
.Ar time
parameter may be specified using one of the formats explained in the
.Sx Time Formats
section below.
.It Fl d
Specify that
.Nm
should not fork and daemonise itself.
.It Fl 6
Force
.Nm
to track IPv6 flows even if the NetFlow export protocol does not support
reporting them.
This is useful for debugging and statistics gathering only.
.It Fl D
Places
.Nm
in a debugging mode.
This implies the
.Fl d
and
.Fl 6
flags and turns on additional debugging output.
.It Fl h
Display command-line usage information.
.It Fl L Ar hoplimit
Set the IPv4 TTL or the IPv6 hop limit to
.Ar hoplimit .
.Nm
will use the default system TTL when exporting flows to a unicast host.
When exporting to a multicast group, the default TTL will be 1
(i.e. link-local).
.It Fl T Ar track_level
Specify which flow elements
.Nm
should be used to define a flow.
.Ar track_level
may be one of:
.Dq full
(track everything in the flow, the default),
.Dq proto
(track source and destination addresses and protocol), or
.Dq ip
(only track source and destination addresses).
Selecting either of the latter options will produce flows with less information
in them (e.g. TCP/UDP ports will not be recorded).
This will cause flows to be consolidated, reducing the quantity of output
and CPU load that
.Nm
will place on the system at the cost of some detail being lost.
.It Fl v Ar netflow_version
Specify which version of the NetFlow(tm) protocol
.Nm
should use for export of the flow data.
Supported versions are 1, 5 and 9.
Default is version 5.
.It Fl s Ar sampling_rate
Specify periodical sampling rate (denominator).
.El
.Pp
Any further command-line arguments will be concatenated together and
applied as a
.Xr bpf 4
packet filter.
This filter will cause
.Nm
to ignore the specified traffic.
.Ss Timeouts
.Pp
.Nm
will expire quiescent flows after user-configurable periods.
The exact timeout used depends on the nature of the flow.
The various timeouts that may be set from the command-line (using the
.Fl t
option) and their meanings are:
.Bl -tag -width Ds
.It Ar general
This is the general timeout applied to all traffic unless overridden by
one of the other timeouts.
.It Ar tcp
This is the general TCP timeout, applied to open TCP connections.
.It Ar tcp.rst
This timeout is applied to a TCP connection when a RST packet has been
sent by one or both endpoints.
.It Ar tcp.fin
This timeout is applied to a TCP connection when a FIN packet has been
sent by both endpoints.
.It Ar udp
This is the general UDP timeout, applied to all UDP connections.
.It Ar maxlife
This is the maximum lifetime that a flow may exist for.
All flows are forcibly expired when they pass
.Ar maxlife
seconds.
To disable this feature, specify a
.Ar maxlife
of 0.
.It Ar expint
Specify the interval between expiry checks.
Increase this to group more flows into a NetFlow packet.
To disable this feature, specify a
.Ar expint
of 0.
.El
.Pp
Flows may also be expired if there are not enough flow entries to hold them
or if their traffic exceeds 2 Gib in either direction.
.Xr softflowctl 8
may be used to print information on the average lifetimes of flows and
the reasons for their expiry.
.Ss Time Formats
.Pp
.Nm
command-line arguments that specify time may be expressed using a sequence
of the form:
.Sm off
.Ar time Op Ar qualifier ,
.Sm on
where
.Ar time
is a positive integer value and
.Ar qualifier
is one of the following:
.Pp
.Bl -tag -width Ds -compact -offset indent
.It Cm <none>
seconds
.It Cm s | Cm S
seconds
.It Cm m | Cm M
minutes
.It Cm h | Cm H
hours
.It Cm d | Cm D
days
.It Cm w | Cm W
weeks
.El
.Pp
Each member of the sequence is added together to calculate the total time value.
.Pp
Time format examples:
.Pp
.Bl -tag -width Ds -compact -offset indent
.It 600
600 seconds (10 minutes)
.It 10m
10 minutes
.It 1h30m
1 hour 30 minutes (90 minutes)
.El
.Ss Run-time Control
.Pp
A daemonised
.Nm
instance may be controlled using the
.Xr softflowctl 8
command.
This interface allows one to shut down the daemon, force expiry of
all tracked flows and extract debugging and summary data.
Also, receipt of a
.Dv SIGTERM
or
.Dv SIGINT
will cause
.Nm
to exit, after expiring all flows (and thus sending flow export packets
if
.Fl n
was specified on the command-line).
If you do not want to export flows upon shutdown, clear them first with
.Xr softflowctl 8
or use
.Xr softflowctl 8 's
.Dq exit
command.
.Sh EXAMPLES
.Bl -tag -width Ds
.It softflowd -i fxp0
This command-line will cause
.Nm
to listen on interface fxp0 and to run in statistics gathering mode
only (i.e. no NetFlow data export).
.It softflowd -i fxp0 -n 10.1.0.2:4432
This command-line will cause
.Nm
to listen on interface fxp0 and to export NetFlow v.5 datagrams on flow
expiry to a flow collector running on 10.1.0.2 port 4432.
.It softflowd -v 5 -i fxp0 -n 10.1.0.2:4432 -m 65536 -t udp=1m30s
This command-line increases the number of concurrent flows that
.Nm
will track to 65536 and increases the timeout for UDP flows to 90 seconds.
.It softflowd -v 9 -i fxp0 -n 224.0.1.20:4432 -L 64
This command-line will export NetFlow v.9 flows to the multicast group
224.0.1.20.
The export datagrams will have their TTL set to 64, so multicast receivers
can be many hops away.
.It softflowd -i fxp0 -p /var/run/sfd.pid.fxp0 -c /var/run/sfd.ctl.fxp0
This command-line specifies alternate locations for the control socket
and pid file.
Similar command-lines are useful when running multiple
instances of
.Nm
on a single machine.
.El
.Sh FILES
.Bl -tag -width Ds
.It Pa /var/run/softflowd.pid
This file stores the process ID when
.Nm
is in daemon mode.
This location may be overridden using the
.Fl p
command-line option.
.It Pa /var/run/softflowd.ctl
This is the remote control socket.
.Nm
listens on this socket for commands from
.Xr softflowctl 8 .
This location may be overridden using the
.Fl c
command-line option.
.El
.Sh BUGS
Currently
.Nm
does not handle maliciously fragmented packets properly, i.e. packets
fragemented such that the UDP or TCP header does not fit into the first
fragment.
It will product correct traffic counts when presented with maliciously
fragmented packets, but will not record TCP or UDP port information.
.Sh AUTHORS
.An Damien Miller Aq djm@mindrot.org
.Sh SEE ALSO
.Xr softflowctl 8 ,
.Xr tcpdump 8 ,
.Xr pcap 3 ,
.Xr bpf 4
.Bd -literal
http://www.ietf.org/rfc/rfc3954.txt
.br
http://www.cisco.com/en/US/products/sw/netmgtsw/ps1964/products_implementation_design_guide09186a00800d6a11.html
.Ed