1.\" $OpenBSD: pfsync.4,v 1.28 2009/02/17 10:05:18 dlg Exp $ 2.\" 3.\" Copyright (c) 2002 Michael Shalayeff 4.\" Copyright (c) 2003-2004 Ryan McBride 5.\" All rights reserved. 6.\" 7.\" Redistribution and use in source and binary forms, with or without 8.\" modification, are permitted provided that the following conditions 9.\" are met: 10.\" 1. Redistributions of source code must retain the above copyright 11.\" notice, this list of conditions and the following disclaimer. 12.\" 2. Redistributions in binary form must reproduce the above copyright 13.\" notice, this list of conditions and the following disclaimer in the 14.\" documentation and/or other materials provided with the distribution. 15.\" 16.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR 17.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 18.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 19.\" IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, 20.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 21.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF MIND, 22.\" USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 23.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 24.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF 25.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 26.\" 27.\" $FreeBSD$ 28.\" 29.Dd December 6, 2018 30.Dt PFSYNC 4 31.Os 32.Sh NAME 33.Nm pfsync 34.Nd packet filter state table sychronisation interface 35.Sh SYNOPSIS 36.Cd "device pfsync" 37.Sh DESCRIPTION 38The 39.Nm 40interface is a pseudo-device which exposes certain changes to the state 41table used by 42.Xr pf 4 . 43State changes can be viewed by invoking 44.Xr tcpdump 1 45on the 46.Nm 47interface. 48If configured with a physical synchronisation interface, 49.Nm 50will also send state changes out on that interface, 51and insert state changes received on that interface from other systems 52into the state table. 53.Pp 54By default, all local changes to the state table are exposed via 55.Nm . 56State changes from packets received by 57.Nm 58over the network are not rebroadcast. 59Updates to states created by a rule marked with the 60.Ar no-sync 61keyword are ignored by the 62.Nm 63interface (see 64.Xr pf.conf 5 65for details). 66.Pp 67The 68.Nm 69interface will attempt to collapse multiple state updates into a single 70packet where possible. 71The maximum number of times a single state can be updated before a 72.Nm 73packet will be sent out is controlled by the 74.Ar maxupd 75parameter to ifconfig 76(see 77.Xr ifconfig 8 78and the example below for more details). 79The sending out of a 80.Nm 81packet will be delayed by a maximum of one second. 82.Sh NETWORK SYNCHRONISATION 83States can be synchronised between two or more firewalls using this 84interface, by specifying a synchronisation interface using 85.Xr ifconfig 8 . 86For example, the following command sets fxp0 as the synchronisation 87interface: 88.Bd -literal -offset indent 89# ifconfig pfsync0 syncdev fxp0 90.Ed 91.Pp 92By default, state change messages are sent out on the synchronisation 93interface using IP multicast packets to the 224.0.0.240 group address. 94An alternative destination address for 95.Nm 96packets can be specified using the 97.Ic syncpeer 98keyword. 99This can be used in combination with 100.Xr ipsec 4 101to protect the synchronisation traffic. 102In such a configuration, the syncdev should be set to the 103.Xr enc 4 104interface, as this is where the traffic arrives when it is decapsulated, 105e.g.: 106.Bd -literal -offset indent 107# ifconfig pfsync0 syncpeer 10.0.0.2 syncdev enc0 108.Ed 109.Pp 110It is important that the pfsync traffic be well secured 111as there is no authentication on the protocol and it would 112be trivial to spoof packets which create states, bypassing the pf ruleset. 113Either run the pfsync protocol on a trusted network \- ideally a network 114dedicated to pfsync messages such as a crossover cable between two firewalls, 115or specify a peer address and protect the traffic with 116.Xr ipsec 4 . 117.Pp 118.Nm 119has the following 120.Xr sysctl 8 121tunables: 122.Bl -tag -width ".Va net.pfsync" 123.It Va net.pfsync.carp_demotion_factor 124Value added to 125.Va net.inet.carp.demotion 126while 127.Nm 128tries to perform its bulk update. 129See 130.Xr carp 4 131for more information. 132Default value is 240. 133.It Va net.pfsync.pfsync_buckets 134The number of 135.Nm 136buckets. 137This affects the performance and memory tradeoff. 138Defaults to twice the number of CPUs. 139Change only if benchmarks show this helps on your workload. 140.El 141.Sh EXAMPLES 142.Nm 143and 144.Xr carp 4 145can be used together to provide automatic failover of a pair of firewalls 146configured in parallel. 147One firewall will handle all traffic until it dies, is shut down, or is 148manually demoted, at which point the second firewall will take over 149automatically. 150.Pp 151Both firewalls in this example have three 152.Xr sis 4 153interfaces. 154sis0 is the external interface, on the 10.0.0.0/24 subnet; sis1 is the 155internal interface, on the 192.168.0.0/24 subnet; and sis2 is the 156.Nm 157interface, using the 192.168.254.0/24 subnet. 158A crossover cable connects the two firewalls via their sis2 interfaces. 159On all three interfaces, firewall A uses the .254 address, while firewall B 160uses .253. 161The interfaces are configured as follows (firewall A unless otherwise 162indicated): 163.Pp 164Interfaces configuration in 165.Pa /etc/rc.conf : 166.Bd -literal -offset indent 167network_interfaces="lo0 sis0 sis1 sis2" 168ifconfig_sis0="10.0.0.254/24" 169ifconfig_sis0_alias0="inet 10.0.0.1/24 vhid 1 pass foo" 170ifconfig_sis1="192.168.0.254/24" 171ifconfig_sis1_alias0="inet 192.168.0.1/24 vhid 2 pass bar" 172ifconfig_sis2="192.168.254.254/24" 173pfsync_enable="YES" 174pfsync_syncdev="sis2" 175.Ed 176.Pp 177.Xr pf 4 178must also be configured to allow 179.Nm 180and 181.Xr carp 4 182traffic through. 183The following should be added to the top of 184.Pa /etc/pf.conf : 185.Bd -literal -offset indent 186pass quick on { sis2 } proto pfsync keep state (no-sync) 187pass on { sis0 sis1 } proto carp keep state (no-sync) 188.Ed 189.Pp 190It is preferable that one firewall handle the forwarding of all the traffic, 191therefore the 192.Ar advskew 193on the backup firewall's 194.Xr carp 4 195vhids should be set to something higher than 196the primary's. 197For example, if firewall B is the backup, its 198carp1 configuration would look like this: 199.Bd -literal -offset indent 200ifconfig_sis1_alias0="inet 192.168.0.1/24 vhid 2 pass bar advskew 100" 201.Ed 202.Pp 203The following must also be added to 204.Pa /etc/sysctl.conf : 205.Bd -literal -offset indent 206net.inet.carp.preempt=1 207.Ed 208.Sh SEE ALSO 209.Xr tcpdump 1 , 210.Xr bpf 4 , 211.Xr carp 4 , 212.Xr enc 4 , 213.Xr inet 4 , 214.Xr inet6 4 , 215.Xr ipsec 4 , 216.Xr netintro 4 , 217.Xr pf 4 , 218.Xr pf.conf 5 , 219.Xr protocols 5 , 220.Xr rc.conf 5 , 221.Xr ifconfig 8 222.Sh HISTORY 223The 224.Nm 225device first appeared in 226.Ox 3.3 . 227It was first imported to 228.Fx 5.3 . 229.Pp 230The 231.Nm 232protocol and kernel implementation were significantly modified in 233.Fx 9.0 . 234The newer protocol is not compatible with older one and will not interoperate 235with it. 236