1.\" $FreeBSD$ 2.\" $KAME: ipsec.4,v 1.17 2001/06/27 15:25:10 itojun Exp $ 3.\" 4.\" Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. 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.\" 3. Neither the name of the project nor the names of its contributors 16.\" may be used to endorse or promote products derived from this software 17.\" without specific prior written permission. 18.\" 19.\" THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND 20.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 21.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 22.\" ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE 23.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 24.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 25.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 26.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 27.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 28.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 29.\" SUCH DAMAGE. 30.\" 31.Dd January 29, 1999 32.Dt IPSEC 4 33.Os 34.Sh NAME 35.Nm ipsec 36.Nd IP security protocol 37.Sh SYNOPSIS 38.In sys/types.h 39.In netinet/in.h 40.In netinet6/ipsec.h 41.Sh DESCRIPTION 42.Nm 43is a security protocol in Internet Protocol layer. 44.Nm 45is defined for both IPv4 and IPv6 46.Xr ( inet 4 47and 48.Xr inet6 4 ) . 49.Nm 50consists of two sub-protocols, namely 51ESP 52(encapsulated security payload) 53and AH 54(authentication header). 55ESP protects IP payload from wire-tapping by encrypting it by 56secret key cryptography algorithms. 57AH guarantees integrity of IP packet 58and protects it from intermediate alteration or impersonation, 59by attaching cryptographic checksum computed by one-way hash functions. 60.Nm 61has two operation modes: transport mode and tunnel mode. 62Transport mode is for protecting peer-to-peer communication between end nodes. 63Tunnel mode includes IP-in-IP encapsulation operation 64and is designed for security gateways, like VPN configurations. 65.\" 66.Ss Kernel interface 67.Nm 68is controlled by key management engine and policy engine, 69in the operating system kernel. 70.Pp 71Key management engine can be accessed from the userland by using 72.Dv PF_KEY 73sockets. 74The 75.Dv PF_KEY 76socket API is defined in RFC2367. 77.Pp 78Policy engine can be controlled by extended part of 79.Dv PF_KEY 80API, 81.Xr setsockopt 2 82operations, and 83.Xr sysctl 3 84interface. 85The kernel implements 86extended version of 87.Dv PF_KEY 88interface, and allows you to define IPsec policy like per-packet filters. 89.Xr setsockopt 2 90interface is used to define per-socket behavior, and 91.Xr sysctl 3 92interface is used to define host-wide default behavior. 93.Pp 94The kernel code does not implement dynamic encryption key exchange protocol 95like IKE 96(Internet Key Exchange). 97That should be implemented as userland programs 98(usually as daemons), 99by using the above described APIs. 100.\" 101.Ss Policy management 102The kernel implements experimental policy management code. 103You can manage the IPsec policy in two ways. 104One is to configure per-socket policy using 105.Xr setsockopt 2 . 106The other is to configure kernel packet filter-based policy using 107.Dv PF_KEY 108interface, via 109.Xr setkey 8 . 110In both cases, IPsec policy must be specified with syntax described in 111.Xr ipsec_set_policy 3 . 112.Pp 113With 114.Xr setsockopt 2 , 115you can define IPsec policy in per-socket basis. 116You can enforce particular IPsec policy onto packets that go through 117particular socket. 118.Pp 119With 120.Xr setkey 8 121you can define IPsec policy against packets, 122using sort of packet filtering rule. 123Refer to 124.Xr setkey 8 125on how to use it. 126.Pp 127In the latter case, 128.Dq Li default 129policy is allowed for use with 130.Xr setkey 8 . 131By configuring policy to 132.Li default , 133you can refer system-wide 134.Xr sysctl 8 135variable for default settings. 136The following variables are available. 137.Li 1 138means 139.Dq Li use , 140and 141.Li 2 142means 143.Dq Li require 144in the syntax. 145.Bl -column net.inet6.ipsec6.esp_trans_deflev integerxxx 146.It Sy "Name Type Changeable" 147.It "net.inet.ipsec.esp_trans_deflev integer yes" 148.It "net.inet.ipsec.esp_net_deflev integer yes" 149.It "net.inet.ipsec.ah_trans_deflev integer yes" 150.It "net.inet.ipsec.ah_net_deflev integer yes" 151.It "net.inet6.ipsec6.esp_trans_deflev integer yes" 152.It "net.inet6.ipsec6.esp_net_deflev integer yes" 153.It "net.inet6.ipsec6.ah_trans_deflev integer yes" 154.It "net.inet6.ipsec6.ah_net_deflev integer yes" 155.El 156.Pp 157If kernel finds no matching policy system wide default value is applied. 158System wide default is specified by the following 159.Xr sysctl 8 160variables. 161.Li 0 162means 163.Dq Li discard 164which asks the kernel to drop the packet. 165.Li 1 166means 167.Dq Li none . 168.Bl -column net.inet6.ipsec6.def_policy integerxxx 169.It Sy "Name Type Changeable" 170.It "net.inet.ipsec.def_policy integer yes" 171.It "net.inet6.ipsec6.def_policy integer yes" 172.El 173.\" 174.Ss Miscellaneous sysctl variables 175The following variables are accessible via 176.Xr sysctl 8 , 177for tweaking kernel IPsec behavior: 178.Bl -column net.inet6.ipsec6.inbonud_call_ike integerxxx 179.It Sy "Name Type Changeable" 180.It "net.inet.ipsec.ah_cleartos integer yes" 181.It "net.inet.ipsec.ah_offsetmask integer yes" 182.It "net.inet.ipsec.dfbit integer yes" 183.It "net.inet.ipsec.ecn integer yes" 184.It "net.inet.ipsec.debug integer yes" 185.It "net.inet6.ipsec6.ecn integer yes" 186.It "net.inet6.ipsec6.debug integer yes" 187.El 188.Pp 189The variables are interpreted as follows: 190.Bl -tag -width 6n 191.It Li ipsec.ah_cleartos 192If set to non-zero, the kernel clears type-of-service field in the IPv4 header 193during AH authentication data computation. 194The variable is for tweaking AH behavior to interoperate with devices that 195implement RFC1826 AH. 196It should be set to non-zero 197(clear the type-of-service field) 198for RFC2402 conformance. 199.It Li ipsec.ah_offsetmask 200During AH authentication data computation, the kernel will include 20116bit fragment offset field 202(including flag bits) 203in IPv4 header, after computing logical AND with the variable. 204The variable is for tweaking AH behavior to interoperate with devices that 205implement RFC1826 AH. 206It should be set to zero 207(clear the fragment offset field during computation) 208for RFC2402 conformance. 209.It Li ipsec.dfbit 210The variable configures the kernel behavior on IPv4 IPsec tunnel encapsulation. 211If set to 0, DF bit on the outer IPv4 header will be cleared. 2121 means that the outer DF bit is set regardless from the inner DF bit. 2132 means that the DF bit is copied from the inner header to the outer. 214The variable is supplied to conform to RFC2401 chapter 6.1. 215.It Li ipsec.ecn 216If set to non-zero, IPv4 IPsec tunnel encapsulation/decapsulation behavior will 217be friendly to ECN 218(explicit congestion notification), 219as documented in 220.Li draft-ietf-ipsec-ecn-02.txt . 221.Xr gif 4 222talks more about the behavior. 223.It Li ipsec.debug 224If set to non-zero, debug messages will be generated via 225.Xr syslog 3 . 226.El 227.Pp 228Variables under 229.Li net.inet6.ipsec6 230tree has similar meaning as the 231.Li net.inet.ipsec 232counterpart. 233.\" 234.Sh PROTOCOLS 235The 236.Nm 237protocol works like plug-in to 238.Xr inet 4 239and 240.Xr inet6 4 241protocols. 242Therefore, 243.Nm 244supports most of the protocols defined upon those IP-layer protocols. 245Some of the protocols, like 246.Xr icmp 4 247or 248.Xr icmp6 4 , 249may behave differently with 250.Nm . 251This is because 252.Nm 253can prevent 254.Xr icmp 4 255or 256.Xr icmp6 4 257routines from looking into IP payload. 258.\" 259.Sh SEE ALSO 260.Xr ioctl 2 , 261.Xr socket 2 , 262.Xr ipsec_set_policy 3 , 263.Xr icmp6 4 , 264.Xr intro 4 , 265.Xr ip6 4 , 266.Xr setkey 8 , 267.Xr sysctl 8 268.\".Xr racoon 8 269.Sh STANDARDS 270.Rs 271.%A Daniel L. McDonald 272.%A Craig Metz 273.%A Bao G. Phan 274.%T "PF_KEY Key Management API, Version 2" 275.%R RFC 276.%N 2367 277.Re 278.Pp 279.Rs 280.%A "D. L. McDonald" 281.%T "A Simple IP Security API Extension to BSD Sockets" 282.%R internet draft 283.%N "draft-mcdonald-simple-ipsec-api-03.txt" 284.%O work in progress material 285.Re 286.Sh HISTORY 287The implementation described herein appeared in WIDE/KAME IPv6/IPsec stack. 288.Sh BUGS 289The IPsec support is subject to change as the IPsec protocols develop. 290.Pp 291There is no single standard for policy engine API, 292so the policy engine API described herein is just for KAME implementation. 293.Pp 294AH and tunnel mode encapsulation may not work as you might expect. 295If you configure inbound 296.Dq require 297policy against AH tunnel or any IPsec encapsulating policy with AH 298(like 299.Dq Li esp/tunnel/A-B/use ah/transport/A-B/require ) , 300tunnelled packets will be rejected. 301This is because we enforce policy check on inner packet on reception, 302and AH authenticates encapsulating 303(outer) 304packet, not the encapsulated 305(inner) 306packet 307(so for the receiving kernel there's no sign of authenticity). 308The issue will be solved when we revamp our policy engine to keep all the 309packet decapsulation history. 310.Pp 311Under certain condition, 312truncated result may be raised from the kernel 313against 314.Dv SADB_DUMP 315and 316.Dv SADB_SPDDUMP 317operation on 318.Dv PF_KEY 319socket. 320This occurs if there are too many database entries in the kernel 321and socket buffer for the 322.Dv PF_KEY 323socket is insufficient. 324If you manipulate many IPsec key/policy database entries, 325increase the size of socket buffer. 326